Abstract
Many hospitals and orthopaedic surgery teams across Canada have instituted quality improvement (QI) programs for hip and knee arthroplasty. One of the common goals is to reduce hospital length of stay (LOS) in order to improve operational efficiency, patient flow and, by achieving this, provide improved access for patients to arthroplasty surgery. A common concern among surgeons and care providers is that hospital readmission rates will increase if LOS is significantly reduced. This study assesses the relationship between LOS and readmission rates in Alberta over a six year period during a focused QI initiative targeting LOS.
Data from all patients undergoing primary elective total hip or knee arthroplasty in Alberta between 2010 and 2015 was captured through a provincial QI program. Patient characteristics captured included age, gender, joint replaced, and pre-surgical co-morbidities. Patient LOS and all-cause hospital readmissions within thirty days from the initial discharge were captured through provincial data repositories, including the Discharge Abstract Database (DAD), operating room information systems, electronic medical records, and comorbidity risk grouper (CRG) data. Three longitudinal analyses were performed: 1) the crude and risk adjusted length of stay and 30-day readmission rates were calculated, 2) the population was grouped into two 3-year subsets and compared using t-test (acute LOS) and chi-square (30-day readmission), and 3) a multivariable regression analyses was performed to determine the rate of change and statistical significance in acute LOS and 30-day readmission between the two time periods.
The number of patients undergoing elective lower extremity arthroplasty in the province during the six-year study period (2010–2015) was 48,760 patients. Fifty-nine percent were female and forty-one percent were male. Mean age of the cohort was 66.9 years. Thirty-nine percent of patients had a total hip arthroplasty and 61% had a total knee arthroplasty. Forty-five percent of patients had no pre-surgical risk factors, 27% had one risk factor, and 28% of the patients had 2 or more risk factors. During the quality improvement program risk-adjusted length of stay improved from a mean of 4.82 days (in 2010–2012) to 3.90 days (in 2013–2015) (p<0.01). Controlling for differences in age, sex, joint replaced, and pre-surgery risk factors, the acute LOS declined by 0.32 days between the two time periods (p<0.001).
Quality improvement programs that target reduced LOS can avoid increasing 30-day hospital readmission rates. This has significant implications for inpatient resource utilisation for lower extremity arthroplasty surgery and for improving patient flow.
