INTERNAÇÕES PEDIÁTRICAS POR CONDIÇÕES SENSÍVEIS À ATENÇÃO PRIMÁRIA NO BRASIL E INDICADORES DO PMAQ-AB 2012-2018: ESTUDO ECOLÓGICO MULTINÍVEL.

Abstract

Introduction: Hospitalizations for conditions sensitive to primary care (ACSC) are preventable hospitalizations with effective actions of primary health care (PHC). It is an indicator of the quality of primary care. Objectives: To analyze, through multilevel modeling, the structure of the Basic Health Units (BHU) and the work process of PHC teams (eAPS) related to child care and other factors associated with the occurrence of ACSC. Methods: Longitudinal ecological study, with secondary data from the three cycles of the PMAQ-AB; the Hospitalization System; the 2010 demographic census of the Brazilian Institute of Geography and Statistics; and the stratification categories of the municipalities. 68,320 BHU participated in the study. SPSS software was used to perform multilevel modeling on two levels, using the mixed command with fixed slope and random intercept. The first level includes the outcome (ACSC rates for children under five years old) and the explanatory variables: structural and procedural indicators, aggregated by BHU. In the second level are the social determinants of the municipalities. Average percentages of each indicator were obtained and the t test with Bonferroni correction was used to compare the averages of the indicators between regions. Correlation coefficients (β) and confidence intervals (95% CI) were estimated by multilevel linear regression. Results: Seven structure indicators and thirteen work process indicators were built. The structural indicators with the highest percentages (2017/2018) were: operation of the unit (99%), equipment / materials (82%) and vaccines (74%). The lowest percentages among the procedural indicators were: covered population (32%) and scheduling for specialists (52%). The ACSC rate for children under five years old was 62.78 / 100 thousand inhabitants per potential area of BHU. They were directly associated with the outcome: PMAQ-AB cycle (β = 2.21); team planning (β = 3.33); special schedule (β = 2.94); unit dependencies (β = 5.23); and vaccines (β = 4.57). They were inversely associated: equipment / materials (β = 8.08); and the dummies of social determinants (β = 28.33 to 40.36 in small municipalities; and β = 46.57 to 63.97 in mediumsized municipalities). Conclusion: The work process indicators showed better average percentages, but only the equipment / materials structural indicator and social determinants were associated with the reduction of ACSC.