SWAT 119: the effectiveness of a Thank You card to improve trial follow-up; a randomised study within a trial (SWAT)

Retention in randomised controlled trials is crucial to maximising study power and generalisability. A Study Within A Trial (SWAT) evaluated if sending a Thank You card improved 6-month questionnaire return rates in the SWHSI-2 trial.

A two-arm SWAT, using 1:1 (intervention:control) allocation, embedded within the SWHSI-2 trial. The primary outcome was the difference in the return rate of the 6-month questionnaire. Secondary outcomes were the difference in return rate of the 12-month questionnaire, questionnaire completeness, need for a reminder, and cost. The primary analysis was conducted using mixed-effects logistic regression adjusted for main trial allocation as a fixed effect and site as a random effect. Random-effects meta-analysis combined all available data for this intervention.

A total of 560 participants were included in the SWAT. There was no difference in the 6-month questionnaire return rate between the Thank You card group and the no Thank You card group (OR 1.04, 95% CI 0.73 to 1.50, p = 0.87). The 12-month retention rate was slightly higher in the Thank You card group, but the difference was not statistically significant (OR 1.15, 95% CI 0.81 to 1.62, p = 0.43). Findings were very similar in sensitivity analyses accounting for intervention participants who did not receive their card. There was no evidence of difference for any of the remaining secondary outcomes. Meta-analysis of the 12-month return rate suggests that Thank You cards may provide slight improvements in questionnaire response rates; however there is uncertainty in this estimate (OR 1.07, 95% CI 0.79 to 1.45).

It remains unclear if a Thank You card increased the rate of 6-month follow-up questionnaire completion in the SWHSI-2 trial. This is further amplified by the limited number of SWAT replications completed to date and included in the meta-analysis (n=2). The SWATs to date have primarily been undertaken in trials with a predominantly older, white, male, population. Further SWAT replications are therefore required in other populations to ensure generalisability of a cumulative SWAT finding.

Host Trial: ISRCTN26277546. Prospectively registered on 25^th March 2019. SWAT: MRC Hub for Trials Methodology Research SWAT repository #119 can be found at: https://www.qub.ac.uk/sites/TheNorthernIrelandNetworkforTrialsMethodologyResearch/FileStore/Filetoupload,959362,en.pdf .