Regional variation in the use of revascularisation for peripheral artery disease in Japan: a cross-sectional observational study

OBJECTIVES: We aimed to investigate the regional variations in the number of interventions and surgeries for peripheral artery disease (PAD) and explore the major determinants of the variations. DESIGN: Cross-sectional study. SETTING: The Japanese Ministry of Health, Labour and Welfare National Database and Diagnostic Procedure Combination database in 2018. DATA: The rates of endovascular treatment (EVT), bypass surgery per 100 000 individuals in the population were calculated for all 47 prefectures in Japan. The total annual changes in the rates of EVT and bypass surgery in Japan from 2012 to 2019 were calculated. ANALYSIS: A linear regression model was developed with rates of EVT and bypass surgery as dependent variables and regional medical supply in each prefecture as explanatory variables. These regional factors included the rate of percutaneous coronary intervention (PCI) for angina, the numbers of cardiovascular specialists, specialists in cardiac surgery, interventional radiology (IVR) training facilities and cardiovascular surgery training facilities, per 100 000, respectively. RESULTS: There was a 5.7-fold difference (143 and 25 per 100 000 individuals aged ≥40 years) in the highest and lowest EVT rates. The highest and lowest rates of bypass surgery were 34 and <10 per 100 000 individuals aged ≥40 years in a prefecture, respectively. The rate of PCI contributed most significantly positive to the rate of EVT (p<0.001). However, the numbers of IVR and cardiovascular surgery training facilities had significant positive and negative relationships, respectively, with the rate of EVT. The numbers of specialists in cardiac surgery and cardiovascular specialists had significant positive (p=0.01) and negative (p=0.01) correlations, respectively, with the rate of bypass surgery. CONCLUSIONS: Considerable regional variations in the rates of revascularisation for PAD were found. Unbalanced presence of medical resources, preference of suppliers and the training system had larger effects on the regional variation in Japan.