Olsen, Catherine M.; Neale, Rachel E.; Cicero, Rebekah A.; Jackman, Lea M.; O'Brien, Suzanne M.; Perry, Susan L.; Ranieri, Barbara A.; Whiteman, David C.
Population Health Department Queensland Institute of Medical Research Queensland, Australia Catherine.Olsen@qimr.edu.au
Supplemental digital content is available through direct URL citations in the HTML and PDF versions of this article (www.epidem.com). This content is not peer-reviewed or copy-edited; it is the sole responsibility of the author.
To the Editor:
Maximizing participation in epidemiologic surveys improves efficiency, generalizability, and study validity. Charities often use leaflets that describe personal experiences of patients in soliciting donations. Such leaflets are increasingly being used in epidemiologic research in an effort to increase study participation, despite a lack of evidence regarding effectiveness. We conducted a randomized trial to test whether including leaflet with the personal stories of patients in a study recruitment pack would increase participation in a prospective study of cancer.
The QSkin Sun and Health Study is a large cohort study of Australian men and women to investigate the role of environment and host/genetic characteristics in the etiology of cutaneous melanoma and other cancers of the skin. We randomly sampled 193,344 men and women 40–69 years of age from the Australian Electoral Roll and contacted them between November 2010 and November 2011. Potentially eligible persons were mailed an invitation pack containing an information sheet, the study survey and consent forms, and a reply-paid envelope. Participants joined the study by completing and returning the hard-copy survey and consent forms or by completing the forms online. We sent a single reminder card to all persons who did not respond to the initial invitation within approximately 2.5 weeks.
In one recruitment mail-out for the QSkin Sun and Health Study cohort, we randomized 31,114 potential participants to either intervention or control groups, balanced within strata of age group (<55; 55+ years) and sex. The intervention group received, in addition to the standard invitation pack, a leaflet that included a family photograph and personal story written by a young wife and mother whose husband had died of melanoma (eAppendix, http://links.lww.com/EDE/A601). The leaflet was designed in consultation with Melanoma Patients Australia, a national advocacy group for melanoma patients and their families, and was circulated in-house among the investigators and research staff for testing and to gauge emotive response. All other aspects of recruitment, follow-up, and participation were the same for both groups. We compared the participation fraction in each of the study arms using the χ2 tests of 2 proportions.
We found no appreciable difference in the participation fractions between the intervention (22.9%) and control groups (22.0%; P = 0.12) (eTable, http://links.lww.com/EDE/A601). There was no evidence of interaction between the intervention and either sex (P = 0.45) or age-group (6 levels; P = 0.62). Respondents in the leaflet group did not return their surveys more quickly; median response time was 48 days for the leaflet group and 45.5 days for the no-leaflet group (Figure).
Figure. Cumulative r...Image Tools
These findings do not support the use of emotive leaflets as a means to increase participation in epidemiologic studies. To our knowledge, this is the first study to evaluate such an intervention. Although a recent Cochrane review reported a slight increase in response for mailed surveys that stressed the benefits of the research to society, there was little evidence for a pooled intervention effect (odds ratio = 1.06 [95% confidence interval = 0.89–1.28]) with significant between-study heterogeneity.1 Moreover, the included studies were conducted in nonhealth disciplines, and thus the relevance of the review for health research has been questioned.2
According to survey and marketing research, the 3 main reasons for participating in mailed surveys are altruism (ie, a willingness to pay a personal cost to benefit others), survey-related reasons, and egoistic reasons.3 Our methodology was designed to elicit an emotive response and appeal to altruism. Our data suggest this approach provides little benefit in an epidemiologic context and is unlikely to justify the expense.
Catherine M. Olsen
Rachel E. Neale
Rebekah A. Cicero
Lea M. Jackman
Suzanne M. O'Brien
Susan L. Perry
Barbara A. Ranieri
David C. Whiteman
for the QSkin Study
Population Health Department Queensland Institute of Medical Research
1. Edwards PJ, Roberts I, Clarke MJ, et al.. Methods to increase response to postal and electronic questionnaires. Cochrane Database Syst Rev 2009:MR000008.
2. Smeeth L, Fletcher AE. Improving the response rates to questionnaires. BMJ. 2002;324:1168–1169.
3. Singer E, Bossarte RM. Incentives for survey participation when are they “coercive”? Am J Prev Med. 2006;31:411–418.
© 2012 Lippincott Williams & Wilkins, Inc.