Background: Hospital length of stay (LOS) is an important indicator of the quality, safety, efficiency, and cost of inpatient care. Hospitalists are often charged with identifying the root causes for excess LOS in order to guide improvement efforts. Our institution noted an increase in the average inpatient LOS for general medicine patients in 2014. We sought to determine whether prolonged hospitalizations had also increased and to characterize this patient subset.
Methods: We conducted a retrospective cohort study using administrative data of general medicine patients discharged from inpatient status from our urban, quaternary-care academic medical center between September 1, 2011, and June 30, 2014. Prolonged hospitalization was defined as LOS > 30 days. We performed a multivariable logistic regression analysis to assess the association between year of discharge, socio-demographic characteristics, and clinical variables with prolonged LOS.
Results: Of 16,984 discharges, 184 (1.1%) demonstrated prolonged LOS. Prolonged hospitalizations contributed 0.9 days to overall LOS (5.3 days) from January through June 2014 and 0.4 days to overall LOS (4.5 days) from September 2011 through December 2013. Prolonged hospitalizations were more likely to occur in 2014 compared to prior years (REF = 2014; 2011 OR 0.45, 95% CI 0.22-0.95; 2012 OR 0.25, 95% CI 0.14-0.44; and 2013 OR 0.56, 95% CI 0.34-0.91). Patients with prolonged LOS were more likely to be younger (OR 0.73 per 10 years increase in age, 95% CI 0.64-0.84) and have Medicaid insurance (OR 4.88, 95% CI 2.52-9.47, REF = Medicare), but not more likely to be indigent/self-pay. Compared to patients without prolonged LOS, those with prolonged LOS were more likely to require an ICU stay (OR 4.98, 95% CI 3.21-7.72), undergo a surgical procedure (OR 3.14, 95% CI 1.79-5.52), receive a palliative care consult (OR 2.40, 95% CI 1.27-4.54), and be discharged to a post-acute care facility (OR 18.08, 95% CI 10.46-31.24). The top 5 primary diagnoses by ICD-9 code for patients with prolonged LOS were unspecified sepsis, alcoholic cirrhosis, sickle cell disease with crisis, methicillin-sensitive Staphylococcus aureus septicemia, and systemic lupus erythematosis.
Conclusions: Prolonged hospitalizations in a small subset of patients can be an important contributor to overall LOS. Prolonged hospitalizations for general medicine patients have increased in 2014 and particularly affect Medicaid enrollees with complex hospital stays who cannot be discharged home. Further studies are needed to determine the reasons for excess LOS in this population, which may include delays in discharge planning or lack of bed availability at post-acute care facilities. The impact of Medicaid expansion under the Affordable Care Act on prolonged LOS also merits further investigation as healthcare systems seek to streamline transitions across care settings.
To cite this abstract:Anderson M, Glasheen J, Anoff D, Pierce R, Capp R, Jones C. Impact of Prolonged Hospitalizations on Overall Length of Stay for General Medicine Patients. Abstract published at Hospital Medicine 2015, March 29-April 1, National Harbor, Md. Abstract 43. Journal of Hospital Medicine. 2015; 10 (suppl 2). https://www.shmabstracts.com/abstract/impact-of-prolonged-hospitalizations-on-overall-length-of-stay-for-general-medicine-patients/. Accessed July 22, 2019.