An exercise ankle-brachial index (ABI) test can provide further insight into the functional significance of peripheral artery disease (PAD). The variability in its use, associated patient factors and its relation to patients' symptoms are unknown. From the international PORTRAIT registry, we identified 1131 patients with PAD. We fit a hierarchical logistic regression model, adjusting for patient factors, country and site, to examine predictors of and variation in ordering exercise ABI testing. We also examined the associations between test components and health status as quantified by the Peripheral Artery Questionnaire (PAQ) using semi-parametric regression methods. Testing was ordered in 22% in the United States versus 80% in the Netherlands and 90% in Australia. Testing was likely to be performed if the patient was male, younger, had typical symptoms and a higher resting ABI, with substantial variability across sites (median odds ratio=5.9, 95% CI: 3.2-19.5). Adjusting for country and site, the resting ABI and all exercise ABI metrics were associated with the PAQ Physical Limitation score. In addition, important components of the test, namely time to onset of claudication, pain-free walking distance (PFWD), and maximum walking distance (MWD), were also associated with PAQ Symptoms and Summary scores. More importantly, even after adjusting for resting ABI, a patient with a post-exercise ABI of 0.29 (25th percentile), compared to 0.61 (75th percentile), achieved 4.4 (95% CI: 0.4-8.4, p=0.031) points less on the PAQ Physical Limitation score. Exercise ABI test use is remarkably variable, and less used in the United States. Its data, specifically PFWD and MWD, might help in objectively assessing the impact of PAD on patients' functioning and quality of life.