Objective: To evaluate the extent to which patient-reported outcomes (PROs) (eg, health-related quality of life) are distinct from clinical outcomes following bariatric surgery.
Background: Hospital quality measurement often focuses on traditional clinical outcomes (eg, complications). However, PROs may provide a unique perspective regarding performance, particularly for common, low-risk procedures.
Methods: We used data from 11,420 patients who underwent bariatric surgery (2008-2012) from the Michigan Bariatric Surgery Collaborative (39 hospitals). We included both short-term (30-day complication rates) and long-term (1-year weight loss and comorbidity resolution) outcomes. For PROs, we used health-related quality of life assessed by the Health and Activities Limitations Index (HALex) and Bariatric Quality of Life (BQL) index preoperatively and at 1 year. We used multivariable linear regression to determine the association between these PROs and both short and long-term clinical outcomes, adjusting for patient factors and the type of surgical procedure.
Results: After adjustment for risk factors and surgical procedure, hospital rankings based on PROs (either the average change in HALex or BQL scores) were not correlated with hospital rankings based on complications. In contrast, both PRO measures were correlated with weight loss. Specifically, the average change in HALex score (R = 0.24, P < 0.002) and average change in BQL score (R = 0.44, P < 0.001) were correlated with hospital average percent excess. One PRO measure-BQL score-was correlated with a decline in the need for medications due to associated comorbidities (R = 0.16, P < 0.01). After accounting for short and long-term clinical outcomes, between 15% and 44% of the variation in PROs remained unexplained at the hospital level.
Conclusions: Patient-reported outcomes are not correlated with early perioperative events, but are correlated with measures of clinical effectiveness after bariatric surgery. A comprehensive approach to surgical quality should incorporate both clinical events and self-reported measures of health status throughout the short and long-term recovery period.