Background: Readmission and length of stay are two hospital level metrics commonly used to assess the performance of hospitalist groups. Health care systems implement strategies aimed at reducing both. It is possible however that tactics aimed at improving one measure in individual patients may adversely impact the other. Studies in focused populations of patients hospitalized with a specific diagnosis have found association between individual patient length of stay and risk of readmission. We sought to analyze the impact of length of stay on readmission risk in a hospitalized general medical population to see whether patients with lower length of stays were readmitted more frequently to the hospital.
Methods: We performed a retrospective analysis of records from patient admitted to our institution between 1/2016 and 12/2018. We selected only inpatients. We recorded demographic variables and the outcomes length of stay (LOS) and 30 day readmission. We defined elevated LOS as greater than median. We excluded patients who expired, left against medical advice, or were transferred to other facilities. We performed both univariate and multivariate analyses.
Results: 78,826 patients were initially identified in the analysis, of which 3,876 were excluded resulting in 74,950 patients included in the final analysis. 56% of patients were female. 25.3 % were African American, 16% Hispanics and 53% Caucasian. The average age was 55 (SD 19). 10,268 patients were readmitted within 30 days. The rate of readmission decreased significantly over the years from 14.5% (2016) to 14.1% (2017) and 12.5% (2018) (p <0.001). LOS varied from 5.36 (SD 7.5) in 2016 to 5.24 (SD 7.1) in 2017 to 5.51 (SD 7.4) in 2018 (P<0.001). Using a multivariate regression analysis, we found that higher LOS was associated with higher readmission risk (1.72 [1.63-1.81], P <0.001). This association was sustained in patients discharged with medical and surgical diagnoses (p<0.001 for both). Readmitted patients had a significantly higher LOS: 6.64 (SD 8) vs. 5.16 (8) P<0.001. Among readmitted patients, 6,180 patients had a high LOS (60.2 %) while 4,088 had a low LOS (39.8%) (P<0.001).
Conclusions: Contrary to our initial hypothesis, we found that patients with higher LOS had a higher likelihood to be readmitted to the hospital after adjusting for other variables. It is possible that factors not captured in the current dataset may help explain both the increase in Los and readmission risk.