Alicia Ault

A set of process measures established by the Joint Commission on Accreditation of Healthcare Organizations has helped hospitals to improve performance, according to a study of the first 2 years of implementation.

The study, by Scott C. Williams, Psy.D., and his colleagues at the commission, found that improvements were made in 15 of 18 standardized measures and that there was no deterioration of quality in any of those areas (N. Engl. J. Med. 2005;353:255–64).

In 2002, the commission began measuring performance in the 18 measures at 3,377 of 4,644 hospitals accredited by the organization. Nonparticipating hospitals either did not offer the services being measured or had an average daily census of fewer than 10 patients. The facilities could choose to submit data on at least two of the following conditions: acute MI, heart failure, pneumonia, and pregnancy and related conditions.

They did not track the pregnancy measures, as two of the measures applied to rare events, and the third, vaginal birth after cesarean section, is controversial, the authors wrote.

The study covered hospitals that submitted data from the third quarter of 2002 to the second quarter of 2004—a total of 3,087 out of the 3,377 hospitals initially identified as study participants. Of those, 1,473 submitted data on heart attack measures, 1,946 on heart failure, and 1,797 on pneumonia.

One of the measures looked at death in the hospital after acute myocardial infarction, and the other 17 assessed processes of care.

There was no improvement in the death measure, but the authors said most of the improvements in the process measures being assessed would not have had an impact on mortality. And there was no significant improvement in the mean time to thrombolysis for patients with acute MI or in mean time to administration of antibiotics for pneumonia.

Hospitals that started with a lower baseline performance improved more quickly than did those with higher initial ratings, but this result was not necessarily expected, the authors noted. “Such hospitals may be less likely to focus on quality or make an effort to improve performance than their counterparts with a higher level of performance,” they wrote.

For acute MI, researchers looked at measures such as whether aspirin was given within 24 hours of admission and prescribed at discharge, whether an ACE inhibitor was prescribed at discharge for patients with left ventricular systolic dysfunction, and the mean time from arrival to thrombolysis or percutaneous coronary intervention.

For heart failure, hospitals were tracked on whether they had given patients smoking cessation counseling and discharge instructions on medication, diet, weight, and worsening of symptoms, and whether an ACE inhibitor was prescribed at discharge for patients with left ventricular systolic dysfunction.

For pneumonia, the commission monitored whether there was an oxygenation assessment within 24 hours of admission and whether pneumococcal screening, vaccination, or both had been given at discharge, or if blood specimens were cultured before starting an antibiotic.

By the end of the study, more than 90% of MI patients at most hospitals received aspirin at admission. Although only 74% of patients received ACE inhibitors at discharge at the lowest performing hospitals, 83% received them at the highest performing facilities.

The biggest improvement was seen in offering smoking cessation counseling. Rates went from a range of 1%–7% at the lowest performing hospitals at baseline to a range of 57%–68% at the study's end. At high-performing facilities, however, rates dropped from an 80%–98% range at baseline to a range of 74%–85% at the end.

Even after improvement, pneumococcal vaccination rates were still low, ranging from 35% in the lowest performing hospitals to 66% at the highest performing facilities.

The authors noted that one potential drawback of the study was its reliance on self-reported data, which could introduce bias. And, they said, the data should not be viewed as static. The picture could change as public reporting of hospital data becomes more prevalent and pay for performance spreads.

A set of process measures established by the Joint Commission on Accreditation of Healthcare Organizations has helped hospitals to improve performance, according to a study of the first 2 years of implementation.

The study, by Scott C. Williams, Psy.D., and his colleagues at the commission, found that improvements were made in 15 of 18 standardized measures and that there was no deterioration of quality in any of those areas (N. Engl. J. Med. 2005;353:255–64).

In 2002, the commission began measuring performance in the 18 measures at 3,377 of 4,644 hospitals accredited by the organization. Nonparticipating hospitals either did not offer the services being measured or had an average daily census of fewer than 10 patients. The facilities could choose to submit data on at least two of the following conditions: acute MI, heart failure, pneumonia, and pregnancy and related conditions.

They did not track the pregnancy measures, as two of the measures applied to rare events, and the third, vaginal birth after cesarean section, is controversial, the authors wrote.

The study covered hospitals that submitted data from the third quarter of 2002 to the second quarter of 2004—a total of 3,087 out of the 3,377 hospitals initially identified as study participants. Of those, 1,473 submitted data on heart attack measures, 1,946 on heart failure, and 1,797 on pneumonia.

One of the measures looked at death in the hospital after acute myocardial infarction, and the other 17 assessed processes of care.

There was no improvement in the death measure, but the authors said most of the improvements in the process measures being assessed would not have had an impact on mortality. And there was no significant improvement in the mean time to thrombolysis for patients with acute MI or in mean time to administration of antibiotics for pneumonia.

Hospitals that started with a lower baseline performance improved more quickly than did those with higher initial ratings, but this result was not necessarily expected, the authors noted. “Such hospitals may be less likely to focus on quality or make an effort to improve performance than their counterparts with a higher level of performance,” they wrote.

For acute MI, researchers looked at measures such as whether aspirin was given within 24 hours of admission and prescribed at discharge, whether an ACE inhibitor was prescribed at discharge for patients with left ventricular systolic dysfunction, and the mean time from arrival to thrombolysis or percutaneous coronary intervention.

For heart failure, hospitals were tracked on whether they had given patients smoking cessation counseling and discharge instructions on medication, diet, weight, and worsening of symptoms, and whether an ACE inhibitor was prescribed at discharge for patients with left ventricular systolic dysfunction.

For pneumonia, the commission monitored whether there was an oxygenation assessment within 24 hours of admission and whether pneumococcal screening, vaccination, or both had been given at discharge, or if blood specimens were cultured before starting an antibiotic.

By the end of the study, more than 90% of MI patients at most hospitals received aspirin at admission. Although only 74% of patients received ACE inhibitors at discharge at the lowest performing hospitals, 83% received them at the highest performing facilities.

The biggest improvement was seen in offering smoking cessation counseling. Rates went from a range of 1%–7% at the lowest performing hospitals at baseline to a range of 57%–68% at the study's end. At high-performing facilities, however, rates dropped from an 80%–98% range at baseline to a range of 74%–85% at the end.

Even after improvement, pneumococcal vaccination rates were still low, ranging from 35% in the lowest performing hospitals to 66% at the highest performing facilities.

The authors noted that one potential drawback of the study was its reliance on self-reported data, which could introduce bias. And, they said, the data should not be viewed as static. The picture could change as public reporting of hospital data becomes more prevalent and pay for performance spreads.

A set of process measures established by the Joint Commission on Accreditation of Healthcare Organizations has helped hospitals to improve performance, according to a study of the first 2 years of implementation.

The study, by Scott C. Williams, Psy.D., and his colleagues at the commission, found that improvements were made in 15 of 18 standardized measures and that there was no deterioration of quality in any of those areas (N. Engl. J. Med. 2005;353:255–64).

In 2002, the commission began measuring performance in the 18 measures at 3,377 of 4,644 hospitals accredited by the organization. Nonparticipating hospitals either did not offer the services being measured or had an average daily census of fewer than 10 patients. The facilities could choose to submit data on at least two of the following conditions: acute MI, heart failure, pneumonia, and pregnancy and related conditions.

They did not track the pregnancy measures, as two of the measures applied to rare events, and the third, vaginal birth after cesarean section, is controversial, the authors wrote.

The study covered hospitals that submitted data from the third quarter of 2002 to the second quarter of 2004—a total of 3,087 out of the 3,377 hospitals initially identified as study participants. Of those, 1,473 submitted data on heart attack measures, 1,946 on heart failure, and 1,797 on pneumonia.

One of the measures looked at death in the hospital after acute myocardial infarction, and the other 17 assessed processes of care.

There was no improvement in the death measure, but the authors said most of the improvements in the process measures being assessed would not have had an impact on mortality. And there was no significant improvement in the mean time to thrombolysis for patients with acute MI or in mean time to administration of antibiotics for pneumonia.

Hospitals that started with a lower baseline performance improved more quickly than did those with higher initial ratings, but this result was not necessarily expected, the authors noted. “Such hospitals may be less likely to focus on quality or make an effort to improve performance than their counterparts with a higher level of performance,” they wrote.

For acute MI, researchers looked at measures such as whether aspirin was given within 24 hours of admission and prescribed at discharge, whether an ACE inhibitor was prescribed at discharge for patients with left ventricular systolic dysfunction, and the mean time from arrival to thrombolysis or percutaneous coronary intervention.

For heart failure, hospitals were tracked on whether they had given patients smoking cessation counseling and discharge instructions on medication, diet, weight, and worsening of symptoms, and whether an ACE inhibitor was prescribed at discharge for patients with left ventricular systolic dysfunction.

For pneumonia, the commission monitored whether there was an oxygenation assessment within 24 hours of admission and whether pneumococcal screening, vaccination, or both had been given at discharge, or if blood specimens were cultured before starting an antibiotic.

By the end of the study, more than 90% of MI patients at most hospitals received aspirin at admission. Although only 74% of patients received ACE inhibitors at discharge at the lowest performing hospitals, 83% received them at the highest performing facilities.

The biggest improvement was seen in offering smoking cessation counseling. Rates went from a range of 1%–7% at the lowest performing hospitals at baseline to a range of 57%–68% at the study's end. At high-performing facilities, however, rates dropped from an 80%–98% range at baseline to a range of 74%–85% at the end.

Even after improvement, pneumococcal vaccination rates were still low, ranging from 35% in the lowest performing hospitals to 66% at the highest performing facilities.

The authors noted that one potential drawback of the study was its reliance on self-reported data, which could introduce bias. And, they said, the data should not be viewed as static. The picture could change as public reporting of hospital data becomes more prevalent and pay for performance spreads.

A set of process measures established by the Joint Commission on Accreditation of Healthcare Organizations has helped hospitals to improve performance, according to a study of the first 2 years of implementation.

The study, by Scott C. Williams, Psy.D., and his colleagues at the commission, found that improvements were made in 15 of 18 standardized measures and that there was no deterioration of quality in any of those areas (N. Engl. J. Med. 2005;353:255–64).

In 2002, the commission began measuring performance in the 18 measures at 3,377 of 4,644 hospitals accredited by the organization. Nonparticipating hospitals either did not offer the services being measured or had an average daily census of fewer than 10 patients. The facilities could choose to submit data on at least two of the following conditions: acute MI, heart failure, pneumonia, and pregnancy and related conditions.

They did not track the pregnancy measures, as two of the measures applied to rare events, and the third, vaginal birth after cesarean section, is controversial, the authors wrote.

The study covered hospitals that submitted data from the third quarter of 2002 to the second quarter of 2004—a total of 3,087 out of the 3,377 hospitals initially identified as study participants.

Of those, 1,473 submitted data on heart attack measures, 1,946 on heart failure, and 1,797 on pneumonia.

One of the measures looked at death in the hospital after acute myocardial infarction, and the other 17 assessed processes of care.

There was no improvement in the death measure, but the authors said most of the improvements in the process measures being assessed would not have had an impact on mortality. And there was no significant improvement in the mean time to thrombolysis for patients with acute MI or in mean time to administration of antibiotics for pneumonia.

For acute MI, researchers looked at measures such as whether aspirin was given within 24 hours of admission and prescribed at discharge, if an ACE inhibitor was prescribed at discharge for those with left ventricular systolic dysfunction, and the mean time from arrival to thrombolysis or percutaneous coronary intervention.

For heart failure, the hospitals were tracked on whether they had given patients smoking cessation counseling and discharge instructions on medication, diet, weight, and worsening of symptoms. The hospitals were also tracked on the prescription of an ACE inhibitor at discharge for patients with left ventricular systolic dysfunction.

For pneumonia, the commission monitored whether there was an oxygenation assessment within 24 hours of admission and whether pneumococcal screening, vaccination, or both had been given at discharge, or if blood specimens were cultured before starting an antibiotic.

By the end of the study period, more than 90% of MI patients at most hospitals received aspirin at admission. Although only 74% of patients received ACE inhibitors at discharge at the lowest performing hospitals, 83% received them at the highest performing facilities.

The biggest improvement was seen in offering smoking cessation counseling. Rates went from a range of 1%–7% at the lowest performing hospitals at baseline to a range of 57%–68% at the study's end. At high-performing facilities, however, rates dropped from an 80%–98% range at baseline to a range of 74%–85% at the end.

Even after improvement, pneumococcal vaccination rates were still low, ranging from 35% in the lowest performing hospitals to a high of 66% at the highest performing facilities.

The authors noted that one potential drawback of the study was its reliance on self-reported data, which could introduce bias.

They also said that the data should not be viewed as static. The picture could change as public reporting of hospital data becomes more prevalent and pay for performance spreads.

A set of process measures established by the Joint Commission on Accreditation of Healthcare Organizations has helped hospitals to improve performance, according to a study of the first 2 years of implementation.

The study, by Scott C. Williams, Psy.D., and his colleagues at the commission, found that improvements were made in 15 of 18 standardized measures and that there was no deterioration of quality in any of those areas (N. Engl. J. Med. 2005;353:255–64).

In 2002, the commission began measuring performance in the 18 measures at 3,377 of 4,644 hospitals accredited by the organization. Nonparticipating hospitals either did not offer the services being measured or had an average daily census of fewer than 10 patients. The facilities could choose to submit data on at least two of the following conditions: acute MI, heart failure, pneumonia, and pregnancy and related conditions.

They did not track the pregnancy measures, as two of the measures applied to rare events, and the third, vaginal birth after cesarean section, is controversial, the authors wrote.

The study covered hospitals that submitted data from the third quarter of 2002 to the second quarter of 2004—a total of 3,087 out of the 3,377 hospitals initially identified as study participants.

Of those, 1,473 submitted data on heart attack measures, 1,946 on heart failure, and 1,797 on pneumonia.

One of the measures looked at death in the hospital after acute myocardial infarction, and the other 17 assessed processes of care.

There was no improvement in the death measure, but the authors said most of the improvements in the process measures being assessed would not have had an impact on mortality. And there was no significant improvement in the mean time to thrombolysis for patients with acute MI or in mean time to administration of antibiotics for pneumonia.

For acute MI, researchers looked at measures such as whether aspirin was given within 24 hours of admission and prescribed at discharge, if an ACE inhibitor was prescribed at discharge for those with left ventricular systolic dysfunction, and the mean time from arrival to thrombolysis or percutaneous coronary intervention.

For heart failure, the hospitals were tracked on whether they had given patients smoking cessation counseling and discharge instructions on medication, diet, weight, and worsening of symptoms. The hospitals were also tracked on the prescription of an ACE inhibitor at discharge for patients with left ventricular systolic dysfunction.

For pneumonia, the commission monitored whether there was an oxygenation assessment within 24 hours of admission and whether pneumococcal screening, vaccination, or both had been given at discharge, or if blood specimens were cultured before starting an antibiotic.

By the end of the study period, more than 90% of MI patients at most hospitals received aspirin at admission. Although only 74% of patients received ACE inhibitors at discharge at the lowest performing hospitals, 83% received them at the highest performing facilities.

The biggest improvement was seen in offering smoking cessation counseling. Rates went from a range of 1%–7% at the lowest performing hospitals at baseline to a range of 57%–68% at the study's end. At high-performing facilities, however, rates dropped from an 80%–98% range at baseline to a range of 74%–85% at the end.

Even after improvement, pneumococcal vaccination rates were still low, ranging from 35% in the lowest performing hospitals to a high of 66% at the highest performing facilities.

The authors noted that one potential drawback of the study was its reliance on self-reported data, which could introduce bias.

They also said that the data should not be viewed as static. The picture could change as public reporting of hospital data becomes more prevalent and pay for performance spreads.

A set of process measures established by the Joint Commission on Accreditation of Healthcare Organizations has helped hospitals to improve performance, according to a study of the first 2 years of implementation.

The study, by Scott C. Williams, Psy.D., and his colleagues at the commission, found that improvements were made in 15 of 18 standardized measures and that there was no deterioration of quality in any of those areas (N. Engl. J. Med. 2005;353:255–64).

In 2002, the commission began measuring performance in the 18 measures at 3,377 of 4,644 hospitals accredited by the organization. Nonparticipating hospitals either did not offer the services being measured or had an average daily census of fewer than 10 patients. The facilities could choose to submit data on at least two of the following conditions: acute MI, heart failure, pneumonia, and pregnancy and related conditions.

They did not track the pregnancy measures, as two of the measures applied to rare events, and the third, vaginal birth after cesarean section, is controversial, the authors wrote.

The study covered hospitals that submitted data from the third quarter of 2002 to the second quarter of 2004—a total of 3,087 out of the 3,377 hospitals initially identified as study participants.

Of those, 1,473 submitted data on heart attack measures, 1,946 on heart failure, and 1,797 on pneumonia.

One of the measures looked at death in the hospital after acute myocardial infarction, and the other 17 assessed processes of care.

There was no improvement in the death measure, but the authors said most of the improvements in the process measures being assessed would not have had an impact on mortality. And there was no significant improvement in the mean time to thrombolysis for patients with acute MI or in mean time to administration of antibiotics for pneumonia.

For acute MI, researchers looked at measures such as whether aspirin was given within 24 hours of admission and prescribed at discharge, if an ACE inhibitor was prescribed at discharge for those with left ventricular systolic dysfunction, and the mean time from arrival to thrombolysis or percutaneous coronary intervention.

For heart failure, the hospitals were tracked on whether they had given patients smoking cessation counseling and discharge instructions on medication, diet, weight, and worsening of symptoms. The hospitals were also tracked on the prescription of an ACE inhibitor at discharge for patients with left ventricular systolic dysfunction.

For pneumonia, the commission monitored whether there was an oxygenation assessment within 24 hours of admission and whether pneumococcal screening, vaccination, or both had been given at discharge, or if blood specimens were cultured before starting an antibiotic.

By the end of the study period, more than 90% of MI patients at most hospitals received aspirin at admission. Although only 74% of patients received ACE inhibitors at discharge at the lowest performing hospitals, 83% received them at the highest performing facilities.

The biggest improvement was seen in offering smoking cessation counseling. Rates went from a range of 1%–7% at the lowest performing hospitals at baseline to a range of 57%–68% at the study's end. At high-performing facilities, however, rates dropped from an 80%–98% range at baseline to a range of 74%–85% at the end.

Even after improvement, pneumococcal vaccination rates were still low, ranging from 35% in the lowest performing hospitals to a high of 66% at the highest performing facilities.

The authors noted that one potential drawback of the study was its reliance on self-reported data, which could introduce bias.

They also said that the data should not be viewed as static. The picture could change as public reporting of hospital data becomes more prevalent and pay for performance spreads.

Recent demands for disclosure of data on hospital infection rates have spurred efforts not only to measure and publicize the numbers but also to demonstrate progress in controlling infections. With many insurers, state and federal regulatory agencies, employers, and consumer groups viewing infection control as a proxy for quality and patient safety, the pressure is on to improve performance.

During the past year, 39 states introduced legislation and 6 states passed laws requiring hospitals to disclose nosocomial infections to the state, and—in many instances—also to the public (N. Engl. J. Med. 2005;353:225–7). But most surveillance and measurement efforts have been made behind closed doors. For example, the Centers for Disease Control and Prevention lets hospitals compare infection rates with other hospitals through the National Healthcare Safety Network, but this information is not available to the public.

The Joint Commission on Accreditation of Healthcare Organizations says it does not maintain statistics on hospital infections, although it recently published a study tracking how well hospitals did in giving antibiotics to pneumonia patients, among other quality of care measures (N. Engl. J. Med. 2005;353:255–64). And the Centers for Medicare and Medicaid Services (CMS) sponsored a pilot project in which hospitals that focused on improving infection control were able to decrease the overall infection rate by 27%.

Robert A. Weinstein, M.D., recently said that those efforts are “a reality” and could lead to improved performance (N. Engl. J. Med. 2005;353:225–7). But to allow for meaningful comparisons among facilities and to spur better quality care, the measures should include such assessments as timely administration of perioperative antibiotic prophylaxis, vascular catheter insertion practices, and hand hygiene, said Dr. Weinstein, chairman of infectious diseases at the John H. Stroger Hospital of Cook County (Ill.). Infection control report cards should also track rates of infection in the ICU that are associated with central vascular catheters; reoperation or rehospitalizations for surgical site infections; rates of nosocomial influenza; and infections caused by multidrug-resistant organisms, he added.

Focus Brings Improvement

Many of those suggested measures were used to track performance in a group of hospitals that participated in the National Surgical Infection Prevention Collaborative. The collaborative was sponsored by CMS and managed by Qualis Health, a Medicare Quality Improvement organization that monitors quality for Washington state, Idaho, and Alaska. Of the 56 hospitals that joined the 12-month project, 44 reported enough data to draw conclusions (Am. J. Surg. 2005;190:9–15).

At each hospital, a team identified a limited set of surgical procedures or surgeons and tracked them for at least 30 days post procedure to determine the proportion of patients getting prophylactic drugs within an hour before the incision, the proportion getting appropriate agents, and the proportion who had prophylaxis discontinued within 24 hours. After identifying the procedures or surgeons to be monitored and gauging a baseline rate for each process to be measured, interdisciplinary teams worked on instituting ways to improve processes.

Over the year, 35,543 patients were tracked. The infection rate was 2.3% in the first 3 months (215 infections among 9,435 cases during that time period); it fell to 1.7% by the last 3 reporting months, constituting a 28% reduction.

Lead author E. Patchen Dellinger, M.D., chief of the division of general surgery at the University of Washington's Eastside Specialty Center, said in an interview that the collaborative focused partly on getting hospitals to more closely identify and monitor infections such as having a nurse check on a patient a set number of times post procedure, or conducting telephone follow-up to ask about problems with wound healing.

The hospitals did not receive any financial assistance for participation; for most, it cost the equivalent of a full-time nurse for the year, estimated Dr. Dellinger.

Getting initial progress is the easy part, he said. “The hardest thing is spreading the gains beyond the pilot population and then holding the gains and not backsliding.”

Spreading and Holding the Gains

Evergreen Hospital is one facility that has managed to keep improving, said Stuart Schrader, R.N., director of surgical services for the 244-bed community facility in Kirkland, Wash.

The hospital did not have a grasp on baseline infection rates, although they appeared to be low (about 0.25% in 1999), Mr. Schrader said in an interview. But the rate climbed each year, hitting 0.7% in 2001. After joining the project and learning some new surveillance techniques, the hospital found that its rate was closer to 1.1%.

Since then, the hospital has adopted quality improvement measures, such as using a convective warming blanket on patients preoperatively and requiring the anesthesiologist to shake hands with each patient—the “warm hands” test—to make sure he or she is normothermic during surgery. Patients are kept warm with the same blankets post procedure in order to ensure proper blood flow to the wound area and thus prevent infection. The hospital has also increased the temperature in its eight operating rooms, and purchased jackets and vests with pockets for cold packs to keep the staff and surgeons comfortable, said Mr. Schrader.

To ensure that patients always receive antibiotics an hour before the procedure, the anesthesiologist is required to call the holding room nurse, who administers the antibiotic.

Razors have been mostly replaced with clippers for shaving hair around surgical sites, which reduces nicks that could invite infection, Mr. Schrader said.

Evergreen has cut its infection rate to 0.85%.

Mr. Schrader expects infection control to continue to be an area of focus. “The hard part was putting the energy into making the changes,” he said. “Once the changes were made, it became a normal way of life.”

Recent demands for disclosure of data on hospital infection rates have spurred efforts not only to measure and publicize the numbers but also to demonstrate progress in controlling infections. With many insurers, state and federal regulatory agencies, employers, and consumer groups viewing infection control as a proxy for quality and patient safety, the pressure is on to improve performance.

During the past year, 39 states introduced legislation and 6 states passed laws requiring hospitals to disclose nosocomial infections to the state, and—in many instances—also to the public (N. Engl. J. Med. 2005;353:225–7). But most surveillance and measurement efforts have been made behind closed doors. For example, the Centers for Disease Control and Prevention lets hospitals compare infection rates with other hospitals through the National Healthcare Safety Network, but this information is not available to the public.

The Joint Commission on Accreditation of Healthcare Organizations says it does not maintain statistics on hospital infections, although it recently published a study tracking how well hospitals did in giving antibiotics to pneumonia patients, among other quality of care measures (N. Engl. J. Med. 2005;353:255–64). And the Centers for Medicare and Medicaid Services (CMS) sponsored a pilot project in which hospitals that focused on improving infection control were able to decrease the overall infection rate by 27%.

Robert A. Weinstein, M.D., recently said that those efforts are “a reality” and could lead to improved performance (N. Engl. J. Med. 2005;353:225–7). But to allow for meaningful comparisons among facilities and to spur better quality care, the measures should include such assessments as timely administration of perioperative antibiotic prophylaxis, vascular catheter insertion practices, and hand hygiene, said Dr. Weinstein, chairman of infectious diseases at the John H. Stroger Hospital of Cook County (Ill.). Infection control report cards should also track rates of infection in the ICU that are associated with central vascular catheters; reoperation or rehospitalizations for surgical site infections; rates of nosocomial influenza; and infections caused by multidrug-resistant organisms, he added.

Focus Brings Improvement

Many of those suggested measures were used to track performance in a group of hospitals that participated in the National Surgical Infection Prevention Collaborative. The collaborative was sponsored by CMS and managed by Qualis Health, a Medicare Quality Improvement organization that monitors quality for Washington state, Idaho, and Alaska. Of the 56 hospitals that joined the 12-month project, 44 reported enough data to draw conclusions (Am. J. Surg. 2005;190:9–15).

At each hospital, a team identified a limited set of surgical procedures or surgeons and tracked them for at least 30 days post procedure to determine the proportion of patients getting prophylactic drugs within an hour before the incision, the proportion getting appropriate agents, and the proportion who had prophylaxis discontinued within 24 hours. After identifying the procedures or surgeons to be monitored and gauging a baseline rate for each process to be measured, interdisciplinary teams worked on instituting ways to improve processes.

Over the year, 35,543 patients were tracked. The infection rate was 2.3% in the first 3 months (215 infections among 9,435 cases during that time period); it fell to 1.7% by the last 3 reporting months, constituting a 28% reduction.

Lead author E. Patchen Dellinger, M.D., chief of the division of general surgery at the University of Washington's Eastside Specialty Center, said in an interview that the collaborative focused partly on getting hospitals to more closely identify and monitor infections such as having a nurse check on a patient a set number of times post procedure, or conducting telephone follow-up to ask about problems with wound healing.

The hospitals did not receive any financial assistance for participation; for most, it cost the equivalent of a full-time nurse for the year, estimated Dr. Dellinger.

Getting initial progress is the easy part, he said. “The hardest thing is spreading the gains beyond the pilot population and then holding the gains and not backsliding.”

Spreading and Holding the Gains

Evergreen Hospital is one facility that has managed to keep improving, said Stuart Schrader, R.N., director of surgical services for the 244-bed community facility in Kirkland, Wash.

The hospital did not have a grasp on baseline infection rates, although they appeared to be low (about 0.25% in 1999), Mr. Schrader said in an interview. But the rate climbed each year, hitting 0.7% in 2001. After joining the project and learning some new surveillance techniques, the hospital found that its rate was closer to 1.1%.

Since then, the hospital has adopted quality improvement measures, such as using a convective warming blanket on patients preoperatively and requiring the anesthesiologist to shake hands with each patient—the “warm hands” test—to make sure he or she is normothermic during surgery. Patients are kept warm with the same blankets post procedure in order to ensure proper blood flow to the wound area and thus prevent infection. The hospital has also increased the temperature in its eight operating rooms, and purchased jackets and vests with pockets for cold packs to keep the staff and surgeons comfortable, said Mr. Schrader.

To ensure that patients always receive antibiotics an hour before the procedure, the anesthesiologist is required to call the holding room nurse, who administers the antibiotic.

Razors have been mostly replaced with clippers for shaving hair around surgical sites, which reduces nicks that could invite infection, Mr. Schrader said.

Evergreen has cut its infection rate to 0.85%.

Mr. Schrader expects infection control to continue to be an area of focus. “The hard part was putting the energy into making the changes,” he said. “Once the changes were made, it became a normal way of life.”

Recent demands for disclosure of data on hospital infection rates have spurred efforts not only to measure and publicize the numbers but also to demonstrate progress in controlling infections. With many insurers, state and federal regulatory agencies, employers, and consumer groups viewing infection control as a proxy for quality and patient safety, the pressure is on to improve performance.

During the past year, 39 states introduced legislation and 6 states passed laws requiring hospitals to disclose nosocomial infections to the state, and—in many instances—also to the public (N. Engl. J. Med. 2005;353:225–7). But most surveillance and measurement efforts have been made behind closed doors. For example, the Centers for Disease Control and Prevention lets hospitals compare infection rates with other hospitals through the National Healthcare Safety Network, but this information is not available to the public.

The Joint Commission on Accreditation of Healthcare Organizations says it does not maintain statistics on hospital infections, although it recently published a study tracking how well hospitals did in giving antibiotics to pneumonia patients, among other quality of care measures (N. Engl. J. Med. 2005;353:255–64). And the Centers for Medicare and Medicaid Services (CMS) sponsored a pilot project in which hospitals that focused on improving infection control were able to decrease the overall infection rate by 27%.

Robert A. Weinstein, M.D., recently said that those efforts are “a reality” and could lead to improved performance (N. Engl. J. Med. 2005;353:225–7). But to allow for meaningful comparisons among facilities and to spur better quality care, the measures should include such assessments as timely administration of perioperative antibiotic prophylaxis, vascular catheter insertion practices, and hand hygiene, said Dr. Weinstein, chairman of infectious diseases at the John H. Stroger Hospital of Cook County (Ill.). Infection control report cards should also track rates of infection in the ICU that are associated with central vascular catheters; reoperation or rehospitalizations for surgical site infections; rates of nosocomial influenza; and infections caused by multidrug-resistant organisms, he added.

Focus Brings Improvement

Many of those suggested measures were used to track performance in a group of hospitals that participated in the National Surgical Infection Prevention Collaborative. The collaborative was sponsored by CMS and managed by Qualis Health, a Medicare Quality Improvement organization that monitors quality for Washington state, Idaho, and Alaska. Of the 56 hospitals that joined the 12-month project, 44 reported enough data to draw conclusions (Am. J. Surg. 2005;190:9–15).

At each hospital, a team identified a limited set of surgical procedures or surgeons and tracked them for at least 30 days post procedure to determine the proportion of patients getting prophylactic drugs within an hour before the incision, the proportion getting appropriate agents, and the proportion who had prophylaxis discontinued within 24 hours. After identifying the procedures or surgeons to be monitored and gauging a baseline rate for each process to be measured, interdisciplinary teams worked on instituting ways to improve processes.

Over the year, 35,543 patients were tracked. The infection rate was 2.3% in the first 3 months (215 infections among 9,435 cases during that time period); it fell to 1.7% by the last 3 reporting months, constituting a 28% reduction.

Lead author E. Patchen Dellinger, M.D., chief of the division of general surgery at the University of Washington's Eastside Specialty Center, said in an interview that the collaborative focused partly on getting hospitals to more closely identify and monitor infections such as having a nurse check on a patient a set number of times post procedure, or conducting telephone follow-up to ask about problems with wound healing.

The hospitals did not receive any financial assistance for participation; for most, it cost the equivalent of a full-time nurse for the year, estimated Dr. Dellinger.

Getting initial progress is the easy part, he said. “The hardest thing is spreading the gains beyond the pilot population and then holding the gains and not backsliding.”

Spreading and Holding the Gains

Evergreen Hospital is one facility that has managed to keep improving, said Stuart Schrader, R.N., director of surgical services for the 244-bed community facility in Kirkland, Wash.

The hospital did not have a grasp on baseline infection rates, although they appeared to be low (about 0.25% in 1999), Mr. Schrader said in an interview. But the rate climbed each year, hitting 0.7% in 2001. After joining the project and learning some new surveillance techniques, the hospital found that its rate was closer to 1.1%.

Since then, the hospital has adopted quality improvement measures, such as using a convective warming blanket on patients preoperatively and requiring the anesthesiologist to shake hands with each patient—the “warm hands” test—to make sure he or she is normothermic during surgery. Patients are kept warm with the same blankets post procedure in order to ensure proper blood flow to the wound area and thus prevent infection. The hospital has also increased the temperature in its eight operating rooms, and purchased jackets and vests with pockets for cold packs to keep the staff and surgeons comfortable, said Mr. Schrader.

To ensure that patients always receive antibiotics an hour before the procedure, the anesthesiologist is required to call the holding room nurse, who administers the antibiotic.

Razors have been mostly replaced with clippers for shaving hair around surgical sites, which reduces nicks that could invite infection, Mr. Schrader said.

Evergreen has cut its infection rate to 0.85%.

Mr. Schrader expects infection control to continue to be an area of focus. “The hard part was putting the energy into making the changes,” he said. “Once the changes were made, it became a normal way of life.”

NEW ORLEANS – Stimulation of a specific area of the posterior hypothalamus reduced and reversed aggressive behavior, according to a case series of two patients presented at the annual meeting of the American Association of Neurological Surgeons.

Giovanni Broggi, M.D., professor and chairman of the department of neurosurgery, Istituto Tumori Besta, Milan, Italy, reported results of stimulation in two patients.

Surgery on the posterior hypothalamus was first attempted and reported in the 1960s, said Dr. Broggi. But it was not widely accepted, as there were concerns about inducing irreversible damage. At his institute in Milan, Dr. Broggi and his colleagues have used chronic stimulation of the area to help treat intractable pain in cluster headaches.

Some have observed that patients with these headaches sometimes engage in aggressive behavior, said Dr. Broggi.

Extending the potential use for stimulation, Dr. Broggi and his colleagues implanted stereotactic bilateral electrodes in the medial portion of the triangle of Sano in two patients who had been institutionalized for psychosis with aggressive behavior. Stimulation was begun one day after surgery to implant the electrodes.

One patient, aged 34 years, had intractable epilepsy in addition to aggressive behavior and had been sedated for 4 years.

After surgery and stimulation, the patient had a 50% decrease in seizures and went home to live with his family.

The other patient's aggressive behavior also resolved, said Dr. Broggi.

At 1 year after implantation, both patients continued to do well. There were no acute or long-term side effects, he said.

Dr. Broggi said he would continue to investigate deep-brain stimulation to control aggression, noting that it had been proved as a reversible, safe surgical method in Parkinson's disease.

In discussing the paper at the meeting, Aviva Abosch, M.D., of Emory University, Atlanta, said that stimulation could be an option for truly medically refractory patients. “The changes described are dramatic,” but, she added, the report was lacking many details, such as whether the device was ever turned off, and if so, if the aggression returned. Dr. Broggi also did not report on any monitoring or testing that might describe any side effects, she said.

Use of deep-brain stimulation to control aggression raises many ethical questions because it could be considered psychosurgery, said Dr. Abosch.

NEW ORLEANS – Stimulation of a specific area of the posterior hypothalamus reduced and reversed aggressive behavior, according to a case series of two patients presented at the annual meeting of the American Association of Neurological Surgeons.

Giovanni Broggi, M.D., professor and chairman of the department of neurosurgery, Istituto Tumori Besta, Milan, Italy, reported results of stimulation in two patients.

Surgery on the posterior hypothalamus was first attempted and reported in the 1960s, said Dr. Broggi. But it was not widely accepted, as there were concerns about inducing irreversible damage. At his institute in Milan, Dr. Broggi and his colleagues have used chronic stimulation of the area to help treat intractable pain in cluster headaches.

Some have observed that patients with these headaches sometimes engage in aggressive behavior, said Dr. Broggi.

Extending the potential use for stimulation, Dr. Broggi and his colleagues implanted stereotactic bilateral electrodes in the medial portion of the triangle of Sano in two patients who had been institutionalized for psychosis with aggressive behavior. Stimulation was begun one day after surgery to implant the electrodes.

One patient, aged 34 years, had intractable epilepsy in addition to aggressive behavior and had been sedated for 4 years.

After surgery and stimulation, the patient had a 50% decrease in seizures and went home to live with his family.

The other patient's aggressive behavior also resolved, said Dr. Broggi.

At 1 year after implantation, both patients continued to do well. There were no acute or long-term side effects, he said.

Dr. Broggi said he would continue to investigate deep-brain stimulation to control aggression, noting that it had been proved as a reversible, safe surgical method in Parkinson's disease.

In discussing the paper at the meeting, Aviva Abosch, M.D., of Emory University, Atlanta, said that stimulation could be an option for truly medically refractory patients. “The changes described are dramatic,” but, she added, the report was lacking many details, such as whether the device was ever turned off, and if so, if the aggression returned. Dr. Broggi also did not report on any monitoring or testing that might describe any side effects, she said.

Use of deep-brain stimulation to control aggression raises many ethical questions because it could be considered psychosurgery, said Dr. Abosch.

NEW ORLEANS – Stimulation of a specific area of the posterior hypothalamus reduced and reversed aggressive behavior, according to a case series of two patients presented at the annual meeting of the American Association of Neurological Surgeons.

Giovanni Broggi, M.D., professor and chairman of the department of neurosurgery, Istituto Tumori Besta, Milan, Italy, reported results of stimulation in two patients.

Surgery on the posterior hypothalamus was first attempted and reported in the 1960s, said Dr. Broggi. But it was not widely accepted, as there were concerns about inducing irreversible damage. At his institute in Milan, Dr. Broggi and his colleagues have used chronic stimulation of the area to help treat intractable pain in cluster headaches.

Some have observed that patients with these headaches sometimes engage in aggressive behavior, said Dr. Broggi.

Extending the potential use for stimulation, Dr. Broggi and his colleagues implanted stereotactic bilateral electrodes in the medial portion of the triangle of Sano in two patients who had been institutionalized for psychosis with aggressive behavior. Stimulation was begun one day after surgery to implant the electrodes.

One patient, aged 34 years, had intractable epilepsy in addition to aggressive behavior and had been sedated for 4 years.

After surgery and stimulation, the patient had a 50% decrease in seizures and went home to live with his family.

The other patient's aggressive behavior also resolved, said Dr. Broggi.

At 1 year after implantation, both patients continued to do well. There were no acute or long-term side effects, he said.

Dr. Broggi said he would continue to investigate deep-brain stimulation to control aggression, noting that it had been proved as a reversible, safe surgical method in Parkinson's disease.

In discussing the paper at the meeting, Aviva Abosch, M.D., of Emory University, Atlanta, said that stimulation could be an option for truly medically refractory patients. “The changes described are dramatic,” but, she added, the report was lacking many details, such as whether the device was ever turned off, and if so, if the aggression returned. Dr. Broggi also did not report on any monitoring or testing that might describe any side effects, she said.

Use of deep-brain stimulation to control aggression raises many ethical questions because it could be considered psychosurgery, said Dr. Abosch.

Celecoxib, the only cyclooxygenase-2 inhibitor left on the U.S. market, has won an additional approval from the Food and Drug Administration–but, as expected, the drug also received a black box warning on the increased risk of cardiovascular events.

The Pfizer Inc. drug was approved for ankylosing spondylitis (AS), a connective tissue disorder causing inflammation of the spine and large joints that affects about 400,000 Americans, primarily between the ages of 20 and 40 years.

The new warning, a result of an FDA advisory committee's recommendations in February, says that celecoxib (Celebrex) may increase the risk of “thrombotic events, myocardial infarction, and stroke, which can be fatal.” The risk may increase with duration of use, according to the warning. The drug is also contraindicated for treating post-coronary artery bypass graft surgery pain.

The black box warning also highlights an increased risk of “serious gastrointestinal adverse events, including bleeding, ulceration, and perforation of the stomach or intestines, which can be fatal.”

Rheumatologists said that despite the warnings, they would use celecoxib, when appropriate, in treating AS. According to the drug's label, in two placebo- and active-controlled trials, celecoxib was statistically superior to placebo (at doses of 100 mg twice daily, 200 mg once daily, and 400 mg once daily) in global pain, global disease, and functional impairment.

Theodore Fields, M.D., of Weill Medical College of Cornell University, noted that it's preferable to start patients on analgesics, but that if there's swelling, anti-inflammatories often have to be used. There are a number of “patients with ankylosing spondylitis where it makes sense, in spite of the known risks, to give them NSAIDs or COX-2s,” said Dr. Fields, who is also clinical director of the Early Arthritis Center at the Hospital for Special Surgery in New York.

The struggle sometimes comes in deciding between celecoxib or an NSAID such as naproxen, said physicians.

John Reveille, M.D., professor of medicine and director of rheumatology at the University of Texas Health Science Center in Houston, said he'd use celecoxib in patients who have a gastrointestinal intolerance to other NSAIDs.

Dr. Fields prefers to use celecoxib in patients who have a higher than average risk of GI complications, but lower than average cardiovascular risk.

Theoretically, because AS patients tend to be younger, the risk of cardiovascular complications is lower, Dr. Fields said in an interview. But, there's also some evidence that patients with inflammatory conditions, such as lupus and rheumatoid arthritis, are at higher risk for atherosclerosis, he said, adding that although the same has not been proven for AS, it is hypothetically possible.

Both Dr. Fields and Dr. Reveille said they'd avoid using celecoxib in patients who have cardiovascular disease or who may be at higher risk–those who are older, male, diabetic, or hypertensive. Neither rheumatologist is a paid consultant for any drug makers.

The new Celebrex label recommends that it be prescribed at the lowest effective dose for the shortest duration. For AS, the recommended dose is 200 mg daily; if there is no response after 6 weeks, the dose should be titrated to 400 mg daily, according to Pfizer. If there is still no response after 6 weeks on that dosage, other treatment options should be considered.

Dr. Fields said he starts patients on 200 mg daily, which he says “is a dose that can be anti-inflammatory.” Dr. Reveille begins patients on a dosage of 100 mg daily and moves up to 200 mg if there is no response.

Both say that patients have become more nervous about using celecoxib than have physicians.

That does not mean rheumatologists have no concerns. “There's no question that I'm watching patients more closely, talking to them about it, and asking if anything new has evolved in their history,” Dr. Fields said, adding that he regularly monitors patients for cardiovascular signs and always considers whether it's possible to shorten therapy or make it intermittent.

Dr. Reveille said he monitors patients on any NSAID or COX-2, including running liver function and complete blood counts at least twice a year. “I don't give a patient an NSAID and see them back a year later.”

AS is the sixth approved indication for celecoxib in the United States, but sales have dropped steeply from a year ago. In the first half of 2005, worldwide celecoxib sales totaled $813 million, a decline of 46% from the previous year.

Celecoxib, the only cyclooxygenase-2 inhibitor left on the U.S. market, has won an additional approval from the Food and Drug Administration–but, as expected, the drug also received a black box warning on the increased risk of cardiovascular events.

The Pfizer Inc. drug was approved for ankylosing spondylitis (AS), a connective tissue disorder causing inflammation of the spine and large joints that affects about 400,000 Americans, primarily between the ages of 20 and 40 years.

The new warning, a result of an FDA advisory committee's recommendations in February, says that celecoxib (Celebrex) may increase the risk of “thrombotic events, myocardial infarction, and stroke, which can be fatal.” The risk may increase with duration of use, according to the warning. The drug is also contraindicated for treating post-coronary artery bypass graft surgery pain.

The black box warning also highlights an increased risk of “serious gastrointestinal adverse events, including bleeding, ulceration, and perforation of the stomach or intestines, which can be fatal.”

Rheumatologists said that despite the warnings, they would use celecoxib, when appropriate, in treating AS. According to the drug's label, in two placebo- and active-controlled trials, celecoxib was statistically superior to placebo (at doses of 100 mg twice daily, 200 mg once daily, and 400 mg once daily) in global pain, global disease, and functional impairment.

Theodore Fields, M.D., of Weill Medical College of Cornell University, noted that it's preferable to start patients on analgesics, but that if there's swelling, anti-inflammatories often have to be used. There are a number of “patients with ankylosing spondylitis where it makes sense, in spite of the known risks, to give them NSAIDs or COX-2s,” said Dr. Fields, who is also clinical director of the Early Arthritis Center at the Hospital for Special Surgery in New York.

The struggle sometimes comes in deciding between celecoxib or an NSAID such as naproxen, said physicians.

John Reveille, M.D., professor of medicine and director of rheumatology at the University of Texas Health Science Center in Houston, said he'd use celecoxib in patients who have a gastrointestinal intolerance to other NSAIDs.

Dr. Fields prefers to use celecoxib in patients who have a higher than average risk of GI complications, but lower than average cardiovascular risk.

Theoretically, because AS patients tend to be younger, the risk of cardiovascular complications is lower, Dr. Fields said in an interview. But, there's also some evidence that patients with inflammatory conditions, such as lupus and rheumatoid arthritis, are at higher risk for atherosclerosis, he said, adding that although the same has not been proven for AS, it is hypothetically possible.

Both Dr. Fields and Dr. Reveille said they'd avoid using celecoxib in patients who have cardiovascular disease or who may be at higher risk–those who are older, male, diabetic, or hypertensive. Neither rheumatologist is a paid consultant for any drug makers.

The new Celebrex label recommends that it be prescribed at the lowest effective dose for the shortest duration. For AS, the recommended dose is 200 mg daily; if there is no response after 6 weeks, the dose should be titrated to 400 mg daily, according to Pfizer. If there is still no response after 6 weeks on that dosage, other treatment options should be considered.

Dr. Fields said he starts patients on 200 mg daily, which he says “is a dose that can be anti-inflammatory.” Dr. Reveille begins patients on a dosage of 100 mg daily and moves up to 200 mg if there is no response.

Both say that patients have become more nervous about using celecoxib than have physicians.

That does not mean rheumatologists have no concerns. “There's no question that I'm watching patients more closely, talking to them about it, and asking if anything new has evolved in their history,” Dr. Fields said, adding that he regularly monitors patients for cardiovascular signs and always considers whether it's possible to shorten therapy or make it intermittent.

Dr. Reveille said he monitors patients on any NSAID or COX-2, including running liver function and complete blood counts at least twice a year. “I don't give a patient an NSAID and see them back a year later.”

AS is the sixth approved indication for celecoxib in the United States, but sales have dropped steeply from a year ago. In the first half of 2005, worldwide celecoxib sales totaled $813 million, a decline of 46% from the previous year.

Celecoxib, the only cyclooxygenase-2 inhibitor left on the U.S. market, has won an additional approval from the Food and Drug Administration–but, as expected, the drug also received a black box warning on the increased risk of cardiovascular events.

The Pfizer Inc. drug was approved for ankylosing spondylitis (AS), a connective tissue disorder causing inflammation of the spine and large joints that affects about 400,000 Americans, primarily between the ages of 20 and 40 years.

The new warning, a result of an FDA advisory committee's recommendations in February, says that celecoxib (Celebrex) may increase the risk of “thrombotic events, myocardial infarction, and stroke, which can be fatal.” The risk may increase with duration of use, according to the warning. The drug is also contraindicated for treating post-coronary artery bypass graft surgery pain.

The black box warning also highlights an increased risk of “serious gastrointestinal adverse events, including bleeding, ulceration, and perforation of the stomach or intestines, which can be fatal.”

Rheumatologists said that despite the warnings, they would use celecoxib, when appropriate, in treating AS. According to the drug's label, in two placebo- and active-controlled trials, celecoxib was statistically superior to placebo (at doses of 100 mg twice daily, 200 mg once daily, and 400 mg once daily) in global pain, global disease, and functional impairment.

Theodore Fields, M.D., of Weill Medical College of Cornell University, noted that it's preferable to start patients on analgesics, but that if there's swelling, anti-inflammatories often have to be used. There are a number of “patients with ankylosing spondylitis where it makes sense, in spite of the known risks, to give them NSAIDs or COX-2s,” said Dr. Fields, who is also clinical director of the Early Arthritis Center at the Hospital for Special Surgery in New York.

The struggle sometimes comes in deciding between celecoxib or an NSAID such as naproxen, said physicians.

John Reveille, M.D., professor of medicine and director of rheumatology at the University of Texas Health Science Center in Houston, said he'd use celecoxib in patients who have a gastrointestinal intolerance to other NSAIDs.

Dr. Fields prefers to use celecoxib in patients who have a higher than average risk of GI complications, but lower than average cardiovascular risk.

Theoretically, because AS patients tend to be younger, the risk of cardiovascular complications is lower, Dr. Fields said in an interview. But, there's also some evidence that patients with inflammatory conditions, such as lupus and rheumatoid arthritis, are at higher risk for atherosclerosis, he said, adding that although the same has not been proven for AS, it is hypothetically possible.

Both Dr. Fields and Dr. Reveille said they'd avoid using celecoxib in patients who have cardiovascular disease or who may be at higher risk–those who are older, male, diabetic, or hypertensive. Neither rheumatologist is a paid consultant for any drug makers.

The new Celebrex label recommends that it be prescribed at the lowest effective dose for the shortest duration. For AS, the recommended dose is 200 mg daily; if there is no response after 6 weeks, the dose should be titrated to 400 mg daily, according to Pfizer. If there is still no response after 6 weeks on that dosage, other treatment options should be considered.

Dr. Fields said he starts patients on 200 mg daily, which he says “is a dose that can be anti-inflammatory.” Dr. Reveille begins patients on a dosage of 100 mg daily and moves up to 200 mg if there is no response.

Both say that patients have become more nervous about using celecoxib than have physicians.

That does not mean rheumatologists have no concerns. “There's no question that I'm watching patients more closely, talking to them about it, and asking if anything new has evolved in their history,” Dr. Fields said, adding that he regularly monitors patients for cardiovascular signs and always considers whether it's possible to shorten therapy or make it intermittent.

Dr. Reveille said he monitors patients on any NSAID or COX-2, including running liver function and complete blood counts at least twice a year. “I don't give a patient an NSAID and see them back a year later.”

AS is the sixth approved indication for celecoxib in the United States, but sales have dropped steeply from a year ago. In the first half of 2005, worldwide celecoxib sales totaled $813 million, a decline of 46% from the previous year.

In the wake of what appears to be underreporting of nosocomial infections, a Pennsylvania state agency is poised to use its leverage to force hospitals to divulge more accurate and full information.

This summer, the Pennsylvania Health Care Cost Containment Council issued a report to the public showing that there were 11,668 confirmed hospital-acquired infections among 1.5 million discharges from 173 general acute care hospitals in 2004. The infections were associated with 1,793 deaths, 205,000 extra hospital days, and $2 billion in charges.

Pennsylvania is the first state to make such numbers public.

As firm as the numbers may sound, however, the data are rife with omissions and invalid reports, and several hospitals reported no infections at all, said Marc Volavka, executive director of the council, which is a state-funded, independent agency. What was reported may only be the tip of the iceberg, Mr. Volavka said in an interview.

Since January 2004, hospitals in Pennsylvania have been required to submit quarterly data to the council on surgical site infections in orthopedic surgery, neurosurgery, and surgery related to the circulatory system; device-related infections, including urinary tract infections from Foley catheters; ventilator-associated pneumonia; and bloodstream infections from central lines. Starting in 2006, hospitals will have to submit data on all hospital-acquired infections.

The council has the authority to release data on a hospital-by-hospital basis, but so far, it has not. "As we started down this uncharted path, we said from the beginning that it would take time for Pennsylvania's hospitals to become accustomed to this reporting process," said Mr. Volavka in a statement, adding that many facilities had been given "lenient time frames and extensions."

"Unless reporting gets more accurate and more complete, the council will start to use [its] authority" to name names, he said. Doing that now could potentially harm hospitals that are complying with the reporting guidelines, according to Mr. Volavka. For instance, 17% of hospitals in the state (29 facilities), which account for only 25% of the admissions, reported more than half of the total infections. Sixteen hospitals reported no infections at all.

There was also a glaring discrepancy between the number of infections reported to the state (11,668) and the number billed to payers: 115,631. The council said the higher number indicates there may have been more hospital-acquired infections than were reported to the state.

On the basis of the $29,000 that insurers actually paid for each infected patient, compared with $8,300 for an inpatient without an infection, the council estimated that third-party payments for the 11,668 infections amounted to nearly $350 million.

In a statement, Carolyn Scanlan, president and CEO of the Hospital & Healthsystem Association of Pennsylvania, said the council's data on third-party payments was somewhat misleading because they "made no distinction between infection-related costs and those costs associated with the patient's entire time in the hospital."

Ms. Scanlan also defended hospitals' response to the council's requirements, noting that the number of infections reported had increased each quarter. She said the increase in reporting indicates that hospitals are becoming more familiar with reporting requirements.

The Centers for Disease Control and Prevention has said that 5%–10% of hospitalized patients will acquire an infection, Mr. Volavka pointed out. So, while the numbers reported in Pennsylvania jibe with that CDC estimate, "it frankly ought to be a wake-up call to health care professionals and to purchasers and consumers who ultimately are paying the bill," he said.

If the Pennsylvania data were extrapolated nationally, the figures indicate that at least 100 people per day die from nosocomial infections, at a cost of about $50 billion a year.

In the wake of what appears to be underreporting of nosocomial infections, a Pennsylvania state agency is poised to use its leverage to force hospitals to divulge more accurate and full information.

This summer, the Pennsylvania Health Care Cost Containment Council issued a report to the public showing that there were 11,668 confirmed hospital-acquired infections among 1.5 million discharges from 173 general acute care hospitals in 2004. The infections were associated with 1,793 deaths, 205,000 extra hospital days, and $2 billion in charges.

Pennsylvania is the first state to make such numbers public.

As firm as the numbers may sound, however, the data are rife with omissions and invalid reports, and several hospitals reported no infections at all, said Marc Volavka, executive director of the council, which is a state-funded, independent agency. What was reported may only be the tip of the iceberg, Mr. Volavka said in an interview.

Since January 2004, hospitals in Pennsylvania have been required to submit quarterly data to the council on surgical site infections in orthopedic surgery, neurosurgery, and surgery related to the circulatory system; device-related infections, including urinary tract infections from Foley catheters; ventilator-associated pneumonia; and bloodstream infections from central lines. Starting in 2006, hospitals will have to submit data on all hospital-acquired infections.

The council has the authority to release data on a hospital-by-hospital basis, but so far, it has not. "As we started down this uncharted path, we said from the beginning that it would take time for Pennsylvania's hospitals to become accustomed to this reporting process," said Mr. Volavka in a statement, adding that many facilities had been given "lenient time frames and extensions."

"Unless reporting gets more accurate and more complete, the council will start to use [its] authority" to name names, he said. Doing that now could potentially harm hospitals that are complying with the reporting guidelines, according to Mr. Volavka. For instance, 17% of hospitals in the state (29 facilities), which account for only 25% of the admissions, reported more than half of the total infections. Sixteen hospitals reported no infections at all.

There was also a glaring discrepancy between the number of infections reported to the state (11,668) and the number billed to payers: 115,631. The council said the higher number indicates there may have been more hospital-acquired infections than were reported to the state.

On the basis of the $29,000 that insurers actually paid for each infected patient, compared with $8,300 for an inpatient without an infection, the council estimated that third-party payments for the 11,668 infections amounted to nearly $350 million.

In a statement, Carolyn Scanlan, president and CEO of the Hospital & Healthsystem Association of Pennsylvania, said the council's data on third-party payments was somewhat misleading because they "made no distinction between infection-related costs and those costs associated with the patient's entire time in the hospital."

Ms. Scanlan also defended hospitals' response to the council's requirements, noting that the number of infections reported had increased each quarter. She said the increase in reporting indicates that hospitals are becoming more familiar with reporting requirements.

The Centers for Disease Control and Prevention has said that 5%–10% of hospitalized patients will acquire an infection, Mr. Volavka pointed out. So, while the numbers reported in Pennsylvania jibe with that CDC estimate, "it frankly ought to be a wake-up call to health care professionals and to purchasers and consumers who ultimately are paying the bill," he said.

If the Pennsylvania data were extrapolated nationally, the figures indicate that at least 100 people per day die from nosocomial infections, at a cost of about $50 billion a year.

In the wake of what appears to be underreporting of nosocomial infections, a Pennsylvania state agency is poised to use its leverage to force hospitals to divulge more accurate and full information.

This summer, the Pennsylvania Health Care Cost Containment Council issued a report to the public showing that there were 11,668 confirmed hospital-acquired infections among 1.5 million discharges from 173 general acute care hospitals in 2004. The infections were associated with 1,793 deaths, 205,000 extra hospital days, and $2 billion in charges.

Pennsylvania is the first state to make such numbers public.

As firm as the numbers may sound, however, the data are rife with omissions and invalid reports, and several hospitals reported no infections at all, said Marc Volavka, executive director of the council, which is a state-funded, independent agency. What was reported may only be the tip of the iceberg, Mr. Volavka said in an interview.

Since January 2004, hospitals in Pennsylvania have been required to submit quarterly data to the council on surgical site infections in orthopedic surgery, neurosurgery, and surgery related to the circulatory system; device-related infections, including urinary tract infections from Foley catheters; ventilator-associated pneumonia; and bloodstream infections from central lines. Starting in 2006, hospitals will have to submit data on all hospital-acquired infections.

The council has the authority to release data on a hospital-by-hospital basis, but so far, it has not. "As we started down this uncharted path, we said from the beginning that it would take time for Pennsylvania's hospitals to become accustomed to this reporting process," said Mr. Volavka in a statement, adding that many facilities had been given "lenient time frames and extensions."

"Unless reporting gets more accurate and more complete, the council will start to use [its] authority" to name names, he said. Doing that now could potentially harm hospitals that are complying with the reporting guidelines, according to Mr. Volavka. For instance, 17% of hospitals in the state (29 facilities), which account for only 25% of the admissions, reported more than half of the total infections. Sixteen hospitals reported no infections at all.

There was also a glaring discrepancy between the number of infections reported to the state (11,668) and the number billed to payers: 115,631. The council said the higher number indicates there may have been more hospital-acquired infections than were reported to the state.

On the basis of the $29,000 that insurers actually paid for each infected patient, compared with $8,300 for an inpatient without an infection, the council estimated that third-party payments for the 11,668 infections amounted to nearly $350 million.

In a statement, Carolyn Scanlan, president and CEO of the Hospital & Healthsystem Association of Pennsylvania, said the council's data on third-party payments was somewhat misleading because they "made no distinction between infection-related costs and those costs associated with the patient's entire time in the hospital."

Ms. Scanlan also defended hospitals' response to the council's requirements, noting that the number of infections reported had increased each quarter. She said the increase in reporting indicates that hospitals are becoming more familiar with reporting requirements.

The Centers for Disease Control and Prevention has said that 5%–10% of hospitalized patients will acquire an infection, Mr. Volavka pointed out. So, while the numbers reported in Pennsylvania jibe with that CDC estimate, "it frankly ought to be a wake-up call to health care professionals and to purchasers and consumers who ultimately are paying the bill," he said.

If the Pennsylvania data were extrapolated nationally, the figures indicate that at least 100 people per day die from nosocomial infections, at a cost of about $50 billion a year.

In the wake of what appears to be underreporting of nosocomial infections, a Pennsylvania state agency is poised to use its leverage to force hospitals to divulge more accurate and full information.

This summer, the Pennsylvania Health Care Cost Containment Council issued a report to the public showing that there were 11,668 confirmed hospital-acquired infections among 1.5 million discharges from 173 general acute care hospitals in 2004. The infections were associated with 1,793 deaths, 205,000 extra hospital days, and $2 billion in charges.

Pennsylvania is the first state to make such numbers public.

As firm as the numbers may sound, however, the data are rife with omissions and invalid reports, and several hospitals reported no infections at all, said Marc Volavka, executive director of the council, which is a state-funded, independent agency. What was reported may only be the tip of the iceberg, Mr. Volavka said in an interview.

Since January 2004, hospitals in Pennsylvania have been required to submit quarterly data to the council on surgical site infections in orthopedic surgery, neurosurgery, and surgery related to the circulatory system; device-related infections, including urinary tract infections from Foley catheters; ventilator-associated pneumonia; and bloodstream infections from central lines. Starting in 2006, hospitals will have to submit data on all hospital-acquired infections.

The council has the authority to release data on a hospital-by-hospital basis, but so far, it has not. “As we started down this uncharted path, we said from the beginning that it would take time for Pennsylvania's hospitals to become accustomed to this reporting process,” said Mr. Volavka in a statement, adding that many facilities had been given “lenient time frames and extensions.”

“Unless reporting gets more accurate and more complete, the council will start to use [its] authority” to name names, he said.

Doing that now could potentially harm hospitals that are complying with the reporting guidelines, according to Mr. Volavka. For instance, 17% of hospitals in the state (29 facilities), which account for only 25% of the admissions, reported more than half of the total infections. Sixteen hospitals reported no infections at all.

There was also a glaring discrepancy between the number of infections reported to the state (11,668) and the number billed to payers: 115,631. The council said the higher number indicates there may have been more hospital-acquired infections than were reported to the state.

On the basis of the $29,000 that insurers actually paid for each infected patient, compared with $8,300 for an inpatient without an infection, the council estimated that third-party payments for the 11,668 infections amounted to nearly $350 million.

In a statement, Carolyn Scanlan, president and CEO of the Hospital & Healthsystem Association of Pennsylvania, said the council's data on third-party payments was somewhat misleading because they “made no distinction between infection-related costs and those costs associated with the patient's entire time in the hospital.”

Ms. Scanlan also defended hospitals' response to the council's requirements, noting that the number of infections reported had increased each quarter. She said the increase in reporting indicates that hospitals are becoming more familiar with reporting requirements.

The Centers for Disease Control and Prevention has said that 5%–10% of hospitalized patients will acquire an infection, Mr. Volavka pointed out. So, while the numbers reported in Pennsylvania jibe with that CDC estimate, “it frankly ought to be a wake-up call to health care professionals and to purchasers and consumers who ultimately are paying the bill,” he said.

If the Pennsylvania data were extrapolated nationally, the figures indicate at least 100 people per day die from nosocomial infections, at a cost of about $50 billion a year.

In the wake of what appears to be underreporting of nosocomial infections, a Pennsylvania state agency is poised to use its leverage to force hospitals to divulge more accurate and full information.

This summer, the Pennsylvania Health Care Cost Containment Council issued a report to the public showing that there were 11,668 confirmed hospital-acquired infections among 1.5 million discharges from 173 general acute care hospitals in 2004. The infections were associated with 1,793 deaths, 205,000 extra hospital days, and $2 billion in charges.

Pennsylvania is the first state to make such numbers public.

As firm as the numbers may sound, however, the data are rife with omissions and invalid reports, and several hospitals reported no infections at all, said Marc Volavka, executive director of the council, which is a state-funded, independent agency. What was reported may only be the tip of the iceberg, Mr. Volavka said in an interview.

Since January 2004, hospitals in Pennsylvania have been required to submit quarterly data to the council on surgical site infections in orthopedic surgery, neurosurgery, and surgery related to the circulatory system; device-related infections, including urinary tract infections from Foley catheters; ventilator-associated pneumonia; and bloodstream infections from central lines. Starting in 2006, hospitals will have to submit data on all hospital-acquired infections.

The council has the authority to release data on a hospital-by-hospital basis, but so far, it has not. “As we started down this uncharted path, we said from the beginning that it would take time for Pennsylvania's hospitals to become accustomed to this reporting process,” said Mr. Volavka in a statement, adding that many facilities had been given “lenient time frames and extensions.”

“Unless reporting gets more accurate and more complete, the council will start to use [its] authority” to name names, he said.

Doing that now could potentially harm hospitals that are complying with the reporting guidelines, according to Mr. Volavka. For instance, 17% of hospitals in the state (29 facilities), which account for only 25% of the admissions, reported more than half of the total infections. Sixteen hospitals reported no infections at all.

There was also a glaring discrepancy between the number of infections reported to the state (11,668) and the number billed to payers: 115,631. The council said the higher number indicates there may have been more hospital-acquired infections than were reported to the state.

On the basis of the $29,000 that insurers actually paid for each infected patient, compared with $8,300 for an inpatient without an infection, the council estimated that third-party payments for the 11,668 infections amounted to nearly $350 million.

In a statement, Carolyn Scanlan, president and CEO of the Hospital & Healthsystem Association of Pennsylvania, said the council's data on third-party payments was somewhat misleading because they “made no distinction between infection-related costs and those costs associated with the patient's entire time in the hospital.”

Ms. Scanlan also defended hospitals' response to the council's requirements, noting that the number of infections reported had increased each quarter. She said the increase in reporting indicates that hospitals are becoming more familiar with reporting requirements.

The Centers for Disease Control and Prevention has said that 5%–10% of hospitalized patients will acquire an infection, Mr. Volavka pointed out. So, while the numbers reported in Pennsylvania jibe with that CDC estimate, “it frankly ought to be a wake-up call to health care professionals and to purchasers and consumers who ultimately are paying the bill,” he said.

If the Pennsylvania data were extrapolated nationally, the figures indicate at least 100 people per day die from nosocomial infections, at a cost of about $50 billion a year.

In the wake of what appears to be underreporting of nosocomial infections, a Pennsylvania state agency is poised to use its leverage to force hospitals to divulge more accurate and full information.

This summer, the Pennsylvania Health Care Cost Containment Council issued a report to the public showing that there were 11,668 confirmed hospital-acquired infections among 1.5 million discharges from 173 general acute care hospitals in 2004. The infections were associated with 1,793 deaths, 205,000 extra hospital days, and $2 billion in charges.

Pennsylvania is the first state to make such numbers public.

As firm as the numbers may sound, however, the data are rife with omissions and invalid reports, and several hospitals reported no infections at all, said Marc Volavka, executive director of the council, which is a state-funded, independent agency. What was reported may only be the tip of the iceberg, Mr. Volavka said in an interview.

Since January 2004, hospitals in Pennsylvania have been required to submit quarterly data to the council on surgical site infections in orthopedic surgery, neurosurgery, and surgery related to the circulatory system; device-related infections, including urinary tract infections from Foley catheters; ventilator-associated pneumonia; and bloodstream infections from central lines. Starting in 2006, hospitals will have to submit data on all hospital-acquired infections.

The council has the authority to release data on a hospital-by-hospital basis, but so far, it has not. “As we started down this uncharted path, we said from the beginning that it would take time for Pennsylvania's hospitals to become accustomed to this reporting process,” said Mr. Volavka in a statement, adding that many facilities had been given “lenient time frames and extensions.”

“Unless reporting gets more accurate and more complete, the council will start to use [its] authority” to name names, he said.

Doing that now could potentially harm hospitals that are complying with the reporting guidelines, according to Mr. Volavka. For instance, 17% of hospitals in the state (29 facilities), which account for only 25% of the admissions, reported more than half of the total infections. Sixteen hospitals reported no infections at all.

There was also a glaring discrepancy between the number of infections reported to the state (11,668) and the number billed to payers: 115,631. The council said the higher number indicates there may have been more hospital-acquired infections than were reported to the state.

On the basis of the $29,000 that insurers actually paid for each infected patient, compared with $8,300 for an inpatient without an infection, the council estimated that third-party payments for the 11,668 infections amounted to nearly $350 million.

In a statement, Carolyn Scanlan, president and CEO of the Hospital & Healthsystem Association of Pennsylvania, said the council's data on third-party payments was somewhat misleading because they “made no distinction between infection-related costs and those costs associated with the patient's entire time in the hospital.”

Ms. Scanlan also defended hospitals' response to the council's requirements, noting that the number of infections reported had increased each quarter. She said the increase in reporting indicates that hospitals are becoming more familiar with reporting requirements.

The Centers for Disease Control and Prevention has said that 5%–10% of hospitalized patients will acquire an infection, Mr. Volavka pointed out. So, while the numbers reported in Pennsylvania jibe with that CDC estimate, “it frankly ought to be a wake-up call to health care professionals and to purchasers and consumers who ultimately are paying the bill,” he said.

If the Pennsylvania data were extrapolated nationally, the figures indicate at least 100 people per day die from nosocomial infections, at a cost of about $50 billion a year.

Celecoxib, the only cyclooxygenase-2 inhibitor left on the U.S. market, has won an additional approval from the Food and Drug Administration—but, as expected, the drug also received a black box warning on the increased risk of cardiovascular events.

The Pfizer Inc. drug was approved for ankylosing spondylitis (AS), a connective tissue disorder causing inflammation of the spine and large joints that affects about 400,000 Americans, primarily between the ages of 20 and 40 years.

The new warning, a result of an FDA advisory committee's recommendations in February, says that celecoxib (Celebrex) may increase the risk of “thrombotic events, myocardial infarction, and stroke, which can be fatal.” The risk may increase with duration of use, according to the warning. The drug is also contraindicated for treating post-coronary artery bypass graft surgery pain.

The black box warning also highlights an increased risk of “serious gastrointestinal adverse events, including bleeding, ulceration, and perforation of the stomach or intestines, which can be fatal.”

According to the drug's label, in two placebo- and active-controlled trials, celecoxib was statistically superior to placebo in global pain, global disease, and functional impairment.

Theodore Fields, M.D., of Weill Medical College of Cornell University, noted that it's preferable to start patients on analgesics, but that if there's swelling, anti-inflammatories often have to be used. There are a number of “patients with ankylosing spondylitis where it makes sense, in spite of the known risks, to give them NSAIDs or COX-2s,” said Dr. Fields, who is also clinical director of the Early Arthritis Center at the Hospital for Special Surgery in New York.

The struggle sometimes comes in deciding between celecoxib or an NSAID such as naproxen, said physicians.

John Reveille, M.D., professor of medicine and director of rheumatology at the University of Texas Health Science Center in Houston, said he'd use celecoxib in patients who have a gastrointestinal intolerance to other NSAIDs.

Dr. Fields prefers to use celecoxib in patients who have a higher than average risk of GI complications, but lower than average cardiovascular risk.

Theoretically, because AS patients tend to be younger, the risk of cardiovascular complications is lower, Dr. Fields said in an interview. But, there's also some evidence that patients with inflammatory conditions, such as lupus and rheumatoid arthritis, are at higher risk for atherosclerosis, he said, adding that although the same has not been proven for AS, it is hypothetically possible.

Both Dr. Fields and Dr. Reveille said they'd avoid using celecoxib in patients who have cardiovascular disease or who may be at higher risk—those who are older, male, diabetic, or hypertensive. Neither physician is a paid consultant for any drug makers.

The new Celebrex label recommends that it be prescribed at the lowest effective dose for the shortest duration. For AS, the recommended dose is 200 mg daily; if there is no response after 6 weeks, the dose should be titrated to 400 mg daily, according to Pfizer. If there is still no response after 6 weeks on that dosage, other treatment options should be considered.

Dr. Fields said he starts patients on 200 mg daily, which he says “is a dose that can be anti-inflammatory.” Dr. Reveille begins patients on a dosage of 100 mg daily and moves up to 200 mg if there is no response.

Both say that patients have become more nervous about celecoxib than physicians.

That does not mean physicians have no concerns. “There's no question that I'm watching patients more closely, talking to them about it, and asking if anything new has evolved in their history,” Dr. Fields said, adding that he regularly monitors patients for cardiovascular signs and always considers whether it's possible to shorten therapy or make it intermittent.

Dr. Reveille said he monitors patients on any NSAID or COX-2, including running liver function and complete blood counts at least twice a year.

AS is the sixth approved indication for celecoxib in the United States, but sales have dropped steeply from a year ago. In the first half of 2005, worldwide celecoxib sales totaled $813 million, a decline of 46% from the previous year.

Celecoxib, the only cyclooxygenase-2 inhibitor left on the U.S. market, has won an additional approval from the Food and Drug Administration—but, as expected, the drug also received a black box warning on the increased risk of cardiovascular events.

The Pfizer Inc. drug was approved for ankylosing spondylitis (AS), a connective tissue disorder causing inflammation of the spine and large joints that affects about 400,000 Americans, primarily between the ages of 20 and 40 years.

The new warning, a result of an FDA advisory committee's recommendations in February, says that celecoxib (Celebrex) may increase the risk of “thrombotic events, myocardial infarction, and stroke, which can be fatal.” The risk may increase with duration of use, according to the warning. The drug is also contraindicated for treating post-coronary artery bypass graft surgery pain.

The black box warning also highlights an increased risk of “serious gastrointestinal adverse events, including bleeding, ulceration, and perforation of the stomach or intestines, which can be fatal.”

According to the drug's label, in two placebo- and active-controlled trials, celecoxib was statistically superior to placebo in global pain, global disease, and functional impairment.

Theodore Fields, M.D., of Weill Medical College of Cornell University, noted that it's preferable to start patients on analgesics, but that if there's swelling, anti-inflammatories often have to be used. There are a number of “patients with ankylosing spondylitis where it makes sense, in spite of the known risks, to give them NSAIDs or COX-2s,” said Dr. Fields, who is also clinical director of the Early Arthritis Center at the Hospital for Special Surgery in New York.

The struggle sometimes comes in deciding between celecoxib or an NSAID such as naproxen, said physicians.

John Reveille, M.D., professor of medicine and director of rheumatology at the University of Texas Health Science Center in Houston, said he'd use celecoxib in patients who have a gastrointestinal intolerance to other NSAIDs.

Dr. Fields prefers to use celecoxib in patients who have a higher than average risk of GI complications, but lower than average cardiovascular risk.

Theoretically, because AS patients tend to be younger, the risk of cardiovascular complications is lower, Dr. Fields said in an interview. But, there's also some evidence that patients with inflammatory conditions, such as lupus and rheumatoid arthritis, are at higher risk for atherosclerosis, he said, adding that although the same has not been proven for AS, it is hypothetically possible.

Both Dr. Fields and Dr. Reveille said they'd avoid using celecoxib in patients who have cardiovascular disease or who may be at higher risk—those who are older, male, diabetic, or hypertensive. Neither physician is a paid consultant for any drug makers.

The new Celebrex label recommends that it be prescribed at the lowest effective dose for the shortest duration. For AS, the recommended dose is 200 mg daily; if there is no response after 6 weeks, the dose should be titrated to 400 mg daily, according to Pfizer. If there is still no response after 6 weeks on that dosage, other treatment options should be considered.

Dr. Fields said he starts patients on 200 mg daily, which he says “is a dose that can be anti-inflammatory.” Dr. Reveille begins patients on a dosage of 100 mg daily and moves up to 200 mg if there is no response.

Both say that patients have become more nervous about celecoxib than physicians.

That does not mean physicians have no concerns. “There's no question that I'm watching patients more closely, talking to them about it, and asking if anything new has evolved in their history,” Dr. Fields said, adding that he regularly monitors patients for cardiovascular signs and always considers whether it's possible to shorten therapy or make it intermittent.

Dr. Reveille said he monitors patients on any NSAID or COX-2, including running liver function and complete blood counts at least twice a year.

AS is the sixth approved indication for celecoxib in the United States, but sales have dropped steeply from a year ago. In the first half of 2005, worldwide celecoxib sales totaled $813 million, a decline of 46% from the previous year.

Celecoxib, the only cyclooxygenase-2 inhibitor left on the U.S. market, has won an additional approval from the Food and Drug Administration—but, as expected, the drug also received a black box warning on the increased risk of cardiovascular events.

The Pfizer Inc. drug was approved for ankylosing spondylitis (AS), a connective tissue disorder causing inflammation of the spine and large joints that affects about 400,000 Americans, primarily between the ages of 20 and 40 years.

The new warning, a result of an FDA advisory committee's recommendations in February, says that celecoxib (Celebrex) may increase the risk of “thrombotic events, myocardial infarction, and stroke, which can be fatal.” The risk may increase with duration of use, according to the warning. The drug is also contraindicated for treating post-coronary artery bypass graft surgery pain.

The black box warning also highlights an increased risk of “serious gastrointestinal adverse events, including bleeding, ulceration, and perforation of the stomach or intestines, which can be fatal.”

According to the drug's label, in two placebo- and active-controlled trials, celecoxib was statistically superior to placebo in global pain, global disease, and functional impairment.

Theodore Fields, M.D., of Weill Medical College of Cornell University, noted that it's preferable to start patients on analgesics, but that if there's swelling, anti-inflammatories often have to be used. There are a number of “patients with ankylosing spondylitis where it makes sense, in spite of the known risks, to give them NSAIDs or COX-2s,” said Dr. Fields, who is also clinical director of the Early Arthritis Center at the Hospital for Special Surgery in New York.

The struggle sometimes comes in deciding between celecoxib or an NSAID such as naproxen, said physicians.

John Reveille, M.D., professor of medicine and director of rheumatology at the University of Texas Health Science Center in Houston, said he'd use celecoxib in patients who have a gastrointestinal intolerance to other NSAIDs.

Dr. Fields prefers to use celecoxib in patients who have a higher than average risk of GI complications, but lower than average cardiovascular risk.

Theoretically, because AS patients tend to be younger, the risk of cardiovascular complications is lower, Dr. Fields said in an interview. But, there's also some evidence that patients with inflammatory conditions, such as lupus and rheumatoid arthritis, are at higher risk for atherosclerosis, he said, adding that although the same has not been proven for AS, it is hypothetically possible.

Both Dr. Fields and Dr. Reveille said they'd avoid using celecoxib in patients who have cardiovascular disease or who may be at higher risk—those who are older, male, diabetic, or hypertensive. Neither physician is a paid consultant for any drug makers.

The new Celebrex label recommends that it be prescribed at the lowest effective dose for the shortest duration. For AS, the recommended dose is 200 mg daily; if there is no response after 6 weeks, the dose should be titrated to 400 mg daily, according to Pfizer. If there is still no response after 6 weeks on that dosage, other treatment options should be considered.

Dr. Fields said he starts patients on 200 mg daily, which he says “is a dose that can be anti-inflammatory.” Dr. Reveille begins patients on a dosage of 100 mg daily and moves up to 200 mg if there is no response.

Both say that patients have become more nervous about celecoxib than physicians.

That does not mean physicians have no concerns. “There's no question that I'm watching patients more closely, talking to them about it, and asking if anything new has evolved in their history,” Dr. Fields said, adding that he regularly monitors patients for cardiovascular signs and always considers whether it's possible to shorten therapy or make it intermittent.

Dr. Reveille said he monitors patients on any NSAID or COX-2, including running liver function and complete blood counts at least twice a year.

AS is the sixth approved indication for celecoxib in the United States, but sales have dropped steeply from a year ago. In the first half of 2005, worldwide celecoxib sales totaled $813 million, a decline of 46% from the previous year.

WASHINGTON — Two doses of influenza vaccine were up to 55% effective against influenza-like illness and 85% effective against pneumonia or flu in children, Mandy Allison, M.D., said at the National Immunization Conference sponsored by the Centers for Disease Control and Prevention.

The aim of Dr. Allison and her colleagues at Children's Hospital, Denver, and the University of Colorado Health Sciences Center, was to gather more data on the flu vaccine's effectiveness in children, especially since the CDC's Advisory Committee on Immunization Practices advised in 2004 that the shot should be included as a routine immunization for children aged 6–23 months.

Dr. Allison said that one study in the Journal of the American Medical Association calculated the vaccine's effectiveness at 66% in any given year, but a recent systematic review published in the Lancet found very little data on vaccine efficacy in children under 2 years old (2005;365:773–80).

A study conducted by Kaiser and the CDC found that the flu vaccine was only 25% effective against influenza-like illness, and 49% effective against pneumonia and flu, which is defined as a subset of influenza-like illness (MMWR 2004;53:707–10), she said.

Her group analyzed billing and immunization registry data from 5,913 healthy 6- to 21-month-old children from five Denver area pediatric practices.

ICD-9 codes for office visits between Nov. 1 and Dec. 31 were reviewed to determine the first influenza-like illness; the same nine codes were used in the Denver study as in the Kaiser/CDC study, Dr. Allison said.

Children were dubbed either partially vaccinated—one shot during the current season and 14 days before the first influenza-like illness—or fully vaccinated, which was defined as two shots more than 14 days before the first influenza-like illness.

During Colorado's flu season, which peaked early, 36% of the children were unvaccinated, 24% were partially vaccinated, and 40% were fully vaccinated.

Only 6% were fully vaccinated by Nov. 1, and 36% by Jan. 1, Dr. Allison said.

Twenty-eight percent of children had an influenza-like illness, and 5% pneumonia or flu, during the season.

The researchers also calculated hazard ratios that accounted for age, gender, and immunization status.

They determined that fully vaccinated children were less likely to have influenza-like illness (a ratio of 0.45), when compared with unvaccinated children, which was not surprising.

But partially vaccinated children were more likely to have influenza-like illness, compared with unvaccinated children, Dr. Allison said. She said the researchers weren't sure why one dose seemed to increase the chance of illness, but said there might be something different about those children or families.

She also wasn't certain why the Denver study showed much higher efficacy than the Kaiser/CDC study but noted that it might be that there was a significantly higher vaccination rate in the Denver population.

All the children in the Denver practices came from more affluent socioeconomic groups, which may have made a difference. That also limited the study's conclusions, though, she added.

WASHINGTON — Two doses of influenza vaccine were up to 55% effective against influenza-like illness and 85% effective against pneumonia or flu in children, Mandy Allison, M.D., said at the National Immunization Conference sponsored by the Centers for Disease Control and Prevention.

The aim of Dr. Allison and her colleagues at Children's Hospital, Denver, and the University of Colorado Health Sciences Center, was to gather more data on the flu vaccine's effectiveness in children, especially since the CDC's Advisory Committee on Immunization Practices advised in 2004 that the shot should be included as a routine immunization for children aged 6–23 months.

Dr. Allison said that one study in the Journal of the American Medical Association calculated the vaccine's effectiveness at 66% in any given year, but a recent systematic review published in the Lancet found very little data on vaccine efficacy in children under 2 years old (2005;365:773–80).

A study conducted by Kaiser and the CDC found that the flu vaccine was only 25% effective against influenza-like illness, and 49% effective against pneumonia and flu, which is defined as a subset of influenza-like illness (MMWR 2004;53:707–10), she said.

Her group analyzed billing and immunization registry data from 5,913 healthy 6- to 21-month-old children from five Denver area pediatric practices.

ICD-9 codes for office visits between Nov. 1 and Dec. 31 were reviewed to determine the first influenza-like illness; the same nine codes were used in the Denver study as in the Kaiser/CDC study, Dr. Allison said.

Children were dubbed either partially vaccinated—one shot during the current season and 14 days before the first influenza-like illness—or fully vaccinated, which was defined as two shots more than 14 days before the first influenza-like illness.

During Colorado's flu season, which peaked early, 36% of the children were unvaccinated, 24% were partially vaccinated, and 40% were fully vaccinated.

Only 6% were fully vaccinated by Nov. 1, and 36% by Jan. 1, Dr. Allison said.

Twenty-eight percent of children had an influenza-like illness, and 5% pneumonia or flu, during the season.

The researchers also calculated hazard ratios that accounted for age, gender, and immunization status.

They determined that fully vaccinated children were less likely to have influenza-like illness (a ratio of 0.45), when compared with unvaccinated children, which was not surprising.

But partially vaccinated children were more likely to have influenza-like illness, compared with unvaccinated children, Dr. Allison said. She said the researchers weren't sure why one dose seemed to increase the chance of illness, but said there might be something different about those children or families.

She also wasn't certain why the Denver study showed much higher efficacy than the Kaiser/CDC study but noted that it might be that there was a significantly higher vaccination rate in the Denver population.

All the children in the Denver practices came from more affluent socioeconomic groups, which may have made a difference. That also limited the study's conclusions, though, she added.

WASHINGTON — Two doses of influenza vaccine were up to 55% effective against influenza-like illness and 85% effective against pneumonia or flu in children, Mandy Allison, M.D., said at the National Immunization Conference sponsored by the Centers for Disease Control and Prevention.

The aim of Dr. Allison and her colleagues at Children's Hospital, Denver, and the University of Colorado Health Sciences Center, was to gather more data on the flu vaccine's effectiveness in children, especially since the CDC's Advisory Committee on Immunization Practices advised in 2004 that the shot should be included as a routine immunization for children aged 6–23 months.

Dr. Allison said that one study in the Journal of the American Medical Association calculated the vaccine's effectiveness at 66% in any given year, but a recent systematic review published in the Lancet found very little data on vaccine efficacy in children under 2 years old (2005;365:773–80).

A study conducted by Kaiser and the CDC found that the flu vaccine was only 25% effective against influenza-like illness, and 49% effective against pneumonia and flu, which is defined as a subset of influenza-like illness (MMWR 2004;53:707–10), she said.

Her group analyzed billing and immunization registry data from 5,913 healthy 6- to 21-month-old children from five Denver area pediatric practices.

ICD-9 codes for office visits between Nov. 1 and Dec. 31 were reviewed to determine the first influenza-like illness; the same nine codes were used in the Denver study as in the Kaiser/CDC study, Dr. Allison said.

Children were dubbed either partially vaccinated—one shot during the current season and 14 days before the first influenza-like illness—or fully vaccinated, which was defined as two shots more than 14 days before the first influenza-like illness.

During Colorado's flu season, which peaked early, 36% of the children were unvaccinated, 24% were partially vaccinated, and 40% were fully vaccinated.

Only 6% were fully vaccinated by Nov. 1, and 36% by Jan. 1, Dr. Allison said.

Twenty-eight percent of children had an influenza-like illness, and 5% pneumonia or flu, during the season.

The researchers also calculated hazard ratios that accounted for age, gender, and immunization status.

They determined that fully vaccinated children were less likely to have influenza-like illness (a ratio of 0.45), when compared with unvaccinated children, which was not surprising.

But partially vaccinated children were more likely to have influenza-like illness, compared with unvaccinated children, Dr. Allison said. She said the researchers weren't sure why one dose seemed to increase the chance of illness, but said there might be something different about those children or families.

She also wasn't certain why the Denver study showed much higher efficacy than the Kaiser/CDC study but noted that it might be that there was a significantly higher vaccination rate in the Denver population.

All the children in the Denver practices came from more affluent socioeconomic groups, which may have made a difference. That also limited the study's conclusions, though, she added.

CHICAGO — Contrast-enhanced intraoperative ultrasound identifies more liver tumors than does the imaging technology when it is used in a conventional manner, which may increase the chances for successful surgery or prevent unnecessary surgery, Marco Montorsi, M.D., said at the annual meeting of the Society for Surgery of the Alimentary Tract.

Dr. Montorsi, of the University of Milan, said he and his colleagues hoped to increase the tumor detection rate achieved with intraoperative ultrasound (IOUS), which has 82% sensitivity in detecting liver metastases from colorectal cancer.

From September 2003 to November 2004, 57 patients at the University of Milan had a liver resection and underwent both IOUS and the contrast-enhanced procedure, which involved 2.4 mL IV of sulfur-hexafluoride microbubbles.

Of the 57 patients, 34 had hepatocellular carcinoma (HCC), and 23 had colorectal metastases. Contrast-enhanced ultrasound provided new data about 21 of the 57 patients. For seven patients with HCC, the contrast did not confirm as malignant the new nodules found by conventional imaging. In these cases its use prevented surgery, Dr. Montorsi said. In four patients, contrast-enhanced ultrasound confirmed malignancies detected by IOUS, and in two it fully confirmed as HCC the additional lesions detected by IOUS.

In patients being evaluated for colorectal metastases, the contrast found new lesions missed by IOUS in four patients and confirmed additional lesions in three patients, Dr. Montorsi said.

CHICAGO — Contrast-enhanced intraoperative ultrasound identifies more liver tumors than does the imaging technology when it is used in a conventional manner, which may increase the chances for successful surgery or prevent unnecessary surgery, Marco Montorsi, M.D., said at the annual meeting of the Society for Surgery of the Alimentary Tract.

Dr. Montorsi, of the University of Milan, said he and his colleagues hoped to increase the tumor detection rate achieved with intraoperative ultrasound (IOUS), which has 82% sensitivity in detecting liver metastases from colorectal cancer.

From September 2003 to November 2004, 57 patients at the University of Milan had a liver resection and underwent both IOUS and the contrast-enhanced procedure, which involved 2.4 mL IV of sulfur-hexafluoride microbubbles.

Of the 57 patients, 34 had hepatocellular carcinoma (HCC), and 23 had colorectal metastases. Contrast-enhanced ultrasound provided new data about 21 of the 57 patients. For seven patients with HCC, the contrast did not confirm as malignant the new nodules found by conventional imaging. In these cases its use prevented surgery, Dr. Montorsi said. In four patients, contrast-enhanced ultrasound confirmed malignancies detected by IOUS, and in two it fully confirmed as HCC the additional lesions detected by IOUS.

In patients being evaluated for colorectal metastases, the contrast found new lesions missed by IOUS in four patients and confirmed additional lesions in three patients, Dr. Montorsi said.

CHICAGO — Contrast-enhanced intraoperative ultrasound identifies more liver tumors than does the imaging technology when it is used in a conventional manner, which may increase the chances for successful surgery or prevent unnecessary surgery, Marco Montorsi, M.D., said at the annual meeting of the Society for Surgery of the Alimentary Tract.

Dr. Montorsi, of the University of Milan, said he and his colleagues hoped to increase the tumor detection rate achieved with intraoperative ultrasound (IOUS), which has 82% sensitivity in detecting liver metastases from colorectal cancer.

From September 2003 to November 2004, 57 patients at the University of Milan had a liver resection and underwent both IOUS and the contrast-enhanced procedure, which involved 2.4 mL IV of sulfur-hexafluoride microbubbles.

Of the 57 patients, 34 had hepatocellular carcinoma (HCC), and 23 had colorectal metastases. Contrast-enhanced ultrasound provided new data about 21 of the 57 patients. For seven patients with HCC, the contrast did not confirm as malignant the new nodules found by conventional imaging. In these cases its use prevented surgery, Dr. Montorsi said. In four patients, contrast-enhanced ultrasound confirmed malignancies detected by IOUS, and in two it fully confirmed as HCC the additional lesions detected by IOUS.

In patients being evaluated for colorectal metastases, the contrast found new lesions missed by IOUS in four patients and confirmed additional lesions in three patients, Dr. Montorsi said.