TABLE 3—
Association of Early-Life State Characteristics With Later-Life Health, Stratified by Lifetime Mobility: Alcoa Cohort, United States, 1997–2012
| Characteristic | Full Population, OR (95% CI) | Current State Same as Early Life, OR (95% CI) | Current State Different From Early Life, OR (95% CI) |
| Hypertension | |||
| % non-White | 1.006 (1.001, 1.011) | 0.983 (0.970, 0.996) | 1.006 (0.999, 1.013) |
| % rural | 1.011 (1.005, 1.017) | 1.013 (0.989, 1.037) | 1.010 (1.002, 1.018) |
| % < high school | 0.987 (0.974, 0.999) | 0.960 (0.943, 0.979) | 1.010 (0.991, 1.030) |
| % unemployed | 0.995 (0.991, 0.999) | 0.991 (0.985, 0.997) | 1.001 (0.994, 1.008) |
| Lower median income | 1.002 (0.989, 1.015) | 1.014 (0.996, 1.032) | 0.972 (0.952, 0.993) |
| Gini | 1.013 (1.004, 1.022) | 1.014 (1.002, 1.027) | 0.996 (0.981, 1.011) |
| Diabetes | |||
| % non-White | 0.997 (0.994, 1.000) | 0.997 (0.989, 1.005) | 0.997 (0.993, 1.001) |
| % rural | 1.000 (0.997, 1.004) | 0.991 (0.977, 1.005) | 1.004 (0.999, 1.009) |
| % < high school | 1.003 (0.995, 1.011) | 1.005 (0.994, 1.016) | 1.005 (0.993, 1.017) |
| % unemployed | 1.002 (1.000, 1.004) | 1.004 (1.001, 1.008) | 0.999 (0.994, 1.003) |
| Lower median income | 1.000 (0.993, 1.008) | 1.000 (0.990, 1.011) | 0.996 (0.983, 1.009) |
| Gini | 1.005 (0.999, 1.010) | 1.004 (0.997, 1.011) | 1.003 (0.993, 1.012) |
| Ischemic heart disease | |||
| % non-White | 0.997 (0.995, 0.999) | 0.993 (0.988, 0.998) | 1.000 (0.997, 1.002) |
| % rural | 0.997 (0.995, 0.999) | 0.998 (0.990, 1.007) | 0.999 (0.996, 1.002) |
| % < high school | 1.013 (1.008, 1.017) | 1.009 (1.002, 1.017) | 1.015 (1.008, 1.023) |
| % unemployed | 1.000 (0.999, 1.002) | 1.001 (0.998, 1.003) | 0.999 (0.996, 1.002) |
| Lower median income | 0.993 (0.989, 0.998) | 0.994 (0.987, 1.001) | 0.990 (0.981, 0.998) |
| Gini | 1.002 (0.998, 1.005) | 1.002 (0.998, 1.007) | 0.997 (0.991, 1.003) |
Note. BMI = body mass index; CI = confidence interval; HDL = high-density lipoprotein; LDL = low-density lipoprotein; OR = odds ratio. Each set of 6 rows of ORs per outcome (and 95% CI) are from a single regression model; thus, this table shows results from 9 different regression models (3 models for each of 3 outcomes). All models include random effects for individual and plant location. In addition to the parameter estimates shown here, the models contain the following covariates: age, age-squared, race/ethnicity (Black, Latino, or other), gender, smelter workplace indicator, the 4-category early-life census region, current state characteristics (the same 6 as the early-life characteristics), whether the individual was an hourly worker, and employment grade (same covariates as model 3). The full population model is identical to that shown in model 3, and is repeated here to facilitate comparison of the 2 populations to the full population.