Get 20M+ Full-Text Papers For Less Than $1.50/day. Start a 14-Day Trial for You or Your Team.

Learn More →

Do Hospitalists Improve Quality?

Do Hospitalists Improve Quality? In this issue, in an article examining the association between hospitalists and performance on hospital-level quality indicators, López and colleagues1 found that hospitals with hospitalists have higher “quality of care” scores than those without hospitalists and attribute these improvements to the hospitalists' presence. This article provides interesting data, which unfortunately are not persuasive enough to support that conclusion. This editorial addresses several important issues: the definition of quality, the value of performance measures, the difference between quality and safety, alternative analyses, alternative explanations, and finally, what questions should we ask in 2009 about hospitalists and quality. Hospital medicine continues to thrive, primarily for economic reasons. Several articles have touted the benefits of hospitalists, as referenced in the article by López et al.1 These articles have focused primarily on decreases in length of stay and hospital charges, while suggesting no change in short-term outcomes and readmissions. The article by López et al1 adds to a growing literature that hopes to convince readers that hospitalists improve quality. Quality and performance measurement The word quality in the medical literature these days confuses us. What is quality medical care? Imagine a high-quality physician. What dimensions do you include in your own definition of high-quality medical care? The following represents a nonexhaustive list of quality attributes: bedside manner; accurate, complete history taking; accurate physical examination; laboratory test interpretation; excellent medical knowledge; diagnostic acumen; understanding the natural history of disease; appropriate use of consultation; and evidence-based treatment of existing conditions. The study by López et al1 (and most previous studies) focus entirely on the last of these attributes, which we should call performance measurement rather than quality. Performance measurement generally focuses on core measures, usually vetted by major organizations. We should rephrase the question to ask if hospitalists improve adherence to disease performance measures. As the noted British philosopher Onora O’Neill opined that performance measures are not always relevant to the question we want to answer,2 she also reminds us that measures can lead to perverse incentives. The well-known experience with the 4-hour rule for pneumonia antibiotic therapy supports that concern.3 She further states that Professional performance can be assessed by persons who are able to judge what they are asked to assess, whose institutional position makes them independent of those whose work they are judging, and who can write intelligible narrative accounts. . . . 2 We should heed her point when considering quality. Attempts to simplify quality into scorecards or checklists may seem attractive, but those attempts may well mislead and negatively affect patient care. Hospitalists and impact on performance Performance measurement has the attention of hospitals and insurance companies. While hospital income depends on adherence to these measures, we must understand how best to achieve success in improving the percentage of patients receiving indicated medications or appropriate advice. Given that large caveat, the study by López et al1 does not completely answer the posed question. This study used complex statistical techniques to show that hospitals with hospitalists had better composite performance scores than those hospitals without hospitalists. One should consider several possibilities to explain these findings. The analysis reveals that hospitals with hospitalists differ from hospitals without hospitalists. Given that the findings are correlations, we should consider the likelihood of variables that lead to both improved performance and to hiring hospitalists. A propensity score analysis might better control for these hospital factors. Because we do not have patient data, we do not really understand whether the hospitalists had an impact on performance. Experience suggests that most patients with myocardial infarction have cardiologists determining their discharge medications. Therefore, we must imagine a construct for the hospitalists influencing cardiologist prescription writing. Perhaps hospitalists better understand their local system and can increase the likelihood that a cardiologist's recommendation is captured appropriately so that they (the hospital) get “credit” for the quality care delivered. Without patient and physician level data, the analysis has too many potential confounders. However, the hospitals with hospitalists did not have improved congestive heart failure performance. Again, one cannot easily develop a construct for explaining this discrepancy from the myocardial infarction data. Hospitalist groups and their structure As the article by López et al1 states, hospitals with the highest nursing staff ratios had improved performance on all measures. This result echoes previous such studies. Should we consider the nursing effect differently from the hospitalist effect, or do they both reflect a more important construct. An alternative hypothesis to explain all these findings postulates that during the period of the study, the hospitals that hired hospitalists had better overall systems of care. They had higher nursing staff ratios. Likely, such hospitals had better systems and infrastructure. These speculations are interesting but not essential to examining the hospitalist movement. We do not need articles touting the benefits of hospitalists. We will have hospitalists with or without these articles. Rather, we need articles that explore the organizational issues that affect hospitalist groups. To paraphrase a common saying, when you have seen one hospitalist situation, you have seen one hospitalist situation. Once we learn that a hospital has hospitalists, do we really know anything about that hospital? We should view the study by López et al,1 and all similar studies, as hypothesis generating. While these studies, which use existing data sets, provide interesting analyses, to really understand the postulated issues, we need primary data collection. We need more complete data on hospitalist structure in each hospital and structure comparisons. We should restrict our analyses to focus on patients for whom the hospitalists provide direct care. We should refocus our research away from proving (or disproving) the value of hospitalists and toward understanding how best to structure hospitalist programs. As one talks with hospitalists around the country, one finds that some programs have excellent organization. These programs become integrated into the hospital culture, with hospitalists participating on various committees with administration and nursing leaders. These programs likely make important contributions to their hospitals. Hospitalists in such programs have higher retention rates. These groups become totally integrated with hospital processes, and individual hospitalists become champions for improving hospital functions and enhancing hospital safety. At other institutions, hospitalists are considered “R-7s”—superresidents. In these hospitals, hospitalists rarely stay long because they receive little respect from other physicians or from the hospital administration. They do not join the systematic process of improving hospital care. Hospitalist influence and safety issues We suggest studying hospitalist groups and their structure. Hospitalists can bring their greatest value when they become integrated into the hospital culture. Given the confusing term quality, we recommend focusing on safety issues rather than core disease measures as indicators of hospitalist contributions. Hospitalists have the best opportunity to observe process variations that increase the risk of a serious adverse outcome—or sentinel event. The salient question thus should focus on what do hospitalists do with these observations, or at a higher level, what can they do about dangerous processes. Thus, we should understand the preparedness and ability of hospitalists to assume roles and contribute to safety efforts as well as hospital support and commitment to addressing these important efforts. All physicians who have experienced these lapses in care understand that preventing error recurrence has more meaning and likely more impact than documenting smoking cessation counseling on discharge. While difficult to perform, studies examining truer measures of hospitalist influence will come when we understand their influence on these safety issues. We should not take a database-driven approach but rather a hospital-by-hospital study that documents safety and hospitalist group structure. Rather than “single-site” studies, we need comparisons of multiple sites, studying their structures and impact on safety. Such studies would allow us to understand “best practices” for hospitalist group organization, exploring questions such as the following: What are the likely differences between the good situations and the undesirable situations? Do these differences reflect on the hospital administration or on the hospitalists themselves? Are there hospital characteristics that predict better hospitalist programs, which in turn, may be able to provide safer medical care? As a young field, hospital medicine has strengths and weaknesses. Future investigations should focus on defining the strengths and minimizing the weaknesses. We believe that hospitalists can help decrease hospital errors and improve safety if they are totally integrated with hospital processes and supported as champions for these important efforts. Lumping hospitalists without a consideration of organizational differences could hide the promise of excellent hospitalist groups. The major contribution of hospital medicine should involve system improvement along with excellent bedside care. We must understand the contributors as well as the detractors to excellence for the hospitalist movement to achieve its full potential. Correspondence: Dr Centor, Department of Internal Medicine, University of Alabama, FOT 720, 1530 Third Ave S, Birmingham, AL 35294 (rcentor@uab.edu). Financial Disclosure: None reported. References 1. López LHicks LSCohen AP McKean SWeissman JS Hospitalists and the quality of care in hospitals. Arch Intern Med 2009;169 (15) 1389- 1394Google ScholarCrossref 2. O'Neill O Trust with accountability? J Health Serv Res Policy 2003;8 (1) 3- 4PubMedGoogle ScholarCrossref 3. Welker JAHuston M McCue JD Antibiotic timing and errors in diagnosing pneumonia. Arch Intern Med 2008;168 (4) 351- 356PubMedGoogle ScholarCrossref http://www.deepdyve.com/assets/images/DeepDyve-Logo-lg.png Archives of Internal Medicine American Medical Association

Do Hospitalists Improve Quality?

Loading next page...
 
/lp/american-medical-association/do-hospitalists-improve-quality-0EiMHEekzb

References (3)

Publisher
American Medical Association
Copyright
Copyright © 2009 American Medical Association. All Rights Reserved.
ISSN
0003-9926
eISSN
1538-3679
DOI
10.1001/archinternmed.2009.245
Publisher site
See Article on Publisher Site

Abstract

In this issue, in an article examining the association between hospitalists and performance on hospital-level quality indicators, López and colleagues1 found that hospitals with hospitalists have higher “quality of care” scores than those without hospitalists and attribute these improvements to the hospitalists' presence. This article provides interesting data, which unfortunately are not persuasive enough to support that conclusion. This editorial addresses several important issues: the definition of quality, the value of performance measures, the difference between quality and safety, alternative analyses, alternative explanations, and finally, what questions should we ask in 2009 about hospitalists and quality. Hospital medicine continues to thrive, primarily for economic reasons. Several articles have touted the benefits of hospitalists, as referenced in the article by López et al.1 These articles have focused primarily on decreases in length of stay and hospital charges, while suggesting no change in short-term outcomes and readmissions. The article by López et al1 adds to a growing literature that hopes to convince readers that hospitalists improve quality. Quality and performance measurement The word quality in the medical literature these days confuses us. What is quality medical care? Imagine a high-quality physician. What dimensions do you include in your own definition of high-quality medical care? The following represents a nonexhaustive list of quality attributes: bedside manner; accurate, complete history taking; accurate physical examination; laboratory test interpretation; excellent medical knowledge; diagnostic acumen; understanding the natural history of disease; appropriate use of consultation; and evidence-based treatment of existing conditions. The study by López et al1 (and most previous studies) focus entirely on the last of these attributes, which we should call performance measurement rather than quality. Performance measurement generally focuses on core measures, usually vetted by major organizations. We should rephrase the question to ask if hospitalists improve adherence to disease performance measures. As the noted British philosopher Onora O’Neill opined that performance measures are not always relevant to the question we want to answer,2 she also reminds us that measures can lead to perverse incentives. The well-known experience with the 4-hour rule for pneumonia antibiotic therapy supports that concern.3 She further states that Professional performance can be assessed by persons who are able to judge what they are asked to assess, whose institutional position makes them independent of those whose work they are judging, and who can write intelligible narrative accounts. . . . 2 We should heed her point when considering quality. Attempts to simplify quality into scorecards or checklists may seem attractive, but those attempts may well mislead and negatively affect patient care. Hospitalists and impact on performance Performance measurement has the attention of hospitals and insurance companies. While hospital income depends on adherence to these measures, we must understand how best to achieve success in improving the percentage of patients receiving indicated medications or appropriate advice. Given that large caveat, the study by López et al1 does not completely answer the posed question. This study used complex statistical techniques to show that hospitals with hospitalists had better composite performance scores than those hospitals without hospitalists. One should consider several possibilities to explain these findings. The analysis reveals that hospitals with hospitalists differ from hospitals without hospitalists. Given that the findings are correlations, we should consider the likelihood of variables that lead to both improved performance and to hiring hospitalists. A propensity score analysis might better control for these hospital factors. Because we do not have patient data, we do not really understand whether the hospitalists had an impact on performance. Experience suggests that most patients with myocardial infarction have cardiologists determining their discharge medications. Therefore, we must imagine a construct for the hospitalists influencing cardiologist prescription writing. Perhaps hospitalists better understand their local system and can increase the likelihood that a cardiologist's recommendation is captured appropriately so that they (the hospital) get “credit” for the quality care delivered. Without patient and physician level data, the analysis has too many potential confounders. However, the hospitals with hospitalists did not have improved congestive heart failure performance. Again, one cannot easily develop a construct for explaining this discrepancy from the myocardial infarction data. Hospitalist groups and their structure As the article by López et al1 states, hospitals with the highest nursing staff ratios had improved performance on all measures. This result echoes previous such studies. Should we consider the nursing effect differently from the hospitalist effect, or do they both reflect a more important construct. An alternative hypothesis to explain all these findings postulates that during the period of the study, the hospitals that hired hospitalists had better overall systems of care. They had higher nursing staff ratios. Likely, such hospitals had better systems and infrastructure. These speculations are interesting but not essential to examining the hospitalist movement. We do not need articles touting the benefits of hospitalists. We will have hospitalists with or without these articles. Rather, we need articles that explore the organizational issues that affect hospitalist groups. To paraphrase a common saying, when you have seen one hospitalist situation, you have seen one hospitalist situation. Once we learn that a hospital has hospitalists, do we really know anything about that hospital? We should view the study by López et al,1 and all similar studies, as hypothesis generating. While these studies, which use existing data sets, provide interesting analyses, to really understand the postulated issues, we need primary data collection. We need more complete data on hospitalist structure in each hospital and structure comparisons. We should restrict our analyses to focus on patients for whom the hospitalists provide direct care. We should refocus our research away from proving (or disproving) the value of hospitalists and toward understanding how best to structure hospitalist programs. As one talks with hospitalists around the country, one finds that some programs have excellent organization. These programs become integrated into the hospital culture, with hospitalists participating on various committees with administration and nursing leaders. These programs likely make important contributions to their hospitals. Hospitalists in such programs have higher retention rates. These groups become totally integrated with hospital processes, and individual hospitalists become champions for improving hospital functions and enhancing hospital safety. At other institutions, hospitalists are considered “R-7s”—superresidents. In these hospitals, hospitalists rarely stay long because they receive little respect from other physicians or from the hospital administration. They do not join the systematic process of improving hospital care. Hospitalist influence and safety issues We suggest studying hospitalist groups and their structure. Hospitalists can bring their greatest value when they become integrated into the hospital culture. Given the confusing term quality, we recommend focusing on safety issues rather than core disease measures as indicators of hospitalist contributions. Hospitalists have the best opportunity to observe process variations that increase the risk of a serious adverse outcome—or sentinel event. The salient question thus should focus on what do hospitalists do with these observations, or at a higher level, what can they do about dangerous processes. Thus, we should understand the preparedness and ability of hospitalists to assume roles and contribute to safety efforts as well as hospital support and commitment to addressing these important efforts. All physicians who have experienced these lapses in care understand that preventing error recurrence has more meaning and likely more impact than documenting smoking cessation counseling on discharge. While difficult to perform, studies examining truer measures of hospitalist influence will come when we understand their influence on these safety issues. We should not take a database-driven approach but rather a hospital-by-hospital study that documents safety and hospitalist group structure. Rather than “single-site” studies, we need comparisons of multiple sites, studying their structures and impact on safety. Such studies would allow us to understand “best practices” for hospitalist group organization, exploring questions such as the following: What are the likely differences between the good situations and the undesirable situations? Do these differences reflect on the hospital administration or on the hospitalists themselves? Are there hospital characteristics that predict better hospitalist programs, which in turn, may be able to provide safer medical care? As a young field, hospital medicine has strengths and weaknesses. Future investigations should focus on defining the strengths and minimizing the weaknesses. We believe that hospitalists can help decrease hospital errors and improve safety if they are totally integrated with hospital processes and supported as champions for these important efforts. Lumping hospitalists without a consideration of organizational differences could hide the promise of excellent hospitalist groups. The major contribution of hospital medicine should involve system improvement along with excellent bedside care. We must understand the contributors as well as the detractors to excellence for the hospitalist movement to achieve its full potential. Correspondence: Dr Centor, Department of Internal Medicine, University of Alabama, FOT 720, 1530 Third Ave S, Birmingham, AL 35294 (rcentor@uab.edu). Financial Disclosure: None reported. References 1. López LHicks LSCohen AP McKean SWeissman JS Hospitalists and the quality of care in hospitals. Arch Intern Med 2009;169 (15) 1389- 1394Google ScholarCrossref 2. O'Neill O Trust with accountability? J Health Serv Res Policy 2003;8 (1) 3- 4PubMedGoogle ScholarCrossref 3. Welker JAHuston M McCue JD Antibiotic timing and errors in diagnosing pneumonia. Arch Intern Med 2008;168 (4) 351- 356PubMedGoogle ScholarCrossref

Journal

Archives of Internal MedicineAmerican Medical Association

Published: Aug 10, 2009

Keywords: hospitalists,quality improvement

There are no references for this article.