Response rates in Birmingham and Jefferson County, Tuskegee and Macon County, Hartford and Hartford County, and San Antonio and Bexar County were 70%, 65%, 49%, and 50%, respectively. The overall completion rate (number of completed interviews per number of initiated interviews) exceeded 90% in each city. For San Antonio, the major Spanish-speaking Hispanic population accessed in this survey, 10% of the contacted individuals indicated that they could not participate with the English-language-only instrument. shows the age, gender, education, and income distribution of the 1133 respondents within the 3 racial/ethnic groups.
| TABLE 1Age, Gender, Education, and Income Distribution of Respondents (N = 1113), by Racial/Ethnic Group: Tuskegee Legacy Project Study, 1999–2000 |
To determine if a respondent was aware of the Tuskegee Syphilis Study, the TLP Questionnaire had 2 separate recognition probes. The first recognition probe consisted of the respondents being asked directly whether they had ever heard about the Tuskegee Syphilis Study. The 3 racial/ethnic groups differed markedly on responses to this first recognition probe for the Tuskegee Syphilis Study with 72.6% of Blacks, 55.2% of Whites, and 23.6% of Hispanics answering yes (P < .001, by the χ2 test). Each contrast between any 2 of the racial/ethnic groups was also statistically significant at P < .001. The second probe, which occurred 6 questions later in the interview, was only asked of individuals who either said no to having heard of the Tuskegee Syphilis Study in the first probe or who answered yes to that first probe but could provide no details about that study.
Of the original 1133 respondents who were interviewed, 57.2% (95% CI = 54.3%, 60.1%) indicated that they had heard of the Tuskegee Syphilis Study when the data from both probes were combined. For the 1128 respondents who filtered through the 2 probes with valid responses (i.e., a yes or no response), there were marked differences among the racial/ethnic groups on their final yes or no answer to the inquiry on whether they had ever heard of the Tuskegee Syphilis Study.
Results of an unadjusted bivariate analysis based upon the 2-probe combination showed that 76.4% of the 353 Blacks in the study (95% CI = 71.7%, 80.8%), 56.8% of the 623 Whites in the study (95% CI = 52.8%, 60.8%), and 25.3% of the 157 Hispanics in the study (95% CI = 18.9%, 33.0%) had indicated that they had heard of the Tuskegee Syphilis Study (). This difference was statistically significant across the 3 racial/ethnic groups (P < .001, by the χ2 test), with each 2-way contrast also statistically significant (P < .001, by the χ2 test). Across racial/ethnic groups, a comparison of yes responses to the first probe only versus the yes responses to the 2-probe combination revealed that only a very slight upward correction factor resulted from the use of the second probe for each of the 3 racial/ethnic groups (i.e., an increase of 3.8 percentage points in Blacks, 1.6 percentage points in Whites, and 1.7 percentage points in Hispanics).
also shows the percentage of correct responses to: “Has any US President ever apologized for the Tuskegee Syphilis Study?” and “Which US president?” Of the Blacks in the study, 42.5% had heard of the presidential apology (95% CI = 27.3%, 47.8%) and 34.8% knew that President Clinton had given the apology (95% CI=29.9%, 40.0%). Of the Whites in the study, 28.1% had heard of the presidential apology (95% CI=24.6%, 31.8%) and 24.1% knew that President Clinton had given the apology (95% CI = 20.8%, 27.6%). Of the Hispanics in the study, 8.3% had heard of the presidential apology (95% CI = 4.5%, 13.7%) and 3.2% knew that President Clinton had given the apology (95% CI=1.0%, 7.3%).
shows the multivariate logistic regression analyses of the responses to the same 3 questions adjusted for age, gender, education, income, and city. The adjusted multivariate analysis shows that the odds of hearing of the Tuskegee Syphilis Study were nearly 4 times greater for Blacks than for Whites (odds ratio [OR] = 3.9; 95% CI = 2.6, 5.7). The difference in odds between Hispanics and Whites was not statistically significant (OR = 1.6; 95% CI = 0.93, 2.7). The adjusted multivariate analysis for having heard of the presidential apology showed that whereas the odds for Blacks did not significantly differ from the odds for Whites (OR = 1.6; 95% CI = 0.91, 2.7), the odds of correctly naming Clinton as the President who made the apology were 2 times greater for Blacks than for Whites (OR = 2.3; 95% CI = 1.6, 3.4). Too few Hispanics had heard of the apology to include Hispanics in these latter 2 analyses.
| TABLE 2Logistic Regression Multivariate Analyses for Key Questions on Questionnaire: Tuskegee Legacy Project Study, 1999–2000 |
Given that the ORs for these 3 questions by city () showed a strong effect of city of residence on awareness of the study and of the presidential apology, adjusted ORs were computed to assess the racial difference for each of the 3 questions within each city. Because of the limitations of sample size on the stability and interpretation of data, this further analysis was only conducted for Blacks and Whites in the 3 cities with substantial numbers of these 2 racial groups (i.e., Birmingham, Tuskegee, and Hartford). The results of this additional within-city analysis indicated that the odds of Blacks having ever heard of the Tuskegee Syphilis Study were 4 to 5 times higher than were the odds for Whites in the cities of Hartford and Birmingham (OR = 4.37 [95% CI = 2.40, 7.96] and OR = 5.47 [95% CI = 2.35, 12.71], respectively), but not significantly different from Whites in the city of Tuskegee (OR = 2.37; 95% CI = 0.89, 6.34).
Although there were no statistically significant differences in odds between Blacks and Whites for the question “Has any US President ever apologized for the Tuskegee Syphilis Study?” in any of the 3 cities, statistically significant differences were observed for the third question: “Which US President?” For each of the 3 cities, Blacks were 2.5 to 3 times more likely than were Whites to correctly name Clinton as the president who made the apology (Hartford: OR = 2.81 [95% CI = 1.31, 6.03]; Birmingham: OR = 2.97 [95% CI = 1.44, 6.11]; and Tuskegee: OR = 2.50 [95% CI = 1.33, 4.70]).
Respondents who replied yes to the 2-probe series on ever having heard of the Tuskegee Syphilis Study were then asked the follow-up question: “As a result of what you have heard about the Tuskegee Syphilis Study, how likely are you to participate in a medical research study?” with responses on a 5-point Likert scale ranging from much more likely to much less likely. Because very few Hispanic respondents had responded yes to this 2-probe series (n = 39), this analysis was performed only for Black and White respondents.
shows an unadjusted bivariate analysis that revealed that among those who had heard of the Tuskegee Syphilis Study, the negative influence on the likelihood of participation in future studies was less among Blacks than among Whites; i.e., 50.0% of Blacks and 70.3% of Whites reported that they were less likely to participate as a result of what they had heard about the Tuskegee Syphilis Study (P < .001; Kendall’s tau-B).
Multivariate logistic regression analysis of this question (adjusted for age, gender, education, income, and city) revealed that despite what they had heard about the Tuskegee Syphilis Study, the odds of Blacks indicating a willingness to participate in biomedical studies were nearly 3 times greater than the odds of Whites (OR = 2.9; 95% CI = 1.4, 6.2).
After the respondent was asked “Has any US President ever apologized for the Tuskegee Syphilis Study?” a follow-up question was asked: “Based upon what you heard about the apology, would it influence your decision to join a biomedical research study today? Did that apology make you more or less likely to join a study?” This question also had a 5-point Likert scale of responses ranging from much more likely to much less likely.
The results from an unadjusted bivariate analysis of this follow-up question are shown in and again revealed a much less negative influence on Blacks, with 41.3% (vs 61.8% of Whites) self-reporting that they were less likely to join a biomedical study as a result of the presidential apology (P = .008, by the χ2 test). Multivariate logistic regression analysis of this question (adjusted for age, gender, education, income, and city) revealed that the odds of indicating that they were more likely to participate in biomedical studies as a result of having heard of the presidential apology were more than 2 times greater for Blacks than for Whites (OR = 2.3; 95% CI = 1.4, 3.9).