Table 4.

Multivariate Correlates to Clinical Depressive Symptoms by Two Study Samples

Significant Correlates*South TexasSignificant Correlates*Northern Tamaulipas
OR (95% CI)POR (95% CI)P
Demographic factorsHealth factors
    Gender    Leg discomfort interferes with home activities
        Female3.51 (1.19–10.40).023        Yes3.82 (1.60–9.11).003
        Male1.00        No1.00
    EducationDiabetes symptoms1.66 (1.16–1.59).000
        <High school4.16 (1.55–11.16).005
        ≥High school1.00
Health factors
    High Cholesterol
        Yes0.37 (0.15–0.92).032
    Diabetes symptoms1.27 (1.08–1.50).004
    Leg discomfort interferes with wearing preferred shoes
        Yes3.21 (1.11–9.24).031
    “Testing blood glucose is difficult”
        Agree3.14 (1.14–8.66).027
Healthcare utilization factors
    Number of ER visits
        ≥13.37 (1.02–11.12).046
  • * Only variables that were significant individually were included into the logistic regression model.; P < .05.

  • CI, confidence interval; OR, odds ratio.

    Note: the Nagelkerke R2 was 0.527 for the model of the US sample and 0.514 for the model of the Mexico sample. These values indicate that a substantial portion (over 50%) of variation of the depression variable is explained away by the fitted models for both populations. The Hosmer-Lemeshow goodness-of-fit tests for both models were not significant (P = 0.567 for the US model; P = 0.975 for the Mexico model), which suggest adequate model fitting to the data.