Skip to main content
. 2020 Jun;22(1):221–235.

Table 3.

Factors associated with higher likelihood of lawsuits among Rio Grande do Sul municipalities: Results from fully adjusted multi-level random effects logistic regression model

Variable Coefficient 95% confidence interval p-value
Demographics
 Populationa 1.04 (1.01–1.07) 0.02
 Mortality rateb 1.17 (1.01–1.34) 0.031
Socioeconomic
 Urbanc 1.01 (1.00–1.03) 0.022
 Illiteracy rated 0.97 (0.89–1.06) NS
 Per capita incomee 0.99 (0.86–1.14) NS
Service availability
 Health servicesf 0.56 (0.27–1.17) NS
Public Defender’s Officeg 6.95 (3.2–15.1) <0.0001
Constant 0.09 (0.01–0.56) NS
N 496

Notes: Multivariable multi-level random effects logistic regression model. The independent variable is the presence or absence of medicine-requesting lawsuits in a municipality. All the listed characteristics are included as fixed-effects regressors. The model includes a random intercept for health administrative region (HAR). Coefficients represent expected change in the odds of lawsuits associated with a 1-unit change in the variable, keeping all other characteristics constant. NS: not statistically significant (p-value >0.05). N=496 municipalities.

a Total population, in thousands

b Crude mortality rate: total deaths per 1,000 inhabitants

c Municipality has predominance of urban areas

d Percentage of the adult population that cannot read or write

e Average per capita income in 2010 Brazilian reais

f Health services per 100,000 inhabitants

g Municipality has at least one Public Defender’s Office