Geriatricians and Hospitalists: Opportunities for Partnership

One of the most important recent changes in U.S. health care is the rise in the fraction of hospitalized patients cared for by hospitalists. At the inception of the hospitalist movement, leaders predicted a profound effect on academic medical centers. They also predicted that hospitalists would be able to deliver more-efficient, “high value” care, with value defined as the quality of care divided by its cost.1 The first prediction has come true, with growing hospitalist groups at most major academic medical centers and academic hospitalists becoming leaders in quality improvement. The question remains what the effect has been on the overall quality of care provided to older patients. One large retrospective cohort study of 76,926 patients aged 18 and older used multivariable models to compare length of stay (LOS), rate of death, and costs of patients cared for by hospitalists, family physicians, and general internists. Hospitalists in this study were defined according to physician specialty codes. Patients cared for by hospitalists had a slightly shorter hospital stay (adjusted difference of 0.4 days) than those cared for by general internists and family physicians. Costs were very modestly lower for those cared for by hospitalists than for those cared for by general internists, but there was no significant difference in costs between those cared for by hospitalists and family physicians. Rates of death and of 14-day readmission were similar.2 A plethora of small observational studies have largely had similar findings, with patients being cared for by hospitalists having modest but statistically significantly shorter LOS,3–7 although many of these are small, single-institution studies and had few hospitalists caring for all of the patients in the hospitalist arm. Data from the Multicenter Hospitalist (MCH) Trial, a prospective multicenter observational study of the effect of hospitalist care at academic medical centers, have not indicated differences in LOS between patients cared for by hospitalists and nonhospitalists. For example, a recent analysis of 450 patients with upper gastrointestinal hemorrhage from the MCH Trial showed similar mean LOS for patients cared for by hospitalists and nonhospitalists and higher costs for patients cared for by hospitalists.8 Similarly, data from the MCH Trial did not show any significant differences in LOS or costs between patients with decompensated heart failure cared for by hospitalists and nonhospitalists.9 Even if we accept that hospitalists reduce LOS modestly, is that a clinically significant finding, and what are the implications for elderly patients? Kuo and colleagues have previously reported on the growth in the care of older adults by hospitalists, but few studies have focused on outcomes of hospitalist care in the elderly population.10 One small observational study at a community teaching hospital showed statistically significantly shorter LOS and lower total costs that were most pronounced in patients aged 65 and older.7 There have been a few studies of the effect of hospitalist management or comanagement of elderly patients with hip fractures, with one historical cohort study showing significantly shorter LOS (10.6 vs 8.4 days) in patients with hip fractures comanaged by hospitalists, with no difference in inpatient deaths, 30-day readmission rates, or 1-year mortality.11,12 Why has there been so little research on the effect of hospitalists on elderly patients? Hospital medicine remains a popular career choice for young internists, for many reasons, including the ability to take care of acutely ill patients, the ability to engage in quality improvement projects, and work–life balance, but it is probably safe to say that few hospitalists specifically choose the field for the opportunity to care for older adults and target research and quality improvement initiatives toward that population. Few hospitalists have special training in geriatric medicine, and few hospitalist groups have programs addressing the needs of elderly patients, even though hospitalists care for many elderly patients and may even care for patients in skilled nursing facilities.13 Geriatricians, with a few notable exceptions, have been largely absent from the debate over the merits of the hospitalist system. In this issue of the Journal of the American Geriatrics Society, Kuo and Goodwin14 report results of an ambitious study to understand the health services implications of nationwide hospital care of Medicare-eligible persons by hospitalists and nonhospitalists. Because several studies have shown shorter LOS associated with hospitalist care, the authors further attempt to understand various patient and hospital characteristics that may or may not be associated with this shorter LOS and whether this phenomenon is changing over time. This is a retrospective cohort study using a 5% national sample of Medicare beneficiaries, winnowed down to approximately 300,000 subjects for the characteristics analysis and approximately 2 million for the trend analysis. The latter sample is approximately half of all hospitalizations in the 5% sample. They employ elaborate and appropriate multivariable methods to control for a large number of variables and interaction terms, using a very large sample size to do so. They found that the adjusted LOS was 0.36 days shorter in patients cared for by hospitalists, a finding similar to a previous study.2 Kuo and Goodwin also found a greater difference in LOS in the oldest patients, with a difference of 0.63 days (95% confidence interval (CI)=0.55–0.71) of adjusted LOS for patients aged 85 and older. The shorter LOS with hospitalist care in this study increased over time during the study period, with a difference of 0.02 (95% CI=0.09–0.12) days in 2001/02, 0.22 (95% CI=0.14–0.30) days in 2003/04, and 0.35 (95% CI=0.29–0.41) days in 2005/06 (P<.001 for trend). Even though it is tempting to extrapolate that the shorter LOS represents “efficiency” and even better quality of care, these assertions about efficiency and quality simply cannot be made from the analysis presented. Studying quality outcomes for care of complex, acutely ill older patients with multimorbidity is notoriously difficult, unless the care provided currently by some subgroup is unusually bad, which is unlikely to be the case. A premise repeated in the article a few times is that the authors did (and readers should) concentrate on clinical significance and not statistical significance, because the sample size is so large that statistical significance is common and not always very meaningful—essentially two levels of potential Type I error. Are the findings in this study clinically significant? We would argue that these findings are clinically significant in that geriatricians can plausibly expect their elderly patients to spend a bit less time in the hospital if they are being cared for by hospitalists. Whether this implies better or worse care cannot be ascertained from these data or from any other good-quality published information on the care of elderly patients by hospitalists to date. It is tempting to assume that a shorter LOS would benefit an elderly patient, especially for geriatricians, who are acutely aware of the risk of iatrogenic complications in older adults, and the dramatic functional decline that can occur during a hospitalization,15–18 but older patients are at high risk for hospital readmission and postdischarge adverse events. One study in a similar population of Medicare beneficiaries showed a 19.6% 30-day readmission rate and 34% 90-day readmission rate,19 and another study found that 19% of patients in the sample suffered an adverse event after discharge.20 The relationships between LOS, mortality, rates of in-hospital complications common in older adults, readmission, and postdischarge adverse events are not clear. Future well-done studies on outcomes of hospitalist care in older adults that address readmission rates; mortality; postdischarge adverse events; and physician, patient, and family satisfaction would be welcome. However, even if these studies showed that hospitalist care was detrimental to elderly patients, hospitalists will continue to care for a significant proportion, if not the majority, of elderly hospitalized patients. Given that hospitalists are here to stay, we would like to see hospitalists and geriatricians synergize their energies toward initiatives to improve the care of hospitalized elderly patients. There have already been some important partnerships between geriatricians and hospitalists. For example, geriatricians have partnered with hospitalists to develop innovations to improve transitions of care for elderly patients.21,22 More work in this area is essential, because hospitalists are rarely familiar with the nursing home setting, and transitions in and out of the hospital can be a vulnerable time for elderly patients. The Curriculum for the Hospitalized Aging Medical Patient program, an initiative to improve the teaching of geriatrics by general internists in academic medical centers, is another example of a successful partnership between hospitalists and geriatricians.23 One recent study in this journal showed a lower LOS index and lower hospital costs in elderly patients receiving proactive geriatric consultation in collaboration with hospitalists.24 Patient outcomes other than LOS were not studied, so the effect on quality of care is not clear, but this model requires further study. If these results are sustainable and can be replicated, proactive geriatric consultation for elderly patients being cared for by hospitalists may be a way to ensure that attention is paid to the special needs of elderly patients in the hospital. Geriatricians working in hospitals have conducted many of the landmark studies providing evidence on how to improve the care of elderly adults, the Hospital Elder Life Program and Acute Care of Elders units being two examples.25,26 Neither of these interventions is widely disseminated; there have been some attempts to implement some aspects of these interventions, with varying level of success, and further study is required, but these would also be natural partnerships for geriatricians and hospitalists interested in patient safety and quality improvement.27,28 Many hospitalists seem to enjoy aspects of hospital work that are part of geriatrics, such interdisciplinary care, transition improvement, and palliative medicine. The hospitalist model is far from perfect but is here to stay. Hence, geriatricians should continue to work with hospitalists to improve outcomes for elderly patients. Conflict of Interest: The editor in chief has reviewed the conflict of interest checklist provided by the authors and has determined that the authors have no financial or any other kind of personal conflicts with this paper. Author Contributions: Both authors contributed equally to concept, design, and preparation of this manuscript. Sponsor's Role: None.