Evan L. Thacker and Dr. Ray M. Merrill, Health Science

Randomized clinical trials (RCTs) are the final testing phase for new medical procedures or treatments in human subjects. A common problem in community-based RCTs is low participation. Researchers have sought to identify barriers to RCT recruitment by measuring knowledge and attitudes about RCTs. Most studies focus on subjects already involved in clinical trials or subjects who were invited to participate in clinical trials but declined. Our research addresses recruitment barriers in a broader context, by studying knowledge and attitudes among the general public, including individuals who have never participated in a RCT.

We developed a questionnaire based on a literature review and the input of experts in survey research, as well as feedback from 19 volunteers who tested a prototype version of the questionnaire. The questionnaire contains three sections. Section I consists of eleven true or false questions testing general knowledge of RCT methods and purposes. Section II contains four statements dealing with future RCT participation and eight statements portraying common attitudes about RCTs, each of which is scored on a four point scale ranging from “strongly disagree” to “strongly agree.” Section III contains ten demographic and health questions. We administered the questionnaire in April 2003 at Brigham Young University’s annual Gerontology Conference, a one-day event which draws local university students studying gerontology as well as local seniors who are interested in healthy aging. We offered a one-dollar cash incentive to each person who completed the survey.

We collected completed surveys from 163 people. Survey respondents ranged in age from 18 to 84 years, with a mean age of 45.5 years. The majority of respondents were white women, college educated, with an annual income of $30,000 or more. Only 16.7% of respondents had previous experience as RCT subjects. 30.7% of respondents had a history of chronic illness. We noted several significant differences between the university students and the seniors, including levels of education and income, as well as prior RCT experience and prevalence of chronic illness. We scored each knowledge statement in Section I correct or incorrect, and for each respondent computed a total knowledge score, with 11 points possible. Scores ranged from 5 to 11, with a mean score of 9.48. 22% of respondents answered every knowledge statement correctly.

We scored each attitude statement in Section II from 0 to 3, with 3 indicating the most positive attitude. The majority of respondents expressed positive attitudes regarding willingness to participate in RCTs. When asked to imagine different scenarios, respondents were more willing to participate given that they had a life-threatening or chronic illness versus a less serious illness or no illness at all. Most respondents had positive attitudes toward medical research and researchers, and benefits of RCT participation to self and others. On the other hand, the majority had negative attitudes about random selection of treatment, effectiveness of trial treatments, and personal inconvenience of participating in RCTs.

One of the purposes of this study is to measure the relationship of knowledge and attitudes. We used Spearman’s rho to measure correlations between correct answers for the knowledge statements and positive responses for the attitude statements. We considered correlations significant at the 0.05 significance level. Of the 88 correlations we computed, only 14 (15.9%) are significant. All but one of the significant correlations are negative, indicating a slight negative correlation between knowledge about RCTs and positive attitudes about RCTs.

The second purpose of this study is to predict willingness to participate in RCTs from knowledge, attitudes, prior RCT involvement, and chronic illness status. We used logistic regression to estimate odds ratios for the effects of these variables on willingness to participate. First we modeled willingness to participate in RCTs with a life-threatening or chronic illness. Under this scenario, none of the knowledge statements and only two of the attitude statements significantly predict willingness to participate. We then modeled willingness to participate in a RCT with a less serious illness or when healthy. Under this scenario, only one knowledge statement and two attitude statements predict willingness to participate. Prior RCT experience and chronic illness status have no effect on willingness to participate under either scenario.

We draw three main conclusions from our research. First, based on the high mean knowledge score, (9.48 out of 11), and the high level of positive response to the attitude statements among our sample, we conclude that a well-educated population holds a generally positive view of RCTs and is somewhat interested in future participation in such studies. However, our second conclusion is that specific knowledge of RCT purposes and procedures is in fact negatively associated with attitudes toward RCTs. The statistics we examined show a slight negative correlation between correct knowledge and positive attitudes, suggesting that people are worried by the risks associated with RCTs more than they are interested in the benefits. The correlations we found do not present a clear model for a meaningful relationship between knowledge and attitudes that could aid health educators in promoting positive attitudes through education. Third, willingness to participate in RCTs does not depend on knowledge or attitudes regarding RCTs, but is evidently a product of some other influence. Willingness to participate seems to depend much more on perceived self-risk of illness than on knowledge or attitudes. The results of our study indicate that although people understand RCTs and view medical research in a favorable light, they consider personal involvement in RCTs as somewhat questionable or risky, and they become more willing to take a risk when their personal health is in greater jeopardy.

Further research is needed to reveal true indicators of willingness among the general population to participate in RCTs. Once these indicators are identified, efforts should be made to design and implement community health education programs aimed at overcoming these barriers to participation. Additionally, future work should attempt to measure knowledge, attitudes, and other variables in more varied cross-sections of the population. Finally, continued use of the questionnaire we created will allow us to assess the validity of the questionnaire itself in accurately and consistently measuring knowledge and attitudes, and making further adjustments to the questionnaire to collect the most meaningful and relevant data.

We gratefully acknowledge Rosemary Thackeray, Michael Barnes, Greg Snow, and Sterling Hilton for their expert advice in questionnaire development.