The Patient Advocate Sub-Committee (PASC) of the Cancer and Leukemia Group B's (CALGB) Committee on Advocacy, Research Communication and Ethics (CARE), in collaboration with cancer advocacy organizations, surveyed cancer survivors from July to September 2004 to identify how patients learn about and enroll in trials.

The goal of the survey was to identify opportunities to improve patient enrollment in cancer trials and provide recommendations to improve accrual in CALGB clinical trials. A key challenge to successful cancer clinical trial research is the issue of patient accrual. Poor accrual has been associated with early closure of studies¹ and negative findings.² Therefore, improving trial enrollment is a priority for NCI-funded research groups such as CALGB.

This article presents the findings of the PASC survey and discusses their implications for increasing trial participation among cancer patients.

Survey methods

The survey assessed awareness of, attitudes toward, and sources of knowledge about cancer trials among cancer survivors; documented the factors associated with clinical trial participation and nonparticipation; and assessed how knowledge, attitudes, and willingness to participate in cancer trials are mediated by demographics, such as ethnicity, socioeconomic status, and education.

PASC members developed a survey written at an 8th grade reading level, and the CALGB statistical center staff reviewed it. Multiple choice and open ended questions requiring written comments were included in the survey, and written comments were classified and assigned a number code.

The project was conducted in collaboration with advocacy organizations with which PASC members had ties. Members identified advocacy organizations comprised of cancer survivors who could complete the survey. Each PASC member who identified potential collaborators acted as a liaison to the respective advocacy group. The liaison contacted the advocacy group, explained the project, and assessed the willingness and capacity of the organization to participate. In order to participate, the group needed to inform their membership of the survey and provide access to it.

Liaisons then worked with each organization to identify the most appropriate mechanism to survey their members. Options included posting the survey on the organization's Web site, mailing out the survey to members along with a newsletter, and administering the survey at a scheduled meeting.

The CALGB Web site manager established a link to the survey on each collaborating cancer advocacy organization's Web site. Once posted on the site, cancer survivors could click on the link, complete the survey, and then click on an icon to submit it. Once submitted, online survey responses were emailed to the lead author, a PASC member and project coordinator.

Statistical support staff at Wake Forest University's School of Medicine conducted data analysis. Descriptive analysis was performed on the total sample. Pearson's chi-square tests were used to assess univariate associations of demographic characteristics with clinical trial participants.

Result insights

Sixteen cancer survivor organizations were contacted for their interest in participating in the survey, and 14 participated. Most organizations used their Web site to post the survey and inform their members about the study. One organization sent out a mailing in addition to posting the survey, and one circulated the survey at a scheduled meeting.

A total of 922 surveys were submitted via email, and an additional 35 surveys were mailed in for a total response of 957 surveys over a two-month period. A total of 888 of the submitted surveys contained enough information to analyze, but due to time restrictions, 727 surveys were analyzed. Surveys selected for analysis represented an attempt to maximize demographic diversity, both in type of cancer and ethnicity.