Is Age a Predictor for Length of Stay in an Acute Stroke Unit?

Abstract

Background and Rationale: The average length of stay (LOS) in major Australian hospitals for an acute stroke patient over 65 years is 9.4 days (ACSQHC 2013), but little data exists for LOS on acute stroke units. Increased LOS is associated with increased mortality and morbidity (Ingeman et al. 2011). Calvary Health Care stroke unit has a target LOS of 72 hours but we are unaware how patients perform across different age groups. We hypothesise that age is an independent predictor of increased LOS in the acute stroke unit. Methods: We retrospectively reviewed stroke unit data collected between 01/04/2014 to 31/12/2015. We classified each age group into those below 65 years, 65–74 years, 75–84 years, and 85 years and over. Data for LOS were divided into LOS less than 72 hours and LOS more than 72 hours. Admission NIHSS scores were divided into those with scores less than 4, 5–15 and scores ≥16. Odds ratio was calculated across each age group and LOS after correcting for NIHSS scores. Results: 236 of 351 (67.2%) patients were in the stroke unit for less than72 hours. Of 115 (32.8%) patients staying more than 72 hours; 24 (21%) were under 65, 28 (24%) were between 66–74 years, 38 (33%) were between 75–84 years and 25 (22%) were above 85 years. For NIHSS less than 4; patients in each age group were more likely to have a shorter LOS when compared to older patients more than 85 years (odds ratio 1.88, 1.25, 1.25 respectively) but none were statistically significant. Similar statistically non-significant results were seen with NIHSS 5–15 group (odds ratio 3.125, 1.13, 0.938 respectively) and NIHSS more than 15 group (odds ratio 0.5, 2.0, 0.85 respectively). Conclusion: Our study shows that increasing age is not a predictor for prolonged LOS in an acute stroke unit and only LOS should be used as an independent key performance indicator. We acknowledge low sample size is our study and ongoing study is required to validate these findings.