The full content of Annals is available to subscribers

Subscribe/Learn More  >
Original Research |

Association of Hospitalist Care With Medical Utilization After Discharge: Evidence of Cost Shift From a Cohort Study

Yong-Fang Kuo, PhD; and James S. Goodwin, MD
[+] Article, Author, and Disclosure Information

From the Sealy Center on Aging, University of Texas Medical Branch, Galveston, Texas.

Grant Support: By the National Institute on Aging (grants 1R01-AG033134 and P30AG024832) and the National Cancer Institute (grant K05-CA134923).

Potential Conflicts of Interest: Disclosures can be viewed at www.acponline.org/authors/icmje/ConflictOfInterestForms.do?msNum=M11-0086.

Reproducible Research Statement:Study protocol and data set: Not available. Statistical code: Available from Dr. Kuo (e-mail, yokuo@utmb.edu).

Requests for Single Reprints: Yong-Fang Kuo, PhD, Department of Internal Medicine, University of Texas Medical Branch, 301 University Boulevard, Galveston, TX 77555-0177; e-mail, yokuo@utmb.edu.

Current Author Addresses: Drs. Kuo and Goodwin: Department of Internal Medicine, University of Texas Medical Branch, 301 University Boulevard, Galveston, TX 77555-0177.

Author Contributions: Conception and design: Y.F. Kuo, J.S. Goodwin.

Analysis and interpretation of the data: Y.F. Kuo, J.S. Goodwin.

Drafting of the article: Y.F. Kuo, J.S. Goodwin.

Critical revision of the article for important intellectual content: Y.F. Kuo, J.S. Goodwin.

Final approval of the article: Y.F. Kuo, J.S. Goodwin.

Provision of study materials or patients: Y.F. Kuo, J.S. Goodwin.

Statistical expertise: Y.F. Kuo.

Administrative, technical, or logistic support: Y.F. Kuo, J.S. Goodwin.

Collection and assembly of data: Y.F. Kuo, J.S. Goodwin.

Ann Intern Med. 2011;155(3):152-159. doi:10.7326/0003-4819-155-3-201108020-00005
Text Size: A A A

Background: Hospitalist care has grown rapidly, in part because it is associated with decreased length of stay and hospital costs. No national studies examining the effect of hospitalist care on hospital costs or on medical utilization and costs after discharge have been done.

Objective: To assess the relationship of hospitalist care with hospital length of stay, hospital charges, and medical utilization and Medicare costs after discharge.

Design: Population-based national cohort study.

Setting: Hospital care of Medicare patients.

Patients: A 5% national sample of enrollees in Medicare parts A and B with a primary care physician who were cared for by their primary care physician or a hospitalist during medical hospitalizations from 2001 to 2006.

Measurements: Length of stay, hospital charges, discharge location and physician visits, emergency department visits, rehospitalization, and Medicare spending within 30 days after discharge.

Results: In propensity score analysis, hospital length of stay was 0.64 day less among patients receiving hospitalist care. Hospital charges were $282 lower, whereas Medicare costs in the 30 days after discharge were $332 higher (P < 0.001 for both). Patients cared for by hospitalists were less likely to be discharged to home (odds ratio, 0.82 [95% CI, 0.78 to 0.86]) and were more likely to have emergency department visits (odds ratio, 1.18 [CI, 1.12 to 1.24]) and readmissions (odds ratio, 1.08 [CI, 1.02 to 1.14]) after discharge. They also had fewer visits with their primary care physician and more nursing facility visits after discharge.

Limitation: Observational studies are subject to selection bias.

Conclusion: Decreased length of stay and hospital costs associated with hospitalist care are offset by higher medical utilization and costs after discharge.

Primary Funding Source: National Institute on Aging and National Cancer Institute.


Grahic Jump Location
Appendix Figure.
Distribution of propensity scores for admissions cared for by hospitalists and primary care physicians.

Vertical lines indicate deciles for the propensity score.

Grahic Jump Location




Citing articles are presented as examples only. In non-demo SCM6 implementation, integration with CrossRef’s "Cited By" API will populate this tab (http://www.crossref.org/citedby.html).


Submit a Comment/Letter
Association of Hospitalist Care With Medical Utilization After Discharge: Evidence of Cost Shift From a Cohort Study
Posted on August 5, 2011
Jeffrey L. Schnipper
Brigham and Women's Hospital
Conflict of Interest: None Declared

TO THE EDITOR, I found Kuo and Goodwin's article to be quite interesting. I sincerely hope that this article does not lead to discussions of who is better (PCPs or hospitalists) or whether we should abandon the hospitalist model of care - neither is likely to be productive. Hospitalists provide value in the hospital setting because of their availability and expertise, but their greatest potential value is in their willingness and ability to lead quality improvement efforts in the hospitals in which they work. To date, these efforts have mostly focused on improving the delivery of inpatient medical care. The article by Kuo and Goodwin should remind us that hospitalists need to focus more on the hospital discharge process. But they should not be alone in these efforts. Hospitalists need to work with PCPs and others to ensure that the transition is seamless. For example, while hospitalists can work to improve discharge documentation, patient and caregiver education, and medication reconciliation efforts, PCPs can work to improve availability for post-discharge visits, timely delivery of post-discharge plans of care, and appropriate monitoring. And together, hospitalists and PCPs can work on systems to promote collaborative development of a patient-centered discharge plan, including the most appropriate discharge time and setting. New models of care, including bundled payments, patient-centered medical homes, and accountable care organizations, should encourage the development of such systems. Quality of care is not an inherent property of a certain kind of provider - it only comes with concerted efforts to achieve it.

Conflict of Interest:

None declared

Re:Association of Hospitalist Care With Medical Utilization After Discharge: Evidence of Cost Shift From a Cohort Study
Posted on August 5, 2011
Joshua Freeman
University of Kansas School of Medicine
Conflict of Interest: None Declared

This is a very interesting article, as it looks at post-discharge costs as well as hospital costs, and thus is a step in the right direction. We also need to look at patient-centered issues, such as the loss in quality that might occur because the hospitalist has no prior knowledge of the patient. Historically, patients didn't mind seeing one's partner for a cold or BP check, but when sick enough to be in the hospital were comforted (at least) and benefited by having the involvement of a doctor who knew them. As we segregate primary care doctors into hospitalists and (I hate this term!) "ambulists", this dynamic is lost. Thus "you look better than you did yesterday" is no longer able to be put in the context of "but it is still nowhere near where you were two weeks ago". This is a real cost.

Conflict of Interest:

None declared

Association of Hospitalist Care With Medical Utilization After Discharge: Evidence of Cost Shift From a Cohort Study
Posted on August 7, 2011
Arun K. Raman
Endion Hospitalist systems/Cogent HMG Buffalo NY 14221
Conflict of Interest: None Declared

As a practicing hospitalist I agree with the findings (1) that a seamless PCP driven inpatient care is efficient and cost effective. As noted in the article, Medicare spent $282 less in the hospital for the patient cared by the hospitalists and $332 more after discharge implying a cost shift of about $50. If this is applied to 25% of Medicare admissions, it would amount to $170 million additional cost annually and not $1.1 billion as stated in the article. The intensity of care required for hospitalized patients has been high in recent years. Many PCPs have to be away from their busy practice during the day to see few patients in a distant hospital. Hence, cost effectiveness of this gain of $50 needs further evaluation (2).

In this study, hospitalists cared for more patients with comorbid conditions, low incomes and from nursing homes compared to the PCP cohort. Furthermore, hospitalists in many tertiary care centers accept patients requiring extensive surgical intervention from outlying facilities. Could these be reasons for the higher discharge rate back to nursing facilities? Future studies addressing these issues are warranted at this time.


1. Kuo YF, Goodwin JS. Association of hospitalist care with medical utilization after discharge: evidence of cost shift from a cohort study. Ann Intern Med. 2011;155:152-9. [PMID: 21810708].

2. Centers for Medicare & Medicaid Services. 100% MEDPAR inpatient hospital fiscal year 2008. Short stay inpatient by state. Assessed at www.cms.gov/MedicareFeeforSvcPartsAB/Downloads/DRGState08.pdf on 12 May 2010.

Conflict of Interest:

None declared

Too Many Generalizations
Posted on August 8, 2011
David M. Mitchell
Sibley Memorial Hospital, Washington, D.C.
Conflict of Interest: None Declared

To the Editor:

The article by Kuo and Goodwin (1), suggesting possible cost shifting resulting from hospitalist-based care, strikes at the core of what it means to be a hospitalist. However, I am concerned that too many generalizations have been drawn by the authors from these otherwise valuable results.

First, as a physician who has worked as a hospitalist in 7 states and 12 hospitals over ten years, I think readers need to recognize that hospitalist groups are very heterogeneous, and that such broad-based results are weakest in the sense that they are not capable of evaluating variability between individual hospitalist groups. Variability of practice patterns within individual medical specialties is a well documented phenomenon. Employment model and workload are major factors, in my opinion, in defining how hospitalist-based care is delivered, and such variations could clearly translate into diverse outcomes. Further research could tease out these differences.

Second, we need a more accurate method of comparing severity of illness between groups of hospitalized patients. The finding that hospitalists admitted 73% more nursing home patients than primary care physicians in this study, although statistically adjusted for, demonstrates the tendency of hospitalists to care for sicker patients; and it has been my anecdotal experience that hospitalists generally have significantly higher clinical thresholds for admitting patients to the hospital than primary care physicians, resulting in a greater severity of illness that I believe is not accurately measured by counting comorbid conditions alone, and is likely to result in more nursing home discharges and readmissions.

Third, the study excludes most Medicare patients (72%) and is thereby able to hold up an impractical minority model of care as a standard for comparison. This not only makes extrapolation of results to all Medicare patients, which the authors do multiple times, inaccurate; but also leaves out important patient populations (such as patients with no physician and the non-Medicare populations) as well as other common models of care (such as primary care groups that share rounding responsibilities within their group).

From my view in the trenches, I believe that hospital medicine specialists have a huge (perhaps untapped) potential to improve the quality and cost of inpatient care, and to collaborate effectively with outpatient physicians (much as other specialists do); however, this study does raise the concern that we as a specialty may not have adequately or universally adopted this goal. If hospitalists wish to be respected as true specialists in hospital medicine, clearly we must earn it, and this study, despite its limitations, provides a new impetus to do so.

David M. Mitchell, MD, PhD Sibley Memorial Hospital Washington, D.C. 20016


1. Kuo, YF, Goodwin JS. Association of hospitalist care with medical utilization after discharge: evidence of cost shift from a cohort study. Ann Intern Med. 2011; 155:152-9.

Conflict of Interest:

None declared

Potential for Bias
Posted on August 8, 2011
David O. Meltzer
The University of Chicago
Conflict of Interest: None Declared

The potential for bias in the results of Kuo and Goodwin deserves further discussion. The statistically significant differences they find in the observable attributes of patients of hospitalists and patients of non- hospitalists are consistent with the findings we have published describing determinants of whether care is provided by hospitalist or non- hospitalists. For example, we have found that attributes of the primary care physician practice predict the use of hospitalists (1,2). Kuo and Goodwin conclude that because the differences they observe in patient attributes are "small", concern should be allayed that propensity score analysis cannot control for unobservable differences. However, their discussion does not address the implications of their finding that even small differences in observable characteristics could produce sufficient bias to explain their results on post-discharge resource utilization. Differences in primary care practice attributes between hospitalist and non-hospitalist patients is just one of many possible confounders that could cause the results of Kuo and Goodwin to reflect bias rather than causal effects of hospitalist care. Large randomized or quasi-randomized study designs (such as using instrumental variables) are needed to address these concerns of unmeasured confounders.

Given such concerns about potential bias, it is important to consider the plausibility of some of Kuo and Goodwin's findings. The decrease in the odds of discharge home with hospitalist care (0.82 (95% C.I. 0.78- 0.86) is large. Even with the limited specificity of criteria for appropriateness for post acute care, it seems unlikely that hospitalist care produces changes of this magnitude - a change in care trajectory for nearly 1 in 5 patients who would otherwise be discharged home. Indeed, the estimated decrease in the odds of discharge home with hospitalist care is so large that many of the prior studies of hospitalist care with randomized or quasi-randomized designs are large enough that they could have detected differences of this magnitude if they existed (3). Unfortunately such analyses were not performed. If such analyses are performed and do not suggest these differences, this should heighten concern that Kuo and Goodwin's results about discharge home reflect unmeasured confounding. This would further increase concern that the post- discharge utilization results reflect unmeasured confounding given the authors' analysis that those results are even more easily explained by unobserved confounders. Kuo and Goodwin results could also be assessed for potential bias by examining differences between hospitalist and non- hospitalist resource use and outcomes at multiple points of time after discharge; unobserved confounders are more likely to have lasting association with resources use and outcomes whereas studies of hospital care on outcomes with stronger designs have tended to show effects in the 30-60 days after discharge that then dissipate over time (4,5).

More generally, the results of this study highlight the strengths and weakness of observational versus experimental approaches to comparative effectiveness research. Where small effect sizes may be important, large sample sizes from observational studies may be valuable. Unfortunately, concerns about bias also make studies with small effect sizes especially difficult to rely upon with confidence.


1. Meltzer DO, Chung JW. "U.S. Trends in the Inpatient Activity of Generalist Physicians and the Emergence of Hospitalists." Journal of General Internal Medicine 25(5):453-459, 2010.

2. Meltzer D, Chung J. "Coordination, Switching Costs and the Division of Labor in General Medicine: An Economic Explanation for the Emergence of Hospitalists in the United States." The National Bureau of Economic Research Working Paper Series #16040 May 2010.

3. White HL, Glazier RH. Do hospitalist physicians improve the quality of inpatient care delivery? A systematic review of process, efficiency and outcome measures. BMC Medicine 2011, 9:58.

4. Meltzer, D, Manning W, Morrison J, Shah M, Jin L, Guth T, Levinson W. "Effects of Physician Experience on Costs and Outcomes on an Academic General Medicine Service: Results of a Trial of Hospitalists." Annals of Internal Medicine 137:866-874, 2002.

5. Cutler, D. (1995). "The Incidence of Adverse Medical Outcomes Under Prospective Payment." Econometrica. Vol. 63 (1), (1995), pp.29-50.

Conflict of Interest:

None declared

Mitigating post-discharge costs via better coordinaton
Posted on August 8, 2011
Jonas B. Green
Cedars Sinai Health System
Conflict of Interest: None Declared

The additional post-discharge costs identified by Drs. Kuo and Goodwin (1) may lead some to question the modest measured benefits (2) associated with the hospitalist model. As visit rates to primary care physicians post-discharge were lower among patients cared for by hospitalists, it is worth considering that poor coordination of care may be largely responsible for the higher rates of Emergency visits and readmissions. In addition, poor care coordination is associated with duplication of services,(3) contributing to higher costs.

By improving coordination of care, as we at Cedars Sinai Medicine Intiative and others elsewhere are doing, any additional post-discharge costs associated with hospitalist care may be mitigated, while the inpatient savings would remain intact.


1. Kuo, YF, Goodwin JS. Association of hospitalist care with medical utilization after discharge: evidence of cost shift from a cohort study. Ann Intern Med. 2011; 155:152-9.

2. Meltzer D, Manning WG, Morrison J, Shah MN, Jin L, Guth T, 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;137:866-74.

3. P. T. Huynh, C. Schoen, R. Osborn, and A. L. Holmgren, The U.S. Health Care Divide: Disparities in Primary Care Experiences by Income (New York: The Commonwealth Fund, Apr. 2006).

Conflict of Interest:

None declared

No Title
Posted on August 16, 2011
Jean-Sebastien Rachoin
UMDNJ,RWJ Medical Schoold Cooper University Hospital,Camden NJ
Conflict of Interest: None Declared

Kuo et al. conducted a very interesting study describing outcomes in patients cared for by hospitalist or Primary care physicians. [1]

There are certain points in the complex equation that is the US health care systems that need to be mentioned:

The higher rate of readmissions/cost in patients cared for by a hospitalist could be influenced to a great extent by the underlying characteristics of this patient population (more co-morbid conditions and more from nursing homes [NH]). With the current bed-hold policy by MedicAid , payment to skilled nursing facilities and NH is maintained for 7 to 10 days even if patients are hospitalized .Such a policy has already been shown to be associated with higher rates of hospital readmissions [2].

Furthermore ,as the authors mention the "incentive " for hospitalists to shift costs, it would also be fair to note the "absence of incentive" for NH to reduce readmissions [3] and that it is "easier" for facilities to readmit patients to the hospitals when confronted with questionable symptoms (especially during week-ends or at night).

Additionally, it does not seem clear from the article whether the authors factored the reduction in hospital stay into the cost reduction analysis. Indeed, saving more than half a day per patient stay would not only affect margin, but also result in direct savings (new admissions) and indirect savings (reducing ED wait time, avoiding ED divert )[4] and may improve patient satisfaction and reduce incidence of acquired conditions. Such substantial contributions need to be more highlighted.

Finally, the hospitalist movement has evolved greatly in the last 5 years and a substantial shift has occurred with initiatives to reduce readmission and improve quality of care. The data presented in the article does not unfortunately reflect the current state of mind in the hospitalist world.


1. Kuo YF, Goodwin JS: Association of hospitalist care with medical utilization after discharge: evidence of cost shift from a cohort study. Ann Intern Med 2011, 155(3):152-159.

2. Grabowski DC, Feng Z, Intrator O, Mor V: Medicaid bed-hold policy and Medicare skilled nursing facility rehospitalizations. Health Serv Res 2010, 45(6 Pt 2):1963-1980.

3. Mor V, Intrator O, Feng Z, Grabowski DC: The revolving door of rehospitalization from skilled nursing facilities. Health Aff (Millwood) 2010, 29(1):57-64.

4. Schull MJ, Lazier K, Vermeulen M, Mawhinney S, Morrison LJ: Emergency department contributors to ambulance diversion: a quantitative analysis. Ann Emerg Med 2003, 41(4):467-476.

Conflict of Interest:

None declared

Is hospitalist care also wasteful during during a hospitalization?
Posted on August 16, 2011
Daniel J. Brotman
Johns Hopkins Hospital
Conflict of Interest: None Declared

To the Editor: Kuo and Goodwin provide evidence that hospitalists need to improve how they transition patients back to their primary care providers (PCPs), and that excess post-discharge costs may result from sub -optimal care transitions.(1) However, we are intrigued by the fact that the per-day charges for hospital care varied so dramatically between hospitalized patients cared for by hospitalists versus PCPs. By analyzing length-of-stay (LOS) and total charges from MedPAR files, the authors found that the mean LOS among patients cared for by hospitalists was 5.17 days with mean charges to Medicare of $15,019 per admission, while the mean LOS among patients cared for by PCPs was 5.82 days with mean charges to Medicare of $15,301. If these estimates are precise, then mean hospital charges were lower by less than 2% for patients cared for by hospitalists despite a LOS reduction of over 11%, translating into hospital charges of $2629 per day for PCPs care, and $2905 per day for hospitalist care--a difference of almost 10%. This contrasts sharply with other published data suggesting that LOS reductions parallel cost reductions closely.(2-3) There are two potential explanations for this surprising asymmetry: 1) hospitalists order tests, procedures, and medications less efficiently than PCPs, such that any savings on room, board, and nursing charges associated with shaving off 2/3 of a hospital day are almost completely eroded; 2) the MedPAR charge files used in this study inadequately capture the true cost savings attributable to hospitalists' LOS reduction. This distinction is important because if hospitalists indeed provide such inefficient care, there is an enormous opportunity to increase value that we hope hospitalists will embrace. But if the 0.65 day marginal decrease in average LOS does translate into real cost savings that were not captured by Kuo's and Goodwin's methodology , then better estimates of true costs may find that the financial benefits of hospitalists' inpatient care actually more than offsets increased post- discharge spending, even if the savings ended up on the hospitals' balance sheets due to fixed DRG payments--money that hopefully is reinvested in improving quality, reducing harm, and subsidizing care for the indigent. Regardless, we are optimistic that new payment models that incentivize hospitals to assume financial responsibility for some aspects of outpatient care will help to ensure that patients receive cost-effective care throughout the care continuum.


1. Kuo YF, Goodwin JS. Association of hospitalist care with medical utilization after discharge: Evidence of cost shift from a cohort study. Ann Intern Med. 2011;155(3):152-159.

2. Fine MJ, Pratt HM, Obrosky DS, Lave JR, McIntosh LJ Singer DE, et al. Relation between length of hospital stay and costs of care for patients with community-acquired pneumonia. Am J Med. 2000;109:378-385.

3. Wachter RM, Katz P, Showstack J, Bindman AB, Goldman. Reorganizing an academic medical service: Impact on cost, quality, patient satisfaction, and education. JAMA. 1998;279(19);1560-1565.

Conflict of Interest:

None declared

A generalist by any other name: The need for hospitalist-focused training
Posted on August 27, 2011
Dimitriy Levin
University of Colorado Hospital Medicine Group
Conflict of Interest: None Declared

With great interest, we read the article by Kuo and Goodwin[1] about the potential cost-shifting associated with hospitalist care. The authors' findings of increased healthcare utilization after hospitalist care, exemplified by nursing facility use, emergency department visits, and readmissions, can be summed up as complications of poor transitions of care.

While disappointing, these findings are not entirely surprising. High -quality transitions of care require a set of specific skills that are neither self-evident nor taught in most training programs. Despite the growing need[2], at the beginning of the study period in 2001 there were no hospitalist-focused training programs in the country. In 2004, the University of Colorado School of Medicine established its hospitalist training track which, by the end of the study in 2006, had graduated 12 internal medicine residents with hospitalist-focused training. The majority of the United States hospitalist workforce then, as now, consists overwhelmingly of generalists graduating from undifferentiated internal medicine, family medicine, or pediatric residency programs.

As Kuo and Goodwin so clearly demonstrate, a generalist practicing in the inpatient environment is still a generalist and changing their name to "hospitalist" does not automatically imbue them with skills to, for example, execute high-quality transitions of care. Just as the name "cardiologist" indicates disease-specific training, so should the term "hospitalist" indicate additional proficiency in transitions of care, quality improvement, patient safety, and delivery of cost-effective inpatient care.[3] It is our belief that expanding hospitalist-specific education and teaching to meet defined competencies can result in more safe, effective, and efficient inpatient care without shifting the burden to the outpatient environment.


[1] Kuo YF, Goodwin JS. Association of Hospitalist Care With Medical Utilization After Discharge: Evidence of Cost Shift From a Cohort Study. Ann Intern Med. 2011;155:152-159.

[2] Plauth WH, Pantilat SZ, Wacther RM, Fenton CL. Hospitalists' Perceptions of Their Residency Training Needs: Results of a National Survey. Am J Med. 2001;111:247-254.

[3] Glasheen JJ, Siegal EM, Epstein K, Kutner J, Prochazka AV. Fulfilling the Promise of Hospital Medicine: Tailoring Internal Medicine Training to Address Hospitalists' Needs. J Gen Intern Med. 2008;23:1110- 1115.

Conflict of Interest:

None declared

Author's Response
Posted on September 27, 2011
Yong-Fang Kuo
University of Texas Medical Branch, Galveston, TX 77555
Conflict of Interest: None Declared

We thank the correspondents for their interest in our paper (1). Drs. Mitchell, Meltzer, and Chung note potential issues in our observational study with internal and external validity. We agree that assessing long- term outcomes could be a way to indirectly examine the presence of selection bias, and have used that method in other studies (2). In the Annals paper we controlled for medical utilization and costs in the year prior to the index hospitalization (1). We also included instrumental variable analyses in earlier versions of the paper, but dropped them in favor of conditional models with the non-pooling propensity approach in response to the statistical reviewer's suggestion.

Also, while we agree that randomized, double blinded prospective trials are the gold standard for internal validity, the unique circumstances of the prospective trials of hospitalist care precluded a blinded format. The hospitalists participating in those trials were early adapters of the model. Early adapters of a new methodology are different from other physicians (3). Some were likely "idea champions" (3). Thus, in the prospective trials, the two trial arms are likely to have differed in ways other than presence or absence of the hospitalist model. There are also challenges to the external validity of the prospective trials. All the trials were done in academic hospitals, in which "usual care" was typically not provided by a patient's PCP, but by housestaff supervised by an attending physician. In other words, the comparison was between full time vs. part time hospital physicians, with neither group having any continuity of care with the patient

Meltzer also comments that an 18% reduced odds of discharge home by hospitalists is unfeasibly large. We disagree. There have been very large increases in the past 15 years in discharges to other institutional settings after hospitalization (4). It is not unreasonable to postulate that some of that change has been driven by the growth of hospitalists. We appreciate Dr. Rachoin's comment on the continued evolution in hospitalist care. Evaluations of medical practice often deal with moving targets. In response to Dr. Brotman and Boonyasai on the apparent discrepancy in the magnitudes of the reduction in length of stay vs. hospital cost, his two potential explanations for the discrepancy deserve further study. Our study of stroke patients found some evidence for higher use of consultants in patients receiving hospitalist care (2).

We agree that there are indirect savings to hospitals from shorter length of stay, as noted by Dr. Rachoin. However, there are no cost savings to Medicare due to the fixed DRG payments. That is why the estimate of additional Medicare cost was $1.1 billion, not $179 million as suggested by Dr. Raman. We agree with Drs. Levin, Glasheen, and Mitchell that hospitalists are heterogeneous. We are currently attempting to describe that heterogeneity in processes and outcomes of care among >1000 hospitalists in Texas. We agree with Dr. Levin and Glasheen on the importance of hospitalist-focused training emphasizing proficiency in transitions in care. We appreciate Dr. Freeman's comment on the unmeasured costs from the loss of continuity of care. As we previously reported, for the majority of hospitalized Medicare patients, all the physicians providing care for them in the hospital had never previously cared for them in any setting (5).

Finally, the hospitalist model is not going away. It is important to continually examine the weak points as well as the strengths of this model as an essential step in the design and implementation of interventions to improve hospital care and transitions in care.

Yong-Fang Kuo, Ph.D.

James S. Goodwin, M.D.

University of Texas Medical Branch, Galveston, TX 77555


1. Kuo YF, Goodwin JS. Association of hospitalist care with medical utilization after discharge: evidence of cost shift from a cohort study. Ann Intern Med 2011; 155(3):152-9.

2. Howrey B, Kuo YF, Goodwin JS. Association of care by hospitalists on discharge destination and 30 day outcomes following acute ischemic stroke. Med Care 2011 Aug; 49(8):701-7.

3. Greer AL. The state of the art versus the state of the science. The diffusion of new medical technologies into practice. Int J Technol Assess Health Care 1998; 4(1):5-26.

4. Goodwin JS, Howrey B, Zhang D, Kuo YF. Risk of nursing home utilization after hospitalization among the elderly in the US. J. Gerontol. Med. Sci., in press.

5. Sharma G, Fletcher KE, Zhang DD, Kuo Y-F, Freeman JL, Goodwin JS. Continuity of outpatient and inpatient care for hospitalized older adults. JAMA 2009; 301(16):1671-80.

Conflict of Interest:

None declared

Response to Kuo and Goodwin
Posted on September 29, 2011
Carl H Reynolds
Rochester General Hospital
Conflict of Interest: None Declared

To the Editor:

We congratulate Kuo and Goodwin (1) for trying to test an important hypothesis "that hospitalist care would be associated with cost shifting." However, the conclusions they draw from their data may mislead readers regarding the effects of hospitalists on improving inpatient care and reducing cost.

First, their propensity scoring analysis misses the effect of important confounders. Factors not considered in their propensity scoring include inpatient mortality rates, inpatient ICU utilization rates, admission source (direct from PCP office or via ED), patient education level, secondary health insurance and drug coverage.

To flush out one example: hospitalists at our institution have a ~2% absolute lower inpatient mortality rate than PCPs who attend their own patients, despite a similar case mix index (unpublished data). In the study1, patients who died or utilized the ICU during admission were excluded from analysis. An inpatient who would have died under the PCP's care but survived under a hospitalist's care would thus be treated differently in their analysis: the PCP's patient would be excluded and the hospitalist's patient would be retained and incur post-discharge costs. Hospitalists (in the selected cohort) lost their cost advantage largely due to readmissions and SNF utilization, but these "sickest of the sick" patients are also likely at highest risk for readmission and SNF utilization.

Second, in our area, safety-net clinics without robust infrastructure or resources were among the "early adopters" of hospitalists. This article compares the patients of early adopters to "late/non-adopters": is the difference detected due to the hospitalist or could early adopter PCPs differ from late/non-adopter PCPs in their resources to manage very sick outpatients?

Lastly, Kuo and Goodwin extrapolate from their restricted cohort a total of $1.1 billion in added Medicare costs (Table 3, $332 - $282 = $50/case) associated with hospitalist management. Their larger sample does not support this conclusion: the cost advantage of hospitalists persists, suggesting that Medicare's costs are actually reduced by $1.5 billion (Appendix Table 6, $228 - $296 = ($68)) because of hospitalists.

Physicians, inpatient and office-based, must continue to improve quality, efficiency, access and service. The responsibility of post- discharge care can not only be an issue for hospitalists. A balanced approach with shared responsibility and accountability among inpatient, office-based, and ED providers will be necessary to improve patient care.

Studying episodes of care to find opportunities for improvement is very important. However, we disagree with the conclusions drawn by the authors of this particular analysis.

Carl Reynolds Balazs Zsenits Walter Polashenski


1. Kuo YF, Goodwin JS. Association of hospitalist care with medical utilization after discharge: evidence of cost shift from a cohort study. Ann Intern Med. 2011;155:152-9.

Conflict of Interest:

None declared

Submit a Comment/Letter

Summary for Patients

Clinical Slide Sets

Terms of Use

The In the Clinic® slide sets are owned and copyrighted by the American College of Physicians (ACP). All text, graphics, trademarks, and other intellectual property incorporated into the slide sets remain the sole and exclusive property of the ACP. The slide sets may be used only by the person who downloads or purchases them and only for the purpose of presenting them during not-for-profit educational activities. Users may incorporate the entire slide set or selected individual slides into their own teaching presentations but may not alter the content of the slides in any way or remove the ACP copyright notice. Users may make print copies for use as hand-outs for the audience the user is personally addressing but may not otherwise reproduce or distribute the slides by any means or media, including but not limited to sending them as e-mail attachments, posting them on Internet or Intranet sites, publishing them in meeting proceedings, or making them available for sale or distribution in any unauthorized form, without the express written permission of the ACP. Unauthorized use of the In the Clinic slide sets will constitute copyright infringement.


Buy Now for $32.00

to gain full access to the content and tools.

Want to Subscribe?

Learn more about subscription options

Related Articles
Related Point of Care
Topic Collections
PubMed Articles
Forgot your password?
Enter your username and email address. We'll send you a reminder to the email address on record.