Participant recruitment is a difficult and time consuming aspect of clinical trials, often resulting in delays and budget overruns. Having reached recruitment targets the next challenge is participant retention. Some weight-loss studies have attrition rates around 60% which may introduce bias in the results. It may be possible to reduce attrition rates if known predictors can be found but to date few studies produced consistent results. The aim of this exploratory study was to determine whether start date could be a predictor of attrition for participants involved in the SMART weight loss clinical trial (ACTRN12608000425392). Recruitment for the trial occurred as two cohorts at different times of the year. The relationship between attrition and cohort was determined using a Pearson chi square analysis. At three months 31% of Cohort 1 had withdrawn compared with 18% of Cohort 2. Cohort 1 participants, (enrolled July to November) were significantly more likely to withdraw within 6 months than Cohort 1 (January to March) (P=0.049). In each cohort most withdrawals occurred within the first three months. For Cohort 1, this coincided with the period leading up to Christmas and summer holidays. Changes in routines and increased food temptations at this time of year are known challenges of weight management and this may explain the significant difference between cohort withdrawal rates. The period around Easter and autumn holidays did not seem to have the same effect. Adjusting start dates and developing strategies to minimise the effect of seasonal holidays may help reduce attrition rates..
Available at: http://works.bepress.com/qingsheng_zhang/9/