Have countries offered the best data to the Pan American Health Organization? Evidence of potential inconsistencies found in a study on alcohol policies in Brazil.

OBJECTIVE: The objective of this study was to verify the agreement between the alcohol policies score estimated from documental analysis of Brazilian federal regulatory documents (RD), with primary data collection, and the results previously presented by the Pan American Health Organization (PAHO) in its Alcohol Policy Scoring (APS) report. STUDY DESIGN: Document identification and content analysis. METHODS: Documental research was carried out in two phases: a document identification and content analysis. In the first phase, we carried out the search, identification, and systematization of laws, decrees, and federal ordinances in Brazil, with primary data collection. The second phase consisted of three steps: 1) an RD content analysis and classification into the 10 PAHO/World Health Organization (WHO) policy domains; 2) a score estimation of alcohol policies, based on the APS instrument attached to their report; and 3) comparison of the results for Brazil presented at the APS report and the one estimated by the researchers. RESULTS: The study showed divergences between the results for APS published by PAHO about Brazil and the one achieved with primary data collection. 1146 federal promulgated RD were identified, of which 21 were eligible for content analysis. Only the domains "Community and workplace action" (Domain 3) and "Reducing the public health impact of illicit and informally produced alcohol" (Domain 9) had convergent scores. On the other domains, usually the APS score estimated by PAHO differs from the one estimated with the primary data collection. CONCLUSIONS: We conclude that Brazil is not providing the best data for PAHO/WHO estimate its APS report, leading to the dissemination of imprecise results worldwide.