Proxy-interpreter survey agreement is higher for discrete objective events than for subjective day-to-day-variable self-report domains

When patient-reported outcomes are collected through interpreter proxies versus professional interpreters, method agreement is highest for discrete, objective, memorable events (e.g., readmission, reoperation) and lower for subjective items whose ratings fluctuate day to day or whose definitions are ambiguous (e.g., EQ-5D anxiety/depression, complications) — so residual disagreement reflects the survey items' inherent variability more than the interpreter type.