Rosie J Lacey1, Ross Wilkie1, Gwenllian Wynne-Jones1, Joanne L Jordan1, Emily Wersocki1, John McBeth1,2. 1. Arthritis Research UK Primary Care Centre, Research Institute for Primary Care & Health Sciences, Keele University, Keele, Staffordshire ST5 5BG, UK. 2. Arthritis Research UK Centre for Epidemiology, University of Manchester, Manchester, UK.
Abstract
Background: adults aged ≥65 years are often excluded from health research studies. Lack of representation reduces generalisability of treatments for this age group. Objective: to evaluate the effectiveness of strategies that improve recruitment and retention of adults aged ≥65 in observational studies and randomised controlled trials (RCTs). Methods: searches conducted in 10 databases for RCTs of recruitment and retention strategies in RCTs or observational studies. Two reviewers screened abstracts and full-text articles for eligibility and extracted data. Studies without separate data for adults aged ≥65 were discarded. Risk of bias assessed using the Cochrane Risk of Bias tool. Results were synthesised narratively. Results: thirty-two studies were included in the review (n = 75,444). Twelve studies had low risk of bias, of which 10 had successful strategies including: Opt-out versus opt-in increased recruitment (13.6% (n = 261)-18.7% (n = 36) difference; two studies); Advance notification increased retention (1.6% difference, OR 1.45; 95% CI 1.01, 2.10, one study (n = 2,686); 9.1% difference at 4 months, 1.44; 1.08, 1.92, one study (n = 753)); Hand-delivered versus postal surveys increased response (25.1% difference; X2 = 11.40, P < 0.01; one study (n = 139)); Open randomised design versus blinded RCT increased recruitment (1.56; 1.05, 2.33) and retention (13.9% difference; 3.1%, 24.6%) in one study (n = 538). Risk of bias was high/unclear for studies in which incentives or shorter length questionnaires increased response. Discussion: in low risk of bias studies, few of the strategies that improved participation in older adults had been tested in ≥1 study. Opt-out and advance notification strategies improved recruitment and retention, respectively, although an opt-out approach may have ethical limitations. Evidence from single studies limits the generalisability of other strategies.
Background: adults aged ≥65 years are often excluded from health research studies. Lack of representation reduces generalisability of treatments for this age group. Objective: to evaluate the effectiveness of strategies that improve recruitment and retention of adults aged ≥65 in observational studies and randomised controlled trials (RCTs). Methods: searches conducted in 10 databases for RCTs of recruitment and retention strategies in RCTs or observational studies. Two reviewers screened abstracts and full-text articles for eligibility and extracted data. Studies without separate data for adults aged ≥65 were discarded. Risk of bias assessed using the Cochrane Risk of Bias tool. Results were synthesised narratively. Results: thirty-two studies were included in the review (n = 75,444). Twelve studies had low risk of bias, of which 10 had successful strategies including: Opt-out versus opt-in increased recruitment (13.6% (n = 261)-18.7% (n = 36) difference; two studies); Advance notification increased retention (1.6% difference, OR 1.45; 95% CI 1.01, 2.10, one study (n = 2,686); 9.1% difference at 4 months, 1.44; 1.08, 1.92, one study (n = 753)); Hand-delivered versus postal surveys increased response (25.1% difference; X2 = 11.40, P < 0.01; one study (n = 139)); Open randomised design versus blinded RCT increased recruitment (1.56; 1.05, 2.33) and retention (13.9% difference; 3.1%, 24.6%) in one study (n = 538). Risk of bias was high/unclear for studies in which incentives or shorter length questionnaires increased response. Discussion: in low risk of bias studies, few of the strategies that improved participation in older adults had been tested in ≥1 study. Opt-out and advance notification strategies improved recruitment and retention, respectively, although an opt-out approach may have ethical limitations. Evidence from single studies limits the generalisability of other strategies.
Authors: Thomas M Gill; Joanne M McGloin; Amy Shelton; Luann M Bianco; Eleni A Skokos; Nancy K Latham; David A Ganz; Linda V Nyquist; Robert B Wallace; Martha B Carnie; Patricia C Dykes; Lori A Goehring; Margaret Doyle; Peter A Charpentier; Erich J Greene; Katy L Araujo Journal: J Am Geriatr Soc Date: 2020-03-25 Impact factor: 5.562
Authors: A J Daley; K Jolly; H Bensoussane; N Ives; S A Jebb; S Tearne; S M Greenfield; L Yardley; P Little; N Tyldesley-Marshall; R V Pritchett; E Frew; H M Parretti Journal: Trials Date: 2020-09-01 Impact factor: 2.279
Authors: Nynikka R Palmer; Hala T Borno; Steven E Gregorich; Jennifer Livaudais-Toman; Celia P Kaplan Journal: Cancer Causes Control Date: 2021-06-29 Impact factor: 2.532