Brazil’s more doctors programme and infant health outcomes: a longitudinal analysis

Table 3 Effect of PMM density on primary and intermediate outcomes for subgroups of municipal baseline IMR

	IMR	95% CI	NMR	95% CI	Infants hospitalised	95% CI	≤ 7 prenatal care visits	95% CI	Low birth weight	95% CI
Quintiles of baseline IMR (2007)
Q1 (0 deaths)	− 0.01	− 0.14,0.13	− 0.01	− 0.12,0.09	0.38	− 0.57,1.32	0.09*	0.01,0.17	0.02	− 0.01,0.06
Q2 (1.89–12.2)	− 0.37	− 0.75,0.01	− 0.20	− 0.53,0.13	0.10	− 3.08,3.27	0.60*	0.13,1.08	0.07	− 0.01,0.16
Q3 (12.21–17.54)	0.07	− 0.27,0.42	0.12	− 0.15,0.39	1.36	− 0.51,3.24	0.34	− 0.11,0.79	0.08	− 0.02,0.17
Q4 (17.58–25.23)	0.10	− 0.21,0.41	0.12	− 0.14,0.38	− 0.10	− 2.53,2.32	0.05	− 0.21,0.31	− 0.01	− 0.08,0.05
Q5 (25.24–209.3)	− 0.21*	− 0.38,− 0.03	− 0.14	− 0.30,0.02	− 0.64	− 2.36,1.07	− 0.03	− 0.16,0.11	0.03	− 0.02,0.07

* p < 0.05, ** p < 0.01, *** p < 0.001. All models applied inverse probability of treatment weighting with regression adjustment (IPTW-RA). Primary outcome variables: infant mortality rate (IMR) and neonatal mortality rate (NMR). Intermediate outcome variables: infants hospitalised per 1000 live births; seven or more prenatal care visits; and proportion of infants born with a low birthweight (%). Models adjusted for GDP per capita (BRL), income per capita (BRL), Gini coefficient, proportion of households with inadequate sanitation (%), proportion of households with no electricity (%), proportion of population living in urban areas (%), proportion of population illiterate above the age of 15 (%), proportion of the population with per capita income under 0.25 minimum wage (%), Bolsa Familia stipend (BRL), private health insurance plans per capita, heath expenditure per capita (BRL), hospital beds per 1000 population, nurses per 1000 population, mean municipal mother’s age, proportion of mothers with zero to three years of education (%), proportion of mothers with four to seven years of education (%), proportion of mothers with eight to eleven years of education (%), proportion of mothers with more than twelve years of education (%), and municipality and time fixed effects

ISSN: 1478-4491