Table 2—

Unadjusted and adjusted predicted percents for cost-related medication underuse*

Reported cost-related medication underuse
Unadjusted percent (%)Adjusted predicted percent (95% CI)P value of variable in adjusted model
All participants (n = 5,086)1414
Race/ethnicity0.048
    White (reference) (n = 2,452)(13)(10) (8–12)
    African American (n = 707)1711 (8–13)
    Latino (n = 736)2314 (10–18)
    Asian/Pacific Islander (n = 782)117 (5–10)
    Other (n = 410)1511 (6–15)
Age (years)<0.001
    18–442423 (17–29)
    45–641715 (13–18)
    >65 (reference)(10)(6) (5–7)
Sex<0.001
    Female1812 (10–14)
    Male (reference)(11)8 (7–10)
Education0.28
    Less than high school159 (7–11)
    High school graduate1711 (9–13)
    Some college or higher (reference)(13)(10) (8–12)
Annual household income<0.001
    <$25,0002015 (12–18)
    $25,000 to $49,9991612 (9–14)
    >$50,000 (reference)(7)(5) (4–7)
Health status<0.001
    Excellent/very good (reference)(8)(7) (5–9)
    Good129 (7–11)
    Fair/poor2014 (11–16)
Has prescription drug benefit<0.001
    Yes (reference)(13)(9) (8–11)
    No1913 (10–16)
Number of prescription medications<0.001
    One to five medications (reference)(14)(9) (7–11)
    Six or more medications1511 (9–13)
Aveerage monthly out-of-pocket drug costs<0.001
    ≤$50 per month (reference)(9)(7) (5–8)
    $51 to $100 per month1411 (9–14)
    $101 to $150 per month2016 (12–19)
    > $150 per month2924 (20–29)
  • *

    * The model included race/ethnicity, income, whether a patient had prescription drug benefits, and average monthly out-of-pocket drug costs as main predictors. The remaining variables were included as covariates (age, sex, education, self-reported health status, and the number of prescription medications). Duration of diabetes and four two-way interaction terms (race/ethnicity by age, sex, income, and out-of-pocket drug costs) were assessed but were not significant and were not included in the final model for reasons of parsimony.

  • Statistically significant at P < 0.05.