The effect of primary care policy changes on hospitalisation for ambulatory care sensitive conditions: notes from Brazil

Abstract

Objectives: The aim of this study was to assess the effect of the Primary Health Care Reform and the impact of recent national policy changes on the trend in hospitalisation rates for Ambulatory Conditions Sensitive to Primary Care (ACSC) in the city of Rio de Janeiro, Brazil.Study design: Ecological study with data from the Brazilian National Health Information System.Methods: This study performed an interrupted time-series analysis, comparing three different time periods related to important changes in primary care policies in Brazil: Baseline (2008–2009); Primary Health Care Reform (RCAPS) (2010–2017); and National Primary Health Care Policy (PNAB) (2018–2019). Data included total monthly admissions for ACSC and rates for 19 groups of causes for hospitalisation.Results: There was a non-significant increasing trend in ACSC hospital admissions between January 2008 and December 2009 (β = 4.01, 95% confidence interval [CI] −12.14 to 20.15). The Primary Health Care Reform (RCAPS), which commenced in January 2010, resulted in a significant reduction in ACSC hospital admission trends during this time period (β = −7.97, 95% CI –9.78 to −6.16). Change in the PNAB did not change the declining ACSC hospital admission trend but it did reduce the rate of decline. In addition, the coefficient was no longer significant (β = −1.93, 95% CI –2.46 to 0.59), suggesting that monitoring of the historical series for a long time may show a reversal of the downward trend in some months. The most consistent change in ACSC hospital admission trend occurred between baseline and the RCAPS (β slope = −0.143, P < 0.001) and the level between the RCAPS period and the new PNAB (β step = 7.00, P = 0.008). Trends and differences in the time periods vary in the different ACSC hospitalisation-cause groups.Conclusion: Changes in public health policies impact ACSC hospitalisation rates, which can thus be used as an indicator for monitoring primary health care. We also recommend analysing admission rates by the individual ACSC hospitalisation-cause groups.