By: Jeffrey Maser  Dec. 07, 2017
Recent Dexur analysis of CMS claims data from 1,419 hospitals within the January - December 2016 time period revealed a moderate positive correlation of 0.48 between average length of stay and hospital readmission rate. 265 hospitals with an average length of stay of 3.3 days or more averaged a readmission rate of 19.35% while the 284 hospitals with an average length of stay of 2.29 days or fewer experienced an average readmission rate of 7.94%.
Average Hospital LOS (Days) | Sum of Group's Total DRG-470 Discharges | # of Hospitals in Group | Avg. LOS for Total Hospital Group | Readmission Rate for Total Hospital Group (%) | Correlation between Avg. LOS and Readmission Rate |
---|---|---|---|---|---|
All U.S. Hospitals | 406,360 | 1419 | 2.60 | 10.42 | 0.48 |
> 3.3 | 34,043 | 265 | 3.61 | 19.35 | 0.09 |
2.9 - 3.29 | 81,919 | 324 | 3.06 | 12.10 | 0.11 |
2.6 - 2.89 | 89,215 | 300 | 2.75 | 9.62 | -0.05 |
2.3 - 2.59 | 85,435 | 246 | 2.45 | 9.43 | 0.13 |
< 2.29 | 115,748 | 284 | 1.97 | 7.94 | 0.10 |
The connection between length of stay and readmission rate is a hotly contested topic and studies have yielded varying results. A 2002 study of Norwegian hospitals found that patients who experienced a shorter average LOS had a significantly higher risk for early readmission. On the other hand, a 2015 study on 95,294 bariatric surgery patients found that patients who were hospitalized for 3 days were twice as likely to be readmitted than those hospitalized for one day. Furthermore, patients hospitalized for more than 3 days were four times as likely to be readmitted than those hospitalized for only one day. The Dexur research in this article aligns with the findings from the second study, where excessive length of stay is associated with higher readmission rates.
Dexur analysts filtered on DRG-470 (major joint replacement or reattachment of a lower extremity without major complications) data for all U.S. hospitals with at least 11 instances of readmission. These 1,419 hospitals were then broken up into 5 groups organized by average length of stay. The 0.48 correlation found between average length of stay and readmission rate among the 1,419 hospitals was further compounded by this categorical analysis, as the group with the highest average LOS had a readmission rate more than double the average rate for the 3 groups with the lowest average LOS. An additional analysis was run only on hospitals that had had an average LOS of 3.7 days or more; 99 U.S. hospitals met this criteria and these hospitals accounted for an average readmission rate of 29.13%. Although future research involving mediating factors will continue to shed more light on this apparent correlation, the Dexur research on the subject indicates that as length of stay rises, readmission rates increase as well.