Several studies have found that hospitalists reduce cost and length of stay compared to nonhospitalists, with no adverse effect on quality of care or inpatient outcomes. However reduced length of stay could be accomplished by shifting care to the period after discharge. Our aim was to investigate differences after discharge in hospital readmissions, emergency department (ED) visits, outpatient physician office visits, short‐term nursing facility (SNF) use, and home health agency (HHA) visits between Medicare beneficiaries treated by hospitalists and those treated by nonhospitalists up to 1 year postdischarge.
Our study sample was a cohort of 4485 patients aged 65 and older who were enrolled in a quasi‐randomized trial of hospitalists at a large academic medical center between 2001 and 2003. Hospital administrative and clinical data for these patients were linked to Medicare inpatient, outpatient, HHA, and physician claims data using patient Social Security number with confirmatory match on patient sex (N = 4041 patients, 90% linkage rate). We measured the number of hospital readmissions, ED visits, physician office visits, SNF stays, and HHA visits within 7, 14, 21, 30, 60, 90,180, and 365 days postdischarge. Logistic regression was used to estimate the effect of hospitalists on the log odds of each outcome within 7‐365 days posldischarge. Negative binomial regression was used to estimate the effect of hospitalists on the number of each event within each time interval. Analyses were restricted to patients who were discharged live. All models controlled for patient age; sex; race; admission source; discharge status; diagnosis; Chanson comorbidity score; admission day of week, month, and year; and linear time trend. Robust standard errors were estimated to account for physician‐level clustering.
Hospitalists had no effect on the log odds or number of inpatient readmissions. Hospitalists reduced the likelihood of an ED visit from 7 (OR = 0.78, P = 0.05) through 90 (OR = 0.85, P = 0.05) days postdischarge. Hospitalists lowered the odds of a first physician office visit within 7 (OR = 0.77, P = 0.04) and 14 (OR=0.86, P = 0.02) days postdischarge, and the number of visits from 14 through 365 days (average marginal effect at 365 days = −0.85, P < 0.01). Hospitalists had no effect on the log odds of SNF use at any point. Hospitalists had a positive effect on HHA use within 7 (OR = 1.18, P = 0.03) and 14 (OR = 1.22, P < 0.01) days postdischarge and on the number of HHA visits within 14‐30 days postdischarge.
We found higher HHA utilization and lower ED and physician office utilization among patients treated by hospitalists compared with those treated by nonhospitalists. Future research will explore whether the greater use of HHA by hospitalists is responsible for the lower rates of ED and physician office use and the net effects on total health care costs postdischarge.
D. Meltzer, none; J. Chung, none.
To cite this abstract:Meltzer D, Chung J. Effects of Hospitalists on 1‐Year Postdischarge Resource Utilization By Medicare Beneficiaries. Abstract published at Hospital Medicine 2010, April 8-11, Washington, D.C. Abstract 94. Journal of Hospital Medicine. 2010; 5 (suppl 1). https://www.shmabstracts.com/abstract/effects-of-hospitalists-on-1year-postdischarge-resource-utilization-by-medicare-beneficiaries/. Accessed May 26, 2019.