Table 3—

Bivariate and multivariate associations among patient factors and provider factors and the performance of Pap smears

Bivariate analysisMultivariate analysis
Patient factors
    Age (ref. = 21–39 years)*
        40–490.65 (0.46–0.92)0.97 (0.94–0.99)
        50–700.42 (0.21–0.85)0.94 (0.88–0.98)
    Nonwhite race (ref. = white)1.14 (0.69–1.89)NI
    Education (ref. = <high school)*
        High school/GED1.29 (1.03–1.62)1.01 (0.70–1.46)
        Some college1.66 (1.06–2.62)1.02 (0.49–2.13)
        ≥4 years of college2.15 (1.09–4.25)1.03 (0.34–3.11)
    Income (ref. = <$15,000)*
        $15,000–39,0001.50 (1.19–1.88)1.19 (0.82–1.73)
        $40,000–75,0002.25 (1.42–3.53)1.42 (0.67–2.99)
        >$75,0003.38 (1.68–6.64)1.68 (0.55–5.18)
    Self-reported health status (ref. = excellent)*
        Very good0.74 (0.58–0.93)1.00 (0.71–1.40)
        Good0.55 (0.34–0.86)0.98 (0.50–1.96)
        Fair0.41 (0.20–0.80)0.97 (0.36–2.74)
        Poor0.30 (0.11–0.75)0.96 (0.25–3.84)
    Diabetes duration (ref. = <10 years)0.78 (0.50–1.22)NI
    Diabetes treatment (ref. = diet)NI
        Oral medication0.87 (0.66–1.16)
        Insulin0.76 (0.44–1.34)
        Insulin + oral medication0.66 (0.29–1.56)
    Charlson comorbidity score (ref. = 0–2)*
        3–40.64 (0.43–0.95)0.87 (0.70–1.08)
        >40.41 (0.18–0.90)0.76 (0.49–1.17)
    Diabetes process/quality of care (continuous)*1.23 (1.02–1.49)1.40 (1.10–1.78)
    Insurance type (ref. = Medicaid)1.95 (0.68–5.62)NI
    Any visit to a gynecologist (ref. = 0)*3.86 (1.54–9.68)3.27 (1.05–10.19)
    Total outpatient visits (ref. = 1–10)NI
        11–201.09 (0.81–1.46)
        >201.19 (0.66–2.13)
Provider factors
    Age (continuous)*0.99 (0.97–1.00)1.00 (0.97–1.03)
    Sex (ref. = male)*1.48 (1.09–2.42)1.04 (0.53–2.01)
    Specialty (ref. = family/general practice)0.99 (0.63–1.58)NI
    Years of practice (continuous)0.99 (0.97–1.02)NI
  • Data are OR (95% CI).

  • *

    * Statistically significant bivariate association;

  • statistically significant multivariate association. NI, not included in multivariate analysis because variable was not significant in bivariate analysis.