Establishing a Cohort to Investigate Health Disparities: MATERIALS AND METHODS

25 Oct

This study was approved by the institutional review boards at Vanderbilt University and Meharry Medical College. All participants provided written informed consent and (since April 14, 2003) a signed Health Insurance Portability and Accountability Act (HIPAA) authorization.

Preliminary Work

Before initiating the SCCS, we carried out pilot studies to test various aspects of the full-scale investigation, including recruitment methods and materials, the survey instrument and biospecimen collec­tion. This pilot work, described elsewhere, focused on the feasibility of enrolling subjects from both the general population and from attendees of community health centers (CHCs), which are government-fiinded healthcare facilities that provide basic health services primarily to low-income individuals. A primary finding of our pilot work was that CHC-based in-person recruitment was highly successful in an African-American and low-income population, a specific target of the SCCS. By the end of the pilot study, we had formed relationships with 15 CHCs across the south who supported the study and were willing to act as recruitment sites. In the pilot study, enrollment of participants through sampling driver’s license, voter registration and commercial mailing files was also found to be feasible for broadening the socioeconomic diversity of the cohort. finasteride online

Significant effort was devoted to developing and refining the baseline questionnaire, which elicits information on demographics; anthropometry; tobacco use; diet; alcohol consumption; physical activity; personal medical history; family medical history; reproductive history (for women); medication use; emotional well-being and social support; religion/spirituality; health insurance; use of medical and cancer screening services; occupational history; and miscellaneous factors, such as cell phone use and seat belt use. The largest single portion of the baseline questionnaire concerns the collection of data on usual dietary intake, using an 89-item food frequency questionnaire (FFQ) that was developed specifically for this study population. The questionnaire is comprehensive yet restricted in length to assure a reasonable administration time, both as an in-person interview and paper questionnaire (23 pages in the pilot study). The priorities were to include questions about factors that were: 1) known or suspected to be major contributors to cancer risk (e.g., diet), 2) suspected to play a role in racial disparities (e.g., access to healthcare), or 3) emerging as potentially important risk or protective factors for cancer (e.g., nonsteroidal anti-inflammatory drugs). We also made an effort to utilize many questions (or question groups) used previously in other large epidemiologic studies to facilitate meaningful comparisons in research findings across the different populations. To aid us in refining the final questionnaire, we also conducted two gender- and racially mixed focus groups (one in Nashville, TN, and one in Jacksonville, FL) consisting of 12 and 10 pilot study participants, respectively. The pilot participants provided detailed feedback about the questionnaire, as well as about other printed study materials, recruitment strategies and overall methods.

SCCS Recruitment and Methodology

Study base. The SCCS is seeking to enroll approximately 100,000 male and female participants in the southeastern United States, over two-thirds of whom will be African-American. More than half are expected to be recruited from CHCs and the remainder from stratified random sampling of general population sources. Cohort member enrollment is limited to English speakers between the ages of 40 and 79 who have not been under treatment for cancer (with the exception of nonmelanoma skin cancer) within the past year. Participants are currently being recruited from CHCs in Alabama, Arkansas, Florida, Georgia, Kentucky, Mississippi, North Carolina, South Carolina, Tennessee, Virginia and West Virginia. General population recruitment in these states is slated to begin in 2005. Census statistics indicate that approximately four million blacks aged 40-79 and about five times as many whites reside in these states. Subject enrollment and data collection—CHCs. buy levaquin online

Cohort member enrollment commenced at 14 CHCs in March and April 2002. An additional seven CHCs joined as recruitment sites in March 2003, another four joined in October 2003, and seven more joined in October 2004. CHCs continue to join the study on an ongoing basis, and they are selected to reflect a range of rural and urban areas as well as for the racial makeup of their patient population.

Study interviewers work on-site at each CHC enrolling participants. The interviewers are hired by the CHCs from the local community and are trained by SCCS scientific personnel. To date, 75% of the interviewers have been African-American, in approximate racial concordance with the participating CHCs’ patient population. The interviewers approach entrants to the CHC who appear to be age-eligible and assess their interest and eligibility. Potential participants are typically approached in lobbies and other waiting areas and could be patients, persons accompanying patients, persons using the pharmacy, etc. The interviewers utilize promotional materials placed within the CHC (large color posters, color brochures, small flyers on lobby seats) and sometimes placed outside of the CHC (e.g., brochures left in churches) in order to raise awareness of and interest in the study. The interviewers also typically carry a letter of support signed by a prominent physician or administrator within the CHC to reinforce the legitimacy of the study to potential participants. flomax 0.4 mg

Consent is obtained by the study interviewers, who review the informed consent form with the participants in a private area in the CHC. The interviewers receive training on the protection of human subjects and the administration of the consent form from the SCCS scientific staff, and complete two online tutorials offered through the National Institutes of Health and the Office for Human Research Protections (DHHS) on this subject. When they encounter participants with low literacy skills, they read the consent form aloud to the participant. All participants are given a copy of their signed consent form.

In-person baseline interviews are conducted in a private office in the CHC using a laptop computer and a specially designed computer-assisted personal interview (CAPI) with substantial logic-checking and skip pattern features built into the instrument. The CAPI also accesses a continuously updated participant roster to ensure that persons do not participate more than once. Handcards are used to facilitate several sections of the interview, allowing the participant to read answer choices off the card. Specific questions within the CAPI are updated as needed on an ongoing basis—for example, to reflect new medications on the market. Interviewers also abstract measured weight, height and blood pressure from the participants’ medical records if those meas­urements were taken on the day of study enrollment.

Data collected from the CAPIs are electronically sent to the SCCS data processing center on a daily basis from each CHC. After the uploading and archiving of the data, it is automatically erased from the laptop. This process helps safeguard the security of the data at the CHC, frees space on the laptops, and enables daily updating and checking of the master SCCS database.

With the exception of those who report infection with HIV or hepatitis (because of shipping restrictions), participants are asked to donate a blood (20 ml) or exfoliated buccal cell sample (if blood was refused) at study enrollment. Except for the rare instance where the study interviewer is a phle-botomist, blood samples are collected in the CHC lab by the on-duty phlebotomist. Participants who agree to donate blood are asked a supplemental set of questions about factors that could affect the interpretation of laboratory results or biomarker data (for example, time of last meal; date of last menstrual period; and recent (within the past 24 hours) use of cigarettes, medications and certain foods). Buccal cell samples are collected by the study interviewer, using the swishing method with Scope® mouthwash. All biospecimens are recorded with the date and time of collection, kept refrigerated and normally shipped on the same day of collection for next-morning delivery to Vanderbilt University. The buccal cells are obtained from the mouth rinse, and the blood is separated into its components (plasma, serum, buffy coat, red blood cells and clot) by cen-trifiigation and aliquotted into storage vials, which are stored at -80°C.
zetia 10 mg

Upon completion of the baseline interview and provision of a biospecimen, the participants are given $10 to acknowledge their time and effort.

Subject enrollment and data collection— general population. In order to recruit participants with more diverse income and education into the cohort, we will use government rosters (i.e., driver’s license, Medicare), organizational listings and cornmercial mailing files covering residents of the study states and randomly select an age-, sex- and race-stratified sample of persons aged 40-79 to contact for potential participation. We will strive to achieve the same relative proportion of African Americans to non-African Americans in this group of participants as in the group recruited at the CHCs.

Financial considerations dictate that we utilize a mailed, self-administered questionnaire for data collection in this phase of the study. This paper questionnaire will mirror the CAPI currently in use in the CHCs. Informed consent will be obtained via a consent form in the questionnaire booklet. When these participants return a completed questionnaire and signed consent form, they will be mailed a buccal cell collection kit (with instructions) and asked to self-collect the sample using the swishing method and mail it back to the lab at Vanderbilt. generic levitra online

Cohort follow-up. Cohort member retention is a primary challenge for all prospective studies but can be particularly difficult in minority and less educated populations. To aid us in minimizing loss to follow-up, we collect detailed contact information during the baseline interview, including a home address (and mailing address if different); home, work and cell phone numbers; Social Security number; email address (if available); and the name and phone number of a close friend or relative of the participant. In addition, welcome letters (a simple one-page letter welcoming the new participant to the study) are mailed to participants within two months of their enrollment. Part of the letter confirms their contact information, and a tear-off postcard allows the participant to correct or update their address or phone number as necessary. Newsletters containing information about the study and various health topics are also mailed to participants annually. Participants are also given refrigerator magnets bearing the toll-free number of the study and a reminder to call if their address or phone number changes.

For health outcome ascertainment, we will rely primarily on passive (via registry linkage) follow-up. The major emphasis will be on follow-up for mortality via linkage to the National Death Index and Social Security mortality files, and for incidence via linkage to state cancer registries. In addition, every few years we will attempt to re-contact study subjects and, using a self-administered questionnaire, ask them about interim health outcomes, changes in key exposures originally collected during the baseline interview (for example, health behaviors and insurance status), and new exposures of emerging interest. These methods will allow for the study of both cancer (and other disease) incidence and mortality outcomes.