Table 3.

Adjusted ORs of having a recent Pap test (≤2 vs >2 y or never)

VariableModel 1Model 2
Model 3
Model 1 + culture
Model 1 + insurance
Model 1 + having a provider
Model 1 + communication
OR (95% CI)
Demographic
    Age (y)
        40-49 (vs 18-39)0.78 (0.54-1.12)0.78 (0.54-1.12)0.78 (0.54-1.12)0.75 (0.52-1.08)0.77 (0.53-1.11)0.75 (0.52-1.09)
        ≥50 (vs 18-39)0.47 (0.35-0.62)0.48 (0.36-0.63)0.43 (.32-.57)0.44 (0.33-0.58)0.45 (0.34-0.60)0.41 (0.30-0.54)*
    Education: ≥college (vs <college)1.61 (1.23-2.06)1.56 (1.22-2.00)1.59 (1.24-2.03)1.60 (1.25-2.05)1.62 (1.27-2.08)1.57 (1.22-2.01)*
    Marital status: married (vs nonmarried)2.30 (1.80-2.92)2.30 (1.81-2.92)2.24 (1.76-2.85)2.26 (1.77-2.88)2.31 (1.82-2.95)2.24 (1.76-2.86)*
    Ethnicity: NHW (vs Asians)2.38 (1.67-3.41)2.19 (1.53-3.16)2.38 (1.66-3.41)2.30 (1.60-3.30)2.29 (1.60-3.28)2.09 (1.45-3.02)*
    Health status: excellent/very good (vs good/fair/poor)1.90 (1.50-2.42)1.90 (1.49-2.42)1.82 (1.43-2.32)1.92 (1.51-2.44)1.81 (1.42-2.31)1.77 (1.39-2.27)*
    Eastern cultural views1.10 (1.02-1.18)1.08 (1.00-1.16)
Access factors
    Having health insurance (vs no)2.03 (1.38-2.99)1.70 (1.14-2.54)§
    Having a regular provider (vs no)2.01 (1.48-2.74)1.74 (1.27-2.40)§
    Good communication with providers (vs poor)1.61 (1.21-2.16)1.47 (1.09-1.97)
Provider characteristics
        Female (vs male)1.83 (1.36-2.48)*
        Ethnic concordance (vs discordance)0.95 (0.49-1.83)
  • NOTE: Dashed lines indicate that the variable was not included in the tested model.

    Abbreviation: NHW, non-Hispanic White.

  • * P < 0.001.

  • A sum score on the luck belief and self-care items; high scores mean less Eastern cultural views.

  • P < 0.05.

  • § P < 0.01.

  • The results were based on a post hoc multivariate analysis after we found that having a regular provider predicted the outcome. In the post hoc, gender and ethnic concordance were added to the third model after removing the variable of having a regular provider to avoid the multicolinearity. Women without a regular provider and responses to the questions were coded unknown. The comparisons between unknown and other categories were not shown here. The ORs of the association between Eastern cultural views and the outcome were not changed in the post hoc analysis.