To the Editor: The article by Lindenauer and colleagues (Dec.20 issue)1 compares the outcomes of care by hospitalists withthe outcomes of traditional models of care. As physicians whohave practiced both as hospitalists and as primary care physicians,we believe that the authors failed to consider other pivotalvariables. First, hospitalists are employed by either the hospitalor a private contracting company; this variable affects boththe number of admissions and their appropriateness. Second,the authors did not compare physician experience and the rateof consultations per case when comparing hospitalist and traditionalpractice models. Third, sample homogeneity in regard to patientload per hospitalist per day should also have been examined.Another variable worth examining is the rate of patient satisfaction.Finally, what were the differences in level of service betweenthe hospitalist and traditional practice models?
The above variables are essential for a correct assessment ofthe current practice environment and would help provide guidancein decision making with respect to the optimal setting and structureof a hospitalist service.
Ehab S. Suleiman, M.D. Iowa Health Physicians Waterloo, IA 50703
William P. Jerome, M.D. Trinity Internal Medicine Specialists Bettendorf, IA 52722
References
Lindenauer PK, Rothberg MB, Pekow PS, Kenwood C, Benjamin EM, Auerbach AD. Outcomes of care by hospitalists, general internists, and family physicians. N Engl J Med 2007;357:2589-2600. [Free Full Text]
To the Editor: We believe that the conclusions reached by Lindenauerand colleagues may be based on an underestimate of the effectof hospitalists on outcomes of care. The "hospitalists" identifiedby the authors cared for a median of 70 patients per year withone or more of seven common inpatient medical diagnoses. A typicalhospitalist admits 561 to 748 patients per year.1 Approximately22% of these admissions fall within the seven diagnoses studied.2This represents approximately twice the caseload of the median-volumehospitalists and almost four times the cutoff caseload for thecomparison group of "high-volume" hospitalists (those with casevolumes that met or exceeded the 25th percentile) identifiedin this study.
This discrepancy could have biased the results in two ways.First, the lower-than-expected case volume suggests misclassificationof nonhospitalist physicians as hospitalists, which would biasthe results toward the null hypothesis. Second, on the basisof what is known about the improved outcomes seen when physiciansperform a higher volume of work in a specific area,3,4,5 thestudy might have underestimated the potential beneficial effectsof hospitalists on length of stay and cost of care.
Eugene S. Chu, M.D. Richard K. Albert, M.D. Denver Health Medical Center Denver, CO 80204
Glasheen JJ, Epstein KR, Siegal E, Kutner JS, Prochazka AV. The spectrum of community-based hospitalist practice: a call to tailor internal medicine residency training. Arch Intern Med 2007;167:727-728. [Free Full Text]
Weller WE, Hannan EL. Relationship between provider volume and postoperative complications for bariatric procedures in New York State. J Am Coll Surg 2006;202:753-761. [CrossRef][ISI][Medline]
Hannan EL, Wu C, Walford G, et al. Volume-outcome relationships for percutaneous coronary interventions in the stent era. Circulation 2005;112:1171-1179. [Free Full Text]
Moscucci M, Share D, Smith D, et al. Relationship between operator volume and adverse outcome in contemporary percutaneous coronary intervention practice: an analysis of a quality-controlled multicenter percutaneous coronary intervention clinical database. J Am Coll Cardiol 2005;46:625-632. [Free Full Text]
To the Editor: The study by Lindenauer et al. investigated the outcomes of hospitalist care among 76,926 patients at 45 hospitals.As compared with the outcomes associated with hospitalist carein previous studies based on single hospitals,1,2 the cost savingswere lower and favorable inpatient mortality was not found inthe study by Lindenauer et al. However, I would like to seesome analysis with regard to the homogeneity of the outcomesamong the hospitals in their study. The cost for the same therapycould be quite different among hospitals. Although the authorshave considered some variables, such as number of beds and locationof the hospitals, the outcomes could vary among the hospitalseven after adjustment for these factors. The authors' primaryexplanation for the discrepancy between their findings and theresults of previous studies was that they included many community-basedhospitals in the study. If this is the case, hospitalist caremay have a weak favorable effect, no effect, or a negative effecton the outcomes in community-based hospitals. A subgroup analysiswill be needed.
Ming Wei, M.D. Holy Hospital Mechanicsburg, PA 17050
References
Auerbach AD, Wachter RM, Katz P, Showstack J, Baron RB, Goldman L. Implementation of a voluntary hospitalist service at a community teaching hospital: improved clinical efficiency and patient outcomes. Ann Intern Med 2002;137:859-865. [Free Full Text]
Meltzer D, Manning WG, Morrison J, et al. Effects of physician experience on costs and outcomes on an academic general medicine service: results of a trial of hospitalists. Ann Intern Med 2002;137:866-874. [Free Full Text]
To the Editor: The discussions by Lindenauer et al. and by McMahon,1in the accompanying editorial, present the hospitalist movementas one driven by cost and quality concerns. Having been involvedin physician credentialing at several national health plansover the past dozen years, I see the shift to hospitalists asa cultural one as well on the part of physicians who no longerview the hospital as the axis of their professional activities.I well remember how the request by an allergist to become apart of our plan's panel, yet not maintain hospital privileges,flummoxed our credentialing committee at the time. Now it isroutine to see most primary care physicians and many specialistspresent to the committee without hospital affiliations. Theyno longer see the hospital as offering services or value tothem that justifies joining a medical staff with its attendantobligations. This change has also been facilitated by the explosivegrowth in freestanding ambulatory centers, owned and managedby physicians and usually in direct competition with hospitals,that are supplanting hospital-based services in many communities.
Derek van Amerongen, M.D. Humana of Ohio Cincinnati, OH 45202 dvanamerongen{at}humana.com
References
McMahon LF Jr. The hospitalist movement -- time to move on. N Engl J Med 2007;357:2627-2629. [Free Full Text]
The authors reply: We agree with Al-Shaer and colleagues thatfactors such as employment arrangements for hospitalists (employedby the hospital or by a private group), staffing ratios (e.g.,the ratio of patients to physicians), and number of years inpractice may explain variations in outcome among hospitalistprograms. Similar structural factors are likely to influenceoutcomes of care by family physicians or general internistsas well. However, the aim of our study was to compare outcomesacross representative groups of physicians, not to assess factorsassociated with intraspecialty variation.
Although we acknowledged the risk of misclassification in ourarticle, we disagree with Chu and Albert about the presenceof large differences between the annual volume estimates weproduced and those based on productivity data from the Societyof Hospital Medicine (SHM).1 Our median volume estimates arebased on the number of cases for which each hospitalist wasthe attending physician of record in our data set, whereas SHMdata are derived from surveys, include consultative encounters,and are normalized to a 1.0 full-time-equivalent job description.Despite Chu and Albert's comments about the relationship betweenvolume and outcome, volume was not independently associatedwith any of the outcomes we studied. Although this may seemto be at odds with the experience in cardiac surgery or cardiology,this linkage in hospital medicine has been seen in only onesingle-site, two-hospitalist study,2 and studies of pneumoniaand chronic obstructive pulmonary disease suggest that any associationthat may exist runs counter to that which would be predicted.3,4
Our comment regarding community-based, or nonteaching, hospitalswas made because previous studies examined large academic medicalcenters almost exclusively. Among other factors, we adjustedfor the effects of hospital size, teaching status, region, andpopulation served (urban or rural) in our multivariable modelsand also explored potential interactions between physician specialtyand hospital teaching status. However, there was very littlevariation in the effects of physician specialty in these analyses.Figure 2 of our article shows the relative consistency of ourfindings across all 45 hospitals. We believe these analysesaddress Wei's concern about the need for subgroup analysis.
Although our analysis was limited to the outcomes of care inthe hospital, van Amerongen is correct in noting that the wishesof primary care physicians may be as potent a catalyst for thecontinued growth of the hospitalist model of care as is interestin improving hospital efficiency and quality.5
Peter K. Lindenauer, M.D. Michael B. Rothberg, M.D., M.P.H. Baystate Medical Center Springfield, MA 01199 peter.lindenauer{at}bhs.org
Andrew D. Auerbach, M.D., M.P.H. University of California,San Francisco San Francisco, CA 94143
Lindenauer PK, Behal R, Murray CK, Nsa W, Houck PM, Bratzler DW. Volume, quality of care, and outcome in pneumonia. Ann Intern Med 2006;144:262-269. [Free Full Text]
Lindenauer PK, Pekow P, Gao S, Crawford AS, Gutierrez B, Benjamin EM. Quality of care for patients hospitalized for acute exacerbations of chronic obstructive pulmonary disease. Ann Intern Med 2006;144:894-903. [Free Full Text]
Meltzer D, Manning WG, Morrison J, et al. Effects of physician experience on costs and outcomes on an academic general medicine service: results of a trial of hospitalists. Ann Intern Med 2002;137:866-874. [Free Full Text]
Vasilevskis E, Knebel J, Wachter R, Auerbach A. The rise of the hospitalist in California. Oakland: California Healthcare Foundation, July 2007. (Accessed March 28, 2008, 2008, at http://www.chcf.org/topics/hospitals/index.cfm?itemID=133365.)
Previous
