Table 2.

Regression Coefficients (Unstandardized) and P values for Association between Health Literacy (Coded as a Categorical Variable) and Baseline-Corrected Outcomes (ie, Outcome Value at Follow-up minus Outcome Value at Baseline) in an Unadjusted Model, and a Model Adjusting for Age, Education, and Social Risk

Effect of health literacy (categorical) on baseline-corrected outcomesUnadjusted modelAdjusted model*
Coeff. With Respect to low HL95% Confidence IntervalP valueCoeff.95% Confidence IntervalP value
Clinical
 HbA1cHL Med0.11−0.54 to 0.770.730.09−0.60 to 0.790.79
HL High−0.18−0.71 to 0.340.50−0.25−0.86 to 0.350.40
 BMIHL Med0.01−1.99 to 2.010.990.00−2.11 to 2.120.99
HL High0.15−1.41 to 1.710.850.14−1.64 to 1.920.87
Behavioral
 Medication adherenceHL Med0.25−1.02 to 1.520.700.05−1.29 to 1.400.94
HL High−0.17−1.15 to 0.820.74−0.23−1.37 to 0.900.69
 ExerciseHL Med−0.68−1.62 to 0.270.16−0.74−1.74 to 0.270.15
HL High−0.14−0.87 to 0.590.70−0.21−1.05 to 0.620.62
 Vegetable intakeHL Med−0.09−0.88 to 0.700.830.09−0.75 to −0.930.83
HL High−0.08−0.70 to 0.540.800.27−0.43 to 0.970.45
Fat intakeHL Med0.46−0.19 to 1.110.160.47−0.22 to 1.160.20
HL High0.32−0.19 to 0.830.220.25−0.33 to 0.830.40
 Sweetened beverage intakeHL Med0.26−0.18 to 0.700.250.13−0.33 to 0.600.58
HL High0.17−0.17 to 0.510.33−0.00−0.39 to 0.390.99
  • Note. Performing this analysis with health literacy coded as an ordinal variable produced similar non-significant results in every case.

  • * In the adjusted model, based on a causal DAG (Figure 1), we included the following variables as predictors in the regression model: age, social risks, and education.

  • Please note that the P values shown here are not adjusted for multiple comparisons because none have passed an unadjusted significance threshold.

  • Abbreviations: DAG, directed acyclic graph; HbA1c, hemoglobin A1c; BMI, body mass index; HL, health literacy.