Table 5.

Multivariate logistic regression models of the relationship between perceived ambiguity, mammography-related worry, and perceived breast cancer risk (Maximizing Mammography Participation Trial)

VariablesPrediction 3 model: mammography-related worry (n = 2,868)
Prediction 4 model: perceived breast cancer risk (n = 2,780)
OR (95% CI)P*OR (95% CI)P*
Age0.010.001
    50-591.001.00
    60-690.88 (0.71-1.09)1.00 (0.84-1.19)
    70-800.70 (0.55-0.89)0.72 (0.59-0.88)
Race0.0040.01
    White1.001.00
    Black1.68 (1.01-2.82)0.73 (0.47-1.12)
    Other1.75 (1.17-2.61)0.63 (0.45-0.88)
Education level0.300.85
    Less than high school1.001.00
    Some college0.75 (0.50-1.11)0.91 (0.65-1.27)
    College graduate0.81 (0.53-1.23)0.92 (0.65-1.31)
Income0.480.61
    Under US$30,0001.001.00
    Over US$30,0000.93 (0.76-1.14)1.04 (0.89-1.23)
Prior mammography history0.580.001
    No1.001.00
    Yes0.921.50 (1.18-1.92)
Study intervention<0.00050.10
    None1.001.00
    Reminder postcard1.66 (1.29-2.16)0.83 (0.67-1.03)
    Reminder telephone call1.20 (0.97-1.48)0.86 (0.72-1.02)
Perceived ambiguity<0.00010.87
    Lowest1.001.00
    Low1.47 (1.14-1.90)1.08 (0.88-1.33)
    High2.60 (1.79-3.78)1.03 (0.77-1.36)
    Highest1.53 (0.70-3.34)0.92 (0.50-1.67)
Baseline mammography worry<0.0001
    Low1.00
    High8.02 (6.72-9.58)
Baseline perceived risk<0.0001
    Much lower1.00
    Somewhat lower3.53 (2.90-4.29)
    About the same16.08 (12.89-20.06)
    Somewhat/much higher124.94 (92.67-168.44)
  • NOTE: Total sample N = 3,701 (decreased and unequal n for individual models due to missing data).

    Abbreviations: OR, odds ratio; 95% CI, 95% confidence interval.

  • * P for Wald χ2 test of association.

  • “Somewhat higher” and “Much higher” categories collapsed to accommodate small cell sizes in the model.