Many studies have been published in recent years about the effect of hospitalists on outcomes, efficiency, and cost-effectiveness of care (see also “In the Literature,” p. 30). While the studies have demonstrated varying results, the majority suggest that hospitalists contribute positively to care. And the tremendous growth of hospitalist programs nationwide indicates that hospital administrators and others agree.
Does this mean that there have been enough studies about the cost-effectiveness and efficiency of hospitalists, and about outcomes relating to hospitalists? Also, where should hospitalist research go next?
Looking Back
To date, the results of studies regarding hospitalists and their effect on outcomes and cost-effectiveness have varied. Most suggest positive correlations:
- Diamond, Goldberg, and Janosky demonstrated a 54% decrease in hospital readmission rates and shorter LOS when a community teaching hospital implemented full-time faculty hospitalists.1
- Auerbach, Wachter, and colleagues studied 5,308 patients cared for by hospitalists and community physicians at a community teaching hospital. They found that the voluntary hospitalist service reduced lengths of stay and costs that were statistically significant in the second year the services were used.2
- Bellet and Whitaker compared traditional ward service with a hospitalist system of care at a pediatric teaching hospital and found that the average LOS was a day shorter for the patients care for by hospitalists.3
- A review of five years of evidence-based hospitalist studies showed an average 13.4% cost reduction, as well as a 16.6% LOS reduction.4
- Rifkin, et al, compared treatment provided by hospitalist and primary care physicians among patients with community-acquired pneumonia. The authors found that hospitalists’ patients had shorter LOS and reduced costs.5
- Wachter reviewed the data to date in 2002 and concluded that it supported the hypothesis that hospitalists can lead to improved efficiency without compromising patient outcomes or satisfaction.6
- Meltzer, et al, studied costs and outcomes associated with patients on an academic general medical service cared for by hospitalists and non-hospitalists. They found that the average adjusted costs were similar for hospitalists and non-hospitalists in the first year. However, hospitalist costs were reduced by $782 in year two. The authors also concluded that short-term mortality was lower for hospitalists as well, but, again, only in the second year.7
- Auerbach and Pantilat assessed the effects of hospitalists’ care on communication, care patterns, and outcomes of end-of-life patients. They found that hospitalists documented “substantial efforts” to communicate with their dying patients and their families; and this may have resulted in better care.8
- Hauer, et al, analyzed house staff and student evaluations of their attending physicians and internal medicine ward rotations at two university-affiliated teaching hospitals over a two-year period. They found that trainees reported they received more effective teaching and more satisfying inpatient rotations when supervised by hospitalists.9
Studies that “go under the hood and answer questions about the mechanisms by which hospitalists improve outcomes” also will be useful. “Hospitals need to realize that hospitalists aren’t a magic bullet. It’s not as simple as implementing a hospitalist model of care and costs go down.”
—Peter Lindenauer, MD, MSc
A few studies have indicated that hospitalists may have less impact on costs and outcomes. Among them:
- Smith, Westfall, and Nicholas performed a retrospective chart review of HMO critical care patients and found that the mean charge by primary care physicians ($5,680) was significantly lower than that of the hospitalists ($7,699). The authors suggested that “claims of better and cheaper care by hospitalists need further investigation” and that HMOs should not mandate the use of hospitalists.10
- Kearns, et al, compared clinical outcomes and care costs for patients treated by hospital- and clinic-based attending physicians. The researchers detected no difference in costs or clinical outcomes associated with either type of physician.11