Skip to main content
Open Access Publications from the University of California


UC San Francisco Previously Published Works bannerUCSF

Evaluating patient-reported outcome measures in Peru: a cross-sectional study of satisfaction and net promoter score using the 2016 EnSuSalud survey

Published Web Location
No data is associated with this publication.
Creative Commons 'BY' version 4.0 license


Patient-reported measures attempt to quantify the value health services provide to users. Satisfaction is a common summative measure, but often has limited utility in identifying poor quality care. We compared satisfaction and the net promoter score (NPS), which was developed to help businesses quantify consumer sentiment, in a nationally representative survey in Peru. We aimed to compare NPS and satisfaction as individual ratings of care, assess the relationship of patient-reported experience ratings to these outcome measures and consider the utility of these measures as indicators of facility performance based on reliability within facilities and capacity to discriminate between facilities.


We analysed the 2016 National Survey on User Satisfaction of Health Services, a cross-sectional outpatient exit survey. We assessed ratings by patient characteristics and compared the distributions of satisfaction and NPS categories. We tested the association of patient-reported experience measures with each outcome using multilevel ordinal logistic regression. We used intraclass correlation (ICC) from these models to predict minimum sample for reliable assessment and compared patient-reported experience measures in facilities with average satisfaction but below or above average NPS.


13 434 individuals rated services at 184 facilities. Satisfaction (74% satisfied) and NPS (17% reported at least 9 out of 10) were largely concordant within individuals but weakly correlated (0.37). Ratings varied by individual factors such as age and visit purpose. Most domains of patient-reported experience were associated with both outcomes. Adjusted ICC was higher for NPS (0.26 vs 0.11), requiring a minimum of 7 (vs 20) respondents for adequate reliability. Within the 70% of facilities classified as average based on satisfaction, NPS-based classification revealed systematic differences in patient-reported experience measures.


While satisfaction and NPS were broadly similar at an individual level, this evidence suggests NPS may be useful for benchmarking facility performance as part of national efforts in Peru and throughout Latin America to identify deficits in health service quality.

Many UC-authored scholarly publications are freely available on this site because of the UC's open access policies. Let us know how this access is important for you.

Item not freely available? Link broken?
Report a problem accessing this item