Table 3.
Perceived Risk and Intentions to Have Mammogram, by Condition
| Conditions | |||||||||
|---|---|---|---|---|---|---|---|---|---|
| Comparison: no information |
Comparison: basic info |
Brief intervention + expository |
Brief intervention + untailored exemplars |
Brief intervention + tailored exemplars |
Extended intervention + expository |
Extended intervention + untailored exemplars |
Extended intervention + tailored exemplars |
||
| Participants with an estimated 10-Year Breast Cancer Risk <1.5%a | |||||||||
| F (df) | |||||||||
| n | 194 | 150 | 141 | 169 | 121 | 160 | 154 | 138 | |
| Accuracy of risk (amount of overestimation): Risk measured as Mean (SD) | 17.4 (19.1) | 20.2 (19.4) | 16.7 (20.7) | 12.7 (17.0) | 13.8 (17.5) | 14.7 (19.2) | 13.8 (17.3) | 11.7 (16.1) | 3.39 (7, 1215)** |
| Beta | 0.048 | −0.012 | −0.087*‡‡‡ | −0.057†‡‡ | −0.048‡ | −0.063†‡‡ | −0.096**‡‡‡ | ||
| Accuracy of risk (amount of overestimation): Risk measured as frequency, Mean (SD) | 10.0 (16.7) | 10.5 (18.2) | 11.8 (26.2) | 7.5 (21.1) | 15.3 (32.2) | 9.8 (24.5) | 2.9 (14.3) | 8.8 (24.2) | 3.50 (7, 1212)*** |
| Beta | 0.007 | 0.025 | −0.038 | 0.071*b | −0.003 | −0.105**‡‡ | −0.017 | ||
| X2 (df) | |||||||||
| % intend to wait until age 50 to get mammogram, Mean (SD) | 7.2 (26.0) | 6.7 (25.0) | 18.4 (38.9) | 16.7 (37.4) | 14.0 (34.9) | 19.4 (39.6) | 24.2 (43.0) | 21.0 (40.9) | 31.02 (7)*** |
| OR | 0.91 | 2.89**‡‡ | 2.56**‡‡ | 2.09†‡ | 3.07**‡‡ | 4.08***‡‡‡ | 340***‡‡ | ||
| Participants with an estimated 10-Year Breast Cancer Risk ≥1.5%c | |||||||||
| F (df) | |||||||||
| N | 259 | 196 | 210 | 195 | 200 | 212 | 205 | 214 | |
| Accuracy of risk (amount of overestimation): Risk measured as %, Mean (SD) | 20.2 (21.2) | 23.2 (22.7) | 15.8 (18.4) | 16.0 (19.1) | 14.0 (17.0) | 14.3 (18.4) | 13.9 (18.1) | 13.9 (17.7) | 6.85 (7, 1671)*** |
| Beta | 0.051† | −0.074*‡‡‡ | 0.068*‡‡‡ | −0.102**‡‡‡ | −0.101**‡‡‡ | −0.105***‡‡‡ | −0.108***‡‡‡ | ||
| Accuracy of risk (amount of overestimation): Risk measured as frequency, Mean (SD) | 9.0 (18.7) | 10.6 (19.6) | 10.9 (24.7) | 11.6 (25.3) | 10.9 (25.2) | 9.6 (26.0) | 7.7 (23.7) | 7.9 (23.3) | 0.79 (7, 1666) |
| Beta | 0.022 | 0.026 | 0.035 | 0.026 | 0.008 | −0.018 | −0.015 | ||
| X2 (df) | |||||||||
| % intend to get mammogram at age 40, Mean (SD)d | 72.6 (44.7) | 73.0 (44.5) | 76.7 (42.4) | 77.9 (41.6) | 70.5 (45.7) | 75.9 (42.8) | 76.1 (42.8) | 68.7 (46.5) | 12.92 (7)† |
| OR f | 0.89 | 1.13 | 1.64*‡ | 1.07 | 1.32 | 1.12 | 0.71 | ||
Note. N = 2918. All risk-based intervention conditions are compared to the “Comparison: no information” condition and the “Comparison: basic info” condition using OLS regression to regress accuracy of perceived risk on condition and logistic regression to regress a dichotomous measure of mammography intention on condition. “Accuracy of risk (amount of overestimation): Risk as frequency” refers to the degree to which participants overestimated their objective risk for breast cancer when asked to report perceived risk as a frequency out of 1000. This number was then converted to a 0 to 100 scale so that it could be compared to “Accuracy of risk (amount of overestimation): Risk as %,” which is overestimation when risk was asked as a percentage. The Beta reported above is the standardized regression coefficient associated with each condition compared to “Comparison: no information.” The OR is the odds ratio (the ratio of the odds that the intention is present for participants in a given condition to the odds that the intention is present for those in the comparison condition). To guard against Type I error, individual conditions are only noted as being significantly different from the comparison if the overall test (the F test for OLS regression or the X2 test for logistic regression) for that variable is significant (p < .10). SD = Standard Deviation.
These participants are classified as being at a low risk for breast cancer because they have an estimated 10-year risk for developing breast cancer that is <1.5%, according to the NCI BCRAT [24].
Though significant, this finding is in the unexpected direction, as it represents reduced accuracy (i.e., higher overestimation).
These participants have an estimated 10-year risk for developing breast cancer that is ≥1.5%, according to the NCI Breast Cancer Risk Assessment Tool [24].
Analyses of mammography intentions for women with an estimated 10-year risk ≥1.5% include prior history of mammogram and insurance status as comparison variables.
p < .10
p < .05
p < .01
p < .001 when compared to “Comparison: no information” condition or overall tests (e.g., X2)
†† p < .10
p < .05
p < .01
p < .001 when compared to “Comparison: basic info” condition