In this edition of Journal of General Internal Medicine, we highlight three original investigations1–3 that seem quite different at first glance: reduction of falls in inpatients, understanding reasons for prolonged hospital stay, and forces shaping the quality and quantity of hospitalist services. But what these articles share, other than their high-quality methods and exposition, is their focus on the hospital setting. Thus, their publication indicates the importance JGIM places on creating a scientific and editorial home for the growing field of hospital medicine.
Krauss et al.'s article2 is emblematic of inpatient research by examining inpatient falls from both the patient and systems perspective. Falls are seen as a patient safety target by a number of external agencies, not to mention being of acute importance to clinicians. Thus, any information which can be used to identify those at risk—especially if these risk factors are modifiable—would be highly useful. The risk factors identified by Krauss et al. in their case-control study include clinical factors (e.g., visual, cognitive, and gait disturbances) that have been seen elsewhere.4 While we await inpatient-focused intervention trials,5 Krauss et al.'s results suggest that coordination of efforts to address factors which might otherwise be considered “outpatient” issues may improve care both in the hospital and at home (or skilled nursing facility) after discharge.
Krauss et al.'s findings of a relationship between the number of nurses available to vulnerable patients and the risk of falls are novel in that they suggest that hospital systems are associated with inpatient adverse events. While it is possible that there was bias in the assignment of nurses to patients who did not fall (that is, control patients who were at higher risk may have been preselected for additional nursing attention), their results closely mirror recent work connecting lower nurse staffing ratios with poorer inpatient outcomes, specifically failure to rescue in surgical patients.6,7 It remains unclear whether addressing these risk factors will modify the risks to patients (e.g., would non-RN “sitters” have been adequate to prevent falls? Does use of bed alarms allow for fewer nurses to “cover” more patients?), but results from this study partially connect the dots between lower patient-to-nurse ratios and improved outcomes.
On any day, house officers, social workers, and case managers can readily point, sometimes with exasperation, to a number of patients who are staying in the hospital for nonclinical reasons. Nevertheless, the topic of how many patients are in fact “just waiting for discharge” has not been well addressed by health services researchers. Much of the available data were derived during an era marked by less economic pressure on hospitals,8,9 and before the advent of hospitalists. Carey et al.'s findings that 13% of the hospital days at their site were for nonmedical reasons3 are fairly similar to the 14% to 21% reported during that earlier, perhaps gentler time.8 Again, consistent with previous work, they remain most often related to waiting for procedures, tests, or consultations to take place and waiting for skilled nursing facility beds.9 This is both better and worse news: while it is important to understand the factors contributing to “unnecessary days,” these factors largely defy easy solutions. Carey et al.'s data do not, for example, tell us how many of the patients waiting for a nursing home bed were waiting because of insurance issues (many uninsured and Medicaid patients face this problem regularly), or because in-hospital care coordination was poor (e.g., the case manager did not hear about a potential nursing home referral until the day of planned discharge).10,11 Similarly, it is unclear whether there is a business case for hospital services that are truly available every day, all day, all year. Cardiologists have been able to make a strong case for keeping catheterization facilities open “24-7,” but this argument has been based on the relatively clear benefits of primary angioplasty,12 not bed management or throughput. Carey et al.'s results may help make the case for “off” hours testing, or at least fuel analysis of the clinical and financial implications of the status quo. Future research linking unnecessary days not only to dollars but to higher risk for iatrogenesis and complications would further strengthen the argument to increase inpatient services and coordination of nursing home placement.
The article by Pham et al. points out how both market pressures and hospital requirements are exerting a powerful influence on the development of hospitalist practices.1 Consistent with the best evidence supporting hospitalist practice, most respondents reported that hospitalists are largely viewed as an approach to reducing costs and lengths of stay rather than as a way to improve care. As a result, in regions where hospital beds are in short supply hospitalists are viewed as a way to improve throughput. In addition, hospitalist services were reported to be in demand by primary care physicians, who were increasingly shifting the focus of their practice to the outpatient setting only.
Alignment of provider and institutional goals has been considered a potential adjunctive method for improving quality of care.13,14 Pham et al.'s results suggest how alignment of incentives may be operationalized in the hospital setting; hospitalists who received financial support from their hospital more often performed nonclinical tasks that might be considered aligned with institutional initiatives (e.g., quality improvement). On the other hand, hospitalist programs sponsored by hospitalist practice management groups or insurance plans were less likely to be involved in these nonremunerative activities. Few data exist to describe whether organizational differences between hospital-sponsored hospitalist services and other models produce measurable differences in efficiency (e.g., costs and length of stay) or quality of care, but such studies are clearly needed as the growth of hospitalist physician practice groups expands.
Pham et al.'s observation that growth of hospitalist services is in part driven by primary care physicians' perception of malpractice risk and the need to care for the uninsured is novel. This creates a situation where hospitalists are not only assuming the care of sicker patients, but a higher proportion of financial and legal liability for that care as well. This certainly does not seem a winning combination—one which Pham et al.'s results do not provide a solution for—and is worthy of future research.
Many generalists have viewed the hospitalist movement with interest mixed with angst, because the model violates the longstanding notion of general internists as physicians who provide care across the continuum of care. To this end, it is important to point out that the vast majority of hospitalists are in fact general internists. In one survey, not only were hospitalists generalists (e.g., family physicians, pediatricians, internists), a substantial proportion of hospitalists were primary care generalists before they became hospitalists (L. Wellikson, MD, oral communication, 2004),15 and continued to work in close cooperation with their colleagues even after focusing on inpatient care. While this fact may not be generally known, it provides an important insight into the intellectual roots of hospitalist practice and research. In an era when primary care generalism is under considerable pressure, the rapid growth of hospitalists (more than 10,000 nationwide by some accounts; L. Wellikson, MD, oral communication, 2004) represents a huge opportunity for generalism.
Emergence of an inpatient-focused group of generalists provides an important opportunity to play to the traditional strengths of generalists. In addition, appearance of hospitalists and inpatient generalism has coincided almost perfectly with important new opportunities in clinical care and research. Emblematic of such an opportunity is the drive to improve safety and quality at the institutional level; at many hospitals, hospitalists are the “first call” for these of efforts,15 ranging from implementation of care guidelines to methods for meeting duty hour regulations. This has provided generalism an important foothold in an area of critical importance to institutions and policymakers, and plays to the strengths of generalists as care integrators and cross-specialty collaborators who look at the entire patient, not just a single organ or site of care. We at JGIM look forward to these manuscripts and—by continuing to publish the best of these articles—hope to aid in the dissemination of generalist-derived insights, enhance the field of general internal medicine, and push forward generalism's contribution to hospital care of our patients.