Abstract & Commentary
Hospitalist Workload May Impact Quality of Care
By Jennifer A. Best, MD
Assistant Professor, University of Washington School of Medicine, Seattle, WA
Dr. Best reports no financial relationships in this field of study
Elliott DJ, Young RS, Brice J, Aguiar R, Kolm P. Effect of hospitalist workload on the quality and efficiency of care. JAMA Intern Med 2014;174:786-793
Hospitalists are frequently asked in many systems to work harder and see more patients, yet the effect of the hospitalist workload on the quality and efficiency of care has not been systematically evaluated. In academic institutions, these demands on hospitalists have been exacerbated recently by the shift of patient care from from resident housestaff services to non-teaching hospitalists to meet new graduate medical education duty hour limitations. Anecdotally, a typical hospitalist workload ranges from 10-15 daily encounters, but it has been reported that hospitalists often exceed a workload they feel to be safe.1 This situation is exacerbated in many cases by both personal and system incentives for productivity.
Elliott and colleagues at Christiana Care Health System in Delaware designed a retrospective cohort study of a population of general medicine and step-down patients admitted to a single private hospitalist group. This group is large, composed of 20-35 providers caring for patients independently, without the assistance of housestaff teams. Patient encounters were selected for inclusion in the study if a patient was greater than 18 years old with an admission and discharge bill submitted by a hospitalist from the group. Patients were excluded from study participation if they were admitted to the ICU, admitted on observation status, transferred from an outside facility or had atypically short (0.5d) or long lengths of stay (>30d).
By analyzing daily hospitalist workload during the study period, as determined by relative value units (RVUs) and the total number of patients for whom a bill was submitted, a number of outcomes were evaluated. Regarding efficiency, the primary outcomes in question were length of stay (LOS) and cost of care. Regarding quality, researchers included hospital mortality, rapid response team activation, 7- and 30-day readmission rate, and patient satisfaction (measured by responses to questions about overall hospital rating for the Hospital Consumer Assessment of Healthcare Providers and Services [HCAHPS] survey). Secondary measures of satisfaction included those HCAHPS questions concerning physician interactions — concern, communication, and courtesy. Researchers adjusted all outcomes for both provider and patient characteristics. Provider/hospital characteristics included whether there was continuity of care between providers (by looking at bills submitted over consecutive hospital days); the overall hospital census (with occupancy categorized as low <75%, medium 75-85%, and high >85%); and overall hospital flow (assessed through the surrogate of average time from order to completion of an echocardiogram).
During the study period, the Christiana group managed 33,137 admissions. From this group, 20,241 admissions of 13,916 unique patients met study inclusion criteria. The mean hospitalist daily census was 15.5 encounters, yielding 28.6 RVUs. All outcomes were evaluated across a census scale from 11-22 patients and RVU scale from 15-40; these numbers corresponded with a range between the 5th and 95th percentiles. When hospital occupancy was low, LOS ranged from 5.5 to 7.5 days on the workload continuum from low to high. At a medium occupancy, LOS rose exponentially beyond the mean daily RVU level of 25 and daily census of 15 encounters. When hospital occupancy was high, the correlation of workload with LOS was J-shaped, with an initial decrease, but a later sharp rise again observed with higher workloads. When cost of care was measured after correction for LOS, cost rose by $111 for each unit-rise in RVU and $205 for each unit-increase in patient census. Workload was not found to have significant effects on hospital mortality, RRT activation, readmission, or patient satisfaction.
The study did have some limitations in that data collection was limited to a single private hospitalist group. Results may not generalize to academic or housestaff hospitalist teams. Researchers did not consider the level of provider experience in their analysis, though it stands to reason that more experienced providers may be more efficient and knowledgeable in system navigation.
In summary, hospitalists with higher workloads demonstrated less efficient and more expensive care. Health systems administrators should be aware of this correlation and consider the potential of excessively high workloads to offset potential productivity gains, as a result of longer LOS and higher care costs. Such systems would be wise to consider possible solutions such as increased support for non-clinical tasks and scheduling providers in a manner that maximizes care continuity. Additionally, incentive structures should consider not only patient volumes but care quality. Practicing hospitalists should also be aware of these data, as they may be useful in negotiating employment contracts or incentives.
Reference
- Michtalk HJ, Yeh HC, Pronovost PJ, Brotman DJ. Impact of attending physician workload on patient care: a survey of hospitalists. JAMA Intern Med 2013;173(5):375-377.
