User login
CIPN persists in female cancer survivors
Photo courtesy of NIH
SAN FRANCISCO—A study of female cancer survivors indicates that many still have chemotherapy-induced peripheral neuropathy (CIPN) symptoms years after completing cancer treatment.
In addition, CIPN was associated with worse physical functioning, poorer mobility, and a higher risk of falls.
Although more research is needed, investigators believe these findings may inform rehabilitation and fall prevention interventions for people with CIPN.
The findings were presented at the 2016 Cancer Survivorship Symposium (abstract 130*).
“We can’t dismiss neuropathy as a treatment side effect that goes away because symptoms persist for years in nearly half of women,” said Kerri M. Winters-Stone, PhD, of Oregon Health and Science University in Portland.
“While there are no effective treatments for this side effect, rehabilitative exercise programs may preserve physical functioning and mobility in the presence of neuropathy to help prevent falls and resulting injuries.”
For this study, Dr Winters-Stone and her colleagues assessed data from 512 women enrolled in exercise intervention trials designed to address fractures and falls in female cancer survivors. Most of the women had breast cancer, but there were also cases of lung, colorectal, ovarian, and hematologic cancers.
At an average of 6 years post-cancer diagnosis, 46% of the women (n=238) still reported some symptoms of CIPN, such as loss of feeling in their hands and feet.
The investigators noted significant relationships (P<0.01) between CIPN severity and gait speed, Physical Performance Battery score, self-reported physical functioning, and self-reported disability.
The team also compared measures of physical functioning in the women with CIPN to measures in women without CIPN (n=274). This analysis was adjusted for cancer type and time since diagnosis.
There was a significant difference (P<0.01) between the groups in one measure of lower-extremity fitness but not another. Namely, it took CIPN-positive women significantly longer to rise out of a chair (tested 5 times each). But women in both groups fared similarly on a test measuring maximal leg press strength.
The investigators also tested the women on mobility and physical functioning. The CIPN-positive women fared significantly worse than CIPN-negative women (P<0.01) when it came to walking speed, step number, stride length, percentage of gait cycle in double support, and Physical Performance Battery score. However, there was no significant difference between the groups with regard to base of support.
Finally, CIPN-positive women were significantly more likely than CIPN-negative women to report poor physical function and disability (P<0.01 for both). And CIPN-positive women had a higher rate of falls in the last year (P<0.01).
The investigators said women with CIPN have specific underlying impairments that put them at risk for falls, which may be different from the impairments that occur with other conditions or old age.
For example, CIPN does not cause muscle weakness, but it has a distinct effect on movement and gait patterns.
The team noted that the women with CIPN had difficulty rising from a chair, possibly because their brains do not get enough information from their feet about how quickly or forcefully to stand up.
Based on these findings, the investigators argued that commonly recommended exercise, such as walking, may be safer for women with CIPN when done on a treadmill with handrails because their altered gait puts them at an increased risk of falling.
The team also said that machine-based resistance training may not be beneficial because neuropathy does not appear to decrease leg strength. Instead, rehabilitation efforts should focus on improving balance during upright movement and specific gait training.
Furthermore, the investigators believe that, if the symptoms of CIPN are detected early, cancer treatments could potentially be changed to prevent these debilitating problems or early rehabilitation interventions could be started.
In addition, Dr Winters-Stone and her research team are developing a smartphone-driven device that patients can use to detect and quantify symptoms of neuropathy, such as gait and balance impairments.
*Data in the abstract differ from the presentation.
Photo courtesy of NIH
SAN FRANCISCO—A study of female cancer survivors indicates that many still have chemotherapy-induced peripheral neuropathy (CIPN) symptoms years after completing cancer treatment.
In addition, CIPN was associated with worse physical functioning, poorer mobility, and a higher risk of falls.
Although more research is needed, investigators believe these findings may inform rehabilitation and fall prevention interventions for people with CIPN.
The findings were presented at the 2016 Cancer Survivorship Symposium (abstract 130*).
“We can’t dismiss neuropathy as a treatment side effect that goes away because symptoms persist for years in nearly half of women,” said Kerri M. Winters-Stone, PhD, of Oregon Health and Science University in Portland.
“While there are no effective treatments for this side effect, rehabilitative exercise programs may preserve physical functioning and mobility in the presence of neuropathy to help prevent falls and resulting injuries.”
For this study, Dr Winters-Stone and her colleagues assessed data from 512 women enrolled in exercise intervention trials designed to address fractures and falls in female cancer survivors. Most of the women had breast cancer, but there were also cases of lung, colorectal, ovarian, and hematologic cancers.
At an average of 6 years post-cancer diagnosis, 46% of the women (n=238) still reported some symptoms of CIPN, such as loss of feeling in their hands and feet.
The investigators noted significant relationships (P<0.01) between CIPN severity and gait speed, Physical Performance Battery score, self-reported physical functioning, and self-reported disability.
The team also compared measures of physical functioning in the women with CIPN to measures in women without CIPN (n=274). This analysis was adjusted for cancer type and time since diagnosis.
There was a significant difference (P<0.01) between the groups in one measure of lower-extremity fitness but not another. Namely, it took CIPN-positive women significantly longer to rise out of a chair (tested 5 times each). But women in both groups fared similarly on a test measuring maximal leg press strength.
The investigators also tested the women on mobility and physical functioning. The CIPN-positive women fared significantly worse than CIPN-negative women (P<0.01) when it came to walking speed, step number, stride length, percentage of gait cycle in double support, and Physical Performance Battery score. However, there was no significant difference between the groups with regard to base of support.
Finally, CIPN-positive women were significantly more likely than CIPN-negative women to report poor physical function and disability (P<0.01 for both). And CIPN-positive women had a higher rate of falls in the last year (P<0.01).
The investigators said women with CIPN have specific underlying impairments that put them at risk for falls, which may be different from the impairments that occur with other conditions or old age.
For example, CIPN does not cause muscle weakness, but it has a distinct effect on movement and gait patterns.
The team noted that the women with CIPN had difficulty rising from a chair, possibly because their brains do not get enough information from their feet about how quickly or forcefully to stand up.
Based on these findings, the investigators argued that commonly recommended exercise, such as walking, may be safer for women with CIPN when done on a treadmill with handrails because their altered gait puts them at an increased risk of falling.
The team also said that machine-based resistance training may not be beneficial because neuropathy does not appear to decrease leg strength. Instead, rehabilitation efforts should focus on improving balance during upright movement and specific gait training.
Furthermore, the investigators believe that, if the symptoms of CIPN are detected early, cancer treatments could potentially be changed to prevent these debilitating problems or early rehabilitation interventions could be started.
In addition, Dr Winters-Stone and her research team are developing a smartphone-driven device that patients can use to detect and quantify symptoms of neuropathy, such as gait and balance impairments.
*Data in the abstract differ from the presentation.
Photo courtesy of NIH
SAN FRANCISCO—A study of female cancer survivors indicates that many still have chemotherapy-induced peripheral neuropathy (CIPN) symptoms years after completing cancer treatment.
In addition, CIPN was associated with worse physical functioning, poorer mobility, and a higher risk of falls.
Although more research is needed, investigators believe these findings may inform rehabilitation and fall prevention interventions for people with CIPN.
The findings were presented at the 2016 Cancer Survivorship Symposium (abstract 130*).
“We can’t dismiss neuropathy as a treatment side effect that goes away because symptoms persist for years in nearly half of women,” said Kerri M. Winters-Stone, PhD, of Oregon Health and Science University in Portland.
“While there are no effective treatments for this side effect, rehabilitative exercise programs may preserve physical functioning and mobility in the presence of neuropathy to help prevent falls and resulting injuries.”
For this study, Dr Winters-Stone and her colleagues assessed data from 512 women enrolled in exercise intervention trials designed to address fractures and falls in female cancer survivors. Most of the women had breast cancer, but there were also cases of lung, colorectal, ovarian, and hematologic cancers.
At an average of 6 years post-cancer diagnosis, 46% of the women (n=238) still reported some symptoms of CIPN, such as loss of feeling in their hands and feet.
The investigators noted significant relationships (P<0.01) between CIPN severity and gait speed, Physical Performance Battery score, self-reported physical functioning, and self-reported disability.
The team also compared measures of physical functioning in the women with CIPN to measures in women without CIPN (n=274). This analysis was adjusted for cancer type and time since diagnosis.
There was a significant difference (P<0.01) between the groups in one measure of lower-extremity fitness but not another. Namely, it took CIPN-positive women significantly longer to rise out of a chair (tested 5 times each). But women in both groups fared similarly on a test measuring maximal leg press strength.
The investigators also tested the women on mobility and physical functioning. The CIPN-positive women fared significantly worse than CIPN-negative women (P<0.01) when it came to walking speed, step number, stride length, percentage of gait cycle in double support, and Physical Performance Battery score. However, there was no significant difference between the groups with regard to base of support.
Finally, CIPN-positive women were significantly more likely than CIPN-negative women to report poor physical function and disability (P<0.01 for both). And CIPN-positive women had a higher rate of falls in the last year (P<0.01).
The investigators said women with CIPN have specific underlying impairments that put them at risk for falls, which may be different from the impairments that occur with other conditions or old age.
For example, CIPN does not cause muscle weakness, but it has a distinct effect on movement and gait patterns.
The team noted that the women with CIPN had difficulty rising from a chair, possibly because their brains do not get enough information from their feet about how quickly or forcefully to stand up.
Based on these findings, the investigators argued that commonly recommended exercise, such as walking, may be safer for women with CIPN when done on a treadmill with handrails because their altered gait puts them at an increased risk of falling.
The team also said that machine-based resistance training may not be beneficial because neuropathy does not appear to decrease leg strength. Instead, rehabilitation efforts should focus on improving balance during upright movement and specific gait training.
Furthermore, the investigators believe that, if the symptoms of CIPN are detected early, cancer treatments could potentially be changed to prevent these debilitating problems or early rehabilitation interventions could be started.
In addition, Dr Winters-Stone and her research team are developing a smartphone-driven device that patients can use to detect and quantify symptoms of neuropathy, such as gait and balance impairments.
*Data in the abstract differ from the presentation.
Acupuncture no better than sham for relief of hot flashes
Chinese medicine needle acupuncture was about as effective as a sham blunt needle treatment in the relief of hot flashes, although women reported a 40% drop in symptoms with both treatments.
The findings, published online Jan. 18 in Annals of Internal Medicine, add to a growing, but conflicting body of evidence about the benefits of acupuncture in the treatment of menopause symptoms.
Prior to this study, two trials had demonstrated the effectiveness of acupuncture, compared with self care. And a pilot study had shown the effectiveness of acupuncture, compared with a noninsertive sham control. However, a Cochrane review found that acupuncture was more effective, compared with no treatment, and it had a moderate effect size, but was not effective when compared with a sham control (Cochrane Database Syst Rev. 2013 Jul 30;7:CD007410. doi:10.1002/14651858.CD007410.pub2).
The current trial, conducted at multiple sites in Australia, sought to add to the evidence with an adequately powered trial involving a sham control. But Carolyn Ee and her associates at the University of Melbourne noted that their study did not control for the nonspecific effects of acupuncture, such as regular interaction with a therapist.
The researchers randomly assigned 327 women aged older than 40 years who were in late menopause transition or postmenopause and experiencing at least seven moderate daily hot flashes to receive either a standardized Chinese medicine acupuncture treatment or a noninsertive, blunt needle sham acupuncture treatment. Patients received 10 treatments over 8 weeks, and they were assessed at 4 weeks, at the end of treatment, and at 3 and 6 months after treatment (Ann Intern Med. 2016; Jan 18. doi:10.7326/M15-1380.).
Both groups had about a 40% improvement in their hot flashes at the end of treatment, compared with their mean baseline hot flash score. The improvement was sustained at 3 and 6 months after the trial. In the acupuncture group, the mean hot flash scores at the end of treatment were 15.36, compared with 15.04 in the sham group, which was not statistically different. The researchers also found no advantage for acupuncture in quality of life, anxiety, or depression.
“Unless further high-quality evidence emerges, we cannot recommend skin-penetrating acupuncture as an efficacious treatment of this indication; the effects, if any, of acupuncture on these symptoms seem to be unrelated to needling,” the researchers wrote.
Some of the researchers reported receiving grant, scholarship, or fellowship support from the National Health and Medical Research Council of Australia, which funded the study.
mschneider@frontlinemedcom.com
On Twitter @maryellenny
Chinese medicine needle acupuncture was about as effective as a sham blunt needle treatment in the relief of hot flashes, although women reported a 40% drop in symptoms with both treatments.
The findings, published online Jan. 18 in Annals of Internal Medicine, add to a growing, but conflicting body of evidence about the benefits of acupuncture in the treatment of menopause symptoms.
Prior to this study, two trials had demonstrated the effectiveness of acupuncture, compared with self care. And a pilot study had shown the effectiveness of acupuncture, compared with a noninsertive sham control. However, a Cochrane review found that acupuncture was more effective, compared with no treatment, and it had a moderate effect size, but was not effective when compared with a sham control (Cochrane Database Syst Rev. 2013 Jul 30;7:CD007410. doi:10.1002/14651858.CD007410.pub2).
The current trial, conducted at multiple sites in Australia, sought to add to the evidence with an adequately powered trial involving a sham control. But Carolyn Ee and her associates at the University of Melbourne noted that their study did not control for the nonspecific effects of acupuncture, such as regular interaction with a therapist.
The researchers randomly assigned 327 women aged older than 40 years who were in late menopause transition or postmenopause and experiencing at least seven moderate daily hot flashes to receive either a standardized Chinese medicine acupuncture treatment or a noninsertive, blunt needle sham acupuncture treatment. Patients received 10 treatments over 8 weeks, and they were assessed at 4 weeks, at the end of treatment, and at 3 and 6 months after treatment (Ann Intern Med. 2016; Jan 18. doi:10.7326/M15-1380.).
Both groups had about a 40% improvement in their hot flashes at the end of treatment, compared with their mean baseline hot flash score. The improvement was sustained at 3 and 6 months after the trial. In the acupuncture group, the mean hot flash scores at the end of treatment were 15.36, compared with 15.04 in the sham group, which was not statistically different. The researchers also found no advantage for acupuncture in quality of life, anxiety, or depression.
“Unless further high-quality evidence emerges, we cannot recommend skin-penetrating acupuncture as an efficacious treatment of this indication; the effects, if any, of acupuncture on these symptoms seem to be unrelated to needling,” the researchers wrote.
Some of the researchers reported receiving grant, scholarship, or fellowship support from the National Health and Medical Research Council of Australia, which funded the study.
mschneider@frontlinemedcom.com
On Twitter @maryellenny
Chinese medicine needle acupuncture was about as effective as a sham blunt needle treatment in the relief of hot flashes, although women reported a 40% drop in symptoms with both treatments.
The findings, published online Jan. 18 in Annals of Internal Medicine, add to a growing, but conflicting body of evidence about the benefits of acupuncture in the treatment of menopause symptoms.
Prior to this study, two trials had demonstrated the effectiveness of acupuncture, compared with self care. And a pilot study had shown the effectiveness of acupuncture, compared with a noninsertive sham control. However, a Cochrane review found that acupuncture was more effective, compared with no treatment, and it had a moderate effect size, but was not effective when compared with a sham control (Cochrane Database Syst Rev. 2013 Jul 30;7:CD007410. doi:10.1002/14651858.CD007410.pub2).
The current trial, conducted at multiple sites in Australia, sought to add to the evidence with an adequately powered trial involving a sham control. But Carolyn Ee and her associates at the University of Melbourne noted that their study did not control for the nonspecific effects of acupuncture, such as regular interaction with a therapist.
The researchers randomly assigned 327 women aged older than 40 years who were in late menopause transition or postmenopause and experiencing at least seven moderate daily hot flashes to receive either a standardized Chinese medicine acupuncture treatment or a noninsertive, blunt needle sham acupuncture treatment. Patients received 10 treatments over 8 weeks, and they were assessed at 4 weeks, at the end of treatment, and at 3 and 6 months after treatment (Ann Intern Med. 2016; Jan 18. doi:10.7326/M15-1380.).
Both groups had about a 40% improvement in their hot flashes at the end of treatment, compared with their mean baseline hot flash score. The improvement was sustained at 3 and 6 months after the trial. In the acupuncture group, the mean hot flash scores at the end of treatment were 15.36, compared with 15.04 in the sham group, which was not statistically different. The researchers also found no advantage for acupuncture in quality of life, anxiety, or depression.
“Unless further high-quality evidence emerges, we cannot recommend skin-penetrating acupuncture as an efficacious treatment of this indication; the effects, if any, of acupuncture on these symptoms seem to be unrelated to needling,” the researchers wrote.
Some of the researchers reported receiving grant, scholarship, or fellowship support from the National Health and Medical Research Council of Australia, which funded the study.
mschneider@frontlinemedcom.com
On Twitter @maryellenny
FROM ANNALS OF INTERNAL MEDICINE
Key clinical point: Chinese medicine acupuncture was no better than a sham treatment for the relief of hot flashes.
Major finding: After 8 weeks of treatment, mean hot flash scores were 15.36 in the acupuncture group and 15.04 in the sham treatment group, which was not a statistically significant difference.
Data source: A stratified, blind, parallel, randomized, sham-controlled trial of 327 women in late menopause transition or postmenopause.
Disclosures: Some of the researchers reported receiving grant, scholarship, or fellowship support from the National Health and Medical Research Council of Australia, which funded the study.
Secukinumab receives FDA approval for psoriatic arthritis, ankylosing spondylitis
The Food and Drug Administration approved two new indications for the interleukin-17A inhibitor secukinumab (Cosentyx) – psoriatic arthritis in adults and ankylosing spondylitis in adults – on Jan. 15. These join the approval for moderate to severe plaque psoriasis in adults it received in January 2015, according to an announcement from the drug’s manufacturer, Novartis.
The approvals are based on the efficacy and safety outcomes from four placebo-controlled, phase III studies, which included more than 1,500 adult patients with ankylosing spondylitis (AS) or psoriatic arthritis (PsA) who were biologic treatment naive or had an inadequate response or were intolerant to anti-TNF agents.
Pivotal phase III studies in the secukinumab clinical trial program, which provided key data for the submission, were MEASURE 1 and MEASURE 2 involving 590 patients with AS, and FUTURE 1 and FUTURE 2 involving 1,003 patients with PsA. Novartis continues to investigate the fully human monoclonal antibody against IL-17A for its potential in preventing radiographic progression of spinal and joint structural damage in AS and PsA patients, respectively.
The European Medicines Agency approved secukinumab for PsA and AS in November 2015.
The Food and Drug Administration approved two new indications for the interleukin-17A inhibitor secukinumab (Cosentyx) – psoriatic arthritis in adults and ankylosing spondylitis in adults – on Jan. 15. These join the approval for moderate to severe plaque psoriasis in adults it received in January 2015, according to an announcement from the drug’s manufacturer, Novartis.
The approvals are based on the efficacy and safety outcomes from four placebo-controlled, phase III studies, which included more than 1,500 adult patients with ankylosing spondylitis (AS) or psoriatic arthritis (PsA) who were biologic treatment naive or had an inadequate response or were intolerant to anti-TNF agents.
Pivotal phase III studies in the secukinumab clinical trial program, which provided key data for the submission, were MEASURE 1 and MEASURE 2 involving 590 patients with AS, and FUTURE 1 and FUTURE 2 involving 1,003 patients with PsA. Novartis continues to investigate the fully human monoclonal antibody against IL-17A for its potential in preventing radiographic progression of spinal and joint structural damage in AS and PsA patients, respectively.
The European Medicines Agency approved secukinumab for PsA and AS in November 2015.
The Food and Drug Administration approved two new indications for the interleukin-17A inhibitor secukinumab (Cosentyx) – psoriatic arthritis in adults and ankylosing spondylitis in adults – on Jan. 15. These join the approval for moderate to severe plaque psoriasis in adults it received in January 2015, according to an announcement from the drug’s manufacturer, Novartis.
The approvals are based on the efficacy and safety outcomes from four placebo-controlled, phase III studies, which included more than 1,500 adult patients with ankylosing spondylitis (AS) or psoriatic arthritis (PsA) who were biologic treatment naive or had an inadequate response or were intolerant to anti-TNF agents.
Pivotal phase III studies in the secukinumab clinical trial program, which provided key data for the submission, were MEASURE 1 and MEASURE 2 involving 590 patients with AS, and FUTURE 1 and FUTURE 2 involving 1,003 patients with PsA. Novartis continues to investigate the fully human monoclonal antibody against IL-17A for its potential in preventing radiographic progression of spinal and joint structural damage in AS and PsA patients, respectively.
The European Medicines Agency approved secukinumab for PsA and AS in November 2015.
Ignore your insurance company reminder about healthcare costs
In a letter dated Nov. 20, 2015, a particular insurance company reminded me that I need to think about healthcare costs. I got a breakdown of how often in the quarter I used proprietary vs. generic drugs, how I compared to other rheumatologists, and how I compared to other physicians. I also got a list of the drugs that I used and alternatives that I should be thinking about instead. In principle, this is not a terrible idea. But let me describe some glaring mistakes that show that these letters are in fact a contradiction in themselves. They are a huge waste of resources.
The letter suggested that my number one “prescribed brand drug with potential member savings opportunities” was Uloric, costing an average of $302 per prescription. The suggested “generic” alternative was Colcrys, leading to a “potential annual cost savings” of $600.
Third drug on my list? Colcrys, coming in at $165 per prescription with a potential annual savings of $300. Listed alternative: allopurinol.
We rheumatologists know that Colcrys is not an alternative to Uloric, and allopurinol is not an alternative to Colcrys. Also, suggesting an alternative only to suggest an alternative to that alternative is idiotic. Obviously, the letter is generated by a data-crunching algorithm. But an algorithm can only be as good as the programmer creating it.
As for my other proprietary prescriptions: Lyrica was the second on my list, and Celebrex the fourth. Let me explain why this is both annoying and inefficient. Before I could prescribe those drugs, this insurer made me jump through hoops to get them. In other words, there already exists in their database proof that I had already tried their recommended alternatives.
I can only conclude that within the bowels of health insurance corporate offices, and probably in more places than I care to imagine, there are people who are either incompetent or lazy, or both, making healthcare decisions.
So, to the health insurer: Forgive me if I ignore your reminder. I already know that healthcare costs are bloated. I am already quite conscientious about my prescribing practices. I think I can speak for all rheumatologists who receive these notices: You’re barking up the wrong tree.
Dr. Chan practices rheumatology in Pawtucket, R.I.
In a letter dated Nov. 20, 2015, a particular insurance company reminded me that I need to think about healthcare costs. I got a breakdown of how often in the quarter I used proprietary vs. generic drugs, how I compared to other rheumatologists, and how I compared to other physicians. I also got a list of the drugs that I used and alternatives that I should be thinking about instead. In principle, this is not a terrible idea. But let me describe some glaring mistakes that show that these letters are in fact a contradiction in themselves. They are a huge waste of resources.
The letter suggested that my number one “prescribed brand drug with potential member savings opportunities” was Uloric, costing an average of $302 per prescription. The suggested “generic” alternative was Colcrys, leading to a “potential annual cost savings” of $600.
Third drug on my list? Colcrys, coming in at $165 per prescription with a potential annual savings of $300. Listed alternative: allopurinol.
We rheumatologists know that Colcrys is not an alternative to Uloric, and allopurinol is not an alternative to Colcrys. Also, suggesting an alternative only to suggest an alternative to that alternative is idiotic. Obviously, the letter is generated by a data-crunching algorithm. But an algorithm can only be as good as the programmer creating it.
As for my other proprietary prescriptions: Lyrica was the second on my list, and Celebrex the fourth. Let me explain why this is both annoying and inefficient. Before I could prescribe those drugs, this insurer made me jump through hoops to get them. In other words, there already exists in their database proof that I had already tried their recommended alternatives.
I can only conclude that within the bowels of health insurance corporate offices, and probably in more places than I care to imagine, there are people who are either incompetent or lazy, or both, making healthcare decisions.
So, to the health insurer: Forgive me if I ignore your reminder. I already know that healthcare costs are bloated. I am already quite conscientious about my prescribing practices. I think I can speak for all rheumatologists who receive these notices: You’re barking up the wrong tree.
Dr. Chan practices rheumatology in Pawtucket, R.I.
In a letter dated Nov. 20, 2015, a particular insurance company reminded me that I need to think about healthcare costs. I got a breakdown of how often in the quarter I used proprietary vs. generic drugs, how I compared to other rheumatologists, and how I compared to other physicians. I also got a list of the drugs that I used and alternatives that I should be thinking about instead. In principle, this is not a terrible idea. But let me describe some glaring mistakes that show that these letters are in fact a contradiction in themselves. They are a huge waste of resources.
The letter suggested that my number one “prescribed brand drug with potential member savings opportunities” was Uloric, costing an average of $302 per prescription. The suggested “generic” alternative was Colcrys, leading to a “potential annual cost savings” of $600.
Third drug on my list? Colcrys, coming in at $165 per prescription with a potential annual savings of $300. Listed alternative: allopurinol.
We rheumatologists know that Colcrys is not an alternative to Uloric, and allopurinol is not an alternative to Colcrys. Also, suggesting an alternative only to suggest an alternative to that alternative is idiotic. Obviously, the letter is generated by a data-crunching algorithm. But an algorithm can only be as good as the programmer creating it.
As for my other proprietary prescriptions: Lyrica was the second on my list, and Celebrex the fourth. Let me explain why this is both annoying and inefficient. Before I could prescribe those drugs, this insurer made me jump through hoops to get them. In other words, there already exists in their database proof that I had already tried their recommended alternatives.
I can only conclude that within the bowels of health insurance corporate offices, and probably in more places than I care to imagine, there are people who are either incompetent or lazy, or both, making healthcare decisions.
So, to the health insurer: Forgive me if I ignore your reminder. I already know that healthcare costs are bloated. I am already quite conscientious about my prescribing practices. I think I can speak for all rheumatologists who receive these notices: You’re barking up the wrong tree.
Dr. Chan practices rheumatology in Pawtucket, R.I.
Pruritic Dermatitis Caused by Bird Mite Infestation
To the Editor:
There are a wide variety of zoonotic diseases that can be transmitted from birds to humans. Pigeons, chickens, starlings, canaries, and parakeets are known reservoirs of one particular zoonotic infection caused by the parasitic arthropod Dermanyssus gallinae.1 Dermanyssus gallinae (chicken mite) and Ornithonyssus sylviarum (northern fowl mite) are collectively referred to as bird mites. When these mites are unable to take blood meals from birds, they search out alternative hosts2; in humans, this often leads to the development of pruritic dermatitis.3
A 30-year-old woman presented to our clinic for evaluation of severe generalized pruritus accompanied by a sensation of “bugs on the skin” of 2 weeks’ duration. She noted the pruritus worsened when she was sitting outside on her porch. A few days prior to presentation, she noticed a small, “pinpoint-sized bug” on her arm (<1 mm in size), which she brought in for identification (Figure).
The bug was identified as a bird mite (Dermanyssus gallinae) on light microscopy, which was later confirmed by a medical entomologist. After the diagnosis of bird mite dermatitis was made, the patient noted there was a nest of starlings above the light on her porch. When she later investigated the nest following the current presentation, she noted many small mites crawling around the nest. The nest was removed and her symptoms resolved completely within 2 weeks without treatment.
Bird mites belong to the Arachnida class, under the order Acari. In 1958, Williams4 noted D gallinae’s ability to feed on human blood. Bird mites have 5 stages of development: egg, larva, protonymph, deutonymph, and adult. Protonymphs, deutonymphs, and adults can bite humans for a blood meal.5 Bird mites range from 0.3 to 1 mm in length and have nonsegmented, egg-shaped bodies with 4 pairs of legs. Before taking a blood meal, bird mites generally are a translucent brown color, and appear red when engorged with blood.2 Their small size makes them barely visible to the unaided eye. Of note, D gallinae and O sylviarum can be distinguished from each other based on subtle differences in morphology; for instance, the posterior genitoventral shield of O sylviarum is narrowly rounded, whereas it is broadly rounded in D gallinae. The dorsal shield of O sylviarum abruptly narrows posteriorly but is more smoothly narrowed in D gallinae.6 Additionally, O sylviarum tends to cause more irritating dermatitis in humans than D gallinae.3
Although they can be found worldwide, D gallinae and O sylviarum undergo optimal development at 20°C to 25°C and 70% humidity.3,5,7 Bird mites generally develop over the course of 5 to 12 days; thus, the population of bird mites in a single nest may grow to the tens of thousands before young birds permanently leave. Dermanyssus gallinae can survive for months in abandoned nests without a blood meal, while O sylviarum can survive for several weeks.8 It is important to note that humans are not ideal hosts for bird mites, as they are unable to survive for extended periods of time or reproduce on human hosts.9
When bird mites are no longer able to obtain blood meals from nesting birds, they begin their nocturnal migration to find suitable hosts. Bird nests generally are abandoned in late spring; thus, most patients with bird mite dermatitis present to clinics with bird mite dermatitis in late spring and early summer.10 Mites often travel through cracks in doors, floors, walls, and ceilings but also can gain access to living areas through ventilation ducts and air conditioning units.1 The mite’s bite and crawling on the skin is sometimes noticed by the patient. In general, however, intense itching is not observed until about 1 to 3 days after the mite makes contact with the skin. Patients often report that pruritus is worst at night.9 Papules and vesicles (bite reactions) may accompany the pruritus, and physicians commonly find bloody crust and excoriations in particularly pruritic areas.5 Urticarial plaques and diffuse erythema occasionally also may be present.9 Bird mites sometimes can be scraped from the skin and observed under light microscopy.11 Blood eosinophilia is not found in bird mite dermatitis. On histologic examination, perivascular eosinophilic infiltration can be seen in the upper part of the dermis.12
The differential diagnosis in patients with pruritic dermatitis of unknown origin generally includes scabies, pediculosis, and dermatitis caused by other types of infestation. However, unlike scabies, bird mites do not cause burrows to form on the skin.9 The presence of a bird’s nest near the area where the patient lives places bird mite dermatitis higher in the differential.
Dermanyssus gallinae is a known vector of bacteria (eg, Salmonella, Shigella, Staphylococcus, Spirochaete, Rickettsia, Pasteurella, Chlamydia psittaci, Erysipelothrix rhusiopathiae) as well as the viruses that cause Eastern and Western equine encephalitis and St. Louis encephalitis. Transmission of these bacteria and viruses is known in birds, but transmission to humans has not been reported.2,5,9,13
The management of bird mite dermatitis is straightforward. Usually mites can be successfully removed from the skin simply by bathing. Symptomatic treatment for bites with antihistamines and topical corticosteroids is sometimes but not always necessary.2 Unlike scabies or lice, there is no need for treatment with lindane.1 In terms of the prevention of additional bites, any bird nests located near living areas should be removed. Because bird mites often retreat back to nests between blood meals, insecticide sprays generally are unnecessary in interior spaces. Synthetic pyrethroids (eg, bifenthrin, cyfluthrin, cypermethrin, deltamethrin, cyhalothrin) can be used outside and in attics where nests may be located.2,14,15 However, the ability of bird mites to develop resistance to repeated chemical control could become a future concern.16
Research regarding the true incidence of bird mite dermatitis is lacking. Some researchers believe that the condition is underreported, possibly due to its uncommon environmental origin.3 Reports of bird mite dermatitis in the literature also are scarce. Our case demonstrates the importance of taking a thorough patient history to rule out exposure to bird mites. All patients with pruritic dermatitis of unknown origin should be questioned about possible contact or proximity to bird nests. These simple questions can lead to the correct diagnosis and a treatment plan that will quickly and effectively resolve the pruritic skin eruption.
- Regan AM, Metersky ML, Craven DE. Nosocomial dermatitis and pruritus caused by pigeon mite infestation. Arch Intern Med. 1987;147:2185-2187.
- Collgros H, Iglesias-Sancho M, Aldunce MJ, et al. Dermanyssus gallinae (chicken mite): an underdiagnosed environmental infestation. Clin Exp Dermatol. 2013;38:374-377.
- Bellanger AP, Boris C, Foulet F, et al. Nosocomial dermatitis caused by Dermanyssus gallinae. Infect Cont Hosp Ep. 2008;29:282-283.
- Williams RW. An infestation of a human habitation by Dermanyssus gallinae (de Geer, 1778) (Acarina: Dermanyssidae) in New York resulting in sanguisugent attacks upon the occupants. Am J Trop Med Hyg. 1958;7:627-629.
- Akdemir C, Gülcan E, Tanritanir P. Case report: Dermanyssus gallinae in a patient with pruritus and skin lesions. Turkiye Parazitol Derg. 2009;33:242-244.
- DiPalma A, Giangaspero A, Cafiero MA, et al. A gallery of the key characteristics to ease identification of Dermanyssus gallinae (Acari: Gamasida: Dermanyssidae) and allow differentiation from Ornithonyssus sylviarum (Acari: Gamasida: Macronyssidae). Parasites and Vectors. 2012;5:104.
- Maurer V, Baumgartner J. Temperature influence on life table statistics of the chicken mite Dermanyssus gallinae (Acari: Dermanyssidae). Exp Appl Acarol. 1992;15:27-40.
- Orton DI, Warren LJ, Wilkinson JD. Avian mite dermatitis. Clin Exper Dermatol. 2000;25:129-131.
- Auger P, Nantel J, Meunier N, et al. Skin acariasis caused by Dermanyssus gallinae (de Geer): an in-hospital outbreak. Can Med Assoc J. 1979;120:700-703.
- Kong TK, To WK. Bird mite infestation. N Engl J Med. 2006;354:1728.
- Koh WL, Liu TT, Tay YK. Formication due to true parasitic infection: bird mites. Arch Dermatol. 2011;147:508-509.
- Hidano A, Asanuma K. Letter: Acariasis caused by bird mites. Arch Dermatol. 1976;112:881-882.
- Valiente Moro C, Chauve C, Zenner L. Experimental infection of Salmonella Enteritidis by the poultry red mite, Dermanyssus gallinae. Vet Parasitol. 2007;146:329-336.
- Fletcher MG, Axtell RC. Susceptibilities of northern fowl mite, Ornithonyssus sylviarum (Acarina: Macronyssidae),and chicken mite, Dermanyssus gallinae (Acarina: Dermanyssidae), to selected acaricides. Exp Appl Acarol. 1991;13:137-142.
- Thind BB, Ford HL. Assessment of susceptibility of the poultry red mite Dermanyssus gallinae (Acari: Dermanyssidae) to some acaricides using an adapted filter paper based bioassay. Vet Parasitol. 2007;144:344-348.
- Chauve C. The poultry red mite Dermanyssus gallinae (De Geer, 1778): current situation and future prospects for control. Vet Parasitol. 1998;79:239-245.
To the Editor:
There are a wide variety of zoonotic diseases that can be transmitted from birds to humans. Pigeons, chickens, starlings, canaries, and parakeets are known reservoirs of one particular zoonotic infection caused by the parasitic arthropod Dermanyssus gallinae.1 Dermanyssus gallinae (chicken mite) and Ornithonyssus sylviarum (northern fowl mite) are collectively referred to as bird mites. When these mites are unable to take blood meals from birds, they search out alternative hosts2; in humans, this often leads to the development of pruritic dermatitis.3
A 30-year-old woman presented to our clinic for evaluation of severe generalized pruritus accompanied by a sensation of “bugs on the skin” of 2 weeks’ duration. She noted the pruritus worsened when she was sitting outside on her porch. A few days prior to presentation, she noticed a small, “pinpoint-sized bug” on her arm (<1 mm in size), which she brought in for identification (Figure).
The bug was identified as a bird mite (Dermanyssus gallinae) on light microscopy, which was later confirmed by a medical entomologist. After the diagnosis of bird mite dermatitis was made, the patient noted there was a nest of starlings above the light on her porch. When she later investigated the nest following the current presentation, she noted many small mites crawling around the nest. The nest was removed and her symptoms resolved completely within 2 weeks without treatment.
Bird mites belong to the Arachnida class, under the order Acari. In 1958, Williams4 noted D gallinae’s ability to feed on human blood. Bird mites have 5 stages of development: egg, larva, protonymph, deutonymph, and adult. Protonymphs, deutonymphs, and adults can bite humans for a blood meal.5 Bird mites range from 0.3 to 1 mm in length and have nonsegmented, egg-shaped bodies with 4 pairs of legs. Before taking a blood meal, bird mites generally are a translucent brown color, and appear red when engorged with blood.2 Their small size makes them barely visible to the unaided eye. Of note, D gallinae and O sylviarum can be distinguished from each other based on subtle differences in morphology; for instance, the posterior genitoventral shield of O sylviarum is narrowly rounded, whereas it is broadly rounded in D gallinae. The dorsal shield of O sylviarum abruptly narrows posteriorly but is more smoothly narrowed in D gallinae.6 Additionally, O sylviarum tends to cause more irritating dermatitis in humans than D gallinae.3
Although they can be found worldwide, D gallinae and O sylviarum undergo optimal development at 20°C to 25°C and 70% humidity.3,5,7 Bird mites generally develop over the course of 5 to 12 days; thus, the population of bird mites in a single nest may grow to the tens of thousands before young birds permanently leave. Dermanyssus gallinae can survive for months in abandoned nests without a blood meal, while O sylviarum can survive for several weeks.8 It is important to note that humans are not ideal hosts for bird mites, as they are unable to survive for extended periods of time or reproduce on human hosts.9
When bird mites are no longer able to obtain blood meals from nesting birds, they begin their nocturnal migration to find suitable hosts. Bird nests generally are abandoned in late spring; thus, most patients with bird mite dermatitis present to clinics with bird mite dermatitis in late spring and early summer.10 Mites often travel through cracks in doors, floors, walls, and ceilings but also can gain access to living areas through ventilation ducts and air conditioning units.1 The mite’s bite and crawling on the skin is sometimes noticed by the patient. In general, however, intense itching is not observed until about 1 to 3 days after the mite makes contact with the skin. Patients often report that pruritus is worst at night.9 Papules and vesicles (bite reactions) may accompany the pruritus, and physicians commonly find bloody crust and excoriations in particularly pruritic areas.5 Urticarial plaques and diffuse erythema occasionally also may be present.9 Bird mites sometimes can be scraped from the skin and observed under light microscopy.11 Blood eosinophilia is not found in bird mite dermatitis. On histologic examination, perivascular eosinophilic infiltration can be seen in the upper part of the dermis.12
The differential diagnosis in patients with pruritic dermatitis of unknown origin generally includes scabies, pediculosis, and dermatitis caused by other types of infestation. However, unlike scabies, bird mites do not cause burrows to form on the skin.9 The presence of a bird’s nest near the area where the patient lives places bird mite dermatitis higher in the differential.
Dermanyssus gallinae is a known vector of bacteria (eg, Salmonella, Shigella, Staphylococcus, Spirochaete, Rickettsia, Pasteurella, Chlamydia psittaci, Erysipelothrix rhusiopathiae) as well as the viruses that cause Eastern and Western equine encephalitis and St. Louis encephalitis. Transmission of these bacteria and viruses is known in birds, but transmission to humans has not been reported.2,5,9,13
The management of bird mite dermatitis is straightforward. Usually mites can be successfully removed from the skin simply by bathing. Symptomatic treatment for bites with antihistamines and topical corticosteroids is sometimes but not always necessary.2 Unlike scabies or lice, there is no need for treatment with lindane.1 In terms of the prevention of additional bites, any bird nests located near living areas should be removed. Because bird mites often retreat back to nests between blood meals, insecticide sprays generally are unnecessary in interior spaces. Synthetic pyrethroids (eg, bifenthrin, cyfluthrin, cypermethrin, deltamethrin, cyhalothrin) can be used outside and in attics where nests may be located.2,14,15 However, the ability of bird mites to develop resistance to repeated chemical control could become a future concern.16
Research regarding the true incidence of bird mite dermatitis is lacking. Some researchers believe that the condition is underreported, possibly due to its uncommon environmental origin.3 Reports of bird mite dermatitis in the literature also are scarce. Our case demonstrates the importance of taking a thorough patient history to rule out exposure to bird mites. All patients with pruritic dermatitis of unknown origin should be questioned about possible contact or proximity to bird nests. These simple questions can lead to the correct diagnosis and a treatment plan that will quickly and effectively resolve the pruritic skin eruption.
To the Editor:
There are a wide variety of zoonotic diseases that can be transmitted from birds to humans. Pigeons, chickens, starlings, canaries, and parakeets are known reservoirs of one particular zoonotic infection caused by the parasitic arthropod Dermanyssus gallinae.1 Dermanyssus gallinae (chicken mite) and Ornithonyssus sylviarum (northern fowl mite) are collectively referred to as bird mites. When these mites are unable to take blood meals from birds, they search out alternative hosts2; in humans, this often leads to the development of pruritic dermatitis.3
A 30-year-old woman presented to our clinic for evaluation of severe generalized pruritus accompanied by a sensation of “bugs on the skin” of 2 weeks’ duration. She noted the pruritus worsened when she was sitting outside on her porch. A few days prior to presentation, she noticed a small, “pinpoint-sized bug” on her arm (<1 mm in size), which she brought in for identification (Figure).
The bug was identified as a bird mite (Dermanyssus gallinae) on light microscopy, which was later confirmed by a medical entomologist. After the diagnosis of bird mite dermatitis was made, the patient noted there was a nest of starlings above the light on her porch. When she later investigated the nest following the current presentation, she noted many small mites crawling around the nest. The nest was removed and her symptoms resolved completely within 2 weeks without treatment.
Bird mites belong to the Arachnida class, under the order Acari. In 1958, Williams4 noted D gallinae’s ability to feed on human blood. Bird mites have 5 stages of development: egg, larva, protonymph, deutonymph, and adult. Protonymphs, deutonymphs, and adults can bite humans for a blood meal.5 Bird mites range from 0.3 to 1 mm in length and have nonsegmented, egg-shaped bodies with 4 pairs of legs. Before taking a blood meal, bird mites generally are a translucent brown color, and appear red when engorged with blood.2 Their small size makes them barely visible to the unaided eye. Of note, D gallinae and O sylviarum can be distinguished from each other based on subtle differences in morphology; for instance, the posterior genitoventral shield of O sylviarum is narrowly rounded, whereas it is broadly rounded in D gallinae. The dorsal shield of O sylviarum abruptly narrows posteriorly but is more smoothly narrowed in D gallinae.6 Additionally, O sylviarum tends to cause more irritating dermatitis in humans than D gallinae.3
Although they can be found worldwide, D gallinae and O sylviarum undergo optimal development at 20°C to 25°C and 70% humidity.3,5,7 Bird mites generally develop over the course of 5 to 12 days; thus, the population of bird mites in a single nest may grow to the tens of thousands before young birds permanently leave. Dermanyssus gallinae can survive for months in abandoned nests without a blood meal, while O sylviarum can survive for several weeks.8 It is important to note that humans are not ideal hosts for bird mites, as they are unable to survive for extended periods of time or reproduce on human hosts.9
When bird mites are no longer able to obtain blood meals from nesting birds, they begin their nocturnal migration to find suitable hosts. Bird nests generally are abandoned in late spring; thus, most patients with bird mite dermatitis present to clinics with bird mite dermatitis in late spring and early summer.10 Mites often travel through cracks in doors, floors, walls, and ceilings but also can gain access to living areas through ventilation ducts and air conditioning units.1 The mite’s bite and crawling on the skin is sometimes noticed by the patient. In general, however, intense itching is not observed until about 1 to 3 days after the mite makes contact with the skin. Patients often report that pruritus is worst at night.9 Papules and vesicles (bite reactions) may accompany the pruritus, and physicians commonly find bloody crust and excoriations in particularly pruritic areas.5 Urticarial plaques and diffuse erythema occasionally also may be present.9 Bird mites sometimes can be scraped from the skin and observed under light microscopy.11 Blood eosinophilia is not found in bird mite dermatitis. On histologic examination, perivascular eosinophilic infiltration can be seen in the upper part of the dermis.12
The differential diagnosis in patients with pruritic dermatitis of unknown origin generally includes scabies, pediculosis, and dermatitis caused by other types of infestation. However, unlike scabies, bird mites do not cause burrows to form on the skin.9 The presence of a bird’s nest near the area where the patient lives places bird mite dermatitis higher in the differential.
Dermanyssus gallinae is a known vector of bacteria (eg, Salmonella, Shigella, Staphylococcus, Spirochaete, Rickettsia, Pasteurella, Chlamydia psittaci, Erysipelothrix rhusiopathiae) as well as the viruses that cause Eastern and Western equine encephalitis and St. Louis encephalitis. Transmission of these bacteria and viruses is known in birds, but transmission to humans has not been reported.2,5,9,13
The management of bird mite dermatitis is straightforward. Usually mites can be successfully removed from the skin simply by bathing. Symptomatic treatment for bites with antihistamines and topical corticosteroids is sometimes but not always necessary.2 Unlike scabies or lice, there is no need for treatment with lindane.1 In terms of the prevention of additional bites, any bird nests located near living areas should be removed. Because bird mites often retreat back to nests between blood meals, insecticide sprays generally are unnecessary in interior spaces. Synthetic pyrethroids (eg, bifenthrin, cyfluthrin, cypermethrin, deltamethrin, cyhalothrin) can be used outside and in attics where nests may be located.2,14,15 However, the ability of bird mites to develop resistance to repeated chemical control could become a future concern.16
Research regarding the true incidence of bird mite dermatitis is lacking. Some researchers believe that the condition is underreported, possibly due to its uncommon environmental origin.3 Reports of bird mite dermatitis in the literature also are scarce. Our case demonstrates the importance of taking a thorough patient history to rule out exposure to bird mites. All patients with pruritic dermatitis of unknown origin should be questioned about possible contact or proximity to bird nests. These simple questions can lead to the correct diagnosis and a treatment plan that will quickly and effectively resolve the pruritic skin eruption.
- Regan AM, Metersky ML, Craven DE. Nosocomial dermatitis and pruritus caused by pigeon mite infestation. Arch Intern Med. 1987;147:2185-2187.
- Collgros H, Iglesias-Sancho M, Aldunce MJ, et al. Dermanyssus gallinae (chicken mite): an underdiagnosed environmental infestation. Clin Exp Dermatol. 2013;38:374-377.
- Bellanger AP, Boris C, Foulet F, et al. Nosocomial dermatitis caused by Dermanyssus gallinae. Infect Cont Hosp Ep. 2008;29:282-283.
- Williams RW. An infestation of a human habitation by Dermanyssus gallinae (de Geer, 1778) (Acarina: Dermanyssidae) in New York resulting in sanguisugent attacks upon the occupants. Am J Trop Med Hyg. 1958;7:627-629.
- Akdemir C, Gülcan E, Tanritanir P. Case report: Dermanyssus gallinae in a patient with pruritus and skin lesions. Turkiye Parazitol Derg. 2009;33:242-244.
- DiPalma A, Giangaspero A, Cafiero MA, et al. A gallery of the key characteristics to ease identification of Dermanyssus gallinae (Acari: Gamasida: Dermanyssidae) and allow differentiation from Ornithonyssus sylviarum (Acari: Gamasida: Macronyssidae). Parasites and Vectors. 2012;5:104.
- Maurer V, Baumgartner J. Temperature influence on life table statistics of the chicken mite Dermanyssus gallinae (Acari: Dermanyssidae). Exp Appl Acarol. 1992;15:27-40.
- Orton DI, Warren LJ, Wilkinson JD. Avian mite dermatitis. Clin Exper Dermatol. 2000;25:129-131.
- Auger P, Nantel J, Meunier N, et al. Skin acariasis caused by Dermanyssus gallinae (de Geer): an in-hospital outbreak. Can Med Assoc J. 1979;120:700-703.
- Kong TK, To WK. Bird mite infestation. N Engl J Med. 2006;354:1728.
- Koh WL, Liu TT, Tay YK. Formication due to true parasitic infection: bird mites. Arch Dermatol. 2011;147:508-509.
- Hidano A, Asanuma K. Letter: Acariasis caused by bird mites. Arch Dermatol. 1976;112:881-882.
- Valiente Moro C, Chauve C, Zenner L. Experimental infection of Salmonella Enteritidis by the poultry red mite, Dermanyssus gallinae. Vet Parasitol. 2007;146:329-336.
- Fletcher MG, Axtell RC. Susceptibilities of northern fowl mite, Ornithonyssus sylviarum (Acarina: Macronyssidae),and chicken mite, Dermanyssus gallinae (Acarina: Dermanyssidae), to selected acaricides. Exp Appl Acarol. 1991;13:137-142.
- Thind BB, Ford HL. Assessment of susceptibility of the poultry red mite Dermanyssus gallinae (Acari: Dermanyssidae) to some acaricides using an adapted filter paper based bioassay. Vet Parasitol. 2007;144:344-348.
- Chauve C. The poultry red mite Dermanyssus gallinae (De Geer, 1778): current situation and future prospects for control. Vet Parasitol. 1998;79:239-245.
- Regan AM, Metersky ML, Craven DE. Nosocomial dermatitis and pruritus caused by pigeon mite infestation. Arch Intern Med. 1987;147:2185-2187.
- Collgros H, Iglesias-Sancho M, Aldunce MJ, et al. Dermanyssus gallinae (chicken mite): an underdiagnosed environmental infestation. Clin Exp Dermatol. 2013;38:374-377.
- Bellanger AP, Boris C, Foulet F, et al. Nosocomial dermatitis caused by Dermanyssus gallinae. Infect Cont Hosp Ep. 2008;29:282-283.
- Williams RW. An infestation of a human habitation by Dermanyssus gallinae (de Geer, 1778) (Acarina: Dermanyssidae) in New York resulting in sanguisugent attacks upon the occupants. Am J Trop Med Hyg. 1958;7:627-629.
- Akdemir C, Gülcan E, Tanritanir P. Case report: Dermanyssus gallinae in a patient with pruritus and skin lesions. Turkiye Parazitol Derg. 2009;33:242-244.
- DiPalma A, Giangaspero A, Cafiero MA, et al. A gallery of the key characteristics to ease identification of Dermanyssus gallinae (Acari: Gamasida: Dermanyssidae) and allow differentiation from Ornithonyssus sylviarum (Acari: Gamasida: Macronyssidae). Parasites and Vectors. 2012;5:104.
- Maurer V, Baumgartner J. Temperature influence on life table statistics of the chicken mite Dermanyssus gallinae (Acari: Dermanyssidae). Exp Appl Acarol. 1992;15:27-40.
- Orton DI, Warren LJ, Wilkinson JD. Avian mite dermatitis. Clin Exper Dermatol. 2000;25:129-131.
- Auger P, Nantel J, Meunier N, et al. Skin acariasis caused by Dermanyssus gallinae (de Geer): an in-hospital outbreak. Can Med Assoc J. 1979;120:700-703.
- Kong TK, To WK. Bird mite infestation. N Engl J Med. 2006;354:1728.
- Koh WL, Liu TT, Tay YK. Formication due to true parasitic infection: bird mites. Arch Dermatol. 2011;147:508-509.
- Hidano A, Asanuma K. Letter: Acariasis caused by bird mites. Arch Dermatol. 1976;112:881-882.
- Valiente Moro C, Chauve C, Zenner L. Experimental infection of Salmonella Enteritidis by the poultry red mite, Dermanyssus gallinae. Vet Parasitol. 2007;146:329-336.
- Fletcher MG, Axtell RC. Susceptibilities of northern fowl mite, Ornithonyssus sylviarum (Acarina: Macronyssidae),and chicken mite, Dermanyssus gallinae (Acarina: Dermanyssidae), to selected acaricides. Exp Appl Acarol. 1991;13:137-142.
- Thind BB, Ford HL. Assessment of susceptibility of the poultry red mite Dermanyssus gallinae (Acari: Dermanyssidae) to some acaricides using an adapted filter paper based bioassay. Vet Parasitol. 2007;144:344-348.
- Chauve C. The poultry red mite Dermanyssus gallinae (De Geer, 1778): current situation and future prospects for control. Vet Parasitol. 1998;79:239-245.
Bias and knowing too much about your patient
Years ago, I had a colleague who’d once worked for the prison system, treating people who were some of the more dangerous elements of society.
Once I asked if he’d ever gotten curious about what they were in for. He answered that, while he was always curious, he never asked. He felt as if knowing might prejudice his care. Since a key part of being a doctor is being impartial and objective, he was afraid that knowing about their previous heinous behavior would make him less concerned about treating them properly. And I agree.
When I was a younger doctor, I’d sometimes Google patients. I’d be curious about their backgrounds, or I wanted to see if there was anything on their social media I should be aware of they hadn’t told me. Maybe something like “I scored 20 percs off a neurologist today!”
I stopped after a while, and haven’t done it since. I never saw anything that would affect my treatment plan. I did, however, often learn about their political and religious views, some of which were distasteful to me. I respect anyone’s right to have an opinion, but that doesn’t mean I have to agree with them.
Like I’ve written before, I specifically avoid any discussion of religion or politics with my patients because doing so can lead to antagonism and dislike, with the potential to impact my objectivity.
The same can be said about what else you might learn online: their habits and hobbies, unflattering pictures, stories about their backgrounds, etc. All of those things can, in the right circumstances, lead to a bias against them. Perhaps it may just exist subconsciously, but it’s still there. A recent Medscape report noted the number of physicians who admitted having biases against patients, as well as the things that can trigger our visceral reactions: emotional state, weight, and intelligence, to name a few. We try hard to overcome negative feelings to provide proper care, but are still human and 100% objectivity is often difficult.
To me, Googling a patient became the same thing as asking inmates what they’d been locked up for: You learn things about them that might change how you view and care for them.
The only way to effectively treat patients is to see them as just people, like yourself. Knowing too much about their background that isn’t medically relevant is just asking for trouble.
I’d rather know less and be more objective.
Dr. Block has a solo neurology practice in Scottsdale, Ariz.
Years ago, I had a colleague who’d once worked for the prison system, treating people who were some of the more dangerous elements of society.
Once I asked if he’d ever gotten curious about what they were in for. He answered that, while he was always curious, he never asked. He felt as if knowing might prejudice his care. Since a key part of being a doctor is being impartial and objective, he was afraid that knowing about their previous heinous behavior would make him less concerned about treating them properly. And I agree.
When I was a younger doctor, I’d sometimes Google patients. I’d be curious about their backgrounds, or I wanted to see if there was anything on their social media I should be aware of they hadn’t told me. Maybe something like “I scored 20 percs off a neurologist today!”
I stopped after a while, and haven’t done it since. I never saw anything that would affect my treatment plan. I did, however, often learn about their political and religious views, some of which were distasteful to me. I respect anyone’s right to have an opinion, but that doesn’t mean I have to agree with them.
Like I’ve written before, I specifically avoid any discussion of religion or politics with my patients because doing so can lead to antagonism and dislike, with the potential to impact my objectivity.
The same can be said about what else you might learn online: their habits and hobbies, unflattering pictures, stories about their backgrounds, etc. All of those things can, in the right circumstances, lead to a bias against them. Perhaps it may just exist subconsciously, but it’s still there. A recent Medscape report noted the number of physicians who admitted having biases against patients, as well as the things that can trigger our visceral reactions: emotional state, weight, and intelligence, to name a few. We try hard to overcome negative feelings to provide proper care, but are still human and 100% objectivity is often difficult.
To me, Googling a patient became the same thing as asking inmates what they’d been locked up for: You learn things about them that might change how you view and care for them.
The only way to effectively treat patients is to see them as just people, like yourself. Knowing too much about their background that isn’t medically relevant is just asking for trouble.
I’d rather know less and be more objective.
Dr. Block has a solo neurology practice in Scottsdale, Ariz.
Years ago, I had a colleague who’d once worked for the prison system, treating people who were some of the more dangerous elements of society.
Once I asked if he’d ever gotten curious about what they were in for. He answered that, while he was always curious, he never asked. He felt as if knowing might prejudice his care. Since a key part of being a doctor is being impartial and objective, he was afraid that knowing about their previous heinous behavior would make him less concerned about treating them properly. And I agree.
When I was a younger doctor, I’d sometimes Google patients. I’d be curious about their backgrounds, or I wanted to see if there was anything on their social media I should be aware of they hadn’t told me. Maybe something like “I scored 20 percs off a neurologist today!”
I stopped after a while, and haven’t done it since. I never saw anything that would affect my treatment plan. I did, however, often learn about their political and religious views, some of which were distasteful to me. I respect anyone’s right to have an opinion, but that doesn’t mean I have to agree with them.
Like I’ve written before, I specifically avoid any discussion of religion or politics with my patients because doing so can lead to antagonism and dislike, with the potential to impact my objectivity.
The same can be said about what else you might learn online: their habits and hobbies, unflattering pictures, stories about their backgrounds, etc. All of those things can, in the right circumstances, lead to a bias against them. Perhaps it may just exist subconsciously, but it’s still there. A recent Medscape report noted the number of physicians who admitted having biases against patients, as well as the things that can trigger our visceral reactions: emotional state, weight, and intelligence, to name a few. We try hard to overcome negative feelings to provide proper care, but are still human and 100% objectivity is often difficult.
To me, Googling a patient became the same thing as asking inmates what they’d been locked up for: You learn things about them that might change how you view and care for them.
The only way to effectively treat patients is to see them as just people, like yourself. Knowing too much about their background that isn’t medically relevant is just asking for trouble.
I’d rather know less and be more objective.
Dr. Block has a solo neurology practice in Scottsdale, Ariz.
Patients with Postoperative Myocardial Infarction May Benefit from Higher Transfusion Threshold
Clinical question: Is there an improved 30-day mortality rate if patients receive blood transfusion at higher hematocrit values after postoperative myocardial infarction (MI)?
Background: Prior studies evaluating patients with a history of coronary artery disease (CAD) who undergo non-cardiac surgery have shown similar mortality outcomes with liberal and restrictive transfusion strategies. Data are lacking for transfusion strategies in patients with CAD who experience postoperative MI after non-cardiac surgeries.
Study design: Retrospective cohort.
Setting: Veterans Affairs health system.
Synopsis: The study included 7,361 patients with a history of CAD who underwent non-cardiac surgery whose postoperative hematocrit was between 20% and 30%. Patients were stratified by postoperative hematocrit nadir and presence of postoperative MI. In patients with postoperative MI, transfusion was associated with lower mortality with hematocrit nadir of 20%–24% but not with hematocrit of 24%–27% or 27%–30%. In patients without postoperative MI, transfusion was associated with higher mortality in patients with hematocrit of 27%–30%.
This retrospective study was limited to the VA population of mostly male patients. The sample size was limited. The study was unable to determine if postoperative blood transfusion is a risk for developing MI.
Bottom line: Patients with a history of CAD and MI who have a postoperative MI following non-cardiac surgery may benefit from higher blood transfusion thresholds; however, further controlled studies are needed.
Citation: Hollis RH, Singeltary BA, McMurtrie JT, et al. Blood transfusion and 30-day mortality in patients with coronary artery disease and anemia following noncardiac surgery [published online ahead of print October 7, 2015]. JAMA Surg. doi:10.1001/jamasurg.2015.3420.
Clinical question: Is there an improved 30-day mortality rate if patients receive blood transfusion at higher hematocrit values after postoperative myocardial infarction (MI)?
Background: Prior studies evaluating patients with a history of coronary artery disease (CAD) who undergo non-cardiac surgery have shown similar mortality outcomes with liberal and restrictive transfusion strategies. Data are lacking for transfusion strategies in patients with CAD who experience postoperative MI after non-cardiac surgeries.
Study design: Retrospective cohort.
Setting: Veterans Affairs health system.
Synopsis: The study included 7,361 patients with a history of CAD who underwent non-cardiac surgery whose postoperative hematocrit was between 20% and 30%. Patients were stratified by postoperative hematocrit nadir and presence of postoperative MI. In patients with postoperative MI, transfusion was associated with lower mortality with hematocrit nadir of 20%–24% but not with hematocrit of 24%–27% or 27%–30%. In patients without postoperative MI, transfusion was associated with higher mortality in patients with hematocrit of 27%–30%.
This retrospective study was limited to the VA population of mostly male patients. The sample size was limited. The study was unable to determine if postoperative blood transfusion is a risk for developing MI.
Bottom line: Patients with a history of CAD and MI who have a postoperative MI following non-cardiac surgery may benefit from higher blood transfusion thresholds; however, further controlled studies are needed.
Citation: Hollis RH, Singeltary BA, McMurtrie JT, et al. Blood transfusion and 30-day mortality in patients with coronary artery disease and anemia following noncardiac surgery [published online ahead of print October 7, 2015]. JAMA Surg. doi:10.1001/jamasurg.2015.3420.
Clinical question: Is there an improved 30-day mortality rate if patients receive blood transfusion at higher hematocrit values after postoperative myocardial infarction (MI)?
Background: Prior studies evaluating patients with a history of coronary artery disease (CAD) who undergo non-cardiac surgery have shown similar mortality outcomes with liberal and restrictive transfusion strategies. Data are lacking for transfusion strategies in patients with CAD who experience postoperative MI after non-cardiac surgeries.
Study design: Retrospective cohort.
Setting: Veterans Affairs health system.
Synopsis: The study included 7,361 patients with a history of CAD who underwent non-cardiac surgery whose postoperative hematocrit was between 20% and 30%. Patients were stratified by postoperative hematocrit nadir and presence of postoperative MI. In patients with postoperative MI, transfusion was associated with lower mortality with hematocrit nadir of 20%–24% but not with hematocrit of 24%–27% or 27%–30%. In patients without postoperative MI, transfusion was associated with higher mortality in patients with hematocrit of 27%–30%.
This retrospective study was limited to the VA population of mostly male patients. The sample size was limited. The study was unable to determine if postoperative blood transfusion is a risk for developing MI.
Bottom line: Patients with a history of CAD and MI who have a postoperative MI following non-cardiac surgery may benefit from higher blood transfusion thresholds; however, further controlled studies are needed.
Citation: Hollis RH, Singeltary BA, McMurtrie JT, et al. Blood transfusion and 30-day mortality in patients with coronary artery disease and anemia following noncardiac surgery [published online ahead of print October 7, 2015]. JAMA Surg. doi:10.1001/jamasurg.2015.3420.
Nebulized Hypertonic Saline Does Not Improve Outcomes for Non-ICU Infants with Acute Bronchiolitis
Clinical question: Does the use of nebulized 3% hypertonic saline shorten length of stay (LOS) in infants hospitalized with acute bronchiolitis?
Background: Acute bronchiolitis is a disease primarily of infants and young children, triggered by a viral infection that leads to variable inflammation, edema, and inspissated mucus in the lower airways. Although bronchiolitis is the most common cause of hospitalization in children under the age of two, few interventions have been shown to improve patient-level outcomes.
Hypertonic saline (generally 3%) has been one of the few interventions that has improved outcomes in some studies, leading the most recent American Academy of Pediatrics (AAP) clinical practice guideline (CPG) to state that nebulized hypertonic saline may be considered for infants and children hospitalized for bronchiolitis. The studies cited in this CPG statement were heterogeneous, with many of them performed in Europe, where the LOS for bronchiolitis is generally longer than in the U.S. In addition, most of the studies administered hypertonic saline (HS) with a bronchodilator, confounding the outcomes with an intervention not recommended in the most recent bronchiolitis CPG.
Study design: Prospective, randomized controlled, double-blinded, parallel-group study.
Setting: Urban, tertiary-care, 136-bed children’s hospital.
Synopsis: Infants 4 points received a bronchodilator and were withdrawn from the study.
Of the 227 patients enrolled after application of inclusion and exclusion criteria, 113 were randomized to receive HS and 114 to NS. Twenty patients in the HS group and 17 in the NS group discontinued intervention due to ICU transfer, provider choice to use albuterol, parental request, or protocol deviation, but patients were analyzed by intention-to-treat (ITT) assignments. No significant difference in LOS between the HS and NS groups was found, either by the traditional definition or the treatment-to-discharge order definition. No significant differences were found in secondary outcomes between the two groups, including readmission rates or clinical worsening. In addition, pre- to post-treatment RDAI score changes were not significantly different for HS versus NS.
Bottom line: Treating infants
Citation: Silver AH, Esteban-Cruciani N, Azzarone G, et al. 3% hypertonic saline versus normal saline in inpatient bronchiolitis: a randomized controlled trial. Pediatrics. 2015;136(6):1036-1043. TH
Clinical question: Does the use of nebulized 3% hypertonic saline shorten length of stay (LOS) in infants hospitalized with acute bronchiolitis?
Background: Acute bronchiolitis is a disease primarily of infants and young children, triggered by a viral infection that leads to variable inflammation, edema, and inspissated mucus in the lower airways. Although bronchiolitis is the most common cause of hospitalization in children under the age of two, few interventions have been shown to improve patient-level outcomes.
Hypertonic saline (generally 3%) has been one of the few interventions that has improved outcomes in some studies, leading the most recent American Academy of Pediatrics (AAP) clinical practice guideline (CPG) to state that nebulized hypertonic saline may be considered for infants and children hospitalized for bronchiolitis. The studies cited in this CPG statement were heterogeneous, with many of them performed in Europe, where the LOS for bronchiolitis is generally longer than in the U.S. In addition, most of the studies administered hypertonic saline (HS) with a bronchodilator, confounding the outcomes with an intervention not recommended in the most recent bronchiolitis CPG.
Study design: Prospective, randomized controlled, double-blinded, parallel-group study.
Setting: Urban, tertiary-care, 136-bed children’s hospital.
Synopsis: Infants 4 points received a bronchodilator and were withdrawn from the study.
Of the 227 patients enrolled after application of inclusion and exclusion criteria, 113 were randomized to receive HS and 114 to NS. Twenty patients in the HS group and 17 in the NS group discontinued intervention due to ICU transfer, provider choice to use albuterol, parental request, or protocol deviation, but patients were analyzed by intention-to-treat (ITT) assignments. No significant difference in LOS between the HS and NS groups was found, either by the traditional definition or the treatment-to-discharge order definition. No significant differences were found in secondary outcomes between the two groups, including readmission rates or clinical worsening. In addition, pre- to post-treatment RDAI score changes were not significantly different for HS versus NS.
Bottom line: Treating infants
Citation: Silver AH, Esteban-Cruciani N, Azzarone G, et al. 3% hypertonic saline versus normal saline in inpatient bronchiolitis: a randomized controlled trial. Pediatrics. 2015;136(6):1036-1043. TH
Clinical question: Does the use of nebulized 3% hypertonic saline shorten length of stay (LOS) in infants hospitalized with acute bronchiolitis?
Background: Acute bronchiolitis is a disease primarily of infants and young children, triggered by a viral infection that leads to variable inflammation, edema, and inspissated mucus in the lower airways. Although bronchiolitis is the most common cause of hospitalization in children under the age of two, few interventions have been shown to improve patient-level outcomes.
Hypertonic saline (generally 3%) has been one of the few interventions that has improved outcomes in some studies, leading the most recent American Academy of Pediatrics (AAP) clinical practice guideline (CPG) to state that nebulized hypertonic saline may be considered for infants and children hospitalized for bronchiolitis. The studies cited in this CPG statement were heterogeneous, with many of them performed in Europe, where the LOS for bronchiolitis is generally longer than in the U.S. In addition, most of the studies administered hypertonic saline (HS) with a bronchodilator, confounding the outcomes with an intervention not recommended in the most recent bronchiolitis CPG.
Study design: Prospective, randomized controlled, double-blinded, parallel-group study.
Setting: Urban, tertiary-care, 136-bed children’s hospital.
Synopsis: Infants 4 points received a bronchodilator and were withdrawn from the study.
Of the 227 patients enrolled after application of inclusion and exclusion criteria, 113 were randomized to receive HS and 114 to NS. Twenty patients in the HS group and 17 in the NS group discontinued intervention due to ICU transfer, provider choice to use albuterol, parental request, or protocol deviation, but patients were analyzed by intention-to-treat (ITT) assignments. No significant difference in LOS between the HS and NS groups was found, either by the traditional definition or the treatment-to-discharge order definition. No significant differences were found in secondary outcomes between the two groups, including readmission rates or clinical worsening. In addition, pre- to post-treatment RDAI score changes were not significantly different for HS versus NS.
Bottom line: Treating infants
Citation: Silver AH, Esteban-Cruciani N, Azzarone G, et al. 3% hypertonic saline versus normal saline in inpatient bronchiolitis: a randomized controlled trial. Pediatrics. 2015;136(6):1036-1043. TH
Uncovering the origin of ALCL
In studying a mouse model of anaplastic large cell lymphoma (ALCL), investigators may have discovered how the disease develops.
“The origins of ALCL could be traced to a gene disorder in the development of blood-producing stem cells which are located in the thymus,” explained study author Lukas Kenner, MD, of the Medical University of Vienna in Austria.
He and his colleagues found that ALCL began in early thymocytes before T-cell receptor (TCR) β-rearrangement.
And the spread of ALCL required a major change in the TCR. A TCR was required for thymic emigration and peripheral tumor development, but the TCR had to be downregulated for T-cell lymphomagenesis.
In mice in which ALCL had spread, the TCR was initially required but was then lost from the surface of lymphoma cells.
“This means that the TCR molecule has a strong suppressive effect on tumor development,” Dr Kenner said.
He and his colleagues recounted these findings in Nature Communications.
“We now have a better understanding of the origin of this type of lymphoma and the crucial role played by the major changes to the immune system in the spread of this tumor through the body,” said study author Suzanne Turner, PhD, of the University of Cambridge in the UK.
“With this knowledge, we can better combat the cancer genes which are key to the formation and development of lymphomas and, in the future, develop new treatments which offer a better possibility of finding a long-term cure.”
“Current chemotherapy is particularly exhausting for children and adolescents, especially if a relapse occurs and additional treatment is needed,” Dr Kenner added.
“Our new findings about this lymphoma enable the development of more efficient and less toxic medicines, with which every child will soon be able to return to a normal life after treatment.”
In studying a mouse model of anaplastic large cell lymphoma (ALCL), investigators may have discovered how the disease develops.
“The origins of ALCL could be traced to a gene disorder in the development of blood-producing stem cells which are located in the thymus,” explained study author Lukas Kenner, MD, of the Medical University of Vienna in Austria.
He and his colleagues found that ALCL began in early thymocytes before T-cell receptor (TCR) β-rearrangement.
And the spread of ALCL required a major change in the TCR. A TCR was required for thymic emigration and peripheral tumor development, but the TCR had to be downregulated for T-cell lymphomagenesis.
In mice in which ALCL had spread, the TCR was initially required but was then lost from the surface of lymphoma cells.
“This means that the TCR molecule has a strong suppressive effect on tumor development,” Dr Kenner said.
He and his colleagues recounted these findings in Nature Communications.
“We now have a better understanding of the origin of this type of lymphoma and the crucial role played by the major changes to the immune system in the spread of this tumor through the body,” said study author Suzanne Turner, PhD, of the University of Cambridge in the UK.
“With this knowledge, we can better combat the cancer genes which are key to the formation and development of lymphomas and, in the future, develop new treatments which offer a better possibility of finding a long-term cure.”
“Current chemotherapy is particularly exhausting for children and adolescents, especially if a relapse occurs and additional treatment is needed,” Dr Kenner added.
“Our new findings about this lymphoma enable the development of more efficient and less toxic medicines, with which every child will soon be able to return to a normal life after treatment.”
In studying a mouse model of anaplastic large cell lymphoma (ALCL), investigators may have discovered how the disease develops.
“The origins of ALCL could be traced to a gene disorder in the development of blood-producing stem cells which are located in the thymus,” explained study author Lukas Kenner, MD, of the Medical University of Vienna in Austria.
He and his colleagues found that ALCL began in early thymocytes before T-cell receptor (TCR) β-rearrangement.
And the spread of ALCL required a major change in the TCR. A TCR was required for thymic emigration and peripheral tumor development, but the TCR had to be downregulated for T-cell lymphomagenesis.
In mice in which ALCL had spread, the TCR was initially required but was then lost from the surface of lymphoma cells.
“This means that the TCR molecule has a strong suppressive effect on tumor development,” Dr Kenner said.
He and his colleagues recounted these findings in Nature Communications.
“We now have a better understanding of the origin of this type of lymphoma and the crucial role played by the major changes to the immune system in the spread of this tumor through the body,” said study author Suzanne Turner, PhD, of the University of Cambridge in the UK.
“With this knowledge, we can better combat the cancer genes which are key to the formation and development of lymphomas and, in the future, develop new treatments which offer a better possibility of finding a long-term cure.”
“Current chemotherapy is particularly exhausting for children and adolescents, especially if a relapse occurs and additional treatment is needed,” Dr Kenner added.
“Our new findings about this lymphoma enable the development of more efficient and less toxic medicines, with which every child will soon be able to return to a normal life after treatment.”
Real‐Time Patient Experience Surveys
In 2010, the Centers for Medicare and Medicaid Services implemented value‐based purchasing, a payment model that incentivizes hospitals for reaching certain quality and patient experience thresholds and penalizes those that do not, in part on the basis of patient satisfaction scores.[1] Although low patient satisfaction scores will adversely affect institutions financially, they also reflect patients' perceptions of their care. Some studies suggest that hospitals with higher patient satisfaction scores score higher overall on clinical care processes such as core measures compliance, readmission rates, lower mortality rates, and other quality‐of‐care metrics.[2, 3, 4, 5, 6, 7, 8, 9, 10, 11]
The Hospital Consumer Assessment of Healthcare Providers and Systems (HCAHPS) survey assesses patients' experience following their hospital stay.[1] The percent of top box scores (ie, response of always on a four point scale, or scores of 9 or 10 on a 10‐point scale) are utilized to compare hospitals and determine the reimbursement or penalty a hospital will receive. Although these scores are available to the public on the Hospital Compare website,[12] physicians may not know how their hospital is ranked or how they are individually perceived by their patients. Additionally, these surveys are typically conducted 48 hours to 6 weeks after patients are discharged, and the results are distributed back to the hospitals well after the time that care was provided, thereby offering providers no chance of improving patient satisfaction during a given hospital stay.
Institutions across the country are trying to improve their HCAHPS scores, but there is limited research identifying specific measures providers can implement. Some studies have suggested that utilizing etiquette‐based communication and sitting at the bedside[13, 14] may help improve patient experience with their providers, and more recently, it has been suggested that providing real‐time deidentified patient experience survey results with education and a rewards/emncentive system to residents may help as well.[15]
Surveys conducted during a patient's hospitalization can offer real‐time actionable feedback to providers. We performed a quality‐improvement project that was designed to determine if real‐time feedback to hospitalist physicians, followed by coaching, and revisits to the patients' bedside could improve the results recorded on provider‐specific patient surveys and/or patients' HCAHPS scores or percentile rankings.
METHODS
Design
This was a prospective, randomized quality‐improvement initiative that was approved by the Colorado Multiple Institutional Review Board and conducted at Denver Health, a 525‐bed university‐affiliated public safety net hospital. The initiative was conducted on both teaching and nonteaching general internal medicine services, which typically have a daily census of between 10 and 15 patients. No protocol changes occurred during the study.
Participants
Participants included all English‐ or Spanish‐speaking patients who were hospitalized on a general internal medicine service, had been admitted within the 2 days prior to enrollment, and had a hospitalist as their attending physician. Patients were excluded if they were enrolled in the study during a previous hospitalization, refused to participate, lacked capacity to participate, had hearing or speech impediments precluding regular conversation, were prisoners, if their clinical condition precluded participation, or their attending was an investigator in the project.
Intervention
Participants were prescreened by investigators by reviewing team sign‐outs to determine if patients had any exclusion criteria. Investigators attempted to survey each patient who met inclusion criteria on a daily basis between 9:00 am and 11:00 am. An investigator administered the survey to each patient verbally using scripted language. Patients were asked to rate how well their doctors were listening to them, explaining what they wanted to know, and whether the doctors were being friendly and helpful, all questions taken from a survey that was available on the US Department of Health and Human Services website (to be referred to as here forward daily survey).[16] We converted the original 5‐point Likert scale used in this survey to a 4‐point scale by removing the option of ok, leaving participants the options of poor, fair, good, or great. Patients were also asked to provide any personalized feedback they had, and these comments were recorded in writing by the investigator.
After being surveyed on day 1, patients were randomized to an intervention or control group using an automated randomization module in Research Electronic Data Capture (REDCap).[17] Patients in both groups who did not provide answers to all 3 questions that qualified as being top box (ie, great) were resurveyed on a daily basis until their responses were all top box or they were discharged, met exclusion criteria, or had been surveyed for a total of 4 consecutive days. In the pilot phase of this study, we found that if patients reported all top box scores on the initial survey their responses typically did not change over time, and the patients became frustrated if asked the same questions again when the patient felt there was not room for improvement. Accordingly, we elected to stop surveying patients when all top box responses were reported.
The attending hospitalist caring for each patient in the intervention group was given feedback about their patients' survey results (both their scores and any specific comments) on a daily basis. Feedback was provided in person by 1 of the investigators. The hospitalist also received an automatically generated electronic mail message with the survey results at 11:00 am on each study day. After informing the hospitalists of the patients' scores, the investigator provided a brief education session that included discussing Denver Health's most recent HCAHPS scores, value‐based purchasing, and the financial consequences of poor patient satisfaction scores. The investigator then coached the hospitalist on etiquette‐based communication,[18, 19] suggested that they sit down when communicating with their patients,[19, 20] and then asked the hospitalist to revisit each patient to discuss how the team could improve in any of the 3 areas where the patient did not give a top box score. These educational sessions were conducted in person and lasted a maximum of 5 minutes. An investigator followed up with each hospitalist the following day to determine whether the revisit occurred. Hospitalists caring for patients who were randomized to the control group were not given real‐time feedback or coaching and were not asked to revisit patients.
A random sample of patients surveyed for this initiative also received HCAHPS surveys 48 hours to 6 weeks following their hospital discharge, according to the standard methodology used to acquire HCAHPS data,[21] by an outside vendor contracted by Denver Health. Our vendor conducted these surveys via telephone in English or Spanish.
Outcomes
The primary outcome was the proportion of patients in each group who reported top box scores on the daily surveys. Secondary outcomes included the percent change for the scores recorded for 3 provider‐specific questions from the daily survey, the median top box HCAHPS scores for the 3 provider related questions and overall hospital rating, and the HCAHPS percentiles of top box scores for these questions.
Sample Size
The sample size for this intervention assumed that the proportion of patients whose treating physicians did not receive real‐time feedback who rated their providers as top box would be 75%, and that the effect of providing real‐time feedback would increase this proportion to 85% on the daily surveys. To have 80% power with a type 1 error of 0.05, we estimated a need to enroll 430 patients, 215 in each group.
Statistics
Data were collected and managed using a secure, Web‐based electronic data capture tool hosted at Denver Health (REDCap), which is designed to support data collection for research studies providing: (1) an intuitive interface for validated data entry, (2) audit trails for tracking data manipulation and export procedures, (3) automated export procedures for seamless data downloads to common statistical packages, and (4) procedures for importing data from external sources.[17]
A 2 test was used to compare the proportion of patients in the 2 groups who reported great scores for each question on the study survey on the first and last day. With the intent of providing a framework for understanding the effect real‐time feedback could have on patient experience, a secondary analysis of HCAHPS results was conducted using several different methods.
First, the proportion of patients in the 2 groups who reported scores of 9 or 10 for the overall hospital rating question or reported always for each doctor communication question on the HCHAPS survey was compared using a 2. Second, to allow for detection of differences in a sample with a smaller N, the median overall hospital rating scores from the HCAHPS survey reported by patients in the 2 groups who completed a survey following discharge were compared using a Wilcoxon rank sum test. Lastly, to place changes in proportion into a larger context (ie, how these changes would relate to value‐based purchasing), HCAHPS scores were converted to percentiles of national performance using the 2014 percentile rankings obtained from the external vendor that conducts the HCAHPS surveys for our hospital and compared between the intervention and control groups using a Wilcoxon rank sum test.
All comments collected from patients during their daily surveys were reviewed, and key words were abstracted from each comment. These key words were sorted and reviewed to categorize recurring key words into themes. Exemplars were then selected for each theme derived from patient comments.
RESULTS
From April 14, 2014 to September 19, 2014, we enrolled 227 patients in the control group and 228 in the intervention group (Figure 1). Patient demographics are summarized in Table 1. Of the 132 patients in the intervention group who reported anything less than top box scores for any of the 3 questions (thus prompting a revisit by their provider), 106 (80%) were revisited by their provider at least once during their hospitalization.
| All Patients | HCAHPS Patients | |||
|---|---|---|---|---|
| Control, N = 227 | Intervention, N = 228 | Control, N = 35 | Intervention, N = 30 | |
| ||||
| Age, mean SD | 55 14 | 55 15 | 55 15 | 57 16 |
| Gender | ||||
| Male | 126 (60) | 121 (55) | 20 (57) | 12 (40) |
| Female | 85 (40) | 98 (45) | 15(43) | 18 (60) |
| Race/ethnicity | ||||
| Hispanic | 84 (40) | 90 (41) | 17 (49) | 12 (40) |
| Black | 38 (18) | 28 (13) | 6 (17) | 7 (23) |
| White | 87 (41) | 97 (44) | 12 (34) | 10 (33) |
| Other | 2 (1) | 4 (2) | 0 (0) | 1 (3) |
| Payer | ||||
| Medicare | 65 (29) | 82 (36) | 15 (43) | 12 (40) |
| Medicaid | 122 (54) | 108 (47) | 17 (49) | 14 (47) |
| Commercial | 12 (5) | 15 (7) | 1 (3) | 1 (3) |
| Medically indigent | 4 (2) | 7 (3) | 0 (0) | 3 (10) |
| Self‐pay | 5 (2) | 4 (2) | 1 (3) | 0 (0) |
| Other/unknown | 19 (8) | 12 (5) | 0 (0) | 0 (0) |
| Team | ||||
| Teaching | 187 (82) | 196 (86) | 27 (77) | 24 (80) |
| Nonteaching | 40 (18) | 32 (14) | 8 (23) | 6 (20) |
| Top 5 primary discharge diagnoses* | ||||
| Septicemia | 26 (11) | 34 (15) | 3 (9) | 5 (17) |
| Heart failure | 14 (6) | 13 (6) | 2 (6) | |
| Acute pancreatitis | 12 (5) | 9 (4) | 3 (9) | 2 (7) |
| Diabetes mellitus | 11 (5) | 8 (4) | 2 (6) | |
| Alcohol withdrawal | 9 (4) | |||
| Cellulitis | 7 (3) | 2 (7) | ||
| Pulmonary embolism | 2 (7) | |||
| Chest pain | 2 (7) | |||
| Atrial fibrillation | 2 (6) | |||
| Length of stay, median (IQR) | 3 (2, 5) | 3 (2, 5) | 3 (2, 5) | 3 (2, 4) |
| Charlson Comorbidity Index, median (IQR) | 1 (0, 3) | 2 (0, 3) | 1 (0, 3) | 1.5 (1, 3) |
Daily Surveys
The proportion of patients in both study groups reporting top box scores tended to increase from the first day to the last day of the survey (Figure 2); however, we found no statistically significant differences between the proportion of patients who reported top box scores on first day or last day in the intervention group compared to the control group. The comments made by the patients are summarized in Supporting Table 1 in the online version of this article.
HCAHPS Scores
The proportion of top box scores from the HCAHPS surveys were higher, though not statistically significant, for all 3 provider‐specific questions and for the overall hospital rating for patients whose hospitalists received real‐time feedback (Table 2). The median [interquartile range] score for the overall hospital rating was higher for patients in the intervention group compared with those in the control group, (10 [9, 10] vs 9 [8, 10], P = 0.04]. After converting the HCAHPS scores to percentiles, we found considerably higher rankings for all 3 provider‐related questions and for the overall hospital rating in the intervention group compared to the control group (P = 0.02 for overall differences in percentiles [Table 2]).
| HCAHPS Questions | Proportion Top Box* | Percentile Rank | ||
|---|---|---|---|---|
| Control, N = 35 | Intervention, N = 30 | Control, N = 35 | Intervention, N = 30 | |
| ||||
| Overall hospital rating | 61% | 80% | 6 | 87 |
| Courtesy/respect | 86% | 93% | 23 | 88 |
| Clear communication | 77% | 80% | 39 | 60 |
| Listening | 83% | 90% | 57 | 95 |
No adverse events occurred during the course of the study in either group.
DISCUSSION
The important findings of this study were that (1) daily patient satisfaction scores improved from first day to last day regardless of study group, (2) patients whose providers received real‐time feedback had a trend toward higher HCAHPS proportions for the 3 provider‐related questions as well as the overall rating of the hospital but were not statistically significant, (3) the percentile differences in these 3 questions as well as the overall rating of the hospital were significantly higher in the intervention group as was the median score for the overall hospital rating.
Our original sample size calculation was based upon our own preliminary data, indicating that our baseline top box scores for the daily survey was around 75%. The daily survey top box score on the first day was, however, much lower (Figure 2). Accordingly, although we did not find a significant difference in these daily scores, we were underpowered to find such a difference. Additionally, because only a small percentage of patients are selected for the HCAHPS survey, our ability to detect a difference in this secondary outcome was also limited. We felt that it was important to analyze the percentile comparisons in addition to the proportion of top box scores on the HCAHPS, because the metrics for value‐based purchasing are based upon, in part, how a hospital system compares to other systems. Finally, to improve our power to detect a difference given a small sample size, we converted the scoring system for overall hospital ranking to a continuous variable, which again was noted to be significant.
To our knowledge, this is the first randomized investigation designed to assess the effect of real‐time, patient‐specific feedback to physicians. Real‐time feedback is increasingly being incorporated into medical practice, but there is only limited information available describing how this type of feedback affects outcomes.[22, 23, 24] Banka et al.[15] found that HCAHPS scores improved as a result of real‐time feedback given to residents, but the study was not randomized, utilized a pre‐post design that resulted in there being differences between the patients studied before and after the intervention, and did not provide patient‐specific data to the residents. Tabib et al.[25] found that operating costs decreased 17% after instituting real‐time feedback to providers about these costs. Reeves et al.[26] conducted a cluster randomized trial of a patient feedback survey that was designed to improve nursing care, but the results were reviewed by the nurses several months after patients had been discharged.
The differences in median top box scores and percentile rank that we observed could have resulted from the real‐time feedback, the educational coaching, the fact that the providers revisited the majority of the patients, or a combination of all of the above. Gross et al.[27] found that longer visits lead to higher satisfaction, though others have not found this to necessarily be the case.[28, 29] Lin et al.[30] found that patient satisfaction was affected by the perceived duration of the visit as well as whether expectations on visit length were met and/or exceeded. Brown et al.[31] found that training providers in communication skills improved the providers perception of their communication skills, although patient experience scores did not improve. We feel that the results seen are more likely a combination thereof as opposed to any 1 component of the intervention.
The most commonly reported complaints or concerns in patients' undirected comments often related to communication issues. Comments on subsequent surveys suggested that patient satisfaction improved over time in the intervention group, indicating that perhaps physicians did try to improve in areas that were highlighted by the real‐time feedback, and that patients perceived the physician efforts to do so (eg, They're doing better than the last time you asked. They sat down and talked to me and listened better. They came back and explained to me about my care. They listened better. They should do this survey at the clinic. See Supporting Table 1 in the online version of this article).
Our study has several limitations. First, we did not randomize providers, and many of our providers (approximately 65%) participated in both the control group and also in the intervention group, and thus received real‐time feedback at some point during the study, which could have affected their overall practice and limited our ability to find a difference between the 2 groups. In an attempt to control for this possibility, the study was conducted on an intermittent basis during the study time frame. Furthermore, the proportion of patients who reported top box scores at the beginning of the study did not have a clear trend of change by the end of the study, suggesting that overall clinician practices with respect to patient satisfaction did not change during this short time period.
Second, only a small number of our patients were randomly selected for the HCAHPS survey, which limited our ability to detect significant differences in HCAHPS proportions. Third, the HCAHPS percentiles at our institution at that time were low. Accordingly, the improvements that we observed in patient satisfaction scores might not be reproducible at institutions with higher satisfactions scores. Fourth, time and resources were needed to obtain patient feedback to provide to providers during this study. There are, however, other ways to obtain feedback that are less resource intensive (eg, electronic feedback, the utilization of volunteers, or partnering this with manager rounding). Finally, the study was conducted at a single, university‐affiliated public teaching hospital and was a quality‐improvement initiative, and thus our results are not generalizable to other institutions.
In conclusion, real‐time feedback of patient experience to their providers, coupled with provider education, coaching, and revisits, seems to improve satisfaction of patients hospitalized on general internal medicine units who were cared for by hospitalists.
Acknowledgements
The authors thank Kate Fagan, MPH, for her excellent technical assistance.
Disclosure: Nothing to report.
- HCAHPS Fact Sheet. 2015. Available at: http://www.hcahpsonline.org/Files/HCAHPS_Fact_Sheet_June_2015.pdf. Accessed August 25, 2015.
- , , , . The relationship between commercial website ratings and traditional hospital performance measures in the USA. BMJ Qual Saf. 2013;22:194–202.
- , , , . Patients' perception of hospital care in the United States. N Engl J Med. 2008;359:1921–1931.
- , , , . The relationship between patients' perception of care and measures of hospital quality and safety. Health Serv Res. 2010;45:1024–1040.
- , , , et al. Relationship between quality of diabetes care and patient satisfaction. J Natl Med Assoc. 2003;95:64–70.
- , , , , . Relationship between patient satisfaction with inpatient care and hospital readmission within 30 days. Am J Manag Care. 2011;17:41–48.
- , , . A systematic review of evidence on the links between patient experience and clinical safety and effectiveness. BMJ Open. 2013;3(1).
- , . The association between satisfaction with services provided in primary care and outcomes in type 2 diabetes mellitus. Diabet Med. 2003;20:486–490.
- , , , et al. Associations between Web‐based patient ratings and objective measures of hospital quality. Arch Intern Med. 2012;172:435–436.
- , , , et al. Patient satisfaction and its relationship with clinical quality and inpatient mortality in acute myocardial infarction. Circ Cardiovasc Qual Outcomes. 2010;3:188–195.
- , , , , . Patients' perceptions of care are associated with quality of hospital care: a survey of 4605 hospitals. Am J Med Qual. 2015;30(4):382–388.
- Centers for Medicare 28:908–913.
- , , , , , . Effect of sitting vs. standing on perception of provider time at bedside: a pilot study. Patient Educ Couns. 2012;86:166–171.
- , , , et al. Improving patient satisfaction through physician education, feedback, and incentives. J Hosp Med. 2015;10:497–502.
- US Department of Health and Human Services. Patient satisfaction survey. Available at: http://bphc.hrsa.gov/policiesregulations/performancemeasures/patientsurvey/surveyform.html. Accessed November 15, 2013.
- , , , , , . Research electronic data capture (REDCap)—a metadata‐driven methodology and workflow process for providing translational research informatics support. J Biomed Inform. 2009;42:377–381.
- . The HCAHPS Handbook. Gulf Breeze, FL: Fire Starter; 2010.
- . Etiquette‐based medicine. N Engl J Med. 2008;358:1988–1989.
- . 5 years after the Kahn's etiquette‐based medicine: a brief checklist proposal for a functional second meeting with the patient. Front Psychol. 2013;4:723.
- Frequently Asked Questions. Hospital Value‐Based Purchasing Program. Available at: http://www.cms.gov/Medicare/Quality‐Initiatives‐Patient‐Assessment‐Instruments/hospital‐value‐based‐purchasing/Downloads/FY‐2013‐Program‐Frequently‐Asked‐Questions‐about‐Hospital‐VBP‐3‐9‐12.pdf. Accessed February 8, 2014.
- , , , . Real‐time patient survey data during routine clinical activities for rapid‐cycle quality improvement. JMIR Med Inform. 2015;3:e13.
- . Mount Sinai launches real‐time patient‐feedback survey tool. Healthcare Informatics website. Available at: http://www.healthcare‐informatics.com/news‐item/mount‐sinai‐launches‐real‐time‐patient‐feedback‐survey‐tool. Accessed August 25, 2015.
- , . Hospitals are finally starting to put real‐time data to use. Harvard Business Review website. Available at: https://hbr.org/2014/11/hospitals‐are‐finally‐starting‐to‐put‐real‐time‐data‐to‐use. Published November 12, 2014. Accessed August 25, 2015.
- , , , , . Reducing operating room costs through real‐time cost information feedback: a pilot study. J Endourol. 2015;29:963–968.
- , , . Facilitated patient experience feedback can improve nursing care: a pilot study for a phase III cluster randomised controlled trial. BMC Health Serv Res. 2013;13:259.
- , , , , . Patient satisfaction with time spent with their physician. J Fam Pract. 1998;47:133–137.
- , , , , , . The relationship between time spent communicating and communication outcomes on a hospital medicine service. J Gen Intern Med. 2012;27:185–189.
- , . Cognitive interview techniques reveal specific behaviors and issues that could affect patient satisfaction relative to hospitalists. J Hosp Med. 2009;4:E1–E6.
- , , , et al. Is patients' perception of time spent with the physician a determinant of ambulatory patient satisfaction? Arch Intern Med. 2001;161:1437–1442.
- , , , . Effect of clinician communication skills training on patient satisfaction. A randomized, controlled trial. Ann Intern Med. 1999;131:822–829.
In 2010, the Centers for Medicare and Medicaid Services implemented value‐based purchasing, a payment model that incentivizes hospitals for reaching certain quality and patient experience thresholds and penalizes those that do not, in part on the basis of patient satisfaction scores.[1] Although low patient satisfaction scores will adversely affect institutions financially, they also reflect patients' perceptions of their care. Some studies suggest that hospitals with higher patient satisfaction scores score higher overall on clinical care processes such as core measures compliance, readmission rates, lower mortality rates, and other quality‐of‐care metrics.[2, 3, 4, 5, 6, 7, 8, 9, 10, 11]
The Hospital Consumer Assessment of Healthcare Providers and Systems (HCAHPS) survey assesses patients' experience following their hospital stay.[1] The percent of top box scores (ie, response of always on a four point scale, or scores of 9 or 10 on a 10‐point scale) are utilized to compare hospitals and determine the reimbursement or penalty a hospital will receive. Although these scores are available to the public on the Hospital Compare website,[12] physicians may not know how their hospital is ranked or how they are individually perceived by their patients. Additionally, these surveys are typically conducted 48 hours to 6 weeks after patients are discharged, and the results are distributed back to the hospitals well after the time that care was provided, thereby offering providers no chance of improving patient satisfaction during a given hospital stay.
Institutions across the country are trying to improve their HCAHPS scores, but there is limited research identifying specific measures providers can implement. Some studies have suggested that utilizing etiquette‐based communication and sitting at the bedside[13, 14] may help improve patient experience with their providers, and more recently, it has been suggested that providing real‐time deidentified patient experience survey results with education and a rewards/emncentive system to residents may help as well.[15]
Surveys conducted during a patient's hospitalization can offer real‐time actionable feedback to providers. We performed a quality‐improvement project that was designed to determine if real‐time feedback to hospitalist physicians, followed by coaching, and revisits to the patients' bedside could improve the results recorded on provider‐specific patient surveys and/or patients' HCAHPS scores or percentile rankings.
METHODS
Design
This was a prospective, randomized quality‐improvement initiative that was approved by the Colorado Multiple Institutional Review Board and conducted at Denver Health, a 525‐bed university‐affiliated public safety net hospital. The initiative was conducted on both teaching and nonteaching general internal medicine services, which typically have a daily census of between 10 and 15 patients. No protocol changes occurred during the study.
Participants
Participants included all English‐ or Spanish‐speaking patients who were hospitalized on a general internal medicine service, had been admitted within the 2 days prior to enrollment, and had a hospitalist as their attending physician. Patients were excluded if they were enrolled in the study during a previous hospitalization, refused to participate, lacked capacity to participate, had hearing or speech impediments precluding regular conversation, were prisoners, if their clinical condition precluded participation, or their attending was an investigator in the project.
Intervention
Participants were prescreened by investigators by reviewing team sign‐outs to determine if patients had any exclusion criteria. Investigators attempted to survey each patient who met inclusion criteria on a daily basis between 9:00 am and 11:00 am. An investigator administered the survey to each patient verbally using scripted language. Patients were asked to rate how well their doctors were listening to them, explaining what they wanted to know, and whether the doctors were being friendly and helpful, all questions taken from a survey that was available on the US Department of Health and Human Services website (to be referred to as here forward daily survey).[16] We converted the original 5‐point Likert scale used in this survey to a 4‐point scale by removing the option of ok, leaving participants the options of poor, fair, good, or great. Patients were also asked to provide any personalized feedback they had, and these comments were recorded in writing by the investigator.
After being surveyed on day 1, patients were randomized to an intervention or control group using an automated randomization module in Research Electronic Data Capture (REDCap).[17] Patients in both groups who did not provide answers to all 3 questions that qualified as being top box (ie, great) were resurveyed on a daily basis until their responses were all top box or they were discharged, met exclusion criteria, or had been surveyed for a total of 4 consecutive days. In the pilot phase of this study, we found that if patients reported all top box scores on the initial survey their responses typically did not change over time, and the patients became frustrated if asked the same questions again when the patient felt there was not room for improvement. Accordingly, we elected to stop surveying patients when all top box responses were reported.
The attending hospitalist caring for each patient in the intervention group was given feedback about their patients' survey results (both their scores and any specific comments) on a daily basis. Feedback was provided in person by 1 of the investigators. The hospitalist also received an automatically generated electronic mail message with the survey results at 11:00 am on each study day. After informing the hospitalists of the patients' scores, the investigator provided a brief education session that included discussing Denver Health's most recent HCAHPS scores, value‐based purchasing, and the financial consequences of poor patient satisfaction scores. The investigator then coached the hospitalist on etiquette‐based communication,[18, 19] suggested that they sit down when communicating with their patients,[19, 20] and then asked the hospitalist to revisit each patient to discuss how the team could improve in any of the 3 areas where the patient did not give a top box score. These educational sessions were conducted in person and lasted a maximum of 5 minutes. An investigator followed up with each hospitalist the following day to determine whether the revisit occurred. Hospitalists caring for patients who were randomized to the control group were not given real‐time feedback or coaching and were not asked to revisit patients.
A random sample of patients surveyed for this initiative also received HCAHPS surveys 48 hours to 6 weeks following their hospital discharge, according to the standard methodology used to acquire HCAHPS data,[21] by an outside vendor contracted by Denver Health. Our vendor conducted these surveys via telephone in English or Spanish.
Outcomes
The primary outcome was the proportion of patients in each group who reported top box scores on the daily surveys. Secondary outcomes included the percent change for the scores recorded for 3 provider‐specific questions from the daily survey, the median top box HCAHPS scores for the 3 provider related questions and overall hospital rating, and the HCAHPS percentiles of top box scores for these questions.
Sample Size
The sample size for this intervention assumed that the proportion of patients whose treating physicians did not receive real‐time feedback who rated their providers as top box would be 75%, and that the effect of providing real‐time feedback would increase this proportion to 85% on the daily surveys. To have 80% power with a type 1 error of 0.05, we estimated a need to enroll 430 patients, 215 in each group.
Statistics
Data were collected and managed using a secure, Web‐based electronic data capture tool hosted at Denver Health (REDCap), which is designed to support data collection for research studies providing: (1) an intuitive interface for validated data entry, (2) audit trails for tracking data manipulation and export procedures, (3) automated export procedures for seamless data downloads to common statistical packages, and (4) procedures for importing data from external sources.[17]
A 2 test was used to compare the proportion of patients in the 2 groups who reported great scores for each question on the study survey on the first and last day. With the intent of providing a framework for understanding the effect real‐time feedback could have on patient experience, a secondary analysis of HCAHPS results was conducted using several different methods.
First, the proportion of patients in the 2 groups who reported scores of 9 or 10 for the overall hospital rating question or reported always for each doctor communication question on the HCHAPS survey was compared using a 2. Second, to allow for detection of differences in a sample with a smaller N, the median overall hospital rating scores from the HCAHPS survey reported by patients in the 2 groups who completed a survey following discharge were compared using a Wilcoxon rank sum test. Lastly, to place changes in proportion into a larger context (ie, how these changes would relate to value‐based purchasing), HCAHPS scores were converted to percentiles of national performance using the 2014 percentile rankings obtained from the external vendor that conducts the HCAHPS surveys for our hospital and compared between the intervention and control groups using a Wilcoxon rank sum test.
All comments collected from patients during their daily surveys were reviewed, and key words were abstracted from each comment. These key words were sorted and reviewed to categorize recurring key words into themes. Exemplars were then selected for each theme derived from patient comments.
RESULTS
From April 14, 2014 to September 19, 2014, we enrolled 227 patients in the control group and 228 in the intervention group (Figure 1). Patient demographics are summarized in Table 1. Of the 132 patients in the intervention group who reported anything less than top box scores for any of the 3 questions (thus prompting a revisit by their provider), 106 (80%) were revisited by their provider at least once during their hospitalization.
| All Patients | HCAHPS Patients | |||
|---|---|---|---|---|
| Control, N = 227 | Intervention, N = 228 | Control, N = 35 | Intervention, N = 30 | |
| ||||
| Age, mean SD | 55 14 | 55 15 | 55 15 | 57 16 |
| Gender | ||||
| Male | 126 (60) | 121 (55) | 20 (57) | 12 (40) |
| Female | 85 (40) | 98 (45) | 15(43) | 18 (60) |
| Race/ethnicity | ||||
| Hispanic | 84 (40) | 90 (41) | 17 (49) | 12 (40) |
| Black | 38 (18) | 28 (13) | 6 (17) | 7 (23) |
| White | 87 (41) | 97 (44) | 12 (34) | 10 (33) |
| Other | 2 (1) | 4 (2) | 0 (0) | 1 (3) |
| Payer | ||||
| Medicare | 65 (29) | 82 (36) | 15 (43) | 12 (40) |
| Medicaid | 122 (54) | 108 (47) | 17 (49) | 14 (47) |
| Commercial | 12 (5) | 15 (7) | 1 (3) | 1 (3) |
| Medically indigent | 4 (2) | 7 (3) | 0 (0) | 3 (10) |
| Self‐pay | 5 (2) | 4 (2) | 1 (3) | 0 (0) |
| Other/unknown | 19 (8) | 12 (5) | 0 (0) | 0 (0) |
| Team | ||||
| Teaching | 187 (82) | 196 (86) | 27 (77) | 24 (80) |
| Nonteaching | 40 (18) | 32 (14) | 8 (23) | 6 (20) |
| Top 5 primary discharge diagnoses* | ||||
| Septicemia | 26 (11) | 34 (15) | 3 (9) | 5 (17) |
| Heart failure | 14 (6) | 13 (6) | 2 (6) | |
| Acute pancreatitis | 12 (5) | 9 (4) | 3 (9) | 2 (7) |
| Diabetes mellitus | 11 (5) | 8 (4) | 2 (6) | |
| Alcohol withdrawal | 9 (4) | |||
| Cellulitis | 7 (3) | 2 (7) | ||
| Pulmonary embolism | 2 (7) | |||
| Chest pain | 2 (7) | |||
| Atrial fibrillation | 2 (6) | |||
| Length of stay, median (IQR) | 3 (2, 5) | 3 (2, 5) | 3 (2, 5) | 3 (2, 4) |
| Charlson Comorbidity Index, median (IQR) | 1 (0, 3) | 2 (0, 3) | 1 (0, 3) | 1.5 (1, 3) |
Daily Surveys
The proportion of patients in both study groups reporting top box scores tended to increase from the first day to the last day of the survey (Figure 2); however, we found no statistically significant differences between the proportion of patients who reported top box scores on first day or last day in the intervention group compared to the control group. The comments made by the patients are summarized in Supporting Table 1 in the online version of this article.
HCAHPS Scores
The proportion of top box scores from the HCAHPS surveys were higher, though not statistically significant, for all 3 provider‐specific questions and for the overall hospital rating for patients whose hospitalists received real‐time feedback (Table 2). The median [interquartile range] score for the overall hospital rating was higher for patients in the intervention group compared with those in the control group, (10 [9, 10] vs 9 [8, 10], P = 0.04]. After converting the HCAHPS scores to percentiles, we found considerably higher rankings for all 3 provider‐related questions and for the overall hospital rating in the intervention group compared to the control group (P = 0.02 for overall differences in percentiles [Table 2]).
| HCAHPS Questions | Proportion Top Box* | Percentile Rank | ||
|---|---|---|---|---|
| Control, N = 35 | Intervention, N = 30 | Control, N = 35 | Intervention, N = 30 | |
| ||||
| Overall hospital rating | 61% | 80% | 6 | 87 |
| Courtesy/respect | 86% | 93% | 23 | 88 |
| Clear communication | 77% | 80% | 39 | 60 |
| Listening | 83% | 90% | 57 | 95 |
No adverse events occurred during the course of the study in either group.
DISCUSSION
The important findings of this study were that (1) daily patient satisfaction scores improved from first day to last day regardless of study group, (2) patients whose providers received real‐time feedback had a trend toward higher HCAHPS proportions for the 3 provider‐related questions as well as the overall rating of the hospital but were not statistically significant, (3) the percentile differences in these 3 questions as well as the overall rating of the hospital were significantly higher in the intervention group as was the median score for the overall hospital rating.
Our original sample size calculation was based upon our own preliminary data, indicating that our baseline top box scores for the daily survey was around 75%. The daily survey top box score on the first day was, however, much lower (Figure 2). Accordingly, although we did not find a significant difference in these daily scores, we were underpowered to find such a difference. Additionally, because only a small percentage of patients are selected for the HCAHPS survey, our ability to detect a difference in this secondary outcome was also limited. We felt that it was important to analyze the percentile comparisons in addition to the proportion of top box scores on the HCAHPS, because the metrics for value‐based purchasing are based upon, in part, how a hospital system compares to other systems. Finally, to improve our power to detect a difference given a small sample size, we converted the scoring system for overall hospital ranking to a continuous variable, which again was noted to be significant.
To our knowledge, this is the first randomized investigation designed to assess the effect of real‐time, patient‐specific feedback to physicians. Real‐time feedback is increasingly being incorporated into medical practice, but there is only limited information available describing how this type of feedback affects outcomes.[22, 23, 24] Banka et al.[15] found that HCAHPS scores improved as a result of real‐time feedback given to residents, but the study was not randomized, utilized a pre‐post design that resulted in there being differences between the patients studied before and after the intervention, and did not provide patient‐specific data to the residents. Tabib et al.[25] found that operating costs decreased 17% after instituting real‐time feedback to providers about these costs. Reeves et al.[26] conducted a cluster randomized trial of a patient feedback survey that was designed to improve nursing care, but the results were reviewed by the nurses several months after patients had been discharged.
The differences in median top box scores and percentile rank that we observed could have resulted from the real‐time feedback, the educational coaching, the fact that the providers revisited the majority of the patients, or a combination of all of the above. Gross et al.[27] found that longer visits lead to higher satisfaction, though others have not found this to necessarily be the case.[28, 29] Lin et al.[30] found that patient satisfaction was affected by the perceived duration of the visit as well as whether expectations on visit length were met and/or exceeded. Brown et al.[31] found that training providers in communication skills improved the providers perception of their communication skills, although patient experience scores did not improve. We feel that the results seen are more likely a combination thereof as opposed to any 1 component of the intervention.
The most commonly reported complaints or concerns in patients' undirected comments often related to communication issues. Comments on subsequent surveys suggested that patient satisfaction improved over time in the intervention group, indicating that perhaps physicians did try to improve in areas that were highlighted by the real‐time feedback, and that patients perceived the physician efforts to do so (eg, They're doing better than the last time you asked. They sat down and talked to me and listened better. They came back and explained to me about my care. They listened better. They should do this survey at the clinic. See Supporting Table 1 in the online version of this article).
Our study has several limitations. First, we did not randomize providers, and many of our providers (approximately 65%) participated in both the control group and also in the intervention group, and thus received real‐time feedback at some point during the study, which could have affected their overall practice and limited our ability to find a difference between the 2 groups. In an attempt to control for this possibility, the study was conducted on an intermittent basis during the study time frame. Furthermore, the proportion of patients who reported top box scores at the beginning of the study did not have a clear trend of change by the end of the study, suggesting that overall clinician practices with respect to patient satisfaction did not change during this short time period.
Second, only a small number of our patients were randomly selected for the HCAHPS survey, which limited our ability to detect significant differences in HCAHPS proportions. Third, the HCAHPS percentiles at our institution at that time were low. Accordingly, the improvements that we observed in patient satisfaction scores might not be reproducible at institutions with higher satisfactions scores. Fourth, time and resources were needed to obtain patient feedback to provide to providers during this study. There are, however, other ways to obtain feedback that are less resource intensive (eg, electronic feedback, the utilization of volunteers, or partnering this with manager rounding). Finally, the study was conducted at a single, university‐affiliated public teaching hospital and was a quality‐improvement initiative, and thus our results are not generalizable to other institutions.
In conclusion, real‐time feedback of patient experience to their providers, coupled with provider education, coaching, and revisits, seems to improve satisfaction of patients hospitalized on general internal medicine units who were cared for by hospitalists.
Acknowledgements
The authors thank Kate Fagan, MPH, for her excellent technical assistance.
Disclosure: Nothing to report.
In 2010, the Centers for Medicare and Medicaid Services implemented value‐based purchasing, a payment model that incentivizes hospitals for reaching certain quality and patient experience thresholds and penalizes those that do not, in part on the basis of patient satisfaction scores.[1] Although low patient satisfaction scores will adversely affect institutions financially, they also reflect patients' perceptions of their care. Some studies suggest that hospitals with higher patient satisfaction scores score higher overall on clinical care processes such as core measures compliance, readmission rates, lower mortality rates, and other quality‐of‐care metrics.[2, 3, 4, 5, 6, 7, 8, 9, 10, 11]
The Hospital Consumer Assessment of Healthcare Providers and Systems (HCAHPS) survey assesses patients' experience following their hospital stay.[1] The percent of top box scores (ie, response of always on a four point scale, or scores of 9 or 10 on a 10‐point scale) are utilized to compare hospitals and determine the reimbursement or penalty a hospital will receive. Although these scores are available to the public on the Hospital Compare website,[12] physicians may not know how their hospital is ranked or how they are individually perceived by their patients. Additionally, these surveys are typically conducted 48 hours to 6 weeks after patients are discharged, and the results are distributed back to the hospitals well after the time that care was provided, thereby offering providers no chance of improving patient satisfaction during a given hospital stay.
Institutions across the country are trying to improve their HCAHPS scores, but there is limited research identifying specific measures providers can implement. Some studies have suggested that utilizing etiquette‐based communication and sitting at the bedside[13, 14] may help improve patient experience with their providers, and more recently, it has been suggested that providing real‐time deidentified patient experience survey results with education and a rewards/emncentive system to residents may help as well.[15]
Surveys conducted during a patient's hospitalization can offer real‐time actionable feedback to providers. We performed a quality‐improvement project that was designed to determine if real‐time feedback to hospitalist physicians, followed by coaching, and revisits to the patients' bedside could improve the results recorded on provider‐specific patient surveys and/or patients' HCAHPS scores or percentile rankings.
METHODS
Design
This was a prospective, randomized quality‐improvement initiative that was approved by the Colorado Multiple Institutional Review Board and conducted at Denver Health, a 525‐bed university‐affiliated public safety net hospital. The initiative was conducted on both teaching and nonteaching general internal medicine services, which typically have a daily census of between 10 and 15 patients. No protocol changes occurred during the study.
Participants
Participants included all English‐ or Spanish‐speaking patients who were hospitalized on a general internal medicine service, had been admitted within the 2 days prior to enrollment, and had a hospitalist as their attending physician. Patients were excluded if they were enrolled in the study during a previous hospitalization, refused to participate, lacked capacity to participate, had hearing or speech impediments precluding regular conversation, were prisoners, if their clinical condition precluded participation, or their attending was an investigator in the project.
Intervention
Participants were prescreened by investigators by reviewing team sign‐outs to determine if patients had any exclusion criteria. Investigators attempted to survey each patient who met inclusion criteria on a daily basis between 9:00 am and 11:00 am. An investigator administered the survey to each patient verbally using scripted language. Patients were asked to rate how well their doctors were listening to them, explaining what they wanted to know, and whether the doctors were being friendly and helpful, all questions taken from a survey that was available on the US Department of Health and Human Services website (to be referred to as here forward daily survey).[16] We converted the original 5‐point Likert scale used in this survey to a 4‐point scale by removing the option of ok, leaving participants the options of poor, fair, good, or great. Patients were also asked to provide any personalized feedback they had, and these comments were recorded in writing by the investigator.
After being surveyed on day 1, patients were randomized to an intervention or control group using an automated randomization module in Research Electronic Data Capture (REDCap).[17] Patients in both groups who did not provide answers to all 3 questions that qualified as being top box (ie, great) were resurveyed on a daily basis until their responses were all top box or they were discharged, met exclusion criteria, or had been surveyed for a total of 4 consecutive days. In the pilot phase of this study, we found that if patients reported all top box scores on the initial survey their responses typically did not change over time, and the patients became frustrated if asked the same questions again when the patient felt there was not room for improvement. Accordingly, we elected to stop surveying patients when all top box responses were reported.
The attending hospitalist caring for each patient in the intervention group was given feedback about their patients' survey results (both their scores and any specific comments) on a daily basis. Feedback was provided in person by 1 of the investigators. The hospitalist also received an automatically generated electronic mail message with the survey results at 11:00 am on each study day. After informing the hospitalists of the patients' scores, the investigator provided a brief education session that included discussing Denver Health's most recent HCAHPS scores, value‐based purchasing, and the financial consequences of poor patient satisfaction scores. The investigator then coached the hospitalist on etiquette‐based communication,[18, 19] suggested that they sit down when communicating with their patients,[19, 20] and then asked the hospitalist to revisit each patient to discuss how the team could improve in any of the 3 areas where the patient did not give a top box score. These educational sessions were conducted in person and lasted a maximum of 5 minutes. An investigator followed up with each hospitalist the following day to determine whether the revisit occurred. Hospitalists caring for patients who were randomized to the control group were not given real‐time feedback or coaching and were not asked to revisit patients.
A random sample of patients surveyed for this initiative also received HCAHPS surveys 48 hours to 6 weeks following their hospital discharge, according to the standard methodology used to acquire HCAHPS data,[21] by an outside vendor contracted by Denver Health. Our vendor conducted these surveys via telephone in English or Spanish.
Outcomes
The primary outcome was the proportion of patients in each group who reported top box scores on the daily surveys. Secondary outcomes included the percent change for the scores recorded for 3 provider‐specific questions from the daily survey, the median top box HCAHPS scores for the 3 provider related questions and overall hospital rating, and the HCAHPS percentiles of top box scores for these questions.
Sample Size
The sample size for this intervention assumed that the proportion of patients whose treating physicians did not receive real‐time feedback who rated their providers as top box would be 75%, and that the effect of providing real‐time feedback would increase this proportion to 85% on the daily surveys. To have 80% power with a type 1 error of 0.05, we estimated a need to enroll 430 patients, 215 in each group.
Statistics
Data were collected and managed using a secure, Web‐based electronic data capture tool hosted at Denver Health (REDCap), which is designed to support data collection for research studies providing: (1) an intuitive interface for validated data entry, (2) audit trails for tracking data manipulation and export procedures, (3) automated export procedures for seamless data downloads to common statistical packages, and (4) procedures for importing data from external sources.[17]
A 2 test was used to compare the proportion of patients in the 2 groups who reported great scores for each question on the study survey on the first and last day. With the intent of providing a framework for understanding the effect real‐time feedback could have on patient experience, a secondary analysis of HCAHPS results was conducted using several different methods.
First, the proportion of patients in the 2 groups who reported scores of 9 or 10 for the overall hospital rating question or reported always for each doctor communication question on the HCHAPS survey was compared using a 2. Second, to allow for detection of differences in a sample with a smaller N, the median overall hospital rating scores from the HCAHPS survey reported by patients in the 2 groups who completed a survey following discharge were compared using a Wilcoxon rank sum test. Lastly, to place changes in proportion into a larger context (ie, how these changes would relate to value‐based purchasing), HCAHPS scores were converted to percentiles of national performance using the 2014 percentile rankings obtained from the external vendor that conducts the HCAHPS surveys for our hospital and compared between the intervention and control groups using a Wilcoxon rank sum test.
All comments collected from patients during their daily surveys were reviewed, and key words were abstracted from each comment. These key words were sorted and reviewed to categorize recurring key words into themes. Exemplars were then selected for each theme derived from patient comments.
RESULTS
From April 14, 2014 to September 19, 2014, we enrolled 227 patients in the control group and 228 in the intervention group (Figure 1). Patient demographics are summarized in Table 1. Of the 132 patients in the intervention group who reported anything less than top box scores for any of the 3 questions (thus prompting a revisit by their provider), 106 (80%) were revisited by their provider at least once during their hospitalization.
| All Patients | HCAHPS Patients | |||
|---|---|---|---|---|
| Control, N = 227 | Intervention, N = 228 | Control, N = 35 | Intervention, N = 30 | |
| ||||
| Age, mean SD | 55 14 | 55 15 | 55 15 | 57 16 |
| Gender | ||||
| Male | 126 (60) | 121 (55) | 20 (57) | 12 (40) |
| Female | 85 (40) | 98 (45) | 15(43) | 18 (60) |
| Race/ethnicity | ||||
| Hispanic | 84 (40) | 90 (41) | 17 (49) | 12 (40) |
| Black | 38 (18) | 28 (13) | 6 (17) | 7 (23) |
| White | 87 (41) | 97 (44) | 12 (34) | 10 (33) |
| Other | 2 (1) | 4 (2) | 0 (0) | 1 (3) |
| Payer | ||||
| Medicare | 65 (29) | 82 (36) | 15 (43) | 12 (40) |
| Medicaid | 122 (54) | 108 (47) | 17 (49) | 14 (47) |
| Commercial | 12 (5) | 15 (7) | 1 (3) | 1 (3) |
| Medically indigent | 4 (2) | 7 (3) | 0 (0) | 3 (10) |
| Self‐pay | 5 (2) | 4 (2) | 1 (3) | 0 (0) |
| Other/unknown | 19 (8) | 12 (5) | 0 (0) | 0 (0) |
| Team | ||||
| Teaching | 187 (82) | 196 (86) | 27 (77) | 24 (80) |
| Nonteaching | 40 (18) | 32 (14) | 8 (23) | 6 (20) |
| Top 5 primary discharge diagnoses* | ||||
| Septicemia | 26 (11) | 34 (15) | 3 (9) | 5 (17) |
| Heart failure | 14 (6) | 13 (6) | 2 (6) | |
| Acute pancreatitis | 12 (5) | 9 (4) | 3 (9) | 2 (7) |
| Diabetes mellitus | 11 (5) | 8 (4) | 2 (6) | |
| Alcohol withdrawal | 9 (4) | |||
| Cellulitis | 7 (3) | 2 (7) | ||
| Pulmonary embolism | 2 (7) | |||
| Chest pain | 2 (7) | |||
| Atrial fibrillation | 2 (6) | |||
| Length of stay, median (IQR) | 3 (2, 5) | 3 (2, 5) | 3 (2, 5) | 3 (2, 4) |
| Charlson Comorbidity Index, median (IQR) | 1 (0, 3) | 2 (0, 3) | 1 (0, 3) | 1.5 (1, 3) |
Daily Surveys
The proportion of patients in both study groups reporting top box scores tended to increase from the first day to the last day of the survey (Figure 2); however, we found no statistically significant differences between the proportion of patients who reported top box scores on first day or last day in the intervention group compared to the control group. The comments made by the patients are summarized in Supporting Table 1 in the online version of this article.
HCAHPS Scores
The proportion of top box scores from the HCAHPS surveys were higher, though not statistically significant, for all 3 provider‐specific questions and for the overall hospital rating for patients whose hospitalists received real‐time feedback (Table 2). The median [interquartile range] score for the overall hospital rating was higher for patients in the intervention group compared with those in the control group, (10 [9, 10] vs 9 [8, 10], P = 0.04]. After converting the HCAHPS scores to percentiles, we found considerably higher rankings for all 3 provider‐related questions and for the overall hospital rating in the intervention group compared to the control group (P = 0.02 for overall differences in percentiles [Table 2]).
| HCAHPS Questions | Proportion Top Box* | Percentile Rank | ||
|---|---|---|---|---|
| Control, N = 35 | Intervention, N = 30 | Control, N = 35 | Intervention, N = 30 | |
| ||||
| Overall hospital rating | 61% | 80% | 6 | 87 |
| Courtesy/respect | 86% | 93% | 23 | 88 |
| Clear communication | 77% | 80% | 39 | 60 |
| Listening | 83% | 90% | 57 | 95 |
No adverse events occurred during the course of the study in either group.
DISCUSSION
The important findings of this study were that (1) daily patient satisfaction scores improved from first day to last day regardless of study group, (2) patients whose providers received real‐time feedback had a trend toward higher HCAHPS proportions for the 3 provider‐related questions as well as the overall rating of the hospital but were not statistically significant, (3) the percentile differences in these 3 questions as well as the overall rating of the hospital were significantly higher in the intervention group as was the median score for the overall hospital rating.
Our original sample size calculation was based upon our own preliminary data, indicating that our baseline top box scores for the daily survey was around 75%. The daily survey top box score on the first day was, however, much lower (Figure 2). Accordingly, although we did not find a significant difference in these daily scores, we were underpowered to find such a difference. Additionally, because only a small percentage of patients are selected for the HCAHPS survey, our ability to detect a difference in this secondary outcome was also limited. We felt that it was important to analyze the percentile comparisons in addition to the proportion of top box scores on the HCAHPS, because the metrics for value‐based purchasing are based upon, in part, how a hospital system compares to other systems. Finally, to improve our power to detect a difference given a small sample size, we converted the scoring system for overall hospital ranking to a continuous variable, which again was noted to be significant.
To our knowledge, this is the first randomized investigation designed to assess the effect of real‐time, patient‐specific feedback to physicians. Real‐time feedback is increasingly being incorporated into medical practice, but there is only limited information available describing how this type of feedback affects outcomes.[22, 23, 24] Banka et al.[15] found that HCAHPS scores improved as a result of real‐time feedback given to residents, but the study was not randomized, utilized a pre‐post design that resulted in there being differences between the patients studied before and after the intervention, and did not provide patient‐specific data to the residents. Tabib et al.[25] found that operating costs decreased 17% after instituting real‐time feedback to providers about these costs. Reeves et al.[26] conducted a cluster randomized trial of a patient feedback survey that was designed to improve nursing care, but the results were reviewed by the nurses several months after patients had been discharged.
The differences in median top box scores and percentile rank that we observed could have resulted from the real‐time feedback, the educational coaching, the fact that the providers revisited the majority of the patients, or a combination of all of the above. Gross et al.[27] found that longer visits lead to higher satisfaction, though others have not found this to necessarily be the case.[28, 29] Lin et al.[30] found that patient satisfaction was affected by the perceived duration of the visit as well as whether expectations on visit length were met and/or exceeded. Brown et al.[31] found that training providers in communication skills improved the providers perception of their communication skills, although patient experience scores did not improve. We feel that the results seen are more likely a combination thereof as opposed to any 1 component of the intervention.
The most commonly reported complaints or concerns in patients' undirected comments often related to communication issues. Comments on subsequent surveys suggested that patient satisfaction improved over time in the intervention group, indicating that perhaps physicians did try to improve in areas that were highlighted by the real‐time feedback, and that patients perceived the physician efforts to do so (eg, They're doing better than the last time you asked. They sat down and talked to me and listened better. They came back and explained to me about my care. They listened better. They should do this survey at the clinic. See Supporting Table 1 in the online version of this article).
Our study has several limitations. First, we did not randomize providers, and many of our providers (approximately 65%) participated in both the control group and also in the intervention group, and thus received real‐time feedback at some point during the study, which could have affected their overall practice and limited our ability to find a difference between the 2 groups. In an attempt to control for this possibility, the study was conducted on an intermittent basis during the study time frame. Furthermore, the proportion of patients who reported top box scores at the beginning of the study did not have a clear trend of change by the end of the study, suggesting that overall clinician practices with respect to patient satisfaction did not change during this short time period.
Second, only a small number of our patients were randomly selected for the HCAHPS survey, which limited our ability to detect significant differences in HCAHPS proportions. Third, the HCAHPS percentiles at our institution at that time were low. Accordingly, the improvements that we observed in patient satisfaction scores might not be reproducible at institutions with higher satisfactions scores. Fourth, time and resources were needed to obtain patient feedback to provide to providers during this study. There are, however, other ways to obtain feedback that are less resource intensive (eg, electronic feedback, the utilization of volunteers, or partnering this with manager rounding). Finally, the study was conducted at a single, university‐affiliated public teaching hospital and was a quality‐improvement initiative, and thus our results are not generalizable to other institutions.
In conclusion, real‐time feedback of patient experience to their providers, coupled with provider education, coaching, and revisits, seems to improve satisfaction of patients hospitalized on general internal medicine units who were cared for by hospitalists.
Acknowledgements
The authors thank Kate Fagan, MPH, for her excellent technical assistance.
Disclosure: Nothing to report.
- HCAHPS Fact Sheet. 2015. Available at: http://www.hcahpsonline.org/Files/HCAHPS_Fact_Sheet_June_2015.pdf. Accessed August 25, 2015.
- , , , . The relationship between commercial website ratings and traditional hospital performance measures in the USA. BMJ Qual Saf. 2013;22:194–202.
- , , , . Patients' perception of hospital care in the United States. N Engl J Med. 2008;359:1921–1931.
- , , , . The relationship between patients' perception of care and measures of hospital quality and safety. Health Serv Res. 2010;45:1024–1040.
- , , , et al. Relationship between quality of diabetes care and patient satisfaction. J Natl Med Assoc. 2003;95:64–70.
- , , , , . Relationship between patient satisfaction with inpatient care and hospital readmission within 30 days. Am J Manag Care. 2011;17:41–48.
- , , . A systematic review of evidence on the links between patient experience and clinical safety and effectiveness. BMJ Open. 2013;3(1).
- , . The association between satisfaction with services provided in primary care and outcomes in type 2 diabetes mellitus. Diabet Med. 2003;20:486–490.
- , , , et al. Associations between Web‐based patient ratings and objective measures of hospital quality. Arch Intern Med. 2012;172:435–436.
- , , , et al. Patient satisfaction and its relationship with clinical quality and inpatient mortality in acute myocardial infarction. Circ Cardiovasc Qual Outcomes. 2010;3:188–195.
- , , , , . Patients' perceptions of care are associated with quality of hospital care: a survey of 4605 hospitals. Am J Med Qual. 2015;30(4):382–388.
- Centers for Medicare 28:908–913.
- , , , , , . Effect of sitting vs. standing on perception of provider time at bedside: a pilot study. Patient Educ Couns. 2012;86:166–171.
- , , , et al. Improving patient satisfaction through physician education, feedback, and incentives. J Hosp Med. 2015;10:497–502.
- US Department of Health and Human Services. Patient satisfaction survey. Available at: http://bphc.hrsa.gov/policiesregulations/performancemeasures/patientsurvey/surveyform.html. Accessed November 15, 2013.
- , , , , , . Research electronic data capture (REDCap)—a metadata‐driven methodology and workflow process for providing translational research informatics support. J Biomed Inform. 2009;42:377–381.
- . The HCAHPS Handbook. Gulf Breeze, FL: Fire Starter; 2010.
- . Etiquette‐based medicine. N Engl J Med. 2008;358:1988–1989.
- . 5 years after the Kahn's etiquette‐based medicine: a brief checklist proposal for a functional second meeting with the patient. Front Psychol. 2013;4:723.
- Frequently Asked Questions. Hospital Value‐Based Purchasing Program. Available at: http://www.cms.gov/Medicare/Quality‐Initiatives‐Patient‐Assessment‐Instruments/hospital‐value‐based‐purchasing/Downloads/FY‐2013‐Program‐Frequently‐Asked‐Questions‐about‐Hospital‐VBP‐3‐9‐12.pdf. Accessed February 8, 2014.
- , , , . Real‐time patient survey data during routine clinical activities for rapid‐cycle quality improvement. JMIR Med Inform. 2015;3:e13.
- . Mount Sinai launches real‐time patient‐feedback survey tool. Healthcare Informatics website. Available at: http://www.healthcare‐informatics.com/news‐item/mount‐sinai‐launches‐real‐time‐patient‐feedback‐survey‐tool. Accessed August 25, 2015.
- , . Hospitals are finally starting to put real‐time data to use. Harvard Business Review website. Available at: https://hbr.org/2014/11/hospitals‐are‐finally‐starting‐to‐put‐real‐time‐data‐to‐use. Published November 12, 2014. Accessed August 25, 2015.
- , , , , . Reducing operating room costs through real‐time cost information feedback: a pilot study. J Endourol. 2015;29:963–968.
- , , . Facilitated patient experience feedback can improve nursing care: a pilot study for a phase III cluster randomised controlled trial. BMC Health Serv Res. 2013;13:259.
- , , , , . Patient satisfaction with time spent with their physician. J Fam Pract. 1998;47:133–137.
- , , , , , . The relationship between time spent communicating and communication outcomes on a hospital medicine service. J Gen Intern Med. 2012;27:185–189.
- , . Cognitive interview techniques reveal specific behaviors and issues that could affect patient satisfaction relative to hospitalists. J Hosp Med. 2009;4:E1–E6.
- , , , et al. Is patients' perception of time spent with the physician a determinant of ambulatory patient satisfaction? Arch Intern Med. 2001;161:1437–1442.
- , , , . Effect of clinician communication skills training on patient satisfaction. A randomized, controlled trial. Ann Intern Med. 1999;131:822–829.
- HCAHPS Fact Sheet. 2015. Available at: http://www.hcahpsonline.org/Files/HCAHPS_Fact_Sheet_June_2015.pdf. Accessed August 25, 2015.
- , , , . The relationship between commercial website ratings and traditional hospital performance measures in the USA. BMJ Qual Saf. 2013;22:194–202.
- , , , . Patients' perception of hospital care in the United States. N Engl J Med. 2008;359:1921–1931.
- , , , . The relationship between patients' perception of care and measures of hospital quality and safety. Health Serv Res. 2010;45:1024–1040.
- , , , et al. Relationship between quality of diabetes care and patient satisfaction. J Natl Med Assoc. 2003;95:64–70.
- , , , , . Relationship between patient satisfaction with inpatient care and hospital readmission within 30 days. Am J Manag Care. 2011;17:41–48.
- , , . A systematic review of evidence on the links between patient experience and clinical safety and effectiveness. BMJ Open. 2013;3(1).
- , . The association between satisfaction with services provided in primary care and outcomes in type 2 diabetes mellitus. Diabet Med. 2003;20:486–490.
- , , , et al. Associations between Web‐based patient ratings and objective measures of hospital quality. Arch Intern Med. 2012;172:435–436.
- , , , et al. Patient satisfaction and its relationship with clinical quality and inpatient mortality in acute myocardial infarction. Circ Cardiovasc Qual Outcomes. 2010;3:188–195.
- , , , , . Patients' perceptions of care are associated with quality of hospital care: a survey of 4605 hospitals. Am J Med Qual. 2015;30(4):382–388.
- Centers for Medicare 28:908–913.
- , , , , , . Effect of sitting vs. standing on perception of provider time at bedside: a pilot study. Patient Educ Couns. 2012;86:166–171.
- , , , et al. Improving patient satisfaction through physician education, feedback, and incentives. J Hosp Med. 2015;10:497–502.
- US Department of Health and Human Services. Patient satisfaction survey. Available at: http://bphc.hrsa.gov/policiesregulations/performancemeasures/patientsurvey/surveyform.html. Accessed November 15, 2013.
- , , , , , . Research electronic data capture (REDCap)—a metadata‐driven methodology and workflow process for providing translational research informatics support. J Biomed Inform. 2009;42:377–381.
- . The HCAHPS Handbook. Gulf Breeze, FL: Fire Starter; 2010.
- . Etiquette‐based medicine. N Engl J Med. 2008;358:1988–1989.
- . 5 years after the Kahn's etiquette‐based medicine: a brief checklist proposal for a functional second meeting with the patient. Front Psychol. 2013;4:723.
- Frequently Asked Questions. Hospital Value‐Based Purchasing Program. Available at: http://www.cms.gov/Medicare/Quality‐Initiatives‐Patient‐Assessment‐Instruments/hospital‐value‐based‐purchasing/Downloads/FY‐2013‐Program‐Frequently‐Asked‐Questions‐about‐Hospital‐VBP‐3‐9‐12.pdf. Accessed February 8, 2014.
- , , , . Real‐time patient survey data during routine clinical activities for rapid‐cycle quality improvement. JMIR Med Inform. 2015;3:e13.
- . Mount Sinai launches real‐time patient‐feedback survey tool. Healthcare Informatics website. Available at: http://www.healthcare‐informatics.com/news‐item/mount‐sinai‐launches‐real‐time‐patient‐feedback‐survey‐tool. Accessed August 25, 2015.
- , . Hospitals are finally starting to put real‐time data to use. Harvard Business Review website. Available at: https://hbr.org/2014/11/hospitals‐are‐finally‐starting‐to‐put‐real‐time‐data‐to‐use. Published November 12, 2014. Accessed August 25, 2015.
- , , , , . Reducing operating room costs through real‐time cost information feedback: a pilot study. J Endourol. 2015;29:963–968.
- , , . Facilitated patient experience feedback can improve nursing care: a pilot study for a phase III cluster randomised controlled trial. BMC Health Serv Res. 2013;13:259.
- , , , , . Patient satisfaction with time spent with their physician. J Fam Pract. 1998;47:133–137.
- , , , , , . The relationship between time spent communicating and communication outcomes on a hospital medicine service. J Gen Intern Med. 2012;27:185–189.
- , . Cognitive interview techniques reveal specific behaviors and issues that could affect patient satisfaction relative to hospitalists. J Hosp Med. 2009;4:E1–E6.
- , , , et al. Is patients' perception of time spent with the physician a determinant of ambulatory patient satisfaction? Arch Intern Med. 2001;161:1437–1442.
- , , , . Effect of clinician communication skills training on patient satisfaction. A randomized, controlled trial. Ann Intern Med. 1999;131:822–829.
© 2016 Society of Hospital Medicine