BMJ Open (Mar 2023)
Regional practice variation in pelvic organ prolapse surgery in Tuscany, Italy: a retrospective cohort study on administrative health data
Abstract
Objectives To explore determinants of practice variation in both access, and quality and efficiency of surgical care for pelvic organ prolapse (POP).Design and setting A retrospective cohort study employing administrative health data from the Tuscany region, Italy.Participants All women over 40 years hospitalised for apical/multicompartmental POP reconstructive surgery (excluding anterior/posterior colporrhaphy without concomitant hysterectomy) from January 2017 to December 2019.Outcomes We first computed treatment rates just for women residing in Tuscany (n=2819) and calculated the Systematic Component of Variation (SCV) to explore variation in access to care among health districts. Then, using the full cohort (n=2959), we ran multilevel models for the average length of stay and reoperations, readmissions and complications, and computed the intraclass correlation coefficient to assess the individual and hospital determinants of efficiency and quality of care provided by hospitals.Results The 5.4-fold variation between the lowest-rate (56/100 000 inhabitants) and the highest-rate (302/100 000) districts and the SCV over 10% confirmed high systematic variation in the access to care. Higher treatment rates were driven by greater provisions of robotic and/or laparoscopic interventions, which showed highly variable usage rates. Both individual and hospital factors influenced quality and efficiency provided by hospitals, but just low proportions of variation were explained by hospital and patient characteristics.Conclusions We found high and systematic variation in the access to POP surgical care in Tuscany and in quality and efficiency provided by hospitals. Such a variation may be mainly explained by user and provider preferences, which should be further explored. Also, supply-side factors may be involved, suggesting that wider and more uniform dissemination of robotic/laparoscopic procedures may reduce variation.