Although the term “hospitalist” was coined in 1996 in a New England Journal of Medicine article, the field of HM grew organically from pressure to optimize hospital economics and improve efficiency in economically pressed healthcare markets.¹

Scholarship in HM has also grown and now includes regular publications of investigations exploring optimization of efficiency and quality, many with an emphasis on patient safety. In this way, HM is a unique field, with tools for approaching problems that aren’t commonly used in other branches of medicine.

In parallel to the emergence of HM as a field distinct from general internal medicine (IM), the HM fellowship is similar but distinct. Such fellowships serve multiple purposes.

HM fellowships can add clinical expertise and scholarship skills for a career in HM. While early HM research focused on proving the value of the hospitalist model, the field has expanded greatly for those interested in an academic career. The molding of a safer, more efficient hospital of the future depends on the creativity and scholarship of HM leaders. Further, experts suggest that with its unique emphasis on quality, safety, and efficiency, the field will be a key player in healthcare reform.² Its strength lies in traditional clinical research, as well as further adoption of lessons from other fields including industry, ethnography, and public health.³ As such, fellowships to train future leaders and researchers is essential.

SHM’s website (www.hospitalmedicine.org/fellowships) lists dozens of IM hospitalist fellowships, as well as programs in family practice, pediatrics, and psychiatry. These programs last from one to three years, accept from one to six fellows per year, and exist in locations throughout the U.S. and Canada.

An excellent description of the nature and scope of pediatric HM fellowships was published last year in the Journal of Hospital Medicine.⁴ Descriptions of IM and HM fellowships also have been published.^3,5

Hospitalist fellowships, like IM fellowships, aren’t credentialed by a governing body. In contrast to subspecialty fellowships, no separate specialty board exam is required for admittance to the field after completion of fellowship. HM positions do not require training after residency, and HM job opportunities continue to outpace the available workforce. This is the basis for the most important question confronting anyone considering such a fellowship: How is a fellowship of benefit to a career as a hospitalist?

Program Types

Ranji and colleagues wrote that the “goal of hospital medicine fellowship training is to produce clinicians who are trained explicitly in studying and optimizing medical care of the hospitalized patient and in disseminating that knowledge for the advancement of patient care.”³ A review of information available for the different programs reveals two distinct approaches to this goal, with much overlap but distinct emphases:

Clinical programs usually last one year with a majority of time spent filling clinical responsibilities. In addition to providing focused exposure to HM with an emphasis on the Core Competencies in Hospital Medicine as outlined by SHM, such a program generally expands a trainee’s clinical scope. Additional training in palliative care, the management of neurologic emergencies, and comanagement of surgical patients are likely to be a part of clinical practice but often are underemphasized during residency. Research expectations vary, but most clinical programs allot some time for quality-improvement (QI) projects.

Clinical fellowships also afford more leadership training than most jobs would offer in the period immediately following residency. It also offers the possibility of refining clinical skills and developing a clinical niche.

 

Academic programs last two years and are characterized by two to four months of clinical responsibility per year. They offer a formal teaching curriculum and provide dedicated training in research, health policy, or health economics.³ Research training varies from program to program. Most include basic biostatistics and research-method coursework at a minimum; others offer the option to pursue a graduate degree in clinical research or public health.

Academic programs also offer dedicated research mentorship.

Other Considerations

The value of an HM fellowship lies in career development. The decision to commit to a relatively low-paying fellowship can be a difficult one, especially given the debt burden most graduating residents bear and the abundance of high-paying HM jobs. It also is important for those interested in a career as an academic hospitalist to consider not only HM fellowships, but other programs as well, such as the Robert Wood Johnson Clinical Scholars Program (rwjcsp.unc.edu/about/index.html).

While all of the fellowship programs aren’t geared specifically toward the hospitalist, they often incorporate faculty with expertise that would benefit a future academic hospitalist. Of course, the best fit for an individual depends on their particular interests and needs.

Fellowship in HM can offer training in clinical skills, clinical research, teaching, and quality and patient safety. Anyone interested in an HM career should consider a fellowship an opportunity for career development in a young specialty entrenched in revolutionizing the care of hospitalized patients. Academic HM fellowships hold the promise of empowering tomorrow’s academic leaders with the tools to continue to move the field forward. TH

Dr. Mann is a fellow in the division of hospital medicine, Department of Medicine, at Mount Sinai School of Medicine in New York City. Dr. Markoff is associate division chief and fellowship director in the division of hospital medicine, Department of Medicine, at Mount Sinai School of Medicine in New York City.

References

1. Wachter RM. Reflections: the hospitalist movement a decade later. J Hosp Med. 2006;1(4):248-252.
2. Wachter RM. Keynote presentation. SHM national meeting. National Harbor, Md.: May 2010.
3. Ranji SR, Rosenman DJ, Amin AN, Kripalani S. Hospital medicine fellowships: works in progress. Am J Med. 2006;119(1):72.e1-e7.
4. Freed GL, Dunham KM, Research advisory committee of the American Board of Pediatrics. Characteristics of pediatric hospital medicine fellowships and training programs. J Hosp Med. 2009;4:157-163.
5. Arora V, Fang MC, Kripalani S, Amin AN. Preparing for “diastole”: advanced training opportunities for academic hospitalists. J Hosp Med. 2006;1(6):368-377.

Although the term “hospitalist” was coined in 1996 in a New England Journal of Medicine article, the field of HM grew organically from pressure to optimize hospital economics and improve efficiency in economically pressed healthcare markets.¹

Scholarship in HM has also grown and now includes regular publications of investigations exploring optimization of efficiency and quality, many with an emphasis on patient safety. In this way, HM is a unique field, with tools for approaching problems that aren’t commonly used in other branches of medicine.

In parallel to the emergence of HM as a field distinct from general internal medicine (IM), the HM fellowship is similar but distinct. Such fellowships serve multiple purposes.

HM fellowships can add clinical expertise and scholarship skills for a career in HM. While early HM research focused on proving the value of the hospitalist model, the field has expanded greatly for those interested in an academic career. The molding of a safer, more efficient hospital of the future depends on the creativity and scholarship of HM leaders. Further, experts suggest that with its unique emphasis on quality, safety, and efficiency, the field will be a key player in healthcare reform.² Its strength lies in traditional clinical research, as well as further adoption of lessons from other fields including industry, ethnography, and public health.³ As such, fellowships to train future leaders and researchers is essential.

SHM’s website (www.hospitalmedicine.org/fellowships) lists dozens of IM hospitalist fellowships, as well as programs in family practice, pediatrics, and psychiatry. These programs last from one to three years, accept from one to six fellows per year, and exist in locations throughout the U.S. and Canada.

An excellent description of the nature and scope of pediatric HM fellowships was published last year in the Journal of Hospital Medicine.⁴ Descriptions of IM and HM fellowships also have been published.^3,5

Hospitalist fellowships, like IM fellowships, aren’t credentialed by a governing body. In contrast to subspecialty fellowships, no separate specialty board exam is required for admittance to the field after completion of fellowship. HM positions do not require training after residency, and HM job opportunities continue to outpace the available workforce. This is the basis for the most important question confronting anyone considering such a fellowship: How is a fellowship of benefit to a career as a hospitalist?

Program Types

Ranji and colleagues wrote that the “goal of hospital medicine fellowship training is to produce clinicians who are trained explicitly in studying and optimizing medical care of the hospitalized patient and in disseminating that knowledge for the advancement of patient care.”³ A review of information available for the different programs reveals two distinct approaches to this goal, with much overlap but distinct emphases:

Clinical programs usually last one year with a majority of time spent filling clinical responsibilities. In addition to providing focused exposure to HM with an emphasis on the Core Competencies in Hospital Medicine as outlined by SHM, such a program generally expands a trainee’s clinical scope. Additional training in palliative care, the management of neurologic emergencies, and comanagement of surgical patients are likely to be a part of clinical practice but often are underemphasized during residency. Research expectations vary, but most clinical programs allot some time for quality-improvement (QI) projects.

Clinical fellowships also afford more leadership training than most jobs would offer in the period immediately following residency. It also offers the possibility of refining clinical skills and developing a clinical niche.

 

Academic programs last two years and are characterized by two to four months of clinical responsibility per year. They offer a formal teaching curriculum and provide dedicated training in research, health policy, or health economics.³ Research training varies from program to program. Most include basic biostatistics and research-method coursework at a minimum; others offer the option to pursue a graduate degree in clinical research or public health.

Academic programs also offer dedicated research mentorship.

Other Considerations

The value of an HM fellowship lies in career development. The decision to commit to a relatively low-paying fellowship can be a difficult one, especially given the debt burden most graduating residents bear and the abundance of high-paying HM jobs. It also is important for those interested in a career as an academic hospitalist to consider not only HM fellowships, but other programs as well, such as the Robert Wood Johnson Clinical Scholars Program (rwjcsp.unc.edu/about/index.html).

While all of the fellowship programs aren’t geared specifically toward the hospitalist, they often incorporate faculty with expertise that would benefit a future academic hospitalist. Of course, the best fit for an individual depends on their particular interests and needs.

Fellowship in HM can offer training in clinical skills, clinical research, teaching, and quality and patient safety. Anyone interested in an HM career should consider a fellowship an opportunity for career development in a young specialty entrenched in revolutionizing the care of hospitalized patients. Academic HM fellowships hold the promise of empowering tomorrow’s academic leaders with the tools to continue to move the field forward. TH

Dr. Mann is a fellow in the division of hospital medicine, Department of Medicine, at Mount Sinai School of Medicine in New York City. Dr. Markoff is associate division chief and fellowship director in the division of hospital medicine, Department of Medicine, at Mount Sinai School of Medicine in New York City.

References

1. Wachter RM. Reflections: the hospitalist movement a decade later. J Hosp Med. 2006;1(4):248-252.
2. Wachter RM. Keynote presentation. SHM national meeting. National Harbor, Md.: May 2010.
3. Ranji SR, Rosenman DJ, Amin AN, Kripalani S. Hospital medicine fellowships: works in progress. Am J Med. 2006;119(1):72.e1-e7.
4. Freed GL, Dunham KM, Research advisory committee of the American Board of Pediatrics. Characteristics of pediatric hospital medicine fellowships and training programs. J Hosp Med. 2009;4:157-163.
5. Arora V, Fang MC, Kripalani S, Amin AN. Preparing for “diastole”: advanced training opportunities for academic hospitalists. J Hosp Med. 2006;1(6):368-377.

Although the term “hospitalist” was coined in 1996 in a New England Journal of Medicine article, the field of HM grew organically from pressure to optimize hospital economics and improve efficiency in economically pressed healthcare markets.¹

Scholarship in HM has also grown and now includes regular publications of investigations exploring optimization of efficiency and quality, many with an emphasis on patient safety. In this way, HM is a unique field, with tools for approaching problems that aren’t commonly used in other branches of medicine.

In parallel to the emergence of HM as a field distinct from general internal medicine (IM), the HM fellowship is similar but distinct. Such fellowships serve multiple purposes.

HM fellowships can add clinical expertise and scholarship skills for a career in HM. While early HM research focused on proving the value of the hospitalist model, the field has expanded greatly for those interested in an academic career. The molding of a safer, more efficient hospital of the future depends on the creativity and scholarship of HM leaders. Further, experts suggest that with its unique emphasis on quality, safety, and efficiency, the field will be a key player in healthcare reform.² Its strength lies in traditional clinical research, as well as further adoption of lessons from other fields including industry, ethnography, and public health.³ As such, fellowships to train future leaders and researchers is essential.

SHM’s website (www.hospitalmedicine.org/fellowships) lists dozens of IM hospitalist fellowships, as well as programs in family practice, pediatrics, and psychiatry. These programs last from one to three years, accept from one to six fellows per year, and exist in locations throughout the U.S. and Canada.

An excellent description of the nature and scope of pediatric HM fellowships was published last year in the Journal of Hospital Medicine.⁴ Descriptions of IM and HM fellowships also have been published.^3,5

Hospitalist fellowships, like IM fellowships, aren’t credentialed by a governing body. In contrast to subspecialty fellowships, no separate specialty board exam is required for admittance to the field after completion of fellowship. HM positions do not require training after residency, and HM job opportunities continue to outpace the available workforce. This is the basis for the most important question confronting anyone considering such a fellowship: How is a fellowship of benefit to a career as a hospitalist?

Program Types

Ranji and colleagues wrote that the “goal of hospital medicine fellowship training is to produce clinicians who are trained explicitly in studying and optimizing medical care of the hospitalized patient and in disseminating that knowledge for the advancement of patient care.”³ A review of information available for the different programs reveals two distinct approaches to this goal, with much overlap but distinct emphases:

Clinical programs usually last one year with a majority of time spent filling clinical responsibilities. In addition to providing focused exposure to HM with an emphasis on the Core Competencies in Hospital Medicine as outlined by SHM, such a program generally expands a trainee’s clinical scope. Additional training in palliative care, the management of neurologic emergencies, and comanagement of surgical patients are likely to be a part of clinical practice but often are underemphasized during residency. Research expectations vary, but most clinical programs allot some time for quality-improvement (QI) projects.

Clinical fellowships also afford more leadership training than most jobs would offer in the period immediately following residency. It also offers the possibility of refining clinical skills and developing a clinical niche.

 

Academic programs last two years and are characterized by two to four months of clinical responsibility per year. They offer a formal teaching curriculum and provide dedicated training in research, health policy, or health economics.³ Research training varies from program to program. Most include basic biostatistics and research-method coursework at a minimum; others offer the option to pursue a graduate degree in clinical research or public health.

Academic programs also offer dedicated research mentorship.

Other Considerations

The value of an HM fellowship lies in career development. The decision to commit to a relatively low-paying fellowship can be a difficult one, especially given the debt burden most graduating residents bear and the abundance of high-paying HM jobs. It also is important for those interested in a career as an academic hospitalist to consider not only HM fellowships, but other programs as well, such as the Robert Wood Johnson Clinical Scholars Program (rwjcsp.unc.edu/about/index.html).

While all of the fellowship programs aren’t geared specifically toward the hospitalist, they often incorporate faculty with expertise that would benefit a future academic hospitalist. Of course, the best fit for an individual depends on their particular interests and needs.

Fellowship in HM can offer training in clinical skills, clinical research, teaching, and quality and patient safety. Anyone interested in an HM career should consider a fellowship an opportunity for career development in a young specialty entrenched in revolutionizing the care of hospitalized patients. Academic HM fellowships hold the promise of empowering tomorrow’s academic leaders with the tools to continue to move the field forward. TH

Dr. Mann is a fellow in the division of hospital medicine, Department of Medicine, at Mount Sinai School of Medicine in New York City. Dr. Markoff is associate division chief and fellowship director in the division of hospital medicine, Department of Medicine, at Mount Sinai School of Medicine in New York City.

References

1. Wachter RM. Reflections: the hospitalist movement a decade later. J Hosp Med. 2006;1(4):248-252.
2. Wachter RM. Keynote presentation. SHM national meeting. National Harbor, Md.: May 2010.
3. Ranji SR, Rosenman DJ, Amin AN, Kripalani S. Hospital medicine fellowships: works in progress. Am J Med. 2006;119(1):72.e1-e7.
4. Freed GL, Dunham KM, Research advisory committee of the American Board of Pediatrics. Characteristics of pediatric hospital medicine fellowships and training programs. J Hosp Med. 2009;4:157-163.
5. Arora V, Fang MC, Kripalani S, Amin AN. Preparing for “diastole”: advanced training opportunities for academic hospitalists. J Hosp Med. 2006;1(6):368-377.

Recent media reports about the dangers surrounding unused prescription medications, including abuse by teens and medications finding their way into the water supply, have prompted an increase in inquiries to healthcare providers about disposing of unused medication. These issues are complicated when controlled substances are involved.

Often, providers are unsure how to respond to patient questions about medication disposal. For example, what would you do if a patient requests an alternative medication because of an unwanted side effect and brings the originally prescribed medication back to you? What if the family of a recently expired patient brings unused medication to you and asks you to donate it to other patients? What if you have a colleague who performs mission work; could you accept and donate unused medication for use in another country?

Unfortunately, the Controlled Substances Act (CSA) does not provide a readily available mechanism to accomplish efficient, secure, and environmentally sound methods to collect and use or dispose of unwanted controlled substances. This article explains the rules physicians must adhere to and guidelines for “taking back” controlled substances.

The Legislation

Enacted in 1970, the CSA combined all existing federal drug laws into a single statute. It created five “schedules” in which certain drugs are classified. These “scheduled” drugs are commonly referred to as controlled substances. A drug’s classification depends on its potential for abuse and its currently accepted medical use in the U.S. Additionally, provisions of international treaties impact classification.

Under the classification system, Schedule I drugs have a high potential for abuse and have no currently accepted medical use in treatment in the U.S. In contrast, Schedule V drugs have a low potential for abuse and do have a currently accepted medical use in treatment in the U.S.

The CSA governs the manufacture, import, export, possession, use, and distribution of controlled substances. In doing so, the CSA established a system to register those authorized to handle controlled substances. Manufacturers, dispensers, distributors, and individual practitioners who prescribe controlled substances must be registered with the Drug Enforcement Administration (DEA).

The CSA requires registrants to keep certain records for at least two years related to their handling of controlled substances. For example, physician registrants must keep records of controlled substances in Schedules II, III, IV, and V that are dispensed via methods other than prescribing or administering (e.g., industry samples). Inventories of controlled substances are required. Most notably, physicians generally are not required to keep records of prescribed medications; however, records must be kept if drugs are dispensed or administered. Moreover, there are heightened recordkeeping responsibilities for providers who prescribe, dispense, or administer for maintenance or detoxification.

Controlled Substance “Takeback”

The system of registration established by the CSA prohibits a DEA registrant from acquiring controlled substances from nonregistered entities and, in turn, bars an end-user from distributing pharmaceutical controlled substances to a DEA registrant. In other words, physicians cannot receive controlled substances from anyone who does not also have a registration. Thus, physicians may not “take back” prescribed medications from patients or their family members. Similarly, except in cases of a drug being recalled or a dispensing error, patients are not allowed to return controlled medications to a pharmacy.

Information on how a patient or family member should properly dispose of medication is commonly misunderstood. DEA regulations provide a process for nonregistrants to dispose of unused medication; however, it is cumbersome and meant to be used only when dealing with large quantities of controlled substances (e.g., large quantities of abandoned drugs). In such cases, the DEA special agent in charge (SAC) may instruct on disposal, which may include transfer of the substance to a DEA registrant, delivery to a DEA agent or office, destruction in the presence of an agent of the administration or other authorized person, or by other means. The person must submit a letter to the local SAC, which includes:

 

• Name and address of the person;
• Name and quantity of each controlled substance to be disposed of;
• Explanation of how the applicant obtained the controlled substance, if known; and
• Name, address, and registration number, if known, of the person who possessed the controlled substances prior to the applicant.

Federal legislation also provides a way for the DEA to grant approval to law-enforcement agencies to operate “takeback” programs. The regulation states that “any person in possession of a controlled substance and desiring to dispose of such substance may request assistance from the SAC in the area in which the person is located.” The regulation allows the SAC to authorize and specify the means of disposal to assure that the controlled substances do not become available to unauthorized persons.

State and local government agencies and community associations might hold takeback programs only if law enforcement makes the request, takes custody of the controlled substances, and is responsible for the disposal.

The U.S. Office of National Drug Control Policy has published guidelines for medication disposal. These guidelines advise flushing medications only if the prescription label or accompanying patient information specifically states to do so. Instead of flushing, the guidelines recommend that medications be disposed of through a takeback program or by:

• Taking the prescription drugs out of their original containers;
• Mixing the drugs with an undesirable substance, such as cat litter or used coffee grounds;
• Placing the mixture into a disposable container with a lid, such as an empty margarine tub, or into a sealable bag;
• Concealing or removing personal information, including Rx number, on the empty containers by covering it with black permanent marker or duct tape, or by scratching it off; and
• Placing the sealed container with the mixture, and the empty drug containers, in the trash.

Unused Medication Donation

The rising cost of prescription medication leaves many questioning whether there is a need for a safe method to allow unused medication to be donated to others. At least 10 states have passed laws allowing or encouraging the donation of unused pharmaceutical drugs. Many of these programs involve healthcare facilities, nursing homes, or pharmacies. The CSA and current DEA regulations, however, prohibit patients from delivering or distributing controlled substances to a DEA registrant, even if it is for the purpose of a donation. Moreover, the Food and Drug Administration (FDA) does not permit redistribution of medications, except under limited circumstances.

Consequently, state law may be inconsistent with federal law for donation and reuse of controlled substances.

Conclusion

Physicians who fail to comply with CSA handling requirements are subject to criminal charges, discipline against their DEA registration, and discipline against their license to practice medicine. Consequently, physicians should use caution whenever handling unused medication.

The application of various aspects of the CSA and implementing rules is situation-specific. Moreover, the DEA may issue additional regulations. Accordingly, if you have a question about a specific situation, consult an attorney, or contact your local DEA field division office and ask for the diversion duty agent. TH

Patrick O’Rourke works in the Office of University Counsel, Department of Litigation, University of Colorado Denver.

Recent media reports about the dangers surrounding unused prescription medications, including abuse by teens and medications finding their way into the water supply, have prompted an increase in inquiries to healthcare providers about disposing of unused medication. These issues are complicated when controlled substances are involved.

Often, providers are unsure how to respond to patient questions about medication disposal. For example, what would you do if a patient requests an alternative medication because of an unwanted side effect and brings the originally prescribed medication back to you? What if the family of a recently expired patient brings unused medication to you and asks you to donate it to other patients? What if you have a colleague who performs mission work; could you accept and donate unused medication for use in another country?

Unfortunately, the Controlled Substances Act (CSA) does not provide a readily available mechanism to accomplish efficient, secure, and environmentally sound methods to collect and use or dispose of unwanted controlled substances. This article explains the rules physicians must adhere to and guidelines for “taking back” controlled substances.

The Legislation

Enacted in 1970, the CSA combined all existing federal drug laws into a single statute. It created five “schedules” in which certain drugs are classified. These “scheduled” drugs are commonly referred to as controlled substances. A drug’s classification depends on its potential for abuse and its currently accepted medical use in the U.S. Additionally, provisions of international treaties impact classification.

Under the classification system, Schedule I drugs have a high potential for abuse and have no currently accepted medical use in treatment in the U.S. In contrast, Schedule V drugs have a low potential for abuse and do have a currently accepted medical use in treatment in the U.S.

The CSA governs the manufacture, import, export, possession, use, and distribution of controlled substances. In doing so, the CSA established a system to register those authorized to handle controlled substances. Manufacturers, dispensers, distributors, and individual practitioners who prescribe controlled substances must be registered with the Drug Enforcement Administration (DEA).

The CSA requires registrants to keep certain records for at least two years related to their handling of controlled substances. For example, physician registrants must keep records of controlled substances in Schedules II, III, IV, and V that are dispensed via methods other than prescribing or administering (e.g., industry samples). Inventories of controlled substances are required. Most notably, physicians generally are not required to keep records of prescribed medications; however, records must be kept if drugs are dispensed or administered. Moreover, there are heightened recordkeeping responsibilities for providers who prescribe, dispense, or administer for maintenance or detoxification.

Controlled Substance “Takeback”

The system of registration established by the CSA prohibits a DEA registrant from acquiring controlled substances from nonregistered entities and, in turn, bars an end-user from distributing pharmaceutical controlled substances to a DEA registrant. In other words, physicians cannot receive controlled substances from anyone who does not also have a registration. Thus, physicians may not “take back” prescribed medications from patients or their family members. Similarly, except in cases of a drug being recalled or a dispensing error, patients are not allowed to return controlled medications to a pharmacy.

Information on how a patient or family member should properly dispose of medication is commonly misunderstood. DEA regulations provide a process for nonregistrants to dispose of unused medication; however, it is cumbersome and meant to be used only when dealing with large quantities of controlled substances (e.g., large quantities of abandoned drugs). In such cases, the DEA special agent in charge (SAC) may instruct on disposal, which may include transfer of the substance to a DEA registrant, delivery to a DEA agent or office, destruction in the presence of an agent of the administration or other authorized person, or by other means. The person must submit a letter to the local SAC, which includes:

 

• Name and address of the person;
• Name and quantity of each controlled substance to be disposed of;
• Explanation of how the applicant obtained the controlled substance, if known; and
• Name, address, and registration number, if known, of the person who possessed the controlled substances prior to the applicant.

Federal legislation also provides a way for the DEA to grant approval to law-enforcement agencies to operate “takeback” programs. The regulation states that “any person in possession of a controlled substance and desiring to dispose of such substance may request assistance from the SAC in the area in which the person is located.” The regulation allows the SAC to authorize and specify the means of disposal to assure that the controlled substances do not become available to unauthorized persons.

State and local government agencies and community associations might hold takeback programs only if law enforcement makes the request, takes custody of the controlled substances, and is responsible for the disposal.

The U.S. Office of National Drug Control Policy has published guidelines for medication disposal. These guidelines advise flushing medications only if the prescription label or accompanying patient information specifically states to do so. Instead of flushing, the guidelines recommend that medications be disposed of through a takeback program or by:

• Taking the prescription drugs out of their original containers;
• Mixing the drugs with an undesirable substance, such as cat litter or used coffee grounds;
• Placing the mixture into a disposable container with a lid, such as an empty margarine tub, or into a sealable bag;
• Concealing or removing personal information, including Rx number, on the empty containers by covering it with black permanent marker or duct tape, or by scratching it off; and
• Placing the sealed container with the mixture, and the empty drug containers, in the trash.

Unused Medication Donation

The rising cost of prescription medication leaves many questioning whether there is a need for a safe method to allow unused medication to be donated to others. At least 10 states have passed laws allowing or encouraging the donation of unused pharmaceutical drugs. Many of these programs involve healthcare facilities, nursing homes, or pharmacies. The CSA and current DEA regulations, however, prohibit patients from delivering or distributing controlled substances to a DEA registrant, even if it is for the purpose of a donation. Moreover, the Food and Drug Administration (FDA) does not permit redistribution of medications, except under limited circumstances.

Consequently, state law may be inconsistent with federal law for donation and reuse of controlled substances.

Conclusion

Physicians who fail to comply with CSA handling requirements are subject to criminal charges, discipline against their DEA registration, and discipline against their license to practice medicine. Consequently, physicians should use caution whenever handling unused medication.

The application of various aspects of the CSA and implementing rules is situation-specific. Moreover, the DEA may issue additional regulations. Accordingly, if you have a question about a specific situation, consult an attorney, or contact your local DEA field division office and ask for the diversion duty agent. TH

Patrick O’Rourke works in the Office of University Counsel, Department of Litigation, University of Colorado Denver.

Recent media reports about the dangers surrounding unused prescription medications, including abuse by teens and medications finding their way into the water supply, have prompted an increase in inquiries to healthcare providers about disposing of unused medication. These issues are complicated when controlled substances are involved.

Often, providers are unsure how to respond to patient questions about medication disposal. For example, what would you do if a patient requests an alternative medication because of an unwanted side effect and brings the originally prescribed medication back to you? What if the family of a recently expired patient brings unused medication to you and asks you to donate it to other patients? What if you have a colleague who performs mission work; could you accept and donate unused medication for use in another country?

Unfortunately, the Controlled Substances Act (CSA) does not provide a readily available mechanism to accomplish efficient, secure, and environmentally sound methods to collect and use or dispose of unwanted controlled substances. This article explains the rules physicians must adhere to and guidelines for “taking back” controlled substances.

The Legislation

Enacted in 1970, the CSA combined all existing federal drug laws into a single statute. It created five “schedules” in which certain drugs are classified. These “scheduled” drugs are commonly referred to as controlled substances. A drug’s classification depends on its potential for abuse and its currently accepted medical use in the U.S. Additionally, provisions of international treaties impact classification.

Under the classification system, Schedule I drugs have a high potential for abuse and have no currently accepted medical use in treatment in the U.S. In contrast, Schedule V drugs have a low potential for abuse and do have a currently accepted medical use in treatment in the U.S.

The CSA governs the manufacture, import, export, possession, use, and distribution of controlled substances. In doing so, the CSA established a system to register those authorized to handle controlled substances. Manufacturers, dispensers, distributors, and individual practitioners who prescribe controlled substances must be registered with the Drug Enforcement Administration (DEA).

The CSA requires registrants to keep certain records for at least two years related to their handling of controlled substances. For example, physician registrants must keep records of controlled substances in Schedules II, III, IV, and V that are dispensed via methods other than prescribing or administering (e.g., industry samples). Inventories of controlled substances are required. Most notably, physicians generally are not required to keep records of prescribed medications; however, records must be kept if drugs are dispensed or administered. Moreover, there are heightened recordkeeping responsibilities for providers who prescribe, dispense, or administer for maintenance or detoxification.

Controlled Substance “Takeback”

The system of registration established by the CSA prohibits a DEA registrant from acquiring controlled substances from nonregistered entities and, in turn, bars an end-user from distributing pharmaceutical controlled substances to a DEA registrant. In other words, physicians cannot receive controlled substances from anyone who does not also have a registration. Thus, physicians may not “take back” prescribed medications from patients or their family members. Similarly, except in cases of a drug being recalled or a dispensing error, patients are not allowed to return controlled medications to a pharmacy.

Information on how a patient or family member should properly dispose of medication is commonly misunderstood. DEA regulations provide a process for nonregistrants to dispose of unused medication; however, it is cumbersome and meant to be used only when dealing with large quantities of controlled substances (e.g., large quantities of abandoned drugs). In such cases, the DEA special agent in charge (SAC) may instruct on disposal, which may include transfer of the substance to a DEA registrant, delivery to a DEA agent or office, destruction in the presence of an agent of the administration or other authorized person, or by other means. The person must submit a letter to the local SAC, which includes:

 

• Name and address of the person;
• Name and quantity of each controlled substance to be disposed of;
• Explanation of how the applicant obtained the controlled substance, if known; and
• Name, address, and registration number, if known, of the person who possessed the controlled substances prior to the applicant.

Federal legislation also provides a way for the DEA to grant approval to law-enforcement agencies to operate “takeback” programs. The regulation states that “any person in possession of a controlled substance and desiring to dispose of such substance may request assistance from the SAC in the area in which the person is located.” The regulation allows the SAC to authorize and specify the means of disposal to assure that the controlled substances do not become available to unauthorized persons.

State and local government agencies and community associations might hold takeback programs only if law enforcement makes the request, takes custody of the controlled substances, and is responsible for the disposal.

The U.S. Office of National Drug Control Policy has published guidelines for medication disposal. These guidelines advise flushing medications only if the prescription label or accompanying patient information specifically states to do so. Instead of flushing, the guidelines recommend that medications be disposed of through a takeback program or by:

• Taking the prescription drugs out of their original containers;
• Mixing the drugs with an undesirable substance, such as cat litter or used coffee grounds;
• Placing the mixture into a disposable container with a lid, such as an empty margarine tub, or into a sealable bag;
• Concealing or removing personal information, including Rx number, on the empty containers by covering it with black permanent marker or duct tape, or by scratching it off; and
• Placing the sealed container with the mixture, and the empty drug containers, in the trash.

Unused Medication Donation

The rising cost of prescription medication leaves many questioning whether there is a need for a safe method to allow unused medication to be donated to others. At least 10 states have passed laws allowing or encouraging the donation of unused pharmaceutical drugs. Many of these programs involve healthcare facilities, nursing homes, or pharmacies. The CSA and current DEA regulations, however, prohibit patients from delivering or distributing controlled substances to a DEA registrant, even if it is for the purpose of a donation. Moreover, the Food and Drug Administration (FDA) does not permit redistribution of medications, except under limited circumstances.

Consequently, state law may be inconsistent with federal law for donation and reuse of controlled substances.

Conclusion

Physicians who fail to comply with CSA handling requirements are subject to criminal charges, discipline against their DEA registration, and discipline against their license to practice medicine. Consequently, physicians should use caution whenever handling unused medication.

The application of various aspects of the CSA and implementing rules is situation-specific. Moreover, the DEA may issue additional regulations. Accordingly, if you have a question about a specific situation, consult an attorney, or contact your local DEA field division office and ask for the diversion duty agent. TH

Patrick O’Rourke works in the Office of University Counsel, Department of Litigation, University of Colorado Denver.

From the outset, HM has been about efficiency. And there was nothing wrong with that, for value is quality divided by cost. But in our story, we found that mere efficiency was not enough: The lowering of the denominator (cost) had to be met with an escalation of the numerator (quality) to ensure value.

And see us as being born in the right place at the right time. For with the national focus turning to the need for quality and patient safety, hospital medicine was in the right place and the right time to heed the call to action: appropriately stepping up to enact efforts to make the slope of the line (on a chart of quality vs. cost) “STEEEPER” … finding systems innovations to make care Safe, Timely, Efficient, Effective, Equitable, and Patient-centered.

Of course, the story continued with the Affordable Care Act and healthcare reform, greatly accelerating our evolution as change agents. And now we find ourselves fully invested in a “change the system” mentality, perfectly positioned to meaningfully change healthcare for millions of people. But threats loom—specifically, the “R” in the STEEEPER mnemonic: the risks to quality in the face of healthcare reform.

So in the next chapter of our story, I present to you our challenge: how to overcome the threats to quality in the context of healthcare reform. The first three are presented here; in subsequent articles, I will address the remainder. Overcoming all threats will hinge on mastering the four truisms of cultural change:

1. Systems drive function;
2. Every system is perfectly designed to produce the outcomes that it does;
3. This is not an issue of people needing to try harder; and
4. The “no blame” culture begins with a paradigm shift from the “person at fault” to the “system at fault.”

Threat 1: Failure to Fund Quality

SHM elected to merge its annual State of Hospital Medicine survey with the MGMA. Though not without risk, this has resulted in the anticipated benefits. The MGMA collaboration brings greater leverage in working with the C-suite, which is pre-programmed to react to MGMA surveys. From the most recent MGMA survey comes good news: The financial compensation for hospitalists has increased. A sobering insight, however, is that this increase in compensation has been met with a corresponding increase in work intensity—RVUs. Further, the link between RVUs and compensation appears to be tightening, quantifying what has long been of concern: The time devoted to the nonclinical “value added” duties of the hospitalist is shrinking.

The threat to the culture of quality is captured in the single question: How many RVUs is a quality-improvement (QI) project worth? I’m not sure we have that answer. But without an answer, it is difficult to believe that meaningful QI can be expected without time to do so. And again, as the gap between compensation and RVUs narrows, one is left wondering if there will soon be a day where there is no value-added time remaining to perform QI at all.

Fortunately, the Affordable Care Act might provide some movement in the right direction via value-based purchasing. Linking quality outcomes to financial reimbursement is a big step forward in the hospitalist’s quest to leverage the C-suite in trading RVUs for devoted QI time. Although we still are left asking the question of how many RVUs a QI project is worth, value-based purchasing at least sets the stage for the conversation. But in the interim, it is still upon hospitalists to design these QI projects, and to learn the skills necessary to see the design to its fruition.

 

Threat 2: Quality Stops at Core Measures

It is hard to argue that fulfilling “core measures” is a bad thing. Nonetheless, the core measures were not meant to be quality; instead, they were meant as surrogate measures of quality. The presumption of the core-measures initiative is that the system would exist without direct attention to the core measures, operating as it ordinarily would with generic attention to meeting all standards of quality for all diseases. And at some point in time, the core measures would be assessed to give an overall assessment of the system’s quality.

What has evolved, however, is a concerted attention to meeting the core measures, with little regard to the overall culture of quality.

Let’s say you were tasked with improving the public school system in your state. As a measure of the improvement, you choose five of the 1,000 schools as “core measure” schools. The state board of education is told that the performance of these five sample schools will be assessed at the end of the year, and financial support for the system as a whole will hinge on their performance. The intended result is that attention would be paid to improving the performance of every school in the system, and this improvement would be reflected in the performance of the five sample schools. The board of education could take the route of improving all schools, but the more pragmatic route would be to funnel all resources into these five schools, to the detriment of resources for the other 995 schools. The performance of the core measure schools would dramatically improve, and funding would be secured. But ask yourself: Did the performance of the school system as a whole actually improve?

Such is the risk of the core measures in healthcare. The original intent of the core measures was to instill a culture of QI for all points of care. And this has been a valuable contribution to changing the consciousness of the healthcare system. The presumption was that the core measures would be “seeds,” and that by emphasizing these select measures, the QI culture eventually would spread to all aspects of patient care. But this plan hinged on the presumption that that there is an unlimited amount of mental energy and resources to be devoted to all tasks within healthcare. The reality is that there is a fixed amount of intellectual energy and resources to be devoted to the various aspects of healthcare. One wonders if the overemphasis on meeting the core measures might actually have taken the wind out of the sails for QI in other non-core-measure patient care.

The implications are twofold. By definition, a core measure has to be applicable to all healthcare systems, and with a fixed amount of mental energy and resources, there is a real risk that what portion is reserved for QI finds its way only to the core indicators, especially if they are overemphasized in the system. The second implication is captured in our experience with time to antibiotics. With meeting the core indicator as the priority, many systems instituted the “work-around”: Give antibiotics to every patient presenting to the ED, and you will be sure to have met the four-hour window in the core indicator. The result was an exponential increase in inappropriate antibiotic administration and radiographic tests, all because meeting the indicator became more important than the overall goal.

As stewards of the hospital system, it is upon us to ensure that the original intent of core measures remains secure: The core measures seed a culture of quality, but do not become ends in and of themselves. QI apart from the core measures must remain an equal priority, and it is the hospitalist who will be central in ensuring this comes to fruition.

 

Threat 3: Misplaced Incentives

There is an interesting anecdote in Steven Levitt and Stephen Dubner’s book Freakonomics.¹ The story begins with a daycare center struggling with a problem: Some parents are showing up late to pick up their children at the end of the day, and this is costing the center in the way of overtime charges for the staff. To solve the problem, the center elects to institute a financial disincentive: Those showing up late to pick up their children will pay a modest financial penalty.

Fast-forward to months after the policy was put in place. The result? An exponential increase in the number of parents showing up late to pick up their children.

How do you explain worsening performance in spite of a financial disincentive? The answer resides in understanding human behavior. According to the authors, there are three primary motivations in life: financial, social, and moral. As ugly as it sounds, the decisions people make in life are driven by one of these three motivations. There is nothing wrong with providing incentives for behavior; incentives work.

But the danger arises when incentives are mismatched to behaviors. For example, if a financial outcome is the goal, then financial incentives make sense. If a social outcome (people should play better as a team) is the goal, the social incentives make sense (public recognition). But when the incentives get misaligned with their respective goals, trouble results.

What went wrong with the daycare’s plan is simple—most of the parents were motivated to pick up their children on time out of moral (“I gave my word”) or social (“I don’t want to be talked about by other parents”) incentives. But once a financial incentive was offered, the daycare center had essentially given the parents a way out in absolving their social and moral obligations. The parents had essentially cost-adjusted their behavior.

If you think this couldn’t happen to the healthcare system, let me ask you this. As a hospitalist, I see all of my patients early in the morning, because I see it as part of my obligation to the hospital team to discharge patients by 11 a.m. (social motivation).

But what if the CEO released this directive: “You will see all of your patients early in the morning, or you will take a $1,000 a year pay cut.” Is it possible that I might cost-adjust the $1,000 in exchange for sleeping in a little later and not having to deal with the morning traffic? I don’t know.

When it comes to financial incentives, there is a valley in the U-shaped curve. When the financial incentive is trivial, it is disregarded and the social/moral motivations of behavior persist (the kids are picked up on time; I persist in seeing patients early in the morning). When the financial incentive is huge, the financial incentive trumps all social/moral motivations, ensuring compliance with the goal behavior (every kid is picked up on time to avoid a penalty; I see all patients early in the day to avoid a larger penalty).

But in between is the risk zone: When the person feels they are paying an appropriate penance for not complying with the goal behavior, the financial disincentive absolves any social/moral guilt.

Healthcare reform is about incentives—and there is nothing wrong with that. But as the stewards of the inpatient healthcare system, it is upon us as hospitalists to ensure that the incentives remain matched to their intended goals, and that the untoward consequences of the incentives do not adversely affect the quality and safety of a patient’s care.

 

It is safe to say that the Affordable Care Act of 2010 moves us closer to a true environment of quality and patient safety. But it is equally safe to say that meaningful change will require more than what the law can provide. As stewards of the inpatient system, we have a responsibility to ensure that the healthcare system, particularly in how it responds to incentives, evolves to remain patient-centered, effective, and safe.

The next chapter in our story—the hospitalists’ story—will be one of accountability and responsibility. While there are things the government can do, the majority of what needs to be done will come directly from us. TH

Dr. Wiese is president of SHM.

Reference

1. Levitt SD, Dubner SJ. Freakonomics: A Rogue Economist Explores the Hidden Side of Everything. New York City: William Morrow; 2005.

From the outset, HM has been about efficiency. And there was nothing wrong with that, for value is quality divided by cost. But in our story, we found that mere efficiency was not enough: The lowering of the denominator (cost) had to be met with an escalation of the numerator (quality) to ensure value.

And see us as being born in the right place at the right time. For with the national focus turning to the need for quality and patient safety, hospital medicine was in the right place and the right time to heed the call to action: appropriately stepping up to enact efforts to make the slope of the line (on a chart of quality vs. cost) “STEEEPER” … finding systems innovations to make care Safe, Timely, Efficient, Effective, Equitable, and Patient-centered.

Of course, the story continued with the Affordable Care Act and healthcare reform, greatly accelerating our evolution as change agents. And now we find ourselves fully invested in a “change the system” mentality, perfectly positioned to meaningfully change healthcare for millions of people. But threats loom—specifically, the “R” in the STEEEPER mnemonic: the risks to quality in the face of healthcare reform.

So in the next chapter of our story, I present to you our challenge: how to overcome the threats to quality in the context of healthcare reform. The first three are presented here; in subsequent articles, I will address the remainder. Overcoming all threats will hinge on mastering the four truisms of cultural change:

1. Systems drive function;
2. Every system is perfectly designed to produce the outcomes that it does;
3. This is not an issue of people needing to try harder; and
4. The “no blame” culture begins with a paradigm shift from the “person at fault” to the “system at fault.”

Threat 1: Failure to Fund Quality

SHM elected to merge its annual State of Hospital Medicine survey with the MGMA. Though not without risk, this has resulted in the anticipated benefits. The MGMA collaboration brings greater leverage in working with the C-suite, which is pre-programmed to react to MGMA surveys. From the most recent MGMA survey comes good news: The financial compensation for hospitalists has increased. A sobering insight, however, is that this increase in compensation has been met with a corresponding increase in work intensity—RVUs. Further, the link between RVUs and compensation appears to be tightening, quantifying what has long been of concern: The time devoted to the nonclinical “value added” duties of the hospitalist is shrinking.

The threat to the culture of quality is captured in the single question: How many RVUs is a quality-improvement (QI) project worth? I’m not sure we have that answer. But without an answer, it is difficult to believe that meaningful QI can be expected without time to do so. And again, as the gap between compensation and RVUs narrows, one is left wondering if there will soon be a day where there is no value-added time remaining to perform QI at all.

Fortunately, the Affordable Care Act might provide some movement in the right direction via value-based purchasing. Linking quality outcomes to financial reimbursement is a big step forward in the hospitalist’s quest to leverage the C-suite in trading RVUs for devoted QI time. Although we still are left asking the question of how many RVUs a QI project is worth, value-based purchasing at least sets the stage for the conversation. But in the interim, it is still upon hospitalists to design these QI projects, and to learn the skills necessary to see the design to its fruition.

 

Threat 2: Quality Stops at Core Measures

It is hard to argue that fulfilling “core measures” is a bad thing. Nonetheless, the core measures were not meant to be quality; instead, they were meant as surrogate measures of quality. The presumption of the core-measures initiative is that the system would exist without direct attention to the core measures, operating as it ordinarily would with generic attention to meeting all standards of quality for all diseases. And at some point in time, the core measures would be assessed to give an overall assessment of the system’s quality.

What has evolved, however, is a concerted attention to meeting the core measures, with little regard to the overall culture of quality.

Let’s say you were tasked with improving the public school system in your state. As a measure of the improvement, you choose five of the 1,000 schools as “core measure” schools. The state board of education is told that the performance of these five sample schools will be assessed at the end of the year, and financial support for the system as a whole will hinge on their performance. The intended result is that attention would be paid to improving the performance of every school in the system, and this improvement would be reflected in the performance of the five sample schools. The board of education could take the route of improving all schools, but the more pragmatic route would be to funnel all resources into these five schools, to the detriment of resources for the other 995 schools. The performance of the core measure schools would dramatically improve, and funding would be secured. But ask yourself: Did the performance of the school system as a whole actually improve?

Such is the risk of the core measures in healthcare. The original intent of the core measures was to instill a culture of QI for all points of care. And this has been a valuable contribution to changing the consciousness of the healthcare system. The presumption was that the core measures would be “seeds,” and that by emphasizing these select measures, the QI culture eventually would spread to all aspects of patient care. But this plan hinged on the presumption that that there is an unlimited amount of mental energy and resources to be devoted to all tasks within healthcare. The reality is that there is a fixed amount of intellectual energy and resources to be devoted to the various aspects of healthcare. One wonders if the overemphasis on meeting the core measures might actually have taken the wind out of the sails for QI in other non-core-measure patient care.

The implications are twofold. By definition, a core measure has to be applicable to all healthcare systems, and with a fixed amount of mental energy and resources, there is a real risk that what portion is reserved for QI finds its way only to the core indicators, especially if they are overemphasized in the system. The second implication is captured in our experience with time to antibiotics. With meeting the core indicator as the priority, many systems instituted the “work-around”: Give antibiotics to every patient presenting to the ED, and you will be sure to have met the four-hour window in the core indicator. The result was an exponential increase in inappropriate antibiotic administration and radiographic tests, all because meeting the indicator became more important than the overall goal.

As stewards of the hospital system, it is upon us to ensure that the original intent of core measures remains secure: The core measures seed a culture of quality, but do not become ends in and of themselves. QI apart from the core measures must remain an equal priority, and it is the hospitalist who will be central in ensuring this comes to fruition.

 

Threat 3: Misplaced Incentives

There is an interesting anecdote in Steven Levitt and Stephen Dubner’s book Freakonomics.¹ The story begins with a daycare center struggling with a problem: Some parents are showing up late to pick up their children at the end of the day, and this is costing the center in the way of overtime charges for the staff. To solve the problem, the center elects to institute a financial disincentive: Those showing up late to pick up their children will pay a modest financial penalty.

Fast-forward to months after the policy was put in place. The result? An exponential increase in the number of parents showing up late to pick up their children.

How do you explain worsening performance in spite of a financial disincentive? The answer resides in understanding human behavior. According to the authors, there are three primary motivations in life: financial, social, and moral. As ugly as it sounds, the decisions people make in life are driven by one of these three motivations. There is nothing wrong with providing incentives for behavior; incentives work.

But the danger arises when incentives are mismatched to behaviors. For example, if a financial outcome is the goal, then financial incentives make sense. If a social outcome (people should play better as a team) is the goal, the social incentives make sense (public recognition). But when the incentives get misaligned with their respective goals, trouble results.

What went wrong with the daycare’s plan is simple—most of the parents were motivated to pick up their children on time out of moral (“I gave my word”) or social (“I don’t want to be talked about by other parents”) incentives. But once a financial incentive was offered, the daycare center had essentially given the parents a way out in absolving their social and moral obligations. The parents had essentially cost-adjusted their behavior.

If you think this couldn’t happen to the healthcare system, let me ask you this. As a hospitalist, I see all of my patients early in the morning, because I see it as part of my obligation to the hospital team to discharge patients by 11 a.m. (social motivation).

But what if the CEO released this directive: “You will see all of your patients early in the morning, or you will take a $1,000 a year pay cut.” Is it possible that I might cost-adjust the $1,000 in exchange for sleeping in a little later and not having to deal with the morning traffic? I don’t know.

When it comes to financial incentives, there is a valley in the U-shaped curve. When the financial incentive is trivial, it is disregarded and the social/moral motivations of behavior persist (the kids are picked up on time; I persist in seeing patients early in the morning). When the financial incentive is huge, the financial incentive trumps all social/moral motivations, ensuring compliance with the goal behavior (every kid is picked up on time to avoid a penalty; I see all patients early in the day to avoid a larger penalty).

But in between is the risk zone: When the person feels they are paying an appropriate penance for not complying with the goal behavior, the financial disincentive absolves any social/moral guilt.

Healthcare reform is about incentives—and there is nothing wrong with that. But as the stewards of the inpatient healthcare system, it is upon us as hospitalists to ensure that the incentives remain matched to their intended goals, and that the untoward consequences of the incentives do not adversely affect the quality and safety of a patient’s care.

 

It is safe to say that the Affordable Care Act of 2010 moves us closer to a true environment of quality and patient safety. But it is equally safe to say that meaningful change will require more than what the law can provide. As stewards of the inpatient system, we have a responsibility to ensure that the healthcare system, particularly in how it responds to incentives, evolves to remain patient-centered, effective, and safe.

The next chapter in our story—the hospitalists’ story—will be one of accountability and responsibility. While there are things the government can do, the majority of what needs to be done will come directly from us. TH

Dr. Wiese is president of SHM.

Reference

1. Levitt SD, Dubner SJ. Freakonomics: A Rogue Economist Explores the Hidden Side of Everything. New York City: William Morrow; 2005.

From the outset, HM has been about efficiency. And there was nothing wrong with that, for value is quality divided by cost. But in our story, we found that mere efficiency was not enough: The lowering of the denominator (cost) had to be met with an escalation of the numerator (quality) to ensure value.

And see us as being born in the right place at the right time. For with the national focus turning to the need for quality and patient safety, hospital medicine was in the right place and the right time to heed the call to action: appropriately stepping up to enact efforts to make the slope of the line (on a chart of quality vs. cost) “STEEEPER” … finding systems innovations to make care Safe, Timely, Efficient, Effective, Equitable, and Patient-centered.

Of course, the story continued with the Affordable Care Act and healthcare reform, greatly accelerating our evolution as change agents. And now we find ourselves fully invested in a “change the system” mentality, perfectly positioned to meaningfully change healthcare for millions of people. But threats loom—specifically, the “R” in the STEEEPER mnemonic: the risks to quality in the face of healthcare reform.

So in the next chapter of our story, I present to you our challenge: how to overcome the threats to quality in the context of healthcare reform. The first three are presented here; in subsequent articles, I will address the remainder. Overcoming all threats will hinge on mastering the four truisms of cultural change:

1. Systems drive function;
2. Every system is perfectly designed to produce the outcomes that it does;
3. This is not an issue of people needing to try harder; and
4. The “no blame” culture begins with a paradigm shift from the “person at fault” to the “system at fault.”

Threat 1: Failure to Fund Quality

SHM elected to merge its annual State of Hospital Medicine survey with the MGMA. Though not without risk, this has resulted in the anticipated benefits. The MGMA collaboration brings greater leverage in working with the C-suite, which is pre-programmed to react to MGMA surveys. From the most recent MGMA survey comes good news: The financial compensation for hospitalists has increased. A sobering insight, however, is that this increase in compensation has been met with a corresponding increase in work intensity—RVUs. Further, the link between RVUs and compensation appears to be tightening, quantifying what has long been of concern: The time devoted to the nonclinical “value added” duties of the hospitalist is shrinking.

The threat to the culture of quality is captured in the single question: How many RVUs is a quality-improvement (QI) project worth? I’m not sure we have that answer. But without an answer, it is difficult to believe that meaningful QI can be expected without time to do so. And again, as the gap between compensation and RVUs narrows, one is left wondering if there will soon be a day where there is no value-added time remaining to perform QI at all.

Fortunately, the Affordable Care Act might provide some movement in the right direction via value-based purchasing. Linking quality outcomes to financial reimbursement is a big step forward in the hospitalist’s quest to leverage the C-suite in trading RVUs for devoted QI time. Although we still are left asking the question of how many RVUs a QI project is worth, value-based purchasing at least sets the stage for the conversation. But in the interim, it is still upon hospitalists to design these QI projects, and to learn the skills necessary to see the design to its fruition.

 

Threat 2: Quality Stops at Core Measures

It is hard to argue that fulfilling “core measures” is a bad thing. Nonetheless, the core measures were not meant to be quality; instead, they were meant as surrogate measures of quality. The presumption of the core-measures initiative is that the system would exist without direct attention to the core measures, operating as it ordinarily would with generic attention to meeting all standards of quality for all diseases. And at some point in time, the core measures would be assessed to give an overall assessment of the system’s quality.

What has evolved, however, is a concerted attention to meeting the core measures, with little regard to the overall culture of quality.

Let’s say you were tasked with improving the public school system in your state. As a measure of the improvement, you choose five of the 1,000 schools as “core measure” schools. The state board of education is told that the performance of these five sample schools will be assessed at the end of the year, and financial support for the system as a whole will hinge on their performance. The intended result is that attention would be paid to improving the performance of every school in the system, and this improvement would be reflected in the performance of the five sample schools. The board of education could take the route of improving all schools, but the more pragmatic route would be to funnel all resources into these five schools, to the detriment of resources for the other 995 schools. The performance of the core measure schools would dramatically improve, and funding would be secured. But ask yourself: Did the performance of the school system as a whole actually improve?

Such is the risk of the core measures in healthcare. The original intent of the core measures was to instill a culture of QI for all points of care. And this has been a valuable contribution to changing the consciousness of the healthcare system. The presumption was that the core measures would be “seeds,” and that by emphasizing these select measures, the QI culture eventually would spread to all aspects of patient care. But this plan hinged on the presumption that that there is an unlimited amount of mental energy and resources to be devoted to all tasks within healthcare. The reality is that there is a fixed amount of intellectual energy and resources to be devoted to the various aspects of healthcare. One wonders if the overemphasis on meeting the core measures might actually have taken the wind out of the sails for QI in other non-core-measure patient care.

The implications are twofold. By definition, a core measure has to be applicable to all healthcare systems, and with a fixed amount of mental energy and resources, there is a real risk that what portion is reserved for QI finds its way only to the core indicators, especially if they are overemphasized in the system. The second implication is captured in our experience with time to antibiotics. With meeting the core indicator as the priority, many systems instituted the “work-around”: Give antibiotics to every patient presenting to the ED, and you will be sure to have met the four-hour window in the core indicator. The result was an exponential increase in inappropriate antibiotic administration and radiographic tests, all because meeting the indicator became more important than the overall goal.

As stewards of the hospital system, it is upon us to ensure that the original intent of core measures remains secure: The core measures seed a culture of quality, but do not become ends in and of themselves. QI apart from the core measures must remain an equal priority, and it is the hospitalist who will be central in ensuring this comes to fruition.

 

Threat 3: Misplaced Incentives

There is an interesting anecdote in Steven Levitt and Stephen Dubner’s book Freakonomics.¹ The story begins with a daycare center struggling with a problem: Some parents are showing up late to pick up their children at the end of the day, and this is costing the center in the way of overtime charges for the staff. To solve the problem, the center elects to institute a financial disincentive: Those showing up late to pick up their children will pay a modest financial penalty.

Fast-forward to months after the policy was put in place. The result? An exponential increase in the number of parents showing up late to pick up their children.

How do you explain worsening performance in spite of a financial disincentive? The answer resides in understanding human behavior. According to the authors, there are three primary motivations in life: financial, social, and moral. As ugly as it sounds, the decisions people make in life are driven by one of these three motivations. There is nothing wrong with providing incentives for behavior; incentives work.

But the danger arises when incentives are mismatched to behaviors. For example, if a financial outcome is the goal, then financial incentives make sense. If a social outcome (people should play better as a team) is the goal, the social incentives make sense (public recognition). But when the incentives get misaligned with their respective goals, trouble results.

What went wrong with the daycare’s plan is simple—most of the parents were motivated to pick up their children on time out of moral (“I gave my word”) or social (“I don’t want to be talked about by other parents”) incentives. But once a financial incentive was offered, the daycare center had essentially given the parents a way out in absolving their social and moral obligations. The parents had essentially cost-adjusted their behavior.

If you think this couldn’t happen to the healthcare system, let me ask you this. As a hospitalist, I see all of my patients early in the morning, because I see it as part of my obligation to the hospital team to discharge patients by 11 a.m. (social motivation).

But what if the CEO released this directive: “You will see all of your patients early in the morning, or you will take a $1,000 a year pay cut.” Is it possible that I might cost-adjust the $1,000 in exchange for sleeping in a little later and not having to deal with the morning traffic? I don’t know.

When it comes to financial incentives, there is a valley in the U-shaped curve. When the financial incentive is trivial, it is disregarded and the social/moral motivations of behavior persist (the kids are picked up on time; I persist in seeing patients early in the morning). When the financial incentive is huge, the financial incentive trumps all social/moral motivations, ensuring compliance with the goal behavior (every kid is picked up on time to avoid a penalty; I see all patients early in the day to avoid a larger penalty).

But in between is the risk zone: When the person feels they are paying an appropriate penance for not complying with the goal behavior, the financial disincentive absolves any social/moral guilt.

Healthcare reform is about incentives—and there is nothing wrong with that. But as the stewards of the inpatient healthcare system, it is upon us as hospitalists to ensure that the incentives remain matched to their intended goals, and that the untoward consequences of the incentives do not adversely affect the quality and safety of a patient’s care.

 

It is safe to say that the Affordable Care Act of 2010 moves us closer to a true environment of quality and patient safety. But it is equally safe to say that meaningful change will require more than what the law can provide. As stewards of the inpatient system, we have a responsibility to ensure that the healthcare system, particularly in how it responds to incentives, evolves to remain patient-centered, effective, and safe.

The next chapter in our story—the hospitalists’ story—will be one of accountability and responsibility. While there are things the government can do, the majority of what needs to be done will come directly from us. TH

Dr. Wiese is president of SHM.

Reference

1. Levitt SD, Dubner SJ. Freakonomics: A Rogue Economist Explores the Hidden Side of Everything. New York City: William Morrow; 2005.

Several months ago, my toilet broke. You should also know that I’m not particularly handy. So when I first realized that the toilet bowl seemed to fill constantly, I got a little stressed out.

How much was it going cost to call in a plumber on the weekend?

What kind of a water bill was I going to have?

Was this a serious problem?

I took a quick peek in the tank, but that just made me more confused. I was paralyzed by a lack of know-how.

Normally, I would have just Googled a local plumber. But that day, I decided to do something different. Maybe it was because it was the fantasy football offseason. Maybe it was because my wife had started to ask my father-in-law to change light bulbs around the house. Or, maybe, I wanted to learn to actually fix the problem. A few hours later, after an Internet lesson in toilet physiology, a $4.12 trip to Home Depot, and a wet pair of hands, I had replaced my first toilet flapper.

This wasn’t the rebuilding of a car engine, but it was a clear DIY step toward self-improvement. Easily the most memorable moment here was my sense of accomplishment.

I felt empowered.

One Part Science, One Part Art

It’s taken me a while to realize this, but I’ve begun to take advantage of improvement opportunities at work as well. No, I haven’t been moonlighting as a plumber for my hospital. I’ve just been fortunate to be part of a trifecta of rewarding quality-improvement (QI) projects over the past year. Before I’d gotten my hands dirty with these, my understanding of QI was fairly naive. I’d heard about Plan-Do-Study-Act many times. I had listened to a talk at a national conference. And I had kept up with the general medical literature on the subject.

But none of those activities had truly prepared me for experience of actually doing the work on my own.

By taking on a project, an ambitious attempt to reduce continuous pulse oximetry use, I experienced a crash course in both the science and the art of process improvement. I was forced to overcome my “I don’t know how” inertia. And with expert guidance in the form of a clinical safety and effectiveness class, I learned the importance of run charts (science) and a well-crafted multidisciplinary team (art) in changing established but inefficient behavior.

Our rates of continuous pulse oximetry usage dropped by 50%, and cost savings were $12,000 per year on one unit. These results made my prior attempts at change—years of complaining about ingrained nursing culture—look infantile. (OK, maybe it was ineffective, but who hasn’t complained about the overuse of continuous monitoring?)

I haven’t met a pediatric hospitalist who wouldn’t understand the symbolic importance of this success. But I know of many hospitalists who have not yet participated in meaningful QI project. Imagine calling a plumber who grasped the flush and fill mechanism of a toilet but had never touched real porcelain. Here’s an even better analogy: What if doctors could get licensed without having touched real patients?

If pediatric hospitalists are to transform the care delivery of hospitalized children, and quality learning only comes through hands-on training, then we need some more hands in the pot.

Discharge Improvement

On the heels of my first project, I was fortunate enough to augment my education through another effort—this time with a cohort of fellow pediatric hospitalists. This was a national collaborative to improve discharge handoffs, and I will admit that, at the outset, I was as puzzled as the first time I pulled the lid off the tank of the toilet. There were just too many permutations on PCP communication at the participating institutions, and some felt our aim of timely discharge handoffs was unattainable.

 

What carried me through, however, was the collective and infectious DIY—no, QIY (Quality Improve-it-Yourself) attitude of the group. We were all learning, and regular participation in the collaborative essentially guaranteed improvement. We achieved our aim of 90% communication with PCPs within two days of discharge. The secret was simple: The more you do, the more you learn.

Pediatric hospitalists can transform care delivery through a focus on safe and quality care, but the tools to accomplish this must come through post-residency, on-the-job learning. This QI know-how must efficiently spread among our ranks through practical and project-based educational efforts. It’s “see one, do one, teach one,” but we’re not talking about lumbar punctures anymore.

This is a journey in which we all take on the responsibility of rolling up our sleeves and simply learn by doing. And here is where the third leg of my as-yet-unfinished QI course unfolds.

Through my involvement with the Value in Inpatient Pediatrics (VIP) Network, I’ve gained a newfound vision for what the future might hold. VIP has evolved from a benchmarking project focused on bronchiolitis to an improvement network that will incorporate projects similar to the discharge handoff collaborative above.

In the process, a model for how to rapidly spread QI learning has emerged. The capacity lies in the network’s rapidly growing connectivity. The power comes from the individuals: motivated, card-carrying pediatric hospitalists from a wide array of sites. Collaborative learning harbors the potential to exponentially increase the pace at which we improve.

The future of our quality care is bright. I see an open network of improvement doers and learners. I see collaboration on quality and safety initiatives in all manner of hospitals and communities. I see that this will all be built upon a foundation of hard work and a QIY attitude.

You, too, will play a role.

Just don’t be afraid to get your hands a little dirty. TH

Dr. Shen is medical director of hospital medicine at Dell Children’s Medical Center in Austin, Texas. He is pediatric editor of The Hospitalist.

Several months ago, my toilet broke. You should also know that I’m not particularly handy. So when I first realized that the toilet bowl seemed to fill constantly, I got a little stressed out.

How much was it going cost to call in a plumber on the weekend?

What kind of a water bill was I going to have?

Was this a serious problem?

I took a quick peek in the tank, but that just made me more confused. I was paralyzed by a lack of know-how.

Normally, I would have just Googled a local plumber. But that day, I decided to do something different. Maybe it was because it was the fantasy football offseason. Maybe it was because my wife had started to ask my father-in-law to change light bulbs around the house. Or, maybe, I wanted to learn to actually fix the problem. A few hours later, after an Internet lesson in toilet physiology, a $4.12 trip to Home Depot, and a wet pair of hands, I had replaced my first toilet flapper.

This wasn’t the rebuilding of a car engine, but it was a clear DIY step toward self-improvement. Easily the most memorable moment here was my sense of accomplishment.

I felt empowered.

One Part Science, One Part Art

It’s taken me a while to realize this, but I’ve begun to take advantage of improvement opportunities at work as well. No, I haven’t been moonlighting as a plumber for my hospital. I’ve just been fortunate to be part of a trifecta of rewarding quality-improvement (QI) projects over the past year. Before I’d gotten my hands dirty with these, my understanding of QI was fairly naive. I’d heard about Plan-Do-Study-Act many times. I had listened to a talk at a national conference. And I had kept up with the general medical literature on the subject.

But none of those activities had truly prepared me for experience of actually doing the work on my own.

By taking on a project, an ambitious attempt to reduce continuous pulse oximetry use, I experienced a crash course in both the science and the art of process improvement. I was forced to overcome my “I don’t know how” inertia. And with expert guidance in the form of a clinical safety and effectiveness class, I learned the importance of run charts (science) and a well-crafted multidisciplinary team (art) in changing established but inefficient behavior.

Our rates of continuous pulse oximetry usage dropped by 50%, and cost savings were $12,000 per year on one unit. These results made my prior attempts at change—years of complaining about ingrained nursing culture—look infantile. (OK, maybe it was ineffective, but who hasn’t complained about the overuse of continuous monitoring?)

I haven’t met a pediatric hospitalist who wouldn’t understand the symbolic importance of this success. But I know of many hospitalists who have not yet participated in meaningful QI project. Imagine calling a plumber who grasped the flush and fill mechanism of a toilet but had never touched real porcelain. Here’s an even better analogy: What if doctors could get licensed without having touched real patients?

If pediatric hospitalists are to transform the care delivery of hospitalized children, and quality learning only comes through hands-on training, then we need some more hands in the pot.

Discharge Improvement

On the heels of my first project, I was fortunate enough to augment my education through another effort—this time with a cohort of fellow pediatric hospitalists. This was a national collaborative to improve discharge handoffs, and I will admit that, at the outset, I was as puzzled as the first time I pulled the lid off the tank of the toilet. There were just too many permutations on PCP communication at the participating institutions, and some felt our aim of timely discharge handoffs was unattainable.

 

What carried me through, however, was the collective and infectious DIY—no, QIY (Quality Improve-it-Yourself) attitude of the group. We were all learning, and regular participation in the collaborative essentially guaranteed improvement. We achieved our aim of 90% communication with PCPs within two days of discharge. The secret was simple: The more you do, the more you learn.

Pediatric hospitalists can transform care delivery through a focus on safe and quality care, but the tools to accomplish this must come through post-residency, on-the-job learning. This QI know-how must efficiently spread among our ranks through practical and project-based educational efforts. It’s “see one, do one, teach one,” but we’re not talking about lumbar punctures anymore.

This is a journey in which we all take on the responsibility of rolling up our sleeves and simply learn by doing. And here is where the third leg of my as-yet-unfinished QI course unfolds.

Through my involvement with the Value in Inpatient Pediatrics (VIP) Network, I’ve gained a newfound vision for what the future might hold. VIP has evolved from a benchmarking project focused on bronchiolitis to an improvement network that will incorporate projects similar to the discharge handoff collaborative above.

In the process, a model for how to rapidly spread QI learning has emerged. The capacity lies in the network’s rapidly growing connectivity. The power comes from the individuals: motivated, card-carrying pediatric hospitalists from a wide array of sites. Collaborative learning harbors the potential to exponentially increase the pace at which we improve.

The future of our quality care is bright. I see an open network of improvement doers and learners. I see collaboration on quality and safety initiatives in all manner of hospitals and communities. I see that this will all be built upon a foundation of hard work and a QIY attitude.

You, too, will play a role.

Just don’t be afraid to get your hands a little dirty. TH

Dr. Shen is medical director of hospital medicine at Dell Children’s Medical Center in Austin, Texas. He is pediatric editor of The Hospitalist.

Several months ago, my toilet broke. You should also know that I’m not particularly handy. So when I first realized that the toilet bowl seemed to fill constantly, I got a little stressed out.

How much was it going cost to call in a plumber on the weekend?

What kind of a water bill was I going to have?

Was this a serious problem?

I took a quick peek in the tank, but that just made me more confused. I was paralyzed by a lack of know-how.

Normally, I would have just Googled a local plumber. But that day, I decided to do something different. Maybe it was because it was the fantasy football offseason. Maybe it was because my wife had started to ask my father-in-law to change light bulbs around the house. Or, maybe, I wanted to learn to actually fix the problem. A few hours later, after an Internet lesson in toilet physiology, a $4.12 trip to Home Depot, and a wet pair of hands, I had replaced my first toilet flapper.

This wasn’t the rebuilding of a car engine, but it was a clear DIY step toward self-improvement. Easily the most memorable moment here was my sense of accomplishment.

I felt empowered.

One Part Science, One Part Art

It’s taken me a while to realize this, but I’ve begun to take advantage of improvement opportunities at work as well. No, I haven’t been moonlighting as a plumber for my hospital. I’ve just been fortunate to be part of a trifecta of rewarding quality-improvement (QI) projects over the past year. Before I’d gotten my hands dirty with these, my understanding of QI was fairly naive. I’d heard about Plan-Do-Study-Act many times. I had listened to a talk at a national conference. And I had kept up with the general medical literature on the subject.

But none of those activities had truly prepared me for experience of actually doing the work on my own.

By taking on a project, an ambitious attempt to reduce continuous pulse oximetry use, I experienced a crash course in both the science and the art of process improvement. I was forced to overcome my “I don’t know how” inertia. And with expert guidance in the form of a clinical safety and effectiveness class, I learned the importance of run charts (science) and a well-crafted multidisciplinary team (art) in changing established but inefficient behavior.

Our rates of continuous pulse oximetry usage dropped by 50%, and cost savings were $12,000 per year on one unit. These results made my prior attempts at change—years of complaining about ingrained nursing culture—look infantile. (OK, maybe it was ineffective, but who hasn’t complained about the overuse of continuous monitoring?)

I haven’t met a pediatric hospitalist who wouldn’t understand the symbolic importance of this success. But I know of many hospitalists who have not yet participated in meaningful QI project. Imagine calling a plumber who grasped the flush and fill mechanism of a toilet but had never touched real porcelain. Here’s an even better analogy: What if doctors could get licensed without having touched real patients?

If pediatric hospitalists are to transform the care delivery of hospitalized children, and quality learning only comes through hands-on training, then we need some more hands in the pot.

Discharge Improvement

On the heels of my first project, I was fortunate enough to augment my education through another effort—this time with a cohort of fellow pediatric hospitalists. This was a national collaborative to improve discharge handoffs, and I will admit that, at the outset, I was as puzzled as the first time I pulled the lid off the tank of the toilet. There were just too many permutations on PCP communication at the participating institutions, and some felt our aim of timely discharge handoffs was unattainable.

 

What carried me through, however, was the collective and infectious DIY—no, QIY (Quality Improve-it-Yourself) attitude of the group. We were all learning, and regular participation in the collaborative essentially guaranteed improvement. We achieved our aim of 90% communication with PCPs within two days of discharge. The secret was simple: The more you do, the more you learn.

Pediatric hospitalists can transform care delivery through a focus on safe and quality care, but the tools to accomplish this must come through post-residency, on-the-job learning. This QI know-how must efficiently spread among our ranks through practical and project-based educational efforts. It’s “see one, do one, teach one,” but we’re not talking about lumbar punctures anymore.

This is a journey in which we all take on the responsibility of rolling up our sleeves and simply learn by doing. And here is where the third leg of my as-yet-unfinished QI course unfolds.

Through my involvement with the Value in Inpatient Pediatrics (VIP) Network, I’ve gained a newfound vision for what the future might hold. VIP has evolved from a benchmarking project focused on bronchiolitis to an improvement network that will incorporate projects similar to the discharge handoff collaborative above.

In the process, a model for how to rapidly spread QI learning has emerged. The capacity lies in the network’s rapidly growing connectivity. The power comes from the individuals: motivated, card-carrying pediatric hospitalists from a wide array of sites. Collaborative learning harbors the potential to exponentially increase the pace at which we improve.

The future of our quality care is bright. I see an open network of improvement doers and learners. I see collaboration on quality and safety initiatives in all manner of hospitals and communities. I see that this will all be built upon a foundation of hard work and a QIY attitude.

You, too, will play a role.

Just don’t be afraid to get your hands a little dirty. TH

Dr. Shen is medical director of hospital medicine at Dell Children’s Medical Center in Austin, Texas. He is pediatric editor of The Hospitalist.

Despite never advancing his musical tastes beyond the arena bands of the 1970s and ’80s (think Def Leppard), Mark Williams, MD, FACP, FHM, editor-in-chief of the Journal of Hospital Medicine, has done a great job in securing informative and meaningful research and opinion for the journal. Did you see read the July/August 2010 issue of JHM? It is a great example of content uniquely relevant to hospitalists: several original research articles documenting how hospitalists spend their time. Anyone thinking about the best way to organize and operate a hospitalist practice should read through these studies, along with one published by Kevin O’Leary, MD, and colleagues in the March/April 2006 issue.¹ But as a service, I’ll provide a CliffsNotes version of them, along with some comments here.

Time-Motion Studies

What all the studies demonstrate is that academic hospitalists spend only about 15% to 20% of their time in direct patient care, generally defined as time spent taking a patient’s history and examination, meeting with families, etc. Indirect patient care, such as time spent reviewing records, documenting, and communicating with consultants and other patient care staff, consumes about 60% to 70% of their time. The remainder of time is spent in transit (around 7% of each day) and in personal activities.

Remember, all these studies reported on academic hospitalists in large academic medical centers. As noted in the discussion sections, the results in nonteaching community hospitals might be different. My guess is that community hospitalists spend about the same portion of time in the broad categories above, but the individual activities within each category might differ. So I’m willing to believe that these studies tell us something about the majority of hospitalists who practice outside of academia.

90 Minutes of Doctoring?

While the JHM studies assess hospitalist time in a number of different categories, I think it makes the most sense to divide our time into just two categories: “real doctoring” and other. We’ll probably never see a study that divides hospitalists’ time that way, as there would be endless debate about what is and isn’t real doctoring. But it is worth thinking about your work this way.

A lot of what the studies generally defined as indirect patient care is still “real doctoring.” Things like reviewing old records are critically important and typically can’t be done adequately by a nonclinician. But the 10 minutes you spent to get the CD of outside X-rays to show up on your computer, and rearranging the faxed pages so they’re all oriented the same way and in order, are not a good use of your time; a clerical person could do it.

I periodically have an experience that makes me think I spend too much time on patients’ social issues (e.g. long conversations about why Medicare won’t pay for a patient’s skilled nursing facility stay) and too little on “real doctoring.” One such experience is when I have a patient with an unusual pulmonary infiltrate and the radiologist is able to generate a much more comprehensive differential diagnosis than I can. This is embarrassing. Maybe the radiologist is just smarter than I am, but I think it could be because, compared to me, he spends more of his time every day thinking about “real medicine,” such as pulmonary diseases, and less time dealing with nonclinical issues.

 

Even though we’re paid for a full day’s work, I suspect many hospitalists might spend only about 90 minutes a day immersed in thought about “real medicine,” while doctors in most other specialties probably spend a lot more. If I’m right, then it shouldn’t be a surprise that after practicing for many years, the radiologist who spends several hours a day exercising his fund of medical knowledge probably has more command of some clinical things than a hospitalist who does so only 90 minutes a day. Actively practicing as a hospitalist might not be as effective a method of maintaining proficiency as it is in other specialties. More than many other specialties, we need to rely on self-study and continuing education courses to prevent erosion of our knowledge base.

I’ve just made up this 90-minute figure. I have no idea how accurate it might be, and, the JHM studies don’t offer a lot of insight either. Clearly, it varies a lot by individual doctor and practice setting. How much of your day do you think you spend on “real doctoring” vs. other activities?

What really matters is whether we’ve ended up with too much work that isn’t “real doctoring.” Sure, all of the work needs to be done, but the system isn’t served best when paying a doctor to do work a less expensive person could do.

Max “Doctoring” Time

I think most hospitalists, including me, are stuck spending too much time on activities that don’t add value. For example, while complete and informative documentation is essential, most of us probably spend too much time on it, in part because we’re trying to immunize against lawsuits and ensure our documentation matches the relevant coding regulations.

I think hospitalists have a communication burden that is higher than that of most other specialties. The JHM article by Tipping and colleagues notes that a time-motion study of ED doctors (Ann Emerg Med. 1998:31(1):87-91) found that they spent 13% of their time communicating with other providers and staff, compared with their finding that hospitalists spent 26% of their time communicating.² Only a portion of this communication is real doctoring. Discussing patient management with a surgeon is, but spending 20 minutes figuring out which surgeon is on call and how to reach her isn’t.

Tipping’s study also found that when patient census was above average, hospitalists spent less time communicating and documenting in the electronic record, even though the total time spent working on those days increased. Of course, it is possible that when the patient census is below average, we just work more slowly and let work fill the time available, and the reduced time spent documenting and communicating when busy simply reflects working more efficiently. But I suspect that when our patient census climbs above a certain point, or we’re made less efficient by things like implementing a new technology, we compensate in part by relying on consultants more to do the real doctoring we would otherwise be doing and communicating with them less.

All of us should be thinking about ways to make communication as efficient as possible so that we can spend less time doing it. I’m hopeful that we will figure out new ways to communicate (e-mail, text, IM, etc.) that are quicker and just as effective in certain situations.

Coda

I try to write most of my columns in a way that minimizes the editorializing and maximizes the practical advice. This month is an exception; it’s all editorializing. But I do have some advice for Dr. Williams: Investigate music options other than the arena bands of the 1980s. Try something like Alison Krauss’ live album or Puccini’s opera Gianni Schicchi, which has the beautiful aria O mio babbino caro.

 

Or do what I do: Ask former SHM board member Brad Flansbaum, MD, SFHM, for advice. TH

Dr. Nelson has been a practicing hospitalist since 1988 and is co-founder and past president of SHM. He is a principal in Nelson Flores Hospital Medicine Consultants, a national hospitalist practice management consulting firm (www.nelsonflores.com). He is course co-director and faculty for SHM’s “Best Practices in Managing a Hospital Medicine Program.” This column represents his views and is not intended to reflect an official position of SHM.

References

1. O’Leary KJ, Liebovitz DM, Baker DW. How hospitalists spend their time: insights on efficiency and safety. J Hosp Med. 2006;1(2):88-93.
2. Tipping MD, Forth VE, O’Leary KJ, et al. Where did the day go? A time-motion study of hospitalists. J Hosp Med. 2010;5(6):323-328.

Despite never advancing his musical tastes beyond the arena bands of the 1970s and ’80s (think Def Leppard), Mark Williams, MD, FACP, FHM, editor-in-chief of the Journal of Hospital Medicine, has done a great job in securing informative and meaningful research and opinion for the journal. Did you see read the July/August 2010 issue of JHM? It is a great example of content uniquely relevant to hospitalists: several original research articles documenting how hospitalists spend their time. Anyone thinking about the best way to organize and operate a hospitalist practice should read through these studies, along with one published by Kevin O’Leary, MD, and colleagues in the March/April 2006 issue.¹ But as a service, I’ll provide a CliffsNotes version of them, along with some comments here.

Time-Motion Studies

What all the studies demonstrate is that academic hospitalists spend only about 15% to 20% of their time in direct patient care, generally defined as time spent taking a patient’s history and examination, meeting with families, etc. Indirect patient care, such as time spent reviewing records, documenting, and communicating with consultants and other patient care staff, consumes about 60% to 70% of their time. The remainder of time is spent in transit (around 7% of each day) and in personal activities.

Remember, all these studies reported on academic hospitalists in large academic medical centers. As noted in the discussion sections, the results in nonteaching community hospitals might be different. My guess is that community hospitalists spend about the same portion of time in the broad categories above, but the individual activities within each category might differ. So I’m willing to believe that these studies tell us something about the majority of hospitalists who practice outside of academia.

90 Minutes of Doctoring?

While the JHM studies assess hospitalist time in a number of different categories, I think it makes the most sense to divide our time into just two categories: “real doctoring” and other. We’ll probably never see a study that divides hospitalists’ time that way, as there would be endless debate about what is and isn’t real doctoring. But it is worth thinking about your work this way.

A lot of what the studies generally defined as indirect patient care is still “real doctoring.” Things like reviewing old records are critically important and typically can’t be done adequately by a nonclinician. But the 10 minutes you spent to get the CD of outside X-rays to show up on your computer, and rearranging the faxed pages so they’re all oriented the same way and in order, are not a good use of your time; a clerical person could do it.

I periodically have an experience that makes me think I spend too much time on patients’ social issues (e.g. long conversations about why Medicare won’t pay for a patient’s skilled nursing facility stay) and too little on “real doctoring.” One such experience is when I have a patient with an unusual pulmonary infiltrate and the radiologist is able to generate a much more comprehensive differential diagnosis than I can. This is embarrassing. Maybe the radiologist is just smarter than I am, but I think it could be because, compared to me, he spends more of his time every day thinking about “real medicine,” such as pulmonary diseases, and less time dealing with nonclinical issues.

 

Even though we’re paid for a full day’s work, I suspect many hospitalists might spend only about 90 minutes a day immersed in thought about “real medicine,” while doctors in most other specialties probably spend a lot more. If I’m right, then it shouldn’t be a surprise that after practicing for many years, the radiologist who spends several hours a day exercising his fund of medical knowledge probably has more command of some clinical things than a hospitalist who does so only 90 minutes a day. Actively practicing as a hospitalist might not be as effective a method of maintaining proficiency as it is in other specialties. More than many other specialties, we need to rely on self-study and continuing education courses to prevent erosion of our knowledge base.

I’ve just made up this 90-minute figure. I have no idea how accurate it might be, and, the JHM studies don’t offer a lot of insight either. Clearly, it varies a lot by individual doctor and practice setting. How much of your day do you think you spend on “real doctoring” vs. other activities?

What really matters is whether we’ve ended up with too much work that isn’t “real doctoring.” Sure, all of the work needs to be done, but the system isn’t served best when paying a doctor to do work a less expensive person could do.

Max “Doctoring” Time

I think most hospitalists, including me, are stuck spending too much time on activities that don’t add value. For example, while complete and informative documentation is essential, most of us probably spend too much time on it, in part because we’re trying to immunize against lawsuits and ensure our documentation matches the relevant coding regulations.

I think hospitalists have a communication burden that is higher than that of most other specialties. The JHM article by Tipping and colleagues notes that a time-motion study of ED doctors (Ann Emerg Med. 1998:31(1):87-91) found that they spent 13% of their time communicating with other providers and staff, compared with their finding that hospitalists spent 26% of their time communicating.² Only a portion of this communication is real doctoring. Discussing patient management with a surgeon is, but spending 20 minutes figuring out which surgeon is on call and how to reach her isn’t.

Tipping’s study also found that when patient census was above average, hospitalists spent less time communicating and documenting in the electronic record, even though the total time spent working on those days increased. Of course, it is possible that when the patient census is below average, we just work more slowly and let work fill the time available, and the reduced time spent documenting and communicating when busy simply reflects working more efficiently. But I suspect that when our patient census climbs above a certain point, or we’re made less efficient by things like implementing a new technology, we compensate in part by relying on consultants more to do the real doctoring we would otherwise be doing and communicating with them less.

All of us should be thinking about ways to make communication as efficient as possible so that we can spend less time doing it. I’m hopeful that we will figure out new ways to communicate (e-mail, text, IM, etc.) that are quicker and just as effective in certain situations.

Coda

I try to write most of my columns in a way that minimizes the editorializing and maximizes the practical advice. This month is an exception; it’s all editorializing. But I do have some advice for Dr. Williams: Investigate music options other than the arena bands of the 1980s. Try something like Alison Krauss’ live album or Puccini’s opera Gianni Schicchi, which has the beautiful aria O mio babbino caro.

 

Or do what I do: Ask former SHM board member Brad Flansbaum, MD, SFHM, for advice. TH

Dr. Nelson has been a practicing hospitalist since 1988 and is co-founder and past president of SHM. He is a principal in Nelson Flores Hospital Medicine Consultants, a national hospitalist practice management consulting firm (www.nelsonflores.com). He is course co-director and faculty for SHM’s “Best Practices in Managing a Hospital Medicine Program.” This column represents his views and is not intended to reflect an official position of SHM.

References

1. O’Leary KJ, Liebovitz DM, Baker DW. How hospitalists spend their time: insights on efficiency and safety. J Hosp Med. 2006;1(2):88-93.
2. Tipping MD, Forth VE, O’Leary KJ, et al. Where did the day go? A time-motion study of hospitalists. J Hosp Med. 2010;5(6):323-328.

Despite never advancing his musical tastes beyond the arena bands of the 1970s and ’80s (think Def Leppard), Mark Williams, MD, FACP, FHM, editor-in-chief of the Journal of Hospital Medicine, has done a great job in securing informative and meaningful research and opinion for the journal. Did you see read the July/August 2010 issue of JHM? It is a great example of content uniquely relevant to hospitalists: several original research articles documenting how hospitalists spend their time. Anyone thinking about the best way to organize and operate a hospitalist practice should read through these studies, along with one published by Kevin O’Leary, MD, and colleagues in the March/April 2006 issue.¹ But as a service, I’ll provide a CliffsNotes version of them, along with some comments here.

Time-Motion Studies

What all the studies demonstrate is that academic hospitalists spend only about 15% to 20% of their time in direct patient care, generally defined as time spent taking a patient’s history and examination, meeting with families, etc. Indirect patient care, such as time spent reviewing records, documenting, and communicating with consultants and other patient care staff, consumes about 60% to 70% of their time. The remainder of time is spent in transit (around 7% of each day) and in personal activities.

Remember, all these studies reported on academic hospitalists in large academic medical centers. As noted in the discussion sections, the results in nonteaching community hospitals might be different. My guess is that community hospitalists spend about the same portion of time in the broad categories above, but the individual activities within each category might differ. So I’m willing to believe that these studies tell us something about the majority of hospitalists who practice outside of academia.

90 Minutes of Doctoring?

While the JHM studies assess hospitalist time in a number of different categories, I think it makes the most sense to divide our time into just two categories: “real doctoring” and other. We’ll probably never see a study that divides hospitalists’ time that way, as there would be endless debate about what is and isn’t real doctoring. But it is worth thinking about your work this way.

A lot of what the studies generally defined as indirect patient care is still “real doctoring.” Things like reviewing old records are critically important and typically can’t be done adequately by a nonclinician. But the 10 minutes you spent to get the CD of outside X-rays to show up on your computer, and rearranging the faxed pages so they’re all oriented the same way and in order, are not a good use of your time; a clerical person could do it.

I periodically have an experience that makes me think I spend too much time on patients’ social issues (e.g. long conversations about why Medicare won’t pay for a patient’s skilled nursing facility stay) and too little on “real doctoring.” One such experience is when I have a patient with an unusual pulmonary infiltrate and the radiologist is able to generate a much more comprehensive differential diagnosis than I can. This is embarrassing. Maybe the radiologist is just smarter than I am, but I think it could be because, compared to me, he spends more of his time every day thinking about “real medicine,” such as pulmonary diseases, and less time dealing with nonclinical issues.

 

Even though we’re paid for a full day’s work, I suspect many hospitalists might spend only about 90 minutes a day immersed in thought about “real medicine,” while doctors in most other specialties probably spend a lot more. If I’m right, then it shouldn’t be a surprise that after practicing for many years, the radiologist who spends several hours a day exercising his fund of medical knowledge probably has more command of some clinical things than a hospitalist who does so only 90 minutes a day. Actively practicing as a hospitalist might not be as effective a method of maintaining proficiency as it is in other specialties. More than many other specialties, we need to rely on self-study and continuing education courses to prevent erosion of our knowledge base.

I’ve just made up this 90-minute figure. I have no idea how accurate it might be, and, the JHM studies don’t offer a lot of insight either. Clearly, it varies a lot by individual doctor and practice setting. How much of your day do you think you spend on “real doctoring” vs. other activities?

What really matters is whether we’ve ended up with too much work that isn’t “real doctoring.” Sure, all of the work needs to be done, but the system isn’t served best when paying a doctor to do work a less expensive person could do.

Max “Doctoring” Time

I think most hospitalists, including me, are stuck spending too much time on activities that don’t add value. For example, while complete and informative documentation is essential, most of us probably spend too much time on it, in part because we’re trying to immunize against lawsuits and ensure our documentation matches the relevant coding regulations.

I think hospitalists have a communication burden that is higher than that of most other specialties. The JHM article by Tipping and colleagues notes that a time-motion study of ED doctors (Ann Emerg Med. 1998:31(1):87-91) found that they spent 13% of their time communicating with other providers and staff, compared with their finding that hospitalists spent 26% of their time communicating.² Only a portion of this communication is real doctoring. Discussing patient management with a surgeon is, but spending 20 minutes figuring out which surgeon is on call and how to reach her isn’t.

Tipping’s study also found that when patient census was above average, hospitalists spent less time communicating and documenting in the electronic record, even though the total time spent working on those days increased. Of course, it is possible that when the patient census is below average, we just work more slowly and let work fill the time available, and the reduced time spent documenting and communicating when busy simply reflects working more efficiently. But I suspect that when our patient census climbs above a certain point, or we’re made less efficient by things like implementing a new technology, we compensate in part by relying on consultants more to do the real doctoring we would otherwise be doing and communicating with them less.

All of us should be thinking about ways to make communication as efficient as possible so that we can spend less time doing it. I’m hopeful that we will figure out new ways to communicate (e-mail, text, IM, etc.) that are quicker and just as effective in certain situations.

Coda

I try to write most of my columns in a way that minimizes the editorializing and maximizes the practical advice. This month is an exception; it’s all editorializing. But I do have some advice for Dr. Williams: Investigate music options other than the arena bands of the 1980s. Try something like Alison Krauss’ live album or Puccini’s opera Gianni Schicchi, which has the beautiful aria O mio babbino caro.

 

Or do what I do: Ask former SHM board member Brad Flansbaum, MD, SFHM, for advice. TH

Dr. Nelson has been a practicing hospitalist since 1988 and is co-founder and past president of SHM. He is a principal in Nelson Flores Hospital Medicine Consultants, a national hospitalist practice management consulting firm (www.nelsonflores.com). He is course co-director and faculty for SHM’s “Best Practices in Managing a Hospital Medicine Program.” This column represents his views and is not intended to reflect an official position of SHM.

References

1. O’Leary KJ, Liebovitz DM, Baker DW. How hospitalists spend their time: insights on efficiency and safety. J Hosp Med. 2006;1(2):88-93.
2. Tipping MD, Forth VE, O’Leary KJ, et al. Where did the day go? A time-motion study of hospitalists. J Hosp Med. 2010;5(6):323-328.

If every system is perfectly designed to get the results it gets …

And if 15 million patients are harmed every year from medical care …

And if as many as 98,000 people die every year due to medical errors in hospitals …

Then what does that say about the system we have designed?

A System Designed to Competently Hurt Many

By now you’ve no doubt heard, read, and possibly even uttered the above facts and figures yourself. I think we all have our opinions about the veracity of these numbers, but I don’t think any of us would argue with the sentiment. The U.S. healthcare system comprises the most competent, compassionate, well-meaning, and caring professionals on this planet—who harm, maim, and kill countless people every year.

What a discomforting paradox.

Equity: The Overlooked Quality Domain

Many years back, the Institute of Medicine (IOM) published a list of six “domains” of healthcare quality. You’ve no doubt stumbled across the IOM’s Safe … Timely … Effective … Efficient … Equitable … Patient-Centered mnemonic recipe—STEEEP—for high-quality care.¹ In fact, it’s hard to read a journal, attend a medical presentation, or open a local newspaper without finding reference to these domains. It’s all the rage to talk about wrong-site surgery (safe), access to care (timely), comparative-effectiveness research (effective), lean concepts (efficient), and individualized medicine (patient-centered). However, the sixth domain often seems to get the Jan Brady treatment—minimized, marginalized, and oft-forgotten.

The IOM defines equitable care as that which “does not vary in quality because of personal characteristics such as gender, ethnicity, geographic location, and socioeconomic status.”1 To be sure, there is some hum at the national level about issues of equity, especially around healthcare coverage for all. And this is important. However, what appears lost in the rant surrounding the inherent inequities in our tiered health insurance system is that we have blantant inequalities baked into the everyday machinery of our hospitals. And they affect all, regardless of skin color, gender, or insurance status.

Think for a moment about your hospital. Are the type, level, and access to care equal at all times? Does the level of care change when the streetlights come on? How about on weekends and holidays? Do your operating rooms run on Saturdays and Sundays? Can patients get chemotherapy on the weekends? Does your hospital alter its nursing staff ratios after hours? Can you get an ultrasound at midnight? How about a urology or neurosurgery consult at 2 a.m.? How about getting interventional radiology to place an IVC filter on a holiday?

Now scratch a bit closer to home. Does your hospitalist group downstaff on weekends and holidays, even though the volumes probably warrant more coverage? Are your night providers part of your group, or are they moonlighters? Do your after-hours providers cross-cover and admit a reasonable number of patients, or are they frequently overwhelmed? Do they cover patients or admit for services that they don’t typically care for during the day (e.g. ICU, neurosurgical, subspecialty cardiology or oncology patients)? For the intensely ill patients admitted to U.S. hospitals today, should the type and availability of care differ when it’s delivered at 3 p.m. or 3 a.m., Sunday or Monday?

Disregarding the macro-inequities in our societal approach to healthcare, can we even ensure equitable care within our own hospital walls 24 hours a day, seven days a week?

 

The Answer: An Unfortunate “No”

For the record, I hate working nights, abhor working weekends, and resent working holidays. But the thing I’d hate even more than working nights, weekends, and holidays would be being a patient admitted during a night, weekend, or holiday. To understand why, I have to look no further than my hospital parking lot. During bankers’ hours, I can barely find a parking spot on the top floor of our multilevel parking structure.

Fast-forward to Saturday, and I have my pick of empty football fields’ worth of spots on all floors. Ditto Sundays, nights, and holidays.

Why is it that nationally, a collectively near-trillion-dollar hospital enterprise finds it acceptable to effectively shutter itself for a quarter to a third of the week? Especially when doing so seems to counter their primary mission of providing safe, timely, effective, efficient, equitable, and patient-centered care.

The Weekend Effect

There are, of course, economic and operational reasons to downshift during off hours—some hospitals don’t have the elective procedures to run operating rooms seven days a week, and very few patients want to have their elective colonoscopy at 11 p.m. or their chemotherapy during Thanksgiving dinner. However, in most cases, the reasons for doing so center on hospital staff and physician satisfaction. Most us of just don’t like working off hours. As a result, studies have shown significantly less access to such high-level care as coronary angiography and percutaneous coronary intervention on weekends.^2,3

And the effects of this “weekend effect” can be devastating.

A recent paper in the New England Journal of Medicine reported that for every 1,000 patients admitted with a myocardial infarction on a weekend, nine more would die than a comparable group admitted during the week.⁴ Their offense? Having the misfortune to get ill on a Saturday morning. The authors concluded that this higher mortality was secondary to a lower rate of invasive cardiac procedures, presumably because they were less available. And the weekend effect isn’t just limited to coronary care. Poorer outcomes, including higher mortality rates, have been reported for weekend admissions to the neonatal ICU and adult ICU, as well as admission for epiglottitis, ruptured abdominal aortic aneurysms, and pulmonary embolism.^5,6,7

So let’s connect the dots: A system designed with inequal access to lifesaving therapies and appropriate staffing results in worse outcomes, more harm, more deaths.

To be clear, 98,000 people don’t die every year because of my disdain for working nights and weekends. This is a much deeper problem that hinges on many unsatisfactory systems working unsatisfactorily in tandem (e.g. see the other five IOM domains of quality care). Furthermore, I don’t mean to suggest that hospitalists necessarily have a lot of say in cardiac catheterization schedules. Yet we do control our own systems of care—how many patients we admit and cover during a shift, how strongly we advocate for timely testing and consultation, how we staff weekends and cover patients at night.

And, more and more, we are in a position to enhance the care delivery systems of the hospital and its providers that surround us. With that comes a responsibility to ensure that these systems are highly functioning and equitable, regardless of the time of day, day of the week.

If we are going to fundamentally enhance the quality of care, then we have to design safer systems of care. It will take time and resources to alter many of our bruised systems of care, but we can begin by at least ensuring equity in how we deliver care at our own institutions. That is, unless we are comfortable with a system perfectly designed to harm. TH

 

Dr. Glasheen is associate professor of medicine at the University of Colorado Denver, where he serves as director of the Hospital Medicine Program and the Hospitalist Training Program, and as associate program director of the Internal Medicine Residency Program.

References

1. Institute of Medicine (IOM). Crossing the Quality Chasm: A New Health System for the 21st Century. Washington, D.C.: National Academy Press; 2001.
2. Bell CM, Redelmeier DA. Waiting for urgent procedures on the weekend among emergently hospitalized patients. Am J Med. 2004:117(3):175-181.
3. Magid DJ, Wang Y, Herrin J, et al. Relationship between time of day, day of week, timeliness of reperfusion, and in-hospital mortality for patients with acute ST-segment elevation myocardial infarction. JAMA. 2005;294(7):803-812.
4. Kostis WJ, Demissie K, Marcella SW, Shao YH, Wilson AC, Moreyra AE. Weekend versus weekday admission and mortality from myocardial infarction. N Eng J Med. 2007;356 (11):1099-1109.
5. Hendry RA. The weekend—a dangerous time to be born? Br J Obstet Gynaecol. 1981;88(12):1200-1203.
6. Barnett MJ, Kaboli PJ, Sirio CA, Rosenthal GE. Day of the week of intensive care admission and patient outcomes: a multisite regional evaluation. Med Care. 2002;40(6):530-539.
7. Bell CM, Redelmeier DA. Mortality among patients admitted to hospitals on weekends as compared with weekdays. N Eng J Med. 2001;345(9):663-668.

If every system is perfectly designed to get the results it gets …

And if 15 million patients are harmed every year from medical care …

And if as many as 98,000 people die every year due to medical errors in hospitals …

Then what does that say about the system we have designed?

A System Designed to Competently Hurt Many

By now you’ve no doubt heard, read, and possibly even uttered the above facts and figures yourself. I think we all have our opinions about the veracity of these numbers, but I don’t think any of us would argue with the sentiment. The U.S. healthcare system comprises the most competent, compassionate, well-meaning, and caring professionals on this planet—who harm, maim, and kill countless people every year.

What a discomforting paradox.

Equity: The Overlooked Quality Domain

Many years back, the Institute of Medicine (IOM) published a list of six “domains” of healthcare quality. You’ve no doubt stumbled across the IOM’s Safe … Timely … Effective … Efficient … Equitable … Patient-Centered mnemonic recipe—STEEEP—for high-quality care.¹ In fact, it’s hard to read a journal, attend a medical presentation, or open a local newspaper without finding reference to these domains. It’s all the rage to talk about wrong-site surgery (safe), access to care (timely), comparative-effectiveness research (effective), lean concepts (efficient), and individualized medicine (patient-centered). However, the sixth domain often seems to get the Jan Brady treatment—minimized, marginalized, and oft-forgotten.

The IOM defines equitable care as that which “does not vary in quality because of personal characteristics such as gender, ethnicity, geographic location, and socioeconomic status.”1 To be sure, there is some hum at the national level about issues of equity, especially around healthcare coverage for all. And this is important. However, what appears lost in the rant surrounding the inherent inequities in our tiered health insurance system is that we have blantant inequalities baked into the everyday machinery of our hospitals. And they affect all, regardless of skin color, gender, or insurance status.

Think for a moment about your hospital. Are the type, level, and access to care equal at all times? Does the level of care change when the streetlights come on? How about on weekends and holidays? Do your operating rooms run on Saturdays and Sundays? Can patients get chemotherapy on the weekends? Does your hospital alter its nursing staff ratios after hours? Can you get an ultrasound at midnight? How about a urology or neurosurgery consult at 2 a.m.? How about getting interventional radiology to place an IVC filter on a holiday?

Now scratch a bit closer to home. Does your hospitalist group downstaff on weekends and holidays, even though the volumes probably warrant more coverage? Are your night providers part of your group, or are they moonlighters? Do your after-hours providers cross-cover and admit a reasonable number of patients, or are they frequently overwhelmed? Do they cover patients or admit for services that they don’t typically care for during the day (e.g. ICU, neurosurgical, subspecialty cardiology or oncology patients)? For the intensely ill patients admitted to U.S. hospitals today, should the type and availability of care differ when it’s delivered at 3 p.m. or 3 a.m., Sunday or Monday?

Disregarding the macro-inequities in our societal approach to healthcare, can we even ensure equitable care within our own hospital walls 24 hours a day, seven days a week?

 

The Answer: An Unfortunate “No”

For the record, I hate working nights, abhor working weekends, and resent working holidays. But the thing I’d hate even more than working nights, weekends, and holidays would be being a patient admitted during a night, weekend, or holiday. To understand why, I have to look no further than my hospital parking lot. During bankers’ hours, I can barely find a parking spot on the top floor of our multilevel parking structure.

Fast-forward to Saturday, and I have my pick of empty football fields’ worth of spots on all floors. Ditto Sundays, nights, and holidays.

Why is it that nationally, a collectively near-trillion-dollar hospital enterprise finds it acceptable to effectively shutter itself for a quarter to a third of the week? Especially when doing so seems to counter their primary mission of providing safe, timely, effective, efficient, equitable, and patient-centered care.

The Weekend Effect

There are, of course, economic and operational reasons to downshift during off hours—some hospitals don’t have the elective procedures to run operating rooms seven days a week, and very few patients want to have their elective colonoscopy at 11 p.m. or their chemotherapy during Thanksgiving dinner. However, in most cases, the reasons for doing so center on hospital staff and physician satisfaction. Most us of just don’t like working off hours. As a result, studies have shown significantly less access to such high-level care as coronary angiography and percutaneous coronary intervention on weekends.^2,3

And the effects of this “weekend effect” can be devastating.

A recent paper in the New England Journal of Medicine reported that for every 1,000 patients admitted with a myocardial infarction on a weekend, nine more would die than a comparable group admitted during the week.⁴ Their offense? Having the misfortune to get ill on a Saturday morning. The authors concluded that this higher mortality was secondary to a lower rate of invasive cardiac procedures, presumably because they were less available. And the weekend effect isn’t just limited to coronary care. Poorer outcomes, including higher mortality rates, have been reported for weekend admissions to the neonatal ICU and adult ICU, as well as admission for epiglottitis, ruptured abdominal aortic aneurysms, and pulmonary embolism.^5,6,7

So let’s connect the dots: A system designed with inequal access to lifesaving therapies and appropriate staffing results in worse outcomes, more harm, more deaths.

To be clear, 98,000 people don’t die every year because of my disdain for working nights and weekends. This is a much deeper problem that hinges on many unsatisfactory systems working unsatisfactorily in tandem (e.g. see the other five IOM domains of quality care). Furthermore, I don’t mean to suggest that hospitalists necessarily have a lot of say in cardiac catheterization schedules. Yet we do control our own systems of care—how many patients we admit and cover during a shift, how strongly we advocate for timely testing and consultation, how we staff weekends and cover patients at night.

And, more and more, we are in a position to enhance the care delivery systems of the hospital and its providers that surround us. With that comes a responsibility to ensure that these systems are highly functioning and equitable, regardless of the time of day, day of the week.

If we are going to fundamentally enhance the quality of care, then we have to design safer systems of care. It will take time and resources to alter many of our bruised systems of care, but we can begin by at least ensuring equity in how we deliver care at our own institutions. That is, unless we are comfortable with a system perfectly designed to harm. TH

 

Dr. Glasheen is associate professor of medicine at the University of Colorado Denver, where he serves as director of the Hospital Medicine Program and the Hospitalist Training Program, and as associate program director of the Internal Medicine Residency Program.

References

1. Institute of Medicine (IOM). Crossing the Quality Chasm: A New Health System for the 21st Century. Washington, D.C.: National Academy Press; 2001.
2. Bell CM, Redelmeier DA. Waiting for urgent procedures on the weekend among emergently hospitalized patients. Am J Med. 2004:117(3):175-181.
3. Magid DJ, Wang Y, Herrin J, et al. Relationship between time of day, day of week, timeliness of reperfusion, and in-hospital mortality for patients with acute ST-segment elevation myocardial infarction. JAMA. 2005;294(7):803-812.
4. Kostis WJ, Demissie K, Marcella SW, Shao YH, Wilson AC, Moreyra AE. Weekend versus weekday admission and mortality from myocardial infarction. N Eng J Med. 2007;356 (11):1099-1109.
5. Hendry RA. The weekend—a dangerous time to be born? Br J Obstet Gynaecol. 1981;88(12):1200-1203.
6. Barnett MJ, Kaboli PJ, Sirio CA, Rosenthal GE. Day of the week of intensive care admission and patient outcomes: a multisite regional evaluation. Med Care. 2002;40(6):530-539.
7. Bell CM, Redelmeier DA. Mortality among patients admitted to hospitals on weekends as compared with weekdays. N Eng J Med. 2001;345(9):663-668.

If every system is perfectly designed to get the results it gets …

And if 15 million patients are harmed every year from medical care …

And if as many as 98,000 people die every year due to medical errors in hospitals …

Then what does that say about the system we have designed?

A System Designed to Competently Hurt Many

By now you’ve no doubt heard, read, and possibly even uttered the above facts and figures yourself. I think we all have our opinions about the veracity of these numbers, but I don’t think any of us would argue with the sentiment. The U.S. healthcare system comprises the most competent, compassionate, well-meaning, and caring professionals on this planet—who harm, maim, and kill countless people every year.

What a discomforting paradox.

Equity: The Overlooked Quality Domain

Many years back, the Institute of Medicine (IOM) published a list of six “domains” of healthcare quality. You’ve no doubt stumbled across the IOM’s Safe … Timely … Effective … Efficient … Equitable … Patient-Centered mnemonic recipe—STEEEP—for high-quality care.¹ In fact, it’s hard to read a journal, attend a medical presentation, or open a local newspaper without finding reference to these domains. It’s all the rage to talk about wrong-site surgery (safe), access to care (timely), comparative-effectiveness research (effective), lean concepts (efficient), and individualized medicine (patient-centered). However, the sixth domain often seems to get the Jan Brady treatment—minimized, marginalized, and oft-forgotten.

The IOM defines equitable care as that which “does not vary in quality because of personal characteristics such as gender, ethnicity, geographic location, and socioeconomic status.”1 To be sure, there is some hum at the national level about issues of equity, especially around healthcare coverage for all. And this is important. However, what appears lost in the rant surrounding the inherent inequities in our tiered health insurance system is that we have blantant inequalities baked into the everyday machinery of our hospitals. And they affect all, regardless of skin color, gender, or insurance status.

Think for a moment about your hospital. Are the type, level, and access to care equal at all times? Does the level of care change when the streetlights come on? How about on weekends and holidays? Do your operating rooms run on Saturdays and Sundays? Can patients get chemotherapy on the weekends? Does your hospital alter its nursing staff ratios after hours? Can you get an ultrasound at midnight? How about a urology or neurosurgery consult at 2 a.m.? How about getting interventional radiology to place an IVC filter on a holiday?

Now scratch a bit closer to home. Does your hospitalist group downstaff on weekends and holidays, even though the volumes probably warrant more coverage? Are your night providers part of your group, or are they moonlighters? Do your after-hours providers cross-cover and admit a reasonable number of patients, or are they frequently overwhelmed? Do they cover patients or admit for services that they don’t typically care for during the day (e.g. ICU, neurosurgical, subspecialty cardiology or oncology patients)? For the intensely ill patients admitted to U.S. hospitals today, should the type and availability of care differ when it’s delivered at 3 p.m. or 3 a.m., Sunday or Monday?

Disregarding the macro-inequities in our societal approach to healthcare, can we even ensure equitable care within our own hospital walls 24 hours a day, seven days a week?

 

The Answer: An Unfortunate “No”

For the record, I hate working nights, abhor working weekends, and resent working holidays. But the thing I’d hate even more than working nights, weekends, and holidays would be being a patient admitted during a night, weekend, or holiday. To understand why, I have to look no further than my hospital parking lot. During bankers’ hours, I can barely find a parking spot on the top floor of our multilevel parking structure.

Fast-forward to Saturday, and I have my pick of empty football fields’ worth of spots on all floors. Ditto Sundays, nights, and holidays.

Why is it that nationally, a collectively near-trillion-dollar hospital enterprise finds it acceptable to effectively shutter itself for a quarter to a third of the week? Especially when doing so seems to counter their primary mission of providing safe, timely, effective, efficient, equitable, and patient-centered care.

The Weekend Effect

There are, of course, economic and operational reasons to downshift during off hours—some hospitals don’t have the elective procedures to run operating rooms seven days a week, and very few patients want to have their elective colonoscopy at 11 p.m. or their chemotherapy during Thanksgiving dinner. However, in most cases, the reasons for doing so center on hospital staff and physician satisfaction. Most us of just don’t like working off hours. As a result, studies have shown significantly less access to such high-level care as coronary angiography and percutaneous coronary intervention on weekends.^2,3

And the effects of this “weekend effect” can be devastating.

A recent paper in the New England Journal of Medicine reported that for every 1,000 patients admitted with a myocardial infarction on a weekend, nine more would die than a comparable group admitted during the week.⁴ Their offense? Having the misfortune to get ill on a Saturday morning. The authors concluded that this higher mortality was secondary to a lower rate of invasive cardiac procedures, presumably because they were less available. And the weekend effect isn’t just limited to coronary care. Poorer outcomes, including higher mortality rates, have been reported for weekend admissions to the neonatal ICU and adult ICU, as well as admission for epiglottitis, ruptured abdominal aortic aneurysms, and pulmonary embolism.^5,6,7

So let’s connect the dots: A system designed with inequal access to lifesaving therapies and appropriate staffing results in worse outcomes, more harm, more deaths.

To be clear, 98,000 people don’t die every year because of my disdain for working nights and weekends. This is a much deeper problem that hinges on many unsatisfactory systems working unsatisfactorily in tandem (e.g. see the other five IOM domains of quality care). Furthermore, I don’t mean to suggest that hospitalists necessarily have a lot of say in cardiac catheterization schedules. Yet we do control our own systems of care—how many patients we admit and cover during a shift, how strongly we advocate for timely testing and consultation, how we staff weekends and cover patients at night.

And, more and more, we are in a position to enhance the care delivery systems of the hospital and its providers that surround us. With that comes a responsibility to ensure that these systems are highly functioning and equitable, regardless of the time of day, day of the week.

If we are going to fundamentally enhance the quality of care, then we have to design safer systems of care. It will take time and resources to alter many of our bruised systems of care, but we can begin by at least ensuring equity in how we deliver care at our own institutions. That is, unless we are comfortable with a system perfectly designed to harm. TH

 

Dr. Glasheen is associate professor of medicine at the University of Colorado Denver, where he serves as director of the Hospital Medicine Program and the Hospitalist Training Program, and as associate program director of the Internal Medicine Residency Program.

References

1. Institute of Medicine (IOM). Crossing the Quality Chasm: A New Health System for the 21st Century. Washington, D.C.: National Academy Press; 2001.
2. Bell CM, Redelmeier DA. Waiting for urgent procedures on the weekend among emergently hospitalized patients. Am J Med. 2004:117(3):175-181.
3. Magid DJ, Wang Y, Herrin J, et al. Relationship between time of day, day of week, timeliness of reperfusion, and in-hospital mortality for patients with acute ST-segment elevation myocardial infarction. JAMA. 2005;294(7):803-812.
4. Kostis WJ, Demissie K, Marcella SW, Shao YH, Wilson AC, Moreyra AE. Weekend versus weekday admission and mortality from myocardial infarction. N Eng J Med. 2007;356 (11):1099-1109.
5. Hendry RA. The weekend—a dangerous time to be born? Br J Obstet Gynaecol. 1981;88(12):1200-1203.
6. Barnett MJ, Kaboli PJ, Sirio CA, Rosenthal GE. Day of the week of intensive care admission and patient outcomes: a multisite regional evaluation. Med Care. 2002;40(6):530-539.
7. Bell CM, Redelmeier DA. Mortality among patients admitted to hospitals on weekends as compared with weekdays. N Eng J Med. 2001;345(9):663-668.

While most physicians say patients who discharge from the hospital against medical advice (AMA) put themselves at great risk, little research has been conducted to prove that theory. So two hospitalists at the Montefiore Medical Center in Bronx, N.Y., put it to the test, and they found that discharge AMA could literally be a matter of life or death.

In their award-winning HM10 research poster (PDF), Will Southern, MD, MS, and Julia Arnsten, MD, MPH, discovered that discharge AMA significantly increases the risk for both mortality and readmission in a general medical inpatient population. Their research equally divided more than 7,100 inpatients into two groups that were identical in every measurable domain, including admission laboratory values, acuity of illness, and medical history; one of the two groups chose discharge AMA.

Results showed that the AMA group was at a 45% increased risk of mortality and readmission within 30 days, whereas the non-AMA group was at approximately half the risk, with only a 27% chance of mortality and readmission.

Acknowledging the possibility that patients intuitively know when they are ready to leave and the fundamental medical, behavioral, and psychiatric differences in patients who discharge AMA, Dr. Southern says, "the real question is, 'If I [the hospitalist] intervene and, by whatever means necessary, try to convince the patients not to leave AMA, are their outcomes improved?' "

According to Dr. Southern, outcomes can be improved through patient communication, engagement, and intervention. "If you're interested in leaving AMA, it's OK, it's your choice," he says, "but your chances of being dead within the next 30 days are about double and your chances of being readmitted are about double."

Dr. Southern says that, although their findings are a useful starting point for further disease-centric study, for the practicing hospitalist with a patient considering discharge AMA, "it's about finding out why someone needs to leave: They need to feed their cat, cash their Social Security check, whatever the reason is."

While most physicians say patients who discharge from the hospital against medical advice (AMA) put themselves at great risk, little research has been conducted to prove that theory. So two hospitalists at the Montefiore Medical Center in Bronx, N.Y., put it to the test, and they found that discharge AMA could literally be a matter of life or death.

In their award-winning HM10 research poster (PDF), Will Southern, MD, MS, and Julia Arnsten, MD, MPH, discovered that discharge AMA significantly increases the risk for both mortality and readmission in a general medical inpatient population. Their research equally divided more than 7,100 inpatients into two groups that were identical in every measurable domain, including admission laboratory values, acuity of illness, and medical history; one of the two groups chose discharge AMA.

Results showed that the AMA group was at a 45% increased risk of mortality and readmission within 30 days, whereas the non-AMA group was at approximately half the risk, with only a 27% chance of mortality and readmission.

Acknowledging the possibility that patients intuitively know when they are ready to leave and the fundamental medical, behavioral, and psychiatric differences in patients who discharge AMA, Dr. Southern says, "the real question is, 'If I [the hospitalist] intervene and, by whatever means necessary, try to convince the patients not to leave AMA, are their outcomes improved?' "

According to Dr. Southern, outcomes can be improved through patient communication, engagement, and intervention. "If you're interested in leaving AMA, it's OK, it's your choice," he says, "but your chances of being dead within the next 30 days are about double and your chances of being readmitted are about double."

Dr. Southern says that, although their findings are a useful starting point for further disease-centric study, for the practicing hospitalist with a patient considering discharge AMA, "it's about finding out why someone needs to leave: They need to feed their cat, cash their Social Security check, whatever the reason is."

While most physicians say patients who discharge from the hospital against medical advice (AMA) put themselves at great risk, little research has been conducted to prove that theory. So two hospitalists at the Montefiore Medical Center in Bronx, N.Y., put it to the test, and they found that discharge AMA could literally be a matter of life or death.

In their award-winning HM10 research poster (PDF), Will Southern, MD, MS, and Julia Arnsten, MD, MPH, discovered that discharge AMA significantly increases the risk for both mortality and readmission in a general medical inpatient population. Their research equally divided more than 7,100 inpatients into two groups that were identical in every measurable domain, including admission laboratory values, acuity of illness, and medical history; one of the two groups chose discharge AMA.

Results showed that the AMA group was at a 45% increased risk of mortality and readmission within 30 days, whereas the non-AMA group was at approximately half the risk, with only a 27% chance of mortality and readmission.

Acknowledging the possibility that patients intuitively know when they are ready to leave and the fundamental medical, behavioral, and psychiatric differences in patients who discharge AMA, Dr. Southern says, "the real question is, 'If I [the hospitalist] intervene and, by whatever means necessary, try to convince the patients not to leave AMA, are their outcomes improved?' "

According to Dr. Southern, outcomes can be improved through patient communication, engagement, and intervention. "If you're interested in leaving AMA, it's OK, it's your choice," he says, "but your chances of being dead within the next 30 days are about double and your chances of being readmitted are about double."

Dr. Southern says that, although their findings are a useful starting point for further disease-centric study, for the practicing hospitalist with a patient considering discharge AMA, "it's about finding out why someone needs to leave: They need to feed their cat, cash their Social Security check, whatever the reason is."

Martin Johns, MD, joined Gifford Medical Center, a 25-bed full access hospital in rural Randolph, Vt., five years ago to launch its HM program. The landmark push toward implementing electronic health records (EHR) now has Dr. Johns and his colleagues scrambling.

Dr. Johns, whose program now has four hospitalists and four physician assistants, recently spoke with The Hospitalist eWire to talk about the technology challenges specifically faced by rural hospitalist programs.

Question: What role did EHR play in your HM group when you began the service?

Answer: Five years ago, I was definitely concerned about that. I came from Geisinger Medical Center (in Danville, Pa.). They had Epic embedded into every aspect of documentation. I was very used to the seamless integration of information in transitions of care. I was somewhat spoiled in that regard, but I also realized a small hospital was going to have different types of systems.

Q: How far has your digital record-keeping in the hospital come to date?

A: We currently have a data repository more than EHR. CPSI is the system we use. One of the difficulties in us making the decision to move to an EHR is a very interesting problem for all small hospitals. Because they require so much augmentation after installation, small hospitals can't afford to have your large Epic system, for example, to put in place. They can't have 10 IT people in house running the service. What ends up happening is that the hospital purchases an EHR based on outpatient clinic interest and said EHR usually doesn't really speak to your inpatient system.

Q: How can a smaller institution transition into a comprehensive system?

A: The biggest barrier is there's just not a really great all-level-of-care product that would take you through that in a small hospital. It's just too difficult and too expensive for someone to create that product, or at least it appears that way. An all-encompassing solution for a small hospital—the market is not just there.

Q: So where does that leave the rural hospitalist?

A: At a smaller place, it sets up all the variables for a communication breakdown. Mrs. Jones comes in for cataract surgery and she gets morphine and she has a reaction. That's on her hospital chart, but doesn't make its way onto her clinic chart. Those are the breakdowns that can happen and that's really challenging. … With this EHR push, if it's not done correctly, the patient can suffer more than they would with paper records.

Martin Johns, MD, joined Gifford Medical Center, a 25-bed full access hospital in rural Randolph, Vt., five years ago to launch its HM program. The landmark push toward implementing electronic health records (EHR) now has Dr. Johns and his colleagues scrambling.

Dr. Johns, whose program now has four hospitalists and four physician assistants, recently spoke with The Hospitalist eWire to talk about the technology challenges specifically faced by rural hospitalist programs.

Question: What role did EHR play in your HM group when you began the service?

Answer: Five years ago, I was definitely concerned about that. I came from Geisinger Medical Center (in Danville, Pa.). They had Epic embedded into every aspect of documentation. I was very used to the seamless integration of information in transitions of care. I was somewhat spoiled in that regard, but I also realized a small hospital was going to have different types of systems.

Q: How far has your digital record-keeping in the hospital come to date?

A: We currently have a data repository more than EHR. CPSI is the system we use. One of the difficulties in us making the decision to move to an EHR is a very interesting problem for all small hospitals. Because they require so much augmentation after installation, small hospitals can't afford to have your large Epic system, for example, to put in place. They can't have 10 IT people in house running the service. What ends up happening is that the hospital purchases an EHR based on outpatient clinic interest and said EHR usually doesn't really speak to your inpatient system.

Q: How can a smaller institution transition into a comprehensive system?

A: The biggest barrier is there's just not a really great all-level-of-care product that would take you through that in a small hospital. It's just too difficult and too expensive for someone to create that product, or at least it appears that way. An all-encompassing solution for a small hospital—the market is not just there.

Q: So where does that leave the rural hospitalist?

A: At a smaller place, it sets up all the variables for a communication breakdown. Mrs. Jones comes in for cataract surgery and she gets morphine and she has a reaction. That's on her hospital chart, but doesn't make its way onto her clinic chart. Those are the breakdowns that can happen and that's really challenging. … With this EHR push, if it's not done correctly, the patient can suffer more than they would with paper records.

Martin Johns, MD, joined Gifford Medical Center, a 25-bed full access hospital in rural Randolph, Vt., five years ago to launch its HM program. The landmark push toward implementing electronic health records (EHR) now has Dr. Johns and his colleagues scrambling.

Dr. Johns, whose program now has four hospitalists and four physician assistants, recently spoke with The Hospitalist eWire to talk about the technology challenges specifically faced by rural hospitalist programs.

Question: What role did EHR play in your HM group when you began the service?

Answer: Five years ago, I was definitely concerned about that. I came from Geisinger Medical Center (in Danville, Pa.). They had Epic embedded into every aspect of documentation. I was very used to the seamless integration of information in transitions of care. I was somewhat spoiled in that regard, but I also realized a small hospital was going to have different types of systems.

Q: How far has your digital record-keeping in the hospital come to date?

A: We currently have a data repository more than EHR. CPSI is the system we use. One of the difficulties in us making the decision to move to an EHR is a very interesting problem for all small hospitals. Because they require so much augmentation after installation, small hospitals can't afford to have your large Epic system, for example, to put in place. They can't have 10 IT people in house running the service. What ends up happening is that the hospital purchases an EHR based on outpatient clinic interest and said EHR usually doesn't really speak to your inpatient system.

Q: How can a smaller institution transition into a comprehensive system?

A: The biggest barrier is there's just not a really great all-level-of-care product that would take you through that in a small hospital. It's just too difficult and too expensive for someone to create that product, or at least it appears that way. An all-encompassing solution for a small hospital—the market is not just there.

Q: So where does that leave the rural hospitalist?

A: At a smaller place, it sets up all the variables for a communication breakdown. Mrs. Jones comes in for cataract surgery and she gets morphine and she has a reaction. That's on her hospital chart, but doesn't make its way onto her clinic chart. Those are the breakdowns that can happen and that's really challenging. … With this EHR push, if it's not done correctly, the patient can suffer more than they would with paper records.

You have a 54-year-old black patient with gout, diabetes mellitus, hypertension, and chronic kidney disease (CKD). He has an acute gout flare involving his right knee. In the past year he has had four attacks of gout in the ankles and knees, which you treated with intra-articular glucocorticoid injections. He has been on allopurinol (Zyloprim) 200 mg daily, but his last serum urate level was 9.4 mg/dL (reference range 3.0–8.0). His creatinine clearance is 45 mL/minute (reference range 85–125).

In view of his kidney disease, you are concerned about increasing his dose of allopurinol, but also about the need to treat his frequent attacks. How should you manage this patient?

GOUT IS CHALLENGING TO TREAT IN PATIENTS WITH KIDNEY DISEASE

A major challenge in treating patients with gout is to avoid therapeutic interactions with common comorbidities, including hypertension, insulin resistance, coronary artery disease, heart failure, and especially CKD.¹

In this paper, we discuss approaches to and controversies in the management of gout and hyperuricemia in patients with CKD. Unfortunately, the evidence from clinical trials to guide treatment decisions is limited; therefore, decisions must often be based on experience and pathophysiologic principles.

GENERAL GOALS OF GOUT THERAPY

Depending on the patient and the stage of the disease, the goals in treating patients with gout are to:

• Terminate acute attacks as promptly and safely as possible
• Prevent recurrences of acute gout attacks
• Prevent or reverse complications resulting from deposition of monosodium urate in the joints, in the kidneys, or at other sites.

These goals are more difficult to achieve in patients with CKD because of the potential complications from many of the available drugs.

TERMINATING ACUTE GOUT FLARES

In patients with acute gout, treatment is aimed at quickly resolving pain and inflammation.

Several types of drugs can terminate acute gout flares. The choice in most situations is colchicine (Colcrys); a nonsteroidal anti-inflammatory drug (NSAID); a corticosteroid; or corticotropin (ACTH).

However, in patients with CKD, there are concerns about using colchicine or NSAIDs, and corticotropin is very expensive; thus, corticosteroids are often used.

Colchicine’s clearance is reduced in CKD

Colchicine is somewhat effective in treating acute gout attacks and probably more effective in preventing attacks.

Due to concerns about inappropriate dosing and reported deaths,² the intravenous formulation is not available in many countries, including the United States.

After oral administration, colchicine is rapidly absorbed, with a bioavailability of up to 50%. It undergoes metabolism by the liver, and its metabolites are excreted by renal and biliary-intestinal routes. Up to 20% of the active drug is excreted by the kidneys.³

Colchicine’s clearance is significantly reduced in patients with renal or hepatic insufficiency, and the drug may accumulate in cells, with resultant toxicity.⁴ Colchicine-induced toxicity has been observed when the drug was used for acute treatment, as well as for chronic prophylaxis of gout in patients with CKD; thus, alternative agents for treating acute attacks should be considered.^5,6 With prolonged use, reversible colchicine-induced axonal neuropathy, neutropenia, and vacuolar myopathy can develop in patients with CKD.⁷

In a trial in patients with normal renal function, nearly 100% who received an initial dose of 1 mg followed by 0.5 mg every 2 hours developed diarrhea at a median time of 24 hours.⁸ Emesis may also occur.

A lower dose of 1.8 mg (two 0.6-mg pills followed by one pill an hour later) was well tolerated but only moderately effective in treating acute gout, causing at least a 50% reduction in pain at 24 hours in only 38% of patients.⁹ This study does not clarify the dosage to use to completely resolve attacks. Using additional colchicine likely will increase the response rate, but will also increase side effects. Patients with CKD were not included.

Some patients, as shown in the above trial, can abort attacks by taking only one or two colchicine tablets when they feel the first “twinge” of an attack. This approach is likely to be safe in CKD, but it may be of value to only a few patients.

Nonsteroidal anti-inflammatory drugs can worsen chronic kidney disease

NSAIDs in high doses can effectively treat the pain and inflammation of acute gout. Indomethacin (Indocin) 50 mg three times daily has been standard NSAID therapy.

Other nonselective NSAIDs and NSAIDs that selectively inhibit cyclooxygenase 2 (COX-2) are effective, but all can cause acute renal toxicity or worsen CKD.¹⁰ Renal side effects include salt and water retention, acute tubular necrosis, acute interstitial nephritis, proteinuria, hypertension, hyperkalemia, and chronic renal injury.¹¹

Even short-term use of high-dose NSAIDs should generally be avoided in patients with preexisting CKD, for whom there is no established safe threshold dose. When NSAIDs (including selective COX-2 inhibitors) are used, renal function should be monitored closely and the duration limited as much as possible.

 

 

Corticosteroids are often used to treat acute attacks

Due to the concerns about NSAIDs or colchicine to treat acute gout attacks in patients with CKD, corticosteroids are often used in this setting.

Intra-articular steroid injections are useful in treating acute gout limited to a single joint or bursa.¹² However, one should first make sure that the joint is not infected: septic arthritis should ideally be excluded by arthrocentesis, particularly in immunosuppressed patients¹³ or those with end-stage renal disease, who are predisposed to bacteremia.

Oral, intramuscular, or intravenous steroids can provide complete relief from acute gout, although high doses (eg, prednisone 30–60 mg/day or the equivalent) are often needed. Common errors resulting in inefficacy include using too low a dose or not treating for a sufficient time before tapering or stopping. Groff and colleagues¹⁴ described 13 patients who received oral or intravenous steroids for acute gout. Nine patients received an initial single dose of prednisone ranging from 20 to 50 mg, with tapering over a mean of 10 days. Twelve of the 13 patients had improvement within 48 hours, and the signs and symptoms of acute gout resolved completely within 7 to 10 days.

We often give prednisone 40 mg daily until a day after the acute attack resolves and then taper over another 7 to 10 days. There are no data to guide steroid dosing in an evidence-based way, but we believe too short a course of therapy may result in return of symptoms.

Corticotropin and other agents: Effective but costly

Corticotropin shares the same indications as systemic corticosteroids, being used to treat flares when NSAIDs, intra-articular steroids, and colchicine are contraindicated. However, corticotropin is far more expensive than generic corticosteroids, costing nearly $2,000 for a single 80-IU dose, which may need to be repeated.

Corticotropin is available for subcutaneous or intramuscular injection. A single intramuscular injection of corticotropin gel (H.P. Acthar, 25–80 IU) may terminate an acute gout attack.¹⁵ However, many patients need another injection after 24 to 72 hours, which would require another visit to the physician. This treatment has been touted by some as being more effective than corticosteroid therapy, possibly because of a unique peripheral mechanism of action in addition to stimulating cortisol release.¹⁶

We rarely use corticotropin, in view of its cost as well as concerns about excessive sodium and water retention due to the release of multiple hormones from the adrenal gland. This may be especially deleterious in patients with CKD or congestive heart failure.¹⁷

Parenteral anti-tumor necrosis factor agents or interleukin 1 antagonists can be dramatically effective but are also expensive.^18,19 For example, anakinra (Kineret) 100 mg costs about $73, and multiple daily doses may be necessary.

Under unique conditions in which they can be safely used (eg, patients with CKD, diabetes mellitus, liver disease), they may be cost-effective if they can shorten the stay of a hospitalized patient with acute gout.

PROPHYLACTIC ANTI-INFLAMMATORY THERAPY FOR PATIENTS WITH GOUT

Between attacks, the goal is to prevent new attacks through prophylactic management, which may include anti-inflammatory and hypouricemic therapy along with dietary instruction (such as avoiding excessive beer, liquor, and fructose ingestion).

Colchicine can be used as prophylaxis, with caution and monitoring

Although colchicine is not 100% effective, it markedly reduces the flare rate when started in low doses at the time hypouricemic therapy is initiated.^20,21 (Hypouricemic therapy is discussed below.) We generally try to continue this prophylactic therapy, if the patient tolerates it, for at least 6 months—longer if tophi are still present or if attacks continue to occur.

If renal function is intact, colchicine can be prescribed at a dosage of 0.6 mg orally once or twice daily.²¹ In CKD, since the clearance of colchicine is reduced,⁴ the dosage should be reduced. Patients on colchicine for prophylaxis must be carefully monitored if the glomerular filtration rate is less than 50 mL/minute, or colchicine should be avoided altogether.⁶ Laboratory testing for colchicine levels is not routinely available and may be of limited value in predicting adverse effects; thus, recommendations about dose adjustments in CKD are empiric.

Wallace et al²² recommended a dose of 0.6 mg once daily if the creatinine clearance is 35 to 49 mL/minute and 0.6 mg every 2 to 3 days if it is 10 to 34 mL/minute, but there are no published long-term safety or efficacy data validating these reasonable (based on available information) dosing regimens.

Even with dose adjustment, caution is needed. Low-dose daily colchicine may be associated with reversible neuromyopathy and bone marrow suppression.^7,23 Patients with neuromyopathy may complain of myalgias, proximal muscle weakness, and numbness and may have areflexia and decreased sensation. Laboratory findings include elevated creatine kinase and aminotransferase levels. We regularly check for leukopenia or elevated creatine kinase and aspartate aminotransferase levels in patients with CKD who are receiving colchicine in any dose.

Prolonged colchicine therapy should probably be avoided in patients on hemodialysis, as this drug is not removed by dialysis or by exchange transfusion, and the risk of toxicity under these circumstances may be high.²² When there is no viable alternative and the drug is given, patients should be closely monitored for signs of toxicity.

Concurrent (even short-term) treatment with most macrolide antibiotics, particularly clarithromycin (Biaxin), most statin drugs, ketoconazole (Nizoral), cyclosporine, and likely other drugs predisposes to colchicine toxicity by altering its distribution and elimination, and can in rare cases cause morbidity or death.^24–26

NSAIDs are not optimal as prophylaxis in patients with chronic kidney disease

Little information has been published about using NSAIDs chronically to prevent flares, but they are not the optimal drugs to use in patients with CKD, as discussed above. In patients with end-stage renal disease, there are also concerns about NSAID-induced gastric and intestinal bleeding.

Low-dose steroids may not be effective as prophylaxis

Lower doses of steroids may not be effective as prophylaxis against gout flares, consistent with the common observation that gout flares still occur in organ transplant recipients who are taking maintenance doses of prednisone.¹³

 

PREVENTING FLARES BY LOWERING SERUM URATE LEVELS

If tophi are present, if radiography shows evidence of damage, if attacks are frequent or disabling, or if there are relative contraindications to the drugs that would be needed to treat acute attacks, then hypouricemic therapy should be strongly considered to reduce the burden of urate in the body, resorb tophi, and ultimately reduce the frequency of gout flares.²⁰

Although intermittent therapy for attacks or prolonged prophylactic use of colchicine may prevent recurrent episodes of gouty arthritis and may be reasonable for many patients, this approach does not prevent continued urate deposition, with the potential development of bony erosions, tophaceous deposits, and chronic arthritis.

The definitive therapy for gouty arthritis is to deplete the periarticular deposits of urate by maintaining a low serum urate level. Urate-lowering therapy, when indicated, is almost always lifelong.

Four strategies for lowering serum urate

The serum urate concentration can be lowered in four ways:

• Increasing renal uric acid excretion
• Altering the diet
• Decreasing urate synthesis
• Converting urate to a more soluble metabolite.

Increasing uric acid excretion is rarely effective if renal function is impaired

Probenecid, sulfinpyrazone (Anturane), and losartan (Cozaar) modestly increase uric acid secretion and reduce serum urate levels, but they are rarely effective if the creatinine clearance rate is less than 60 mL/minute, and they require significant fluid intake for maximal efficacy.

Uricosuric drugs probably should be avoided in patients who excrete more than 1,000 mg of uric acid per day on a normal diet, since urinary uric acid stones may form. In practice, however, patients are given losartan to treat hypertension without attention to uric acid excretion.

More-potent urocosuric drugs are being tested in clinical trials.

Altering the diet: Traditional advice confirmed

The Health Professionals Follow-up Study^27,28 prospectively examined the relation between diet and gout over 12 years in 47,150 men. The study confirmed some long-standing beliefs, such as that consuming meat, seafood, beer, and liquor increases the risk. Other risk factors were consumption of sugar-sweetened soft drinks and fructose, adiposity, weight gain, hypertension, and diuretic use. On the other hand, protein, wine, and purine-rich vegetables were not associated with gout flares. Low-fat dairy products may have a protective effect. Weight loss was found to be protective.

Low-purine diets are not very palatable, are difficult to adhere to, and are at best only minimally effective, lowering serum urate by 1 to 2 mg/dL. Low-protein diets designed to slow progression of CKD will likely also have only a slight effect on serum urate. Dietary change alone is not likely to dramatically lower serum urate levels.

Metabolizing urate with exogenous uricase

Rasburicase (Elitek) effectively converts urate to allantoin, which is more soluble, but rasburicase is fraught with allergic reactions and cannot be used as chronic therapy.

A pegylated intravenous uricase²⁹ has just been approved by the US Food and Drug Administration (FDA); the retail cost is not yet known. It is dramatically effective in those patients able to use it chronically, but it has not been fully evaluated in patients with CKD.

Decreasing urate synthesis with allopurinol

Allopurinol acts by competitively inhibiting xanthine oxidase, the enzyme that converts hypoxanthine to xanthine and xanthine to uric acid. The drug, a structural analogue of hypoxanthine, is converted by xanthine oxidase to oxypurinol, which is an even more effective inhibitor of xanthine oxidase than allopurinol.

Allopurinol is metabolized in the liver and has a half-life of 1 to 3 hours, but oxypurinol, which is excreted in the urine, has a half-life of 12 to 17 hours. Because of these pharmacokinetic properties, allopurinol can usually be given once daily, and the dosage required to reduce serum urate levels should in theory be lower in patients with lower glomerular filtration rates.

Allopurinol (100- and 300-mg tablets) is approved by the FDA in doses of up to 800 mg/day to treat hyperuricemia in patients with gout,³⁰ while guidelines from the British Society of Rheumatology advocate a maximum dose of 900 mg/day.³¹ These maximum doses are based on the limited amount of data with higher doses, not on documented toxicity.

Practice survey data in the United States indicate that most physicians prescribe no greater than 300 mg daily, although this dosage is likely to reduce the serum urate to less than 6 mg/dL—the goal level—in fewer than 50% of patients.^20,32 Patients with normal renal function occasionally require more than 1,000 mg daily to reduce the serum urate level to less than 6 mg/dL.

How low should the serum urate level be?

Ideally, therapy should keep the serum urate level significantly below 6.7 mg/dL, the approximate saturation point of urate in physiologic fluids.

Lowering the serum urate level from 10 mg/dL to 7 mg/dL may seem encouraging, and the urate level may be in the laboratory “normal” range; however, urate may continue to precipitate in tissues if the concentration is greater than 6.7 mg/dL. A target of 6 mg/dL, used in clinical studies, is far enough below the saturation level to provide some margin for fluctuations in serum levels. A serum level of 6.0 mg/dL has thus been arbitrarily proposed as a reasonable therapeutic target.

The lower the serum urate level achieved during hypouricemic therapy, the faster the reduction in tophaceous deposits. With adequate urate lowering, tophi can be visibly reduced in less than a year of hypouricemic therapy.^33,34

We have as yet no convincing evidence that lowering the serum urate level to less than 6.0 mg/dL is harmful, despite theoretical concerns that urate is a beneficial circulating antioxidant and epidemiologic observations that urate levels have been inversely correlated with progression of Parkinson disease.

 

 

Start low, go slow to avoid a flare

Rapid reduction of the serum urate level in a patient with chronic hyperuricemia and gout is likely to induce an acute flare.²⁰ We have traditionally used a “start low and increase slowly” approach to escalating hypouricemic therapy in hopes of reducing the likelihood of causing a gout flare.

Without anti-inflammatory prophylaxis, acute flares associated with urate-lowering are extremely likely. In a 28-week trial of allopurinol, febuxostat, and placebo by Schumacher et al,³³ during the first 8 weeks, when prophylaxis against gout flare was provided with either colchicine 0.6 mg once daily or naproxen (Naprosyn) 250 mg twice daily, the proportion of patients requiring treatment of gout flares was still 23% to 46%. When prophylaxis was stopped, the flare rate increased further.³³

The more we acutely lower serum urate levels, the more likely flares are to occur. In the study by Schumacher et al,³³ the percentage of patients needing treatment for gout flares during the first 8 weeks of the study, despite gout flare prophylaxis, was related to the percent reduction in serum urate by week 28 of the trial (Table 2).

IS IT NECESSARY TO ADJUST THE ALLOPURINOL DOSE IN CHRONIC KIDNEY DISEASE?

In 1984, Hande et al³⁵ proposed that allopurinol doses be lower in patients with renal insufficiency, with a dosage scale based on creatinine clearance.

Their thoughtful proposal was based on data from six of their own patients and 72 others with severe allopurinol toxicity, mainly allopurinol hypersensitivity syndrome, reported in the literature.

Perez-Ruiz et al³⁶ noted that patients who had experienced adverse effects from allopurinol in their series were likely to have had received “higher” doses of allopurinol, if the dosage was corrected for reduced oxypurinol elimination based on their estimated creatinine clearance.

However, most of these reactions occurred soon after initiating therapy, a temporal pattern more typical of non-dose-dependent allergic reactions. Additionally, allopurinol hypersensitivity has been linked to T-cell-mediated immune reactions to oxypurinol,³⁷ a mechanism not likely linked to drug levels.

Arguments against dose adjustment

Despite the compelling information that allopurinol reactions are more common in CKD, adjusting the dosage of allopurinol has not been clearly shown to reduce the frequency of these reactions.

In a small retrospective analysis, Vázquez-Mellado et al³⁸ reported that adjusting the allopurinol dosage according to creatinine clearance did not decrease the incidence of allopurinol hypersensitivity.

In a study in 250 patients, Dalbeth et al³⁹ showed that the overall incidence of hypersensitivity reaction was 1.6%, and the incidence of allergic reactions did not decrease when allopurinol was given according to the dosing guidelines proposed by Hande et al.³⁵ However, it is worth noting that, of the patients who received the recommended lower doses, only 19% achieved the target serum urate level of 6 mg/dL.³⁹

Silverberg et al⁴⁰ found that of 15 patients who developed hypersensitivity reactions to allopurinol, 10 had received doses that were low or appropriate according to the guidelines of Hande et al.³⁵

More recently, Stamp et al⁴¹ found that gradually increasing the allopurinol dose above the proposed creatinine clearance-based dose was safe and effective. Thirty-one (89%) of the 35 patients who completed the study achieved the target serum urate level of 6 mg/dL, while only 3 of 45 who started the study developed rashes, which were not serious.

The small number of patients in these studies limits any strong conclusion, but at present there is no interventional study showing that allopurinol dosing adjustment based on glomerular filtration rate is effective or safer than dosing based on the serum urate level.

Our view on allopurinol dosing adjustment

We believe the initial observations of Hande et al³⁵ and the subsequent meticulous data from Perez-Ruiz et al³⁶ suggest a relationship between CKD and the occurrence of severe allopurinol reactions. However, these observations do not prove that dose adjustment will prevent these reactions.

In patients with normal kidney function, the FDA³⁰ and the European League Against Rheumatism (EULAR)⁴² recommend slow upward titration, starting with 100 to 200 mg/day, which we agree should decrease the frequency of acute gout flares. The dose is increased by increments of 100 mg/day at intervals of 1 week (FDA recommendation) or 2 to 4 weeks (EULAR recommendation) until the serum urate level is lower than 6 mg/dL.

We believe the optimal approach to allopurinol dosing in patients with CKD remains uncertain. We generally escalate the dose slowly, with ongoing frequent laboratory and clinical monitoring, and we do not limit the maximal dose as suggested by Hande et al.³⁵

An alternative strategy is to use the newer, far more expensive xanthine oxidase inhibitor febuxostat in patients with CKD, since it is not excreted by the kidney. We usually first try escalating doses of allopurinol.

 

FEBUXOSTAT, AN ALTERNATIVE TO ALLOPURINOL

Febuxostat is an oral nonpurine inhibitor of xanthine oxidase.⁴³ Approved by the FDA in 2009, it is available in 40- and 80-mg tablets.

Unlike allopurinol, febuxostat is metabolized primarily by hepatic glucuronide formation and oxidation and then excreted in stool and urine,⁴⁴ making it in theory an attractive agent in patients with renal insufficiency, bypassing the controversial dose-adjustment issue with allopurinol.

In the Febuxostat Versus Allopurinol Controlled Trial (FACT),²⁰ a 52-week randomized, double-blind study in hyperuricemic patients with gout, serum urate levels were reduced to less than 6.0 mg/dL in over 50% of patients receiving febuxostat 80 mg or 120 mg once daily, while only 21% of patients receiving 300 mg of allopurinol achieved this goal. This does not imply that allopurinol at higher doses, as should be used in clinical practice,⁴⁵ would not be equally effective. Patients with CKD were not included in this trial.

In the study by Schumacher et al,³³ febuxostat 80, 120, or 240 mg once daily reduced serum urate. A small subset (35 patients) had mild to moderate renal insufficiency (serum creatinine 1.5–2 mg/dL).³³ The number of patients with renal insufficiency who achieved the primary end point of a serum urate level lower than 6 mg/dL was 4 (44%) of 9 in the febuxostat 80-mg group, 5 (46%) of 11 in the 120-mg group, and 3 (60%) of 5 in the 240-mg group, while none of the 10 patients in the dose-adjusted allopurinol group achieved the primary end point (P < .05). Of note, 41% of the patients with normal renal function who received allopurinol achieved the primary end point.³³ As proposed above, if the allopurinol dose had been slowly increased in the patients with renal insufficiency, it might have been equally effective.

Febuxostat has not been thoroughly evaluated in patients with severe CKD or in patients on hemodialysis.

A presumed niche indication of febuxostat is in patients allergic to allopurinol, since the drugs are not similar in chemical structure. However, at present, experience with this use is limited. Allopurinol-allergic patients were excluded from the clinical trials; thus, if there is any allergic overlap, it would not likely have been recognized in those studies. The FDA has received reports of patients who were allergic to allopurinol also having reactions to febuxostat, and it is currently evaluating these reports (personal communication).

Concern was raised over cardiovascular adverse events in patients treated with febuxostat during clinical trials. In the FACT trial, two patients died of cardiac causes.²⁰ In the study by Schumacher et al,³³ 11 of 670 patients experienced cardiac adverse events in the febuxostat group vs 3 of 268 in the allopurinol group. Events included atrial fibrillation, chest pain, coronary artery disease, and myocardial infarction. However, this difference was not statistically significant.

Febuxostat costs much more than allopurinol. Currently, patients pay $153.88 for 1 month of febuxostat 40 or 80 mg from Cleveland Clinic pharmacy; 1 month of allopurinol costs $17.45 (300 mg) or $14.00 (100 mg). We believe febuxostat should be reserved for patients with documented intolerance to allopurinol in effective doses.

Monitoring serum urate levels is important in all patients on hypouricemic therapy so that dosage adjustments can be made until the target serum urate concentration is reached. In patients failing to meet target serum urate levels, patient adherence with the prescribed dosing should be specifically addressed because as many as 50% of patients do not adhere to their prescribed regimen.

DOES URATE-LOWERING THERAPY HAVE BENEFITS BEYOND GOUT?

Despite experimental animal data and a strong epidemiologic association between hyperuri-cemia and hypertension,⁴⁶ metabolic syndrome, and rates of cardiovascular and all-cause mortality,⁴⁷ the evidence from interventional trials so far does not support the routine use of hypo-uricemic therapy to prevent these outcomes.

Similarly, hyperuricemia has long been associated with renal disease, and there has been debate as to whether hyperuricemia is a result of kidney dysfunction or a contributing factor.^46,48–51 A few studies have documented improvement of renal function after initiation of hypouricemic therapy.⁵² However, treating asymptomatic hyperuricemia to preserve kidney function remains controversial.

A recent study indicates that lowering the serum urate level with allopurinol can lower the blood pressure in hyperuricemic adolescents who have newly diagnosed primary hypertension.⁵³ This does not indicate, however, that initiating hypouricemic therapy in patients with preexisting, long-standing hypertension will be successful.

RECOMMENDED FOR OUR PATIENT

As for our diabetic patient with an acute gout flare and creatinine clearance rate of 45 mL/minute, we would recommend:

• Aspirating the knee, sending the fluid for bacterial culture, and then treating it with a local glucocorticoid injection
• Starting colchicine 0.6 mg every day, with frequent monitoring for signs of toxicity (muscle pain, weakness, leukopenia, and elevations of creatine kinase and aspartate aminotransferase)
• Increasing his allopurinol dose by 100 mg every 2 to 4 weeks until the target serum urate level of less than 6.0 mg/dL is reached
• If he cannot tolerate allopurinol or if the target serum urate level is not achieved despite adequate doses of allopurinol (about 800 mg), we would switch to febuxostat 40 mg and increase the dose as needed to achieve the desired urate level.

You have a 54-year-old black patient with gout, diabetes mellitus, hypertension, and chronic kidney disease (CKD). He has an acute gout flare involving his right knee. In the past year he has had four attacks of gout in the ankles and knees, which you treated with intra-articular glucocorticoid injections. He has been on allopurinol (Zyloprim) 200 mg daily, but his last serum urate level was 9.4 mg/dL (reference range 3.0–8.0). His creatinine clearance is 45 mL/minute (reference range 85–125).

In view of his kidney disease, you are concerned about increasing his dose of allopurinol, but also about the need to treat his frequent attacks. How should you manage this patient?

GOUT IS CHALLENGING TO TREAT IN PATIENTS WITH KIDNEY DISEASE

A major challenge in treating patients with gout is to avoid therapeutic interactions with common comorbidities, including hypertension, insulin resistance, coronary artery disease, heart failure, and especially CKD.¹

In this paper, we discuss approaches to and controversies in the management of gout and hyperuricemia in patients with CKD. Unfortunately, the evidence from clinical trials to guide treatment decisions is limited; therefore, decisions must often be based on experience and pathophysiologic principles.

GENERAL GOALS OF GOUT THERAPY

Depending on the patient and the stage of the disease, the goals in treating patients with gout are to:

• Terminate acute attacks as promptly and safely as possible
• Prevent recurrences of acute gout attacks
• Prevent or reverse complications resulting from deposition of monosodium urate in the joints, in the kidneys, or at other sites.

These goals are more difficult to achieve in patients with CKD because of the potential complications from many of the available drugs.

TERMINATING ACUTE GOUT FLARES

In patients with acute gout, treatment is aimed at quickly resolving pain and inflammation.

Several types of drugs can terminate acute gout flares. The choice in most situations is colchicine (Colcrys); a nonsteroidal anti-inflammatory drug (NSAID); a corticosteroid; or corticotropin (ACTH).

However, in patients with CKD, there are concerns about using colchicine or NSAIDs, and corticotropin is very expensive; thus, corticosteroids are often used.

Colchicine’s clearance is reduced in CKD

Colchicine is somewhat effective in treating acute gout attacks and probably more effective in preventing attacks.

Due to concerns about inappropriate dosing and reported deaths,² the intravenous formulation is not available in many countries, including the United States.

After oral administration, colchicine is rapidly absorbed, with a bioavailability of up to 50%. It undergoes metabolism by the liver, and its metabolites are excreted by renal and biliary-intestinal routes. Up to 20% of the active drug is excreted by the kidneys.³

Colchicine’s clearance is significantly reduced in patients with renal or hepatic insufficiency, and the drug may accumulate in cells, with resultant toxicity.⁴ Colchicine-induced toxicity has been observed when the drug was used for acute treatment, as well as for chronic prophylaxis of gout in patients with CKD; thus, alternative agents for treating acute attacks should be considered.^5,6 With prolonged use, reversible colchicine-induced axonal neuropathy, neutropenia, and vacuolar myopathy can develop in patients with CKD.⁷

In a trial in patients with normal renal function, nearly 100% who received an initial dose of 1 mg followed by 0.5 mg every 2 hours developed diarrhea at a median time of 24 hours.⁸ Emesis may also occur.

A lower dose of 1.8 mg (two 0.6-mg pills followed by one pill an hour later) was well tolerated but only moderately effective in treating acute gout, causing at least a 50% reduction in pain at 24 hours in only 38% of patients.⁹ This study does not clarify the dosage to use to completely resolve attacks. Using additional colchicine likely will increase the response rate, but will also increase side effects. Patients with CKD were not included.

Some patients, as shown in the above trial, can abort attacks by taking only one or two colchicine tablets when they feel the first “twinge” of an attack. This approach is likely to be safe in CKD, but it may be of value to only a few patients.

Nonsteroidal anti-inflammatory drugs can worsen chronic kidney disease

NSAIDs in high doses can effectively treat the pain and inflammation of acute gout. Indomethacin (Indocin) 50 mg three times daily has been standard NSAID therapy.

Other nonselective NSAIDs and NSAIDs that selectively inhibit cyclooxygenase 2 (COX-2) are effective, but all can cause acute renal toxicity or worsen CKD.¹⁰ Renal side effects include salt and water retention, acute tubular necrosis, acute interstitial nephritis, proteinuria, hypertension, hyperkalemia, and chronic renal injury.¹¹

Even short-term use of high-dose NSAIDs should generally be avoided in patients with preexisting CKD, for whom there is no established safe threshold dose. When NSAIDs (including selective COX-2 inhibitors) are used, renal function should be monitored closely and the duration limited as much as possible.

 

 

Corticosteroids are often used to treat acute attacks

Due to the concerns about NSAIDs or colchicine to treat acute gout attacks in patients with CKD, corticosteroids are often used in this setting.

Intra-articular steroid injections are useful in treating acute gout limited to a single joint or bursa.¹² However, one should first make sure that the joint is not infected: septic arthritis should ideally be excluded by arthrocentesis, particularly in immunosuppressed patients¹³ or those with end-stage renal disease, who are predisposed to bacteremia.

Oral, intramuscular, or intravenous steroids can provide complete relief from acute gout, although high doses (eg, prednisone 30–60 mg/day or the equivalent) are often needed. Common errors resulting in inefficacy include using too low a dose or not treating for a sufficient time before tapering or stopping. Groff and colleagues¹⁴ described 13 patients who received oral or intravenous steroids for acute gout. Nine patients received an initial single dose of prednisone ranging from 20 to 50 mg, with tapering over a mean of 10 days. Twelve of the 13 patients had improvement within 48 hours, and the signs and symptoms of acute gout resolved completely within 7 to 10 days.

We often give prednisone 40 mg daily until a day after the acute attack resolves and then taper over another 7 to 10 days. There are no data to guide steroid dosing in an evidence-based way, but we believe too short a course of therapy may result in return of symptoms.

Corticotropin and other agents: Effective but costly

Corticotropin shares the same indications as systemic corticosteroids, being used to treat flares when NSAIDs, intra-articular steroids, and colchicine are contraindicated. However, corticotropin is far more expensive than generic corticosteroids, costing nearly $2,000 for a single 80-IU dose, which may need to be repeated.

Corticotropin is available for subcutaneous or intramuscular injection. A single intramuscular injection of corticotropin gel (H.P. Acthar, 25–80 IU) may terminate an acute gout attack.¹⁵ However, many patients need another injection after 24 to 72 hours, which would require another visit to the physician. This treatment has been touted by some as being more effective than corticosteroid therapy, possibly because of a unique peripheral mechanism of action in addition to stimulating cortisol release.¹⁶

We rarely use corticotropin, in view of its cost as well as concerns about excessive sodium and water retention due to the release of multiple hormones from the adrenal gland. This may be especially deleterious in patients with CKD or congestive heart failure.¹⁷

Parenteral anti-tumor necrosis factor agents or interleukin 1 antagonists can be dramatically effective but are also expensive.^18,19 For example, anakinra (Kineret) 100 mg costs about $73, and multiple daily doses may be necessary.

Under unique conditions in which they can be safely used (eg, patients with CKD, diabetes mellitus, liver disease), they may be cost-effective if they can shorten the stay of a hospitalized patient with acute gout.

PROPHYLACTIC ANTI-INFLAMMATORY THERAPY FOR PATIENTS WITH GOUT

Between attacks, the goal is to prevent new attacks through prophylactic management, which may include anti-inflammatory and hypouricemic therapy along with dietary instruction (such as avoiding excessive beer, liquor, and fructose ingestion).

Colchicine can be used as prophylaxis, with caution and monitoring

Although colchicine is not 100% effective, it markedly reduces the flare rate when started in low doses at the time hypouricemic therapy is initiated.^20,21 (Hypouricemic therapy is discussed below.) We generally try to continue this prophylactic therapy, if the patient tolerates it, for at least 6 months—longer if tophi are still present or if attacks continue to occur.

If renal function is intact, colchicine can be prescribed at a dosage of 0.6 mg orally once or twice daily.²¹ In CKD, since the clearance of colchicine is reduced,⁴ the dosage should be reduced. Patients on colchicine for prophylaxis must be carefully monitored if the glomerular filtration rate is less than 50 mL/minute, or colchicine should be avoided altogether.⁶ Laboratory testing for colchicine levels is not routinely available and may be of limited value in predicting adverse effects; thus, recommendations about dose adjustments in CKD are empiric.

Wallace et al²² recommended a dose of 0.6 mg once daily if the creatinine clearance is 35 to 49 mL/minute and 0.6 mg every 2 to 3 days if it is 10 to 34 mL/minute, but there are no published long-term safety or efficacy data validating these reasonable (based on available information) dosing regimens.

Even with dose adjustment, caution is needed. Low-dose daily colchicine may be associated with reversible neuromyopathy and bone marrow suppression.^7,23 Patients with neuromyopathy may complain of myalgias, proximal muscle weakness, and numbness and may have areflexia and decreased sensation. Laboratory findings include elevated creatine kinase and aminotransferase levels. We regularly check for leukopenia or elevated creatine kinase and aspartate aminotransferase levels in patients with CKD who are receiving colchicine in any dose.

Prolonged colchicine therapy should probably be avoided in patients on hemodialysis, as this drug is not removed by dialysis or by exchange transfusion, and the risk of toxicity under these circumstances may be high.²² When there is no viable alternative and the drug is given, patients should be closely monitored for signs of toxicity.

Concurrent (even short-term) treatment with most macrolide antibiotics, particularly clarithromycin (Biaxin), most statin drugs, ketoconazole (Nizoral), cyclosporine, and likely other drugs predisposes to colchicine toxicity by altering its distribution and elimination, and can in rare cases cause morbidity or death.^24–26

NSAIDs are not optimal as prophylaxis in patients with chronic kidney disease

Little information has been published about using NSAIDs chronically to prevent flares, but they are not the optimal drugs to use in patients with CKD, as discussed above. In patients with end-stage renal disease, there are also concerns about NSAID-induced gastric and intestinal bleeding.

Low-dose steroids may not be effective as prophylaxis

Lower doses of steroids may not be effective as prophylaxis against gout flares, consistent with the common observation that gout flares still occur in organ transplant recipients who are taking maintenance doses of prednisone.¹³

 

PREVENTING FLARES BY LOWERING SERUM URATE LEVELS

If tophi are present, if radiography shows evidence of damage, if attacks are frequent or disabling, or if there are relative contraindications to the drugs that would be needed to treat acute attacks, then hypouricemic therapy should be strongly considered to reduce the burden of urate in the body, resorb tophi, and ultimately reduce the frequency of gout flares.²⁰

Although intermittent therapy for attacks or prolonged prophylactic use of colchicine may prevent recurrent episodes of gouty arthritis and may be reasonable for many patients, this approach does not prevent continued urate deposition, with the potential development of bony erosions, tophaceous deposits, and chronic arthritis.

The definitive therapy for gouty arthritis is to deplete the periarticular deposits of urate by maintaining a low serum urate level. Urate-lowering therapy, when indicated, is almost always lifelong.

Four strategies for lowering serum urate

The serum urate concentration can be lowered in four ways:

• Increasing renal uric acid excretion
• Altering the diet
• Decreasing urate synthesis
• Converting urate to a more soluble metabolite.

Increasing uric acid excretion is rarely effective if renal function is impaired

Probenecid, sulfinpyrazone (Anturane), and losartan (Cozaar) modestly increase uric acid secretion and reduce serum urate levels, but they are rarely effective if the creatinine clearance rate is less than 60 mL/minute, and they require significant fluid intake for maximal efficacy.

Uricosuric drugs probably should be avoided in patients who excrete more than 1,000 mg of uric acid per day on a normal diet, since urinary uric acid stones may form. In practice, however, patients are given losartan to treat hypertension without attention to uric acid excretion.

More-potent urocosuric drugs are being tested in clinical trials.

Altering the diet: Traditional advice confirmed

The Health Professionals Follow-up Study^27,28 prospectively examined the relation between diet and gout over 12 years in 47,150 men. The study confirmed some long-standing beliefs, such as that consuming meat, seafood, beer, and liquor increases the risk. Other risk factors were consumption of sugar-sweetened soft drinks and fructose, adiposity, weight gain, hypertension, and diuretic use. On the other hand, protein, wine, and purine-rich vegetables were not associated with gout flares. Low-fat dairy products may have a protective effect. Weight loss was found to be protective.

Low-purine diets are not very palatable, are difficult to adhere to, and are at best only minimally effective, lowering serum urate by 1 to 2 mg/dL. Low-protein diets designed to slow progression of CKD will likely also have only a slight effect on serum urate. Dietary change alone is not likely to dramatically lower serum urate levels.

Metabolizing urate with exogenous uricase

Rasburicase (Elitek) effectively converts urate to allantoin, which is more soluble, but rasburicase is fraught with allergic reactions and cannot be used as chronic therapy.

A pegylated intravenous uricase²⁹ has just been approved by the US Food and Drug Administration (FDA); the retail cost is not yet known. It is dramatically effective in those patients able to use it chronically, but it has not been fully evaluated in patients with CKD.

Decreasing urate synthesis with allopurinol

Allopurinol acts by competitively inhibiting xanthine oxidase, the enzyme that converts hypoxanthine to xanthine and xanthine to uric acid. The drug, a structural analogue of hypoxanthine, is converted by xanthine oxidase to oxypurinol, which is an even more effective inhibitor of xanthine oxidase than allopurinol.

Allopurinol is metabolized in the liver and has a half-life of 1 to 3 hours, but oxypurinol, which is excreted in the urine, has a half-life of 12 to 17 hours. Because of these pharmacokinetic properties, allopurinol can usually be given once daily, and the dosage required to reduce serum urate levels should in theory be lower in patients with lower glomerular filtration rates.

Allopurinol (100- and 300-mg tablets) is approved by the FDA in doses of up to 800 mg/day to treat hyperuricemia in patients with gout,³⁰ while guidelines from the British Society of Rheumatology advocate a maximum dose of 900 mg/day.³¹ These maximum doses are based on the limited amount of data with higher doses, not on documented toxicity.

Practice survey data in the United States indicate that most physicians prescribe no greater than 300 mg daily, although this dosage is likely to reduce the serum urate to less than 6 mg/dL—the goal level—in fewer than 50% of patients.^20,32 Patients with normal renal function occasionally require more than 1,000 mg daily to reduce the serum urate level to less than 6 mg/dL.

How low should the serum urate level be?

Ideally, therapy should keep the serum urate level significantly below 6.7 mg/dL, the approximate saturation point of urate in physiologic fluids.

Lowering the serum urate level from 10 mg/dL to 7 mg/dL may seem encouraging, and the urate level may be in the laboratory “normal” range; however, urate may continue to precipitate in tissues if the concentration is greater than 6.7 mg/dL. A target of 6 mg/dL, used in clinical studies, is far enough below the saturation level to provide some margin for fluctuations in serum levels. A serum level of 6.0 mg/dL has thus been arbitrarily proposed as a reasonable therapeutic target.

The lower the serum urate level achieved during hypouricemic therapy, the faster the reduction in tophaceous deposits. With adequate urate lowering, tophi can be visibly reduced in less than a year of hypouricemic therapy.^33,34

We have as yet no convincing evidence that lowering the serum urate level to less than 6.0 mg/dL is harmful, despite theoretical concerns that urate is a beneficial circulating antioxidant and epidemiologic observations that urate levels have been inversely correlated with progression of Parkinson disease.

 

 

Start low, go slow to avoid a flare

Rapid reduction of the serum urate level in a patient with chronic hyperuricemia and gout is likely to induce an acute flare.²⁰ We have traditionally used a “start low and increase slowly” approach to escalating hypouricemic therapy in hopes of reducing the likelihood of causing a gout flare.

Without anti-inflammatory prophylaxis, acute flares associated with urate-lowering are extremely likely. In a 28-week trial of allopurinol, febuxostat, and placebo by Schumacher et al,³³ during the first 8 weeks, when prophylaxis against gout flare was provided with either colchicine 0.6 mg once daily or naproxen (Naprosyn) 250 mg twice daily, the proportion of patients requiring treatment of gout flares was still 23% to 46%. When prophylaxis was stopped, the flare rate increased further.³³

The more we acutely lower serum urate levels, the more likely flares are to occur. In the study by Schumacher et al,³³ the percentage of patients needing treatment for gout flares during the first 8 weeks of the study, despite gout flare prophylaxis, was related to the percent reduction in serum urate by week 28 of the trial (Table 2).

IS IT NECESSARY TO ADJUST THE ALLOPURINOL DOSE IN CHRONIC KIDNEY DISEASE?

In 1984, Hande et al³⁵ proposed that allopurinol doses be lower in patients with renal insufficiency, with a dosage scale based on creatinine clearance.

Their thoughtful proposal was based on data from six of their own patients and 72 others with severe allopurinol toxicity, mainly allopurinol hypersensitivity syndrome, reported in the literature.

Perez-Ruiz et al³⁶ noted that patients who had experienced adverse effects from allopurinol in their series were likely to have had received “higher” doses of allopurinol, if the dosage was corrected for reduced oxypurinol elimination based on their estimated creatinine clearance.

However, most of these reactions occurred soon after initiating therapy, a temporal pattern more typical of non-dose-dependent allergic reactions. Additionally, allopurinol hypersensitivity has been linked to T-cell-mediated immune reactions to oxypurinol,³⁷ a mechanism not likely linked to drug levels.

Arguments against dose adjustment

Despite the compelling information that allopurinol reactions are more common in CKD, adjusting the dosage of allopurinol has not been clearly shown to reduce the frequency of these reactions.

In a small retrospective analysis, Vázquez-Mellado et al³⁸ reported that adjusting the allopurinol dosage according to creatinine clearance did not decrease the incidence of allopurinol hypersensitivity.

In a study in 250 patients, Dalbeth et al³⁹ showed that the overall incidence of hypersensitivity reaction was 1.6%, and the incidence of allergic reactions did not decrease when allopurinol was given according to the dosing guidelines proposed by Hande et al.³⁵ However, it is worth noting that, of the patients who received the recommended lower doses, only 19% achieved the target serum urate level of 6 mg/dL.³⁹

Silverberg et al⁴⁰ found that of 15 patients who developed hypersensitivity reactions to allopurinol, 10 had received doses that were low or appropriate according to the guidelines of Hande et al.³⁵

More recently, Stamp et al⁴¹ found that gradually increasing the allopurinol dose above the proposed creatinine clearance-based dose was safe and effective. Thirty-one (89%) of the 35 patients who completed the study achieved the target serum urate level of 6 mg/dL, while only 3 of 45 who started the study developed rashes, which were not serious.

The small number of patients in these studies limits any strong conclusion, but at present there is no interventional study showing that allopurinol dosing adjustment based on glomerular filtration rate is effective or safer than dosing based on the serum urate level.

Our view on allopurinol dosing adjustment

We believe the initial observations of Hande et al³⁵ and the subsequent meticulous data from Perez-Ruiz et al³⁶ suggest a relationship between CKD and the occurrence of severe allopurinol reactions. However, these observations do not prove that dose adjustment will prevent these reactions.

In patients with normal kidney function, the FDA³⁰ and the European League Against Rheumatism (EULAR)⁴² recommend slow upward titration, starting with 100 to 200 mg/day, which we agree should decrease the frequency of acute gout flares. The dose is increased by increments of 100 mg/day at intervals of 1 week (FDA recommendation) or 2 to 4 weeks (EULAR recommendation) until the serum urate level is lower than 6 mg/dL.

We believe the optimal approach to allopurinol dosing in patients with CKD remains uncertain. We generally escalate the dose slowly, with ongoing frequent laboratory and clinical monitoring, and we do not limit the maximal dose as suggested by Hande et al.³⁵

An alternative strategy is to use the newer, far more expensive xanthine oxidase inhibitor febuxostat in patients with CKD, since it is not excreted by the kidney. We usually first try escalating doses of allopurinol.

 

FEBUXOSTAT, AN ALTERNATIVE TO ALLOPURINOL

Febuxostat is an oral nonpurine inhibitor of xanthine oxidase.⁴³ Approved by the FDA in 2009, it is available in 40- and 80-mg tablets.

Unlike allopurinol, febuxostat is metabolized primarily by hepatic glucuronide formation and oxidation and then excreted in stool and urine,⁴⁴ making it in theory an attractive agent in patients with renal insufficiency, bypassing the controversial dose-adjustment issue with allopurinol.

In the Febuxostat Versus Allopurinol Controlled Trial (FACT),²⁰ a 52-week randomized, double-blind study in hyperuricemic patients with gout, serum urate levels were reduced to less than 6.0 mg/dL in over 50% of patients receiving febuxostat 80 mg or 120 mg once daily, while only 21% of patients receiving 300 mg of allopurinol achieved this goal. This does not imply that allopurinol at higher doses, as should be used in clinical practice,⁴⁵ would not be equally effective. Patients with CKD were not included in this trial.

In the study by Schumacher et al,³³ febuxostat 80, 120, or 240 mg once daily reduced serum urate. A small subset (35 patients) had mild to moderate renal insufficiency (serum creatinine 1.5–2 mg/dL).³³ The number of patients with renal insufficiency who achieved the primary end point of a serum urate level lower than 6 mg/dL was 4 (44%) of 9 in the febuxostat 80-mg group, 5 (46%) of 11 in the 120-mg group, and 3 (60%) of 5 in the 240-mg group, while none of the 10 patients in the dose-adjusted allopurinol group achieved the primary end point (P < .05). Of note, 41% of the patients with normal renal function who received allopurinol achieved the primary end point.³³ As proposed above, if the allopurinol dose had been slowly increased in the patients with renal insufficiency, it might have been equally effective.

Febuxostat has not been thoroughly evaluated in patients with severe CKD or in patients on hemodialysis.

A presumed niche indication of febuxostat is in patients allergic to allopurinol, since the drugs are not similar in chemical structure. However, at present, experience with this use is limited. Allopurinol-allergic patients were excluded from the clinical trials; thus, if there is any allergic overlap, it would not likely have been recognized in those studies. The FDA has received reports of patients who were allergic to allopurinol also having reactions to febuxostat, and it is currently evaluating these reports (personal communication).

Concern was raised over cardiovascular adverse events in patients treated with febuxostat during clinical trials. In the FACT trial, two patients died of cardiac causes.²⁰ In the study by Schumacher et al,³³ 11 of 670 patients experienced cardiac adverse events in the febuxostat group vs 3 of 268 in the allopurinol group. Events included atrial fibrillation, chest pain, coronary artery disease, and myocardial infarction. However, this difference was not statistically significant.

Febuxostat costs much more than allopurinol. Currently, patients pay $153.88 for 1 month of febuxostat 40 or 80 mg from Cleveland Clinic pharmacy; 1 month of allopurinol costs $17.45 (300 mg) or $14.00 (100 mg). We believe febuxostat should be reserved for patients with documented intolerance to allopurinol in effective doses.

Monitoring serum urate levels is important in all patients on hypouricemic therapy so that dosage adjustments can be made until the target serum urate concentration is reached. In patients failing to meet target serum urate levels, patient adherence with the prescribed dosing should be specifically addressed because as many as 50% of patients do not adhere to their prescribed regimen.

DOES URATE-LOWERING THERAPY HAVE BENEFITS BEYOND GOUT?

Despite experimental animal data and a strong epidemiologic association between hyperuri-cemia and hypertension,⁴⁶ metabolic syndrome, and rates of cardiovascular and all-cause mortality,⁴⁷ the evidence from interventional trials so far does not support the routine use of hypo-uricemic therapy to prevent these outcomes.

Similarly, hyperuricemia has long been associated with renal disease, and there has been debate as to whether hyperuricemia is a result of kidney dysfunction or a contributing factor.^46,48–51 A few studies have documented improvement of renal function after initiation of hypouricemic therapy.⁵² However, treating asymptomatic hyperuricemia to preserve kidney function remains controversial.

A recent study indicates that lowering the serum urate level with allopurinol can lower the blood pressure in hyperuricemic adolescents who have newly diagnosed primary hypertension.⁵³ This does not indicate, however, that initiating hypouricemic therapy in patients with preexisting, long-standing hypertension will be successful.

RECOMMENDED FOR OUR PATIENT

As for our diabetic patient with an acute gout flare and creatinine clearance rate of 45 mL/minute, we would recommend:

• Aspirating the knee, sending the fluid for bacterial culture, and then treating it with a local glucocorticoid injection
• Starting colchicine 0.6 mg every day, with frequent monitoring for signs of toxicity (muscle pain, weakness, leukopenia, and elevations of creatine kinase and aspartate aminotransferase)
• Increasing his allopurinol dose by 100 mg every 2 to 4 weeks until the target serum urate level of less than 6.0 mg/dL is reached
• If he cannot tolerate allopurinol or if the target serum urate level is not achieved despite adequate doses of allopurinol (about 800 mg), we would switch to febuxostat 40 mg and increase the dose as needed to achieve the desired urate level.

You have a 54-year-old black patient with gout, diabetes mellitus, hypertension, and chronic kidney disease (CKD). He has an acute gout flare involving his right knee. In the past year he has had four attacks of gout in the ankles and knees, which you treated with intra-articular glucocorticoid injections. He has been on allopurinol (Zyloprim) 200 mg daily, but his last serum urate level was 9.4 mg/dL (reference range 3.0–8.0). His creatinine clearance is 45 mL/minute (reference range 85–125).

In view of his kidney disease, you are concerned about increasing his dose of allopurinol, but also about the need to treat his frequent attacks. How should you manage this patient?

GOUT IS CHALLENGING TO TREAT IN PATIENTS WITH KIDNEY DISEASE

A major challenge in treating patients with gout is to avoid therapeutic interactions with common comorbidities, including hypertension, insulin resistance, coronary artery disease, heart failure, and especially CKD.¹

In this paper, we discuss approaches to and controversies in the management of gout and hyperuricemia in patients with CKD. Unfortunately, the evidence from clinical trials to guide treatment decisions is limited; therefore, decisions must often be based on experience and pathophysiologic principles.

GENERAL GOALS OF GOUT THERAPY

Depending on the patient and the stage of the disease, the goals in treating patients with gout are to:

• Terminate acute attacks as promptly and safely as possible
• Prevent recurrences of acute gout attacks
• Prevent or reverse complications resulting from deposition of monosodium urate in the joints, in the kidneys, or at other sites.

These goals are more difficult to achieve in patients with CKD because of the potential complications from many of the available drugs.

TERMINATING ACUTE GOUT FLARES

In patients with acute gout, treatment is aimed at quickly resolving pain and inflammation.

Several types of drugs can terminate acute gout flares. The choice in most situations is colchicine (Colcrys); a nonsteroidal anti-inflammatory drug (NSAID); a corticosteroid; or corticotropin (ACTH).

However, in patients with CKD, there are concerns about using colchicine or NSAIDs, and corticotropin is very expensive; thus, corticosteroids are often used.

Colchicine’s clearance is reduced in CKD

Colchicine is somewhat effective in treating acute gout attacks and probably more effective in preventing attacks.

Due to concerns about inappropriate dosing and reported deaths,² the intravenous formulation is not available in many countries, including the United States.

After oral administration, colchicine is rapidly absorbed, with a bioavailability of up to 50%. It undergoes metabolism by the liver, and its metabolites are excreted by renal and biliary-intestinal routes. Up to 20% of the active drug is excreted by the kidneys.³

Colchicine’s clearance is significantly reduced in patients with renal or hepatic insufficiency, and the drug may accumulate in cells, with resultant toxicity.⁴ Colchicine-induced toxicity has been observed when the drug was used for acute treatment, as well as for chronic prophylaxis of gout in patients with CKD; thus, alternative agents for treating acute attacks should be considered.^5,6 With prolonged use, reversible colchicine-induced axonal neuropathy, neutropenia, and vacuolar myopathy can develop in patients with CKD.⁷

In a trial in patients with normal renal function, nearly 100% who received an initial dose of 1 mg followed by 0.5 mg every 2 hours developed diarrhea at a median time of 24 hours.⁸ Emesis may also occur.

A lower dose of 1.8 mg (two 0.6-mg pills followed by one pill an hour later) was well tolerated but only moderately effective in treating acute gout, causing at least a 50% reduction in pain at 24 hours in only 38% of patients.⁹ This study does not clarify the dosage to use to completely resolve attacks. Using additional colchicine likely will increase the response rate, but will also increase side effects. Patients with CKD were not included.

Some patients, as shown in the above trial, can abort attacks by taking only one or two colchicine tablets when they feel the first “twinge” of an attack. This approach is likely to be safe in CKD, but it may be of value to only a few patients.

Nonsteroidal anti-inflammatory drugs can worsen chronic kidney disease

NSAIDs in high doses can effectively treat the pain and inflammation of acute gout. Indomethacin (Indocin) 50 mg three times daily has been standard NSAID therapy.

Other nonselective NSAIDs and NSAIDs that selectively inhibit cyclooxygenase 2 (COX-2) are effective, but all can cause acute renal toxicity or worsen CKD.¹⁰ Renal side effects include salt and water retention, acute tubular necrosis, acute interstitial nephritis, proteinuria, hypertension, hyperkalemia, and chronic renal injury.¹¹

Even short-term use of high-dose NSAIDs should generally be avoided in patients with preexisting CKD, for whom there is no established safe threshold dose. When NSAIDs (including selective COX-2 inhibitors) are used, renal function should be monitored closely and the duration limited as much as possible.

 

 

Corticosteroids are often used to treat acute attacks

Due to the concerns about NSAIDs or colchicine to treat acute gout attacks in patients with CKD, corticosteroids are often used in this setting.

Intra-articular steroid injections are useful in treating acute gout limited to a single joint or bursa.¹² However, one should first make sure that the joint is not infected: septic arthritis should ideally be excluded by arthrocentesis, particularly in immunosuppressed patients¹³ or those with end-stage renal disease, who are predisposed to bacteremia.

Oral, intramuscular, or intravenous steroids can provide complete relief from acute gout, although high doses (eg, prednisone 30–60 mg/day or the equivalent) are often needed. Common errors resulting in inefficacy include using too low a dose or not treating for a sufficient time before tapering or stopping. Groff and colleagues¹⁴ described 13 patients who received oral or intravenous steroids for acute gout. Nine patients received an initial single dose of prednisone ranging from 20 to 50 mg, with tapering over a mean of 10 days. Twelve of the 13 patients had improvement within 48 hours, and the signs and symptoms of acute gout resolved completely within 7 to 10 days.

We often give prednisone 40 mg daily until a day after the acute attack resolves and then taper over another 7 to 10 days. There are no data to guide steroid dosing in an evidence-based way, but we believe too short a course of therapy may result in return of symptoms.

Corticotropin and other agents: Effective but costly

Corticotropin shares the same indications as systemic corticosteroids, being used to treat flares when NSAIDs, intra-articular steroids, and colchicine are contraindicated. However, corticotropin is far more expensive than generic corticosteroids, costing nearly $2,000 for a single 80-IU dose, which may need to be repeated.

Corticotropin is available for subcutaneous or intramuscular injection. A single intramuscular injection of corticotropin gel (H.P. Acthar, 25–80 IU) may terminate an acute gout attack.¹⁵ However, many patients need another injection after 24 to 72 hours, which would require another visit to the physician. This treatment has been touted by some as being more effective than corticosteroid therapy, possibly because of a unique peripheral mechanism of action in addition to stimulating cortisol release.¹⁶

We rarely use corticotropin, in view of its cost as well as concerns about excessive sodium and water retention due to the release of multiple hormones from the adrenal gland. This may be especially deleterious in patients with CKD or congestive heart failure.¹⁷

Parenteral anti-tumor necrosis factor agents or interleukin 1 antagonists can be dramatically effective but are also expensive.^18,19 For example, anakinra (Kineret) 100 mg costs about $73, and multiple daily doses may be necessary.

Under unique conditions in which they can be safely used (eg, patients with CKD, diabetes mellitus, liver disease), they may be cost-effective if they can shorten the stay of a hospitalized patient with acute gout.

PROPHYLACTIC ANTI-INFLAMMATORY THERAPY FOR PATIENTS WITH GOUT

Between attacks, the goal is to prevent new attacks through prophylactic management, which may include anti-inflammatory and hypouricemic therapy along with dietary instruction (such as avoiding excessive beer, liquor, and fructose ingestion).

Colchicine can be used as prophylaxis, with caution and monitoring

Although colchicine is not 100% effective, it markedly reduces the flare rate when started in low doses at the time hypouricemic therapy is initiated.^20,21 (Hypouricemic therapy is discussed below.) We generally try to continue this prophylactic therapy, if the patient tolerates it, for at least 6 months—longer if tophi are still present or if attacks continue to occur.

If renal function is intact, colchicine can be prescribed at a dosage of 0.6 mg orally once or twice daily.²¹ In CKD, since the clearance of colchicine is reduced,⁴ the dosage should be reduced. Patients on colchicine for prophylaxis must be carefully monitored if the glomerular filtration rate is less than 50 mL/minute, or colchicine should be avoided altogether.⁶ Laboratory testing for colchicine levels is not routinely available and may be of limited value in predicting adverse effects; thus, recommendations about dose adjustments in CKD are empiric.

Wallace et al²² recommended a dose of 0.6 mg once daily if the creatinine clearance is 35 to 49 mL/minute and 0.6 mg every 2 to 3 days if it is 10 to 34 mL/minute, but there are no published long-term safety or efficacy data validating these reasonable (based on available information) dosing regimens.

Even with dose adjustment, caution is needed. Low-dose daily colchicine may be associated with reversible neuromyopathy and bone marrow suppression.^7,23 Patients with neuromyopathy may complain of myalgias, proximal muscle weakness, and numbness and may have areflexia and decreased sensation. Laboratory findings include elevated creatine kinase and aminotransferase levels. We regularly check for leukopenia or elevated creatine kinase and aspartate aminotransferase levels in patients with CKD who are receiving colchicine in any dose.

Prolonged colchicine therapy should probably be avoided in patients on hemodialysis, as this drug is not removed by dialysis or by exchange transfusion, and the risk of toxicity under these circumstances may be high.²² When there is no viable alternative and the drug is given, patients should be closely monitored for signs of toxicity.

Concurrent (even short-term) treatment with most macrolide antibiotics, particularly clarithromycin (Biaxin), most statin drugs, ketoconazole (Nizoral), cyclosporine, and likely other drugs predisposes to colchicine toxicity by altering its distribution and elimination, and can in rare cases cause morbidity or death.^24–26

NSAIDs are not optimal as prophylaxis in patients with chronic kidney disease

Little information has been published about using NSAIDs chronically to prevent flares, but they are not the optimal drugs to use in patients with CKD, as discussed above. In patients with end-stage renal disease, there are also concerns about NSAID-induced gastric and intestinal bleeding.

Low-dose steroids may not be effective as prophylaxis

Lower doses of steroids may not be effective as prophylaxis against gout flares, consistent with the common observation that gout flares still occur in organ transplant recipients who are taking maintenance doses of prednisone.¹³

 

PREVENTING FLARES BY LOWERING SERUM URATE LEVELS

If tophi are present, if radiography shows evidence of damage, if attacks are frequent or disabling, or if there are relative contraindications to the drugs that would be needed to treat acute attacks, then hypouricemic therapy should be strongly considered to reduce the burden of urate in the body, resorb tophi, and ultimately reduce the frequency of gout flares.²⁰

Although intermittent therapy for attacks or prolonged prophylactic use of colchicine may prevent recurrent episodes of gouty arthritis and may be reasonable for many patients, this approach does not prevent continued urate deposition, with the potential development of bony erosions, tophaceous deposits, and chronic arthritis.

The definitive therapy for gouty arthritis is to deplete the periarticular deposits of urate by maintaining a low serum urate level. Urate-lowering therapy, when indicated, is almost always lifelong.

Four strategies for lowering serum urate

The serum urate concentration can be lowered in four ways:

• Increasing renal uric acid excretion
• Altering the diet
• Decreasing urate synthesis
• Converting urate to a more soluble metabolite.

Increasing uric acid excretion is rarely effective if renal function is impaired

Probenecid, sulfinpyrazone (Anturane), and losartan (Cozaar) modestly increase uric acid secretion and reduce serum urate levels, but they are rarely effective if the creatinine clearance rate is less than 60 mL/minute, and they require significant fluid intake for maximal efficacy.

Uricosuric drugs probably should be avoided in patients who excrete more than 1,000 mg of uric acid per day on a normal diet, since urinary uric acid stones may form. In practice, however, patients are given losartan to treat hypertension without attention to uric acid excretion.

More-potent urocosuric drugs are being tested in clinical trials.

Altering the diet: Traditional advice confirmed

The Health Professionals Follow-up Study^27,28 prospectively examined the relation between diet and gout over 12 years in 47,150 men. The study confirmed some long-standing beliefs, such as that consuming meat, seafood, beer, and liquor increases the risk. Other risk factors were consumption of sugar-sweetened soft drinks and fructose, adiposity, weight gain, hypertension, and diuretic use. On the other hand, protein, wine, and purine-rich vegetables were not associated with gout flares. Low-fat dairy products may have a protective effect. Weight loss was found to be protective.

Low-purine diets are not very palatable, are difficult to adhere to, and are at best only minimally effective, lowering serum urate by 1 to 2 mg/dL. Low-protein diets designed to slow progression of CKD will likely also have only a slight effect on serum urate. Dietary change alone is not likely to dramatically lower serum urate levels.

Metabolizing urate with exogenous uricase

Rasburicase (Elitek) effectively converts urate to allantoin, which is more soluble, but rasburicase is fraught with allergic reactions and cannot be used as chronic therapy.

A pegylated intravenous uricase²⁹ has just been approved by the US Food and Drug Administration (FDA); the retail cost is not yet known. It is dramatically effective in those patients able to use it chronically, but it has not been fully evaluated in patients with CKD.

Decreasing urate synthesis with allopurinol

Allopurinol acts by competitively inhibiting xanthine oxidase, the enzyme that converts hypoxanthine to xanthine and xanthine to uric acid. The drug, a structural analogue of hypoxanthine, is converted by xanthine oxidase to oxypurinol, which is an even more effective inhibitor of xanthine oxidase than allopurinol.

Allopurinol is metabolized in the liver and has a half-life of 1 to 3 hours, but oxypurinol, which is excreted in the urine, has a half-life of 12 to 17 hours. Because of these pharmacokinetic properties, allopurinol can usually be given once daily, and the dosage required to reduce serum urate levels should in theory be lower in patients with lower glomerular filtration rates.

Allopurinol (100- and 300-mg tablets) is approved by the FDA in doses of up to 800 mg/day to treat hyperuricemia in patients with gout,³⁰ while guidelines from the British Society of Rheumatology advocate a maximum dose of 900 mg/day.³¹ These maximum doses are based on the limited amount of data with higher doses, not on documented toxicity.

Practice survey data in the United States indicate that most physicians prescribe no greater than 300 mg daily, although this dosage is likely to reduce the serum urate to less than 6 mg/dL—the goal level—in fewer than 50% of patients.^20,32 Patients with normal renal function occasionally require more than 1,000 mg daily to reduce the serum urate level to less than 6 mg/dL.

How low should the serum urate level be?

Ideally, therapy should keep the serum urate level significantly below 6.7 mg/dL, the approximate saturation point of urate in physiologic fluids.

Lowering the serum urate level from 10 mg/dL to 7 mg/dL may seem encouraging, and the urate level may be in the laboratory “normal” range; however, urate may continue to precipitate in tissues if the concentration is greater than 6.7 mg/dL. A target of 6 mg/dL, used in clinical studies, is far enough below the saturation level to provide some margin for fluctuations in serum levels. A serum level of 6.0 mg/dL has thus been arbitrarily proposed as a reasonable therapeutic target.

The lower the serum urate level achieved during hypouricemic therapy, the faster the reduction in tophaceous deposits. With adequate urate lowering, tophi can be visibly reduced in less than a year of hypouricemic therapy.^33,34

We have as yet no convincing evidence that lowering the serum urate level to less than 6.0 mg/dL is harmful, despite theoretical concerns that urate is a beneficial circulating antioxidant and epidemiologic observations that urate levels have been inversely correlated with progression of Parkinson disease.

 

 

Start low, go slow to avoid a flare

Rapid reduction of the serum urate level in a patient with chronic hyperuricemia and gout is likely to induce an acute flare.²⁰ We have traditionally used a “start low and increase slowly” approach to escalating hypouricemic therapy in hopes of reducing the likelihood of causing a gout flare.

Without anti-inflammatory prophylaxis, acute flares associated with urate-lowering are extremely likely. In a 28-week trial of allopurinol, febuxostat, and placebo by Schumacher et al,³³ during the first 8 weeks, when prophylaxis against gout flare was provided with either colchicine 0.6 mg once daily or naproxen (Naprosyn) 250 mg twice daily, the proportion of patients requiring treatment of gout flares was still 23% to 46%. When prophylaxis was stopped, the flare rate increased further.³³

The more we acutely lower serum urate levels, the more likely flares are to occur. In the study by Schumacher et al,³³ the percentage of patients needing treatment for gout flares during the first 8 weeks of the study, despite gout flare prophylaxis, was related to the percent reduction in serum urate by week 28 of the trial (Table 2).

IS IT NECESSARY TO ADJUST THE ALLOPURINOL DOSE IN CHRONIC KIDNEY DISEASE?

In 1984, Hande et al³⁵ proposed that allopurinol doses be lower in patients with renal insufficiency, with a dosage scale based on creatinine clearance.

Their thoughtful proposal was based on data from six of their own patients and 72 others with severe allopurinol toxicity, mainly allopurinol hypersensitivity syndrome, reported in the literature.

Perez-Ruiz et al³⁶ noted that patients who had experienced adverse effects from allopurinol in their series were likely to have had received “higher” doses of allopurinol, if the dosage was corrected for reduced oxypurinol elimination based on their estimated creatinine clearance.

However, most of these reactions occurred soon after initiating therapy, a temporal pattern more typical of non-dose-dependent allergic reactions. Additionally, allopurinol hypersensitivity has been linked to T-cell-mediated immune reactions to oxypurinol,³⁷ a mechanism not likely linked to drug levels.

Arguments against dose adjustment

Despite the compelling information that allopurinol reactions are more common in CKD, adjusting the dosage of allopurinol has not been clearly shown to reduce the frequency of these reactions.

In a small retrospective analysis, Vázquez-Mellado et al³⁸ reported that adjusting the allopurinol dosage according to creatinine clearance did not decrease the incidence of allopurinol hypersensitivity.

In a study in 250 patients, Dalbeth et al³⁹ showed that the overall incidence of hypersensitivity reaction was 1.6%, and the incidence of allergic reactions did not decrease when allopurinol was given according to the dosing guidelines proposed by Hande et al.³⁵ However, it is worth noting that, of the patients who received the recommended lower doses, only 19% achieved the target serum urate level of 6 mg/dL.³⁹

Silverberg et al⁴⁰ found that of 15 patients who developed hypersensitivity reactions to allopurinol, 10 had received doses that were low or appropriate according to the guidelines of Hande et al.³⁵

More recently, Stamp et al⁴¹ found that gradually increasing the allopurinol dose above the proposed creatinine clearance-based dose was safe and effective. Thirty-one (89%) of the 35 patients who completed the study achieved the target serum urate level of 6 mg/dL, while only 3 of 45 who started the study developed rashes, which were not serious.

The small number of patients in these studies limits any strong conclusion, but at present there is no interventional study showing that allopurinol dosing adjustment based on glomerular filtration rate is effective or safer than dosing based on the serum urate level.

Our view on allopurinol dosing adjustment

We believe the initial observations of Hande et al³⁵ and the subsequent meticulous data from Perez-Ruiz et al³⁶ suggest a relationship between CKD and the occurrence of severe allopurinol reactions. However, these observations do not prove that dose adjustment will prevent these reactions.

In patients with normal kidney function, the FDA³⁰ and the European League Against Rheumatism (EULAR)⁴² recommend slow upward titration, starting with 100 to 200 mg/day, which we agree should decrease the frequency of acute gout flares. The dose is increased by increments of 100 mg/day at intervals of 1 week (FDA recommendation) or 2 to 4 weeks (EULAR recommendation) until the serum urate level is lower than 6 mg/dL.

We believe the optimal approach to allopurinol dosing in patients with CKD remains uncertain. We generally escalate the dose slowly, with ongoing frequent laboratory and clinical monitoring, and we do not limit the maximal dose as suggested by Hande et al.³⁵

An alternative strategy is to use the newer, far more expensive xanthine oxidase inhibitor febuxostat in patients with CKD, since it is not excreted by the kidney. We usually first try escalating doses of allopurinol.

 

FEBUXOSTAT, AN ALTERNATIVE TO ALLOPURINOL

Febuxostat is an oral nonpurine inhibitor of xanthine oxidase.⁴³ Approved by the FDA in 2009, it is available in 40- and 80-mg tablets.

Unlike allopurinol, febuxostat is metabolized primarily by hepatic glucuronide formation and oxidation and then excreted in stool and urine,⁴⁴ making it in theory an attractive agent in patients with renal insufficiency, bypassing the controversial dose-adjustment issue with allopurinol.

In the Febuxostat Versus Allopurinol Controlled Trial (FACT),²⁰ a 52-week randomized, double-blind study in hyperuricemic patients with gout, serum urate levels were reduced to less than 6.0 mg/dL in over 50% of patients receiving febuxostat 80 mg or 120 mg once daily, while only 21% of patients receiving 300 mg of allopurinol achieved this goal. This does not imply that allopurinol at higher doses, as should be used in clinical practice,⁴⁵ would not be equally effective. Patients with CKD were not included in this trial.

In the study by Schumacher et al,³³ febuxostat 80, 120, or 240 mg once daily reduced serum urate. A small subset (35 patients) had mild to moderate renal insufficiency (serum creatinine 1.5–2 mg/dL).³³ The number of patients with renal insufficiency who achieved the primary end point of a serum urate level lower than 6 mg/dL was 4 (44%) of 9 in the febuxostat 80-mg group, 5 (46%) of 11 in the 120-mg group, and 3 (60%) of 5 in the 240-mg group, while none of the 10 patients in the dose-adjusted allopurinol group achieved the primary end point (P < .05). Of note, 41% of the patients with normal renal function who received allopurinol achieved the primary end point.³³ As proposed above, if the allopurinol dose had been slowly increased in the patients with renal insufficiency, it might have been equally effective.

Febuxostat has not been thoroughly evaluated in patients with severe CKD or in patients on hemodialysis.

A presumed niche indication of febuxostat is in patients allergic to allopurinol, since the drugs are not similar in chemical structure. However, at present, experience with this use is limited. Allopurinol-allergic patients were excluded from the clinical trials; thus, if there is any allergic overlap, it would not likely have been recognized in those studies. The FDA has received reports of patients who were allergic to allopurinol also having reactions to febuxostat, and it is currently evaluating these reports (personal communication).

Concern was raised over cardiovascular adverse events in patients treated with febuxostat during clinical trials. In the FACT trial, two patients died of cardiac causes.²⁰ In the study by Schumacher et al,³³ 11 of 670 patients experienced cardiac adverse events in the febuxostat group vs 3 of 268 in the allopurinol group. Events included atrial fibrillation, chest pain, coronary artery disease, and myocardial infarction. However, this difference was not statistically significant.

Febuxostat costs much more than allopurinol. Currently, patients pay $153.88 for 1 month of febuxostat 40 or 80 mg from Cleveland Clinic pharmacy; 1 month of allopurinol costs $17.45 (300 mg) or $14.00 (100 mg). We believe febuxostat should be reserved for patients with documented intolerance to allopurinol in effective doses.

Monitoring serum urate levels is important in all patients on hypouricemic therapy so that dosage adjustments can be made until the target serum urate concentration is reached. In patients failing to meet target serum urate levels, patient adherence with the prescribed dosing should be specifically addressed because as many as 50% of patients do not adhere to their prescribed regimen.

DOES URATE-LOWERING THERAPY HAVE BENEFITS BEYOND GOUT?

Despite experimental animal data and a strong epidemiologic association between hyperuri-cemia and hypertension,⁴⁶ metabolic syndrome, and rates of cardiovascular and all-cause mortality,⁴⁷ the evidence from interventional trials so far does not support the routine use of hypo-uricemic therapy to prevent these outcomes.

Similarly, hyperuricemia has long been associated with renal disease, and there has been debate as to whether hyperuricemia is a result of kidney dysfunction or a contributing factor.^46,48–51 A few studies have documented improvement of renal function after initiation of hypouricemic therapy.⁵² However, treating asymptomatic hyperuricemia to preserve kidney function remains controversial.

A recent study indicates that lowering the serum urate level with allopurinol can lower the blood pressure in hyperuricemic adolescents who have newly diagnosed primary hypertension.⁵³ This does not indicate, however, that initiating hypouricemic therapy in patients with preexisting, long-standing hypertension will be successful.

RECOMMENDED FOR OUR PATIENT

As for our diabetic patient with an acute gout flare and creatinine clearance rate of 45 mL/minute, we would recommend:

• Aspirating the knee, sending the fluid for bacterial culture, and then treating it with a local glucocorticoid injection
• Starting colchicine 0.6 mg every day, with frequent monitoring for signs of toxicity (muscle pain, weakness, leukopenia, and elevations of creatine kinase and aspartate aminotransferase)
• Increasing his allopurinol dose by 100 mg every 2 to 4 weeks until the target serum urate level of less than 6.0 mg/dL is reached
• If he cannot tolerate allopurinol or if the target serum urate level is not achieved despite adequate doses of allopurinol (about 800 mg), we would switch to febuxostat 40 mg and increase the dose as needed to achieve the desired urate level.