Table 4. Results from longitudinal fixed-effects Poisson regressions for mortality by groups of ambulatory-care-sensitive conditions in black/pardo and white populations.
| Group | Variable | Black/pardo group | White group | ||
|---|---|---|---|---|---|
| RR or N | 95% CI | RR or N | 95% CI | ||
| Infectious diseases | ESF coverage | 0.725*** | 0.620, 0.848 | 0.956 | 0.854, 1.070 |
| Year | 0.989 | 0.953, 1.026 | 0.972* | 0.948, 0.997 | |
| N observations | 18,046 | 21,476 | |||
| Total deaths | 35,353 | 31,716 | |||
| Nutritional deficiencies and anaemia | ESF coverage | 0.721** | 0.578, 0.899 | 1.251* | 1.011, 1.548 |
| Year | 0.982 | 0.937, 1.031 | 0.951* | 0.909, 0.994 | |
| N observations | 11,662 | 15,932 | |||
| Total deaths | 5,988 | 5,313 | |||
| COPD and asthma | ESF coverage | 1.072 | 0.939, 1.223 | 0.988 | 0.914, 1.068 |
| Year | 0.959** | 0.933, 0.985 | 0.955*** | 0.939, 0.971 | |
| N observations | 19,880 | 22,120 | |||
| Total deaths | 27,174 | 48,055 | |||
| Cardiovascular | ESF coverage | 0.871** | 0.801, 0.947 | 0.929* | 0.876, 0.985 |
| Year | 0.963*** | 0.948, 0.979 | 0.973*** | 0.962, 0.984 | |
| N observations | 21,853 | 22,652 | |||
| Total deaths | 137,061 | 147,682 | |||
| Diabetes | ESF coverage | 0.807*** | 0.713, 0.912 | 0.932 | 0.849, 1.023 |
| Year | 0.973* | 0.952, 0.994 | 0.987 | 0.971, 1.004 | |
| N observations | 20,244 | 22,526 | |||
| Total deaths | 54,873 | 65,003 | |||
| Epilepsy | ESF coverage | 0.961 | 0.745, 1.240 | 1.017 | 0.806, 1.284 |
| Year | 1.000 | 0.949, 1.054 | 0.962 | 0.921, 1.005 | |
| N observations | 11,578 | 15,848 | |||
| Total deaths | 4,045 | 4,908 | |||
| Gastric ulcers | ESF coverage | 0.962 | 0.915, 1.012 | 0.951** | 0.922, 0.981 |
| Year | 0.884 | 0.697, 1.122 | 0.939 | 0.800, 1.103 | |
| N observations | 13,230 | 18,788 | |||
| Total deaths | 8,542 | 10,798 | |||
Exponentiated coefficients:
* p < 0.05,
** p < 0.01,
*** p < 0.001.
The table shows select results from longitudinal Poisson regressions for groups of ambulatory-care-sensitive conditions for both the black/pardo population and the white population, in addition to the number of deaths for each group of conditions and racial group. The study period was from 2000 to 2013. Robust standard errors were employed. ESF coverage is a 2-y average of within-year municipal ESF coverage and coverage in the year before, and is expressed as percentages and scaled so a one-unit increase represents a 100% increase. Year is a continuous variable and is interpreted as the change in mortality rate for each additional year. Although not reported, all regressions control for Bolsa Família coverage (percent), illiteracy rate of those over 25 y (log-transformed), poverty rate (percent), urbanisation rate (percent), public healthcare spending (R$100s per person), public hospital beds per 1,000 population, private hospital beds per 1,000 population, private healthcare insurance (percent) (log-transformed), GDP per person (R$100s per person) (log-transformed), and the interaction of private healthcare insurance (percent) (log-transformed) and GDP per person (R$100s per person) (log-transformed). Some municipalities and/or year observations are not included for a racial group due to there being no deaths from ambulatory-care-sensitive conditions for that racial group.
95% CI, 95% confidence interval; COPD, chronic obstructive pulmonary diesease; ESF, Estratégia de Saúde da Família; GDP, gross domestic product; R$100s, hundreds of Brazilian reais per person; RR, rate ratio.