Table 3.
Logistic Regression Analysis
| Variable | Coefficient | SE | Effect Size | p-value |
|---|---|---|---|---|
| Proportion encouraged | -1.27 | .39 | -.23 | <.01 |
| Proportion shared status | -1.15 | .53 | -.29 | .04 |
| Proportion family members involved in communication & encouragement | -1.44 | .44 | -.25 | <.01 |
Separate logistic regression models were built for each outcome variable of interest. All analyses control for age (>50), risk status (not at risk), age x risk status, family mutation status, and number of dyads per family match.
Effect sizes were calculated by standardizing the regression coefficients 39, p. 50.
Outcome variable: last colonoscopy (1 = within the past year, 2 = in the past 1-2 years, 3 = in the past 2-3 years, 4 = more than 3 years ago, 5 = never)