Table 3.
Regression coefficients for univariate analysis of state-level leprosy outcome variables
| Outcome | ANCDR | Grade 2 Rate | Grade 2 Fraction |
|---|---|---|---|
| Poverty Index | -3.53 (-7.93, 1.12) | — | 0.06 (-0.002, 0.12) |
| Illiteracy | 4.46 (1.25, 10.3) | 4.75 (0.74, 12.8) | -0.08 (-0.15, -0.03) |
| Unemployment | — | — | -0.06 (-0.14, -0.008) |
| Rural | 0.9 (-0.74, 2.14) | — | — |
| Scaled radiance | 0.01 (0.002, 0.02) | 0.03 (0.02, 0.05) | 2×10−4 (− 2×10−5, 2×10−4) |
| Time | -0.04 (-0.07, -0.02) | 0.2 (0.09, 0.26) | 0.004 (0.002, 0.005) |
| Marginal R2 | 0.04 | 0.01 | 0.02 |
| Conditional R2 | 0.68 | 0.60 | 0.21 |
Models were selected using Akaike Information criterion (AIC) from all subsets of the regressors: poverty index, illiteracy, unemployment, rural population fraction, and scaled radiance (see text for details). All models include calendar time in years, the enhanced case finding covariate, a random slope, and a random intercept. Marginal R2 values indicate the fraction of variance explained by the fixed effects, and conditional R2 indicate the fraction of varianceexplained by both fixed and random effects as described in the text. Confidence intervals derived by spatial block bootstrap (with a radius of 1.5 degrees); see text for details