Measuring and characterizing the quality of child care in Brazilian primary health care: a latent class analysis.

This study aimed to characterize the quality of child health care and explore its relationship with municipal characteristics. Using data from the external assessment of the first cycle of the Brazilian National Program for Access and Quality Improvement in Primary Care (PMAQ-AB, acronym in Portuguese), this cross-sectional study evaluated 16,566 Family Health Strategy teams. In total, nine binary indicators were created based on recommendations from the Brazilian Ministry of Health. We employed latent class analysis and multinomial logistic regression to assess the quality of care and its association with region and the Brazilian Deprivation Index. Three patterns of quality of care were identified: high, intermediate, and low adequacy. The "high adequacy" pattern included 22.5% of teams, "intermediate adequacy" 60.2%, and "low adequacy" only 17.3%. Teams in the Northeast Region were over twice as likely to belong to the "high adequacy" pattern (OR = 2.34; 95%CI: 1.15-4.76) compared with those in the Central-West Region. For teams located in municipalities with moderate and low deprivation, the chance of belonging to the "high adequacy" pattern was 2.04 (95%CI: 1.44-2.89) and 9.08 (95%CI: 4.54-18.14) times higher, respectively, compared with the high deprivation municipalities. This study identified three patterns of quality of child care, with most teams characterized by an "intermediate adequacy" pattern. The quality of care was associated with the municipality's characteristics. The methodology used in this study proved effective in characterizing the quality of care in a more consistent way.