Table 2.
Odds of Participating in Research Derived from Logistic Regression Models
| OR | Patients (95% CI) | P | OR | Staff (95% CI) | P | |
|---|---|---|---|---|---|---|
| Vignette 1. Behavioral Intervention Study | n = 3,169* | n = 1,023* | ||||
| Conducted by state university | 1.4 | 1.1 to 1.8 | .009 | 1.5 | 1.1 to 2.2 | .03 |
| Conducted by federal government | 0.5 | 0.4 to 0.7 | <.001 | 0.5 | 0.3 to 0.8 | .006 |
| Conducted by health care provider | 2.9 | 2.3 to 3.8 | <.001 | 2.0 | 1.4 to 2.9 | <.001 |
| Community involved in developing study | 1.6 | 1.2 to 2.2 | .001 | 1.6 | 1.1 to 2.5 | .02 |
| Community involved in collecting data | 1.5 | 1.1 to 1.9 | .007 | 1.6 | 1.0 to 2.6 | .03 |
| Community involved in interpreting results | 1.5 | 1.1 to 2.0 | .005 | 1.4 | 0.9 to 2.2 | .19 |
| Risk that confidentiality might be broken | 0.2 | 0.1 to 0.2) | <.001 | 0.1 | 0.1 to 0.3 | <.001 |
| Participants will not be paid | 0.7 | 0.5 to 0.9 | .002 | 0.8 | 0.5 to 1.2 | .27 |
| Study is about diabetes | 2.6 | 1.9 to 3.6 | <.001 | 7.2 | 4.0 to 12.7 | <.001 |
| Study is about alcoholism | 2.6 | 1.9 to 3.5 | <.001 | 1.7 | 1.1 to 2.7 | .02 |
| Vignette 2. Genetics Study | n = 3,135* | n = 1,033* | ||||
| Conducted by state university | 1.0 | 0.8 to 1.3 | .97 | 0.9 | 0.7 to 1.3 | .59 |
| Conducted by federal government | 0.4 | 0.3 to 0.5 | <.001 | 0.3 | 0.2 to 0.6 | <.001 |
| Conducted by health care provider | 1.9 | 1.4 to 2.4 | <.001 | 1.3 | 1.0 to 1.8 | .08 |
| Community involved in developing study | 0.8 | 0.6 to 1.0 | .09 | 0.9 | 0.6 to 1.3 | .56 |
| Community involved in collecting data | 0.8 | 0.6 to 1.0 | .06 | 0.7 | 0.5 to 1.0 | .09 |
| Community involved in interpreting results | 0.7 | 0.5 to 0.9 | .009 | 0.8 | 0.5 to 1.1 | .18 |
| Risk that confidentiality might be broken | 0.1 | 0.0 to 0.2 | <.001 | 0.2 | 0.1 to 0.3 | <.001 |
| Participants will not be paid | 0.5 | 0.3 to 0.6 | <.001 | 0.5 | 0.4 to 0.8 | .004 |
| Study is about diabetes | 1.5 | 1.1 to 2.0 | .008 | 2.3 | 1.6 to 3.5 | <.001 |
| Study is about alcoholism | 1.2 | 0.9 to 1.6 | .32 | 1.3 | 0.8 to 1.9 | .28 |
| Vignette 3. Clinical Drug Trial | n = 3,136* | n = 1,034* | ||||
| Conducted by state university | 1.1 | 0.9 to 1.3 | .58 | 1.1 | 0.8 to 1.6 | .48 |
| Conducted by federal government | 0.6 | 0.5 to 0.8 | <.001 | 0.4 | 0.2 to 0.7 | .002 |
| Conducted by health care provider | 1.5 | 1.3 to 1.9 | <.001 | 1.4 | 0.9 to 1.9 | .10 |
| Community involved in developing study | 1.0 | 0.8 to 1.2 | .81 | 0.9 | 0.6 to 1.4 | .74 |
| Community involved in collecting data | 1.1 | 0.9 to 1.3 | .50 | 0.9 | 0.6 to 1.4 | .76 |
| Community involved in interpreting results | 1.1 | 0.9 to 1.4 | .43 | 0.9 | 0.6 to 1.4 | .76 |
| Risk that confidentiality might be broken | 0.3 | 0.2 to 0.4 | <.001 | 0.2 | 0.1 to 0.5 | <.001 |
| Participants will not be paid | 0.8 | 0.6 to 1.0 | .02 | 0.6 | 0.3 to 1.1 | .09 |
| Study is about diabetes | 1.9 | 1.5 to 2.5 | <.001 | 3.4 | 2.0 to 5.6 | <.001 |
| Study is about alcoholism | 1.9 | 1.5 to 2.5 | <.001 | 1.9 | 1.1 to 3.2 | .02 |
*
Because of the special data structure used in the analysis (see section on statistical analyses), n = the number of question responses analyzed in each logistic regression, not the number of respondents
CI, confidence interval; OR, odds ratio