Bulletin of the World Health Organization

Diabetes treatment and control: the effect of public health insurance for the poor in Mexico

Sandra G Sosa-Rubí, Omar Galárraga & Ruy López-Ridaura

Volume 87, Number 7, July 2009, 512-519

Table 1. Sociodemographic characteristics of adults with diabetes enrolled in the Seguro Popular and of uninsured adults, Mexico, 2006

Unmatched samples
Matched samples
Percentage reduction in bias
Seguro Populara n = 430 Uninsureda n = 1061 t-test P‑value Seguro Populara n = 425 Uninsureda n = 1029 t-test P‑value
Individual characteristic
Sex
Maleb 30.9 40.1
Female 69.1 59.9 0.001 68.9 69.9 0.115 89.1
Age, in years
20–39 17.5 16.3 0.558 17.2 15.1 0.709 –69.0
40–59 50.0 48.2 0.518 50.12 55.3 0.002 –179
60–79 29.4 31.8 0.378 29.7 26.6 0.109 –30.7
≥ 80b 3.0 3.9
Marital status
Married 73.1 64.8 0.002 72.9 80.2 0.000 12.5
Unmarriedb 26.7 35.3
Education
Noneb 21.4 21.6
Primary 66.8 57.7 0.001 66.8 64.2 0.018 72.5
Secondary 8.7 12.6 0.032 8.7 8.5 0.166 94.0
High school/college 2.8 8.4 0.000 2.8 3.0 0.000 95.8
Ethnicity (self-identified)
Indigenous 23.8 23.3 0.830 24.1 26.8 0.016 –442
Not indigenousb 76.3 76.7
Family assets
Mean (asset index)c –0.504 –0.235 0.000 –0.495 –0.433 0.000 76.9

Regional characteristic
Area of residence
Metropolitan 20.9 42.0 0.000 20.9 22.1 0.000 94.5
Urban 39.9 32.3 0.005 40.0 37.6 0.049 69.4
Ruralb 39.4 25.8
Region of residence
North 16.4 19.7 0.136 16.5 13.4 0.001 8.4
East central 15.7 26.9 0.000 15.8 15.1 0.002 93.7
West central 39.6 26.3 0.000 39.5 39.8 0.000 98.3
South–southeastb 27.9 26.9


a Values are percentages, unless otherwise noted; percentages may not add up to 100% due to rounding error.
b Omitted (reference) category. Table reports nearest neighbour matching with common support. The t test for each covariable may show statistical significance for the variable evaluated for the entire sample; however, for the propensity score, t tests for insignificant differences were conducted at each propensity score block (results not shown). The percentage of bias reduction reports reduction in absolute bias between the treatment category (Seguro Popular enrollees) and the control category (the uninsured) as a percentage of the square root of the average of the sample variances in the treated and non-treated groups.
31
c The asset index is based upon household infrastructure, building materials, and ownership of certain durable assets, as a proxy for household’s wealth. It may be negative because it is constructed with principal components analysis. A more negative index indicates lower household wealth.Estimations based on data from the 2006 Encuesta Nacional de Salud y Nutrición.
14

[an error occurred while processing this directive]