Derivation and Evaluation of a Risk-Scoring Tool to Predict Participant Attrition in a Lifestyle Intervention Project

Abstract

Participant attrition in clinical trials and community-based interventions is a serious, common, and costly problem. In order to develop a simple predictive scoring system that can quantify the risk of participant attrition in a lifestyle intervention project, we analyzed data from the Special Diabetes Program for Indians Diabetes Prevention Program (SDPI-DP), an evidence-based lifestyle intervention to prevent diabetes in 36 American Indian and Alaska Native communities. SDPI-DP participants were randomly divided into a derivation cohort (n = 1600) and a validation cohort (n = 801). Logistic regressions were used to develop a scoring system from the derivation cohort. The discriminatory power and calibration properties of the system were assessed using the validation cohort. Seven independent factors predicted program attrition: gender, age, household income, comorbidity, chronic pain, site–s user population size, and average age of site staff. Six factors predicted long-term attrition: gender, age, marital status, chronic pain, site–s user population size, and average age of site staff. Each model exhibited moderate to fair discriminatory power (C statistic in the validation set: 0.70 for program attrition, and 0.66 for long-term attrition) and excellent calibration. The resulting scoring system offers a low-technology approach to identify participants at elevated risk for attrition in future similar behavioral modification intervention projects, which may inform appropriate allocation of retention resources. This approach also serves as a model for other efforts to prevent participant attrition.

The marketing plan and outcome indicators for recruiting and retaining parents in the HomeStyles randomized controlled trial

“…Female parents outnumbered male parents in this study by 14 to 1, which is the likely cause of the large CI shown in Table 3 . Others have reported that some of these same factors were predictive of premature termination of participation in intervention studies, including illness [ 92 , 164 ], family stress [ 165 ], and being male [ 120 , 164 ], as well as many other factors that were not predictors in this study. For example, predictors of participant completers in other studies associated with behavioral intervention trials that were not predictors in this study include sociodemographic (e.g., household income, level of education, family size, race/ethnicity), parent intrapersonal (e.g., depression, management skills), child intrapersonal (e.g., perceived susceptibility of child to health risk), and family interpersonal (e.g., family support) characteristics, as well as neighborhood conditions [ 22 , 24 , 26 , 33 , 92 , 129 – 134 , 138 , 164 , 166 , 167 ].…”

“…Additionally, this is among the first papers to report recruitment and retention strategies for an online, community-based, parent-driven childhood obesity prevention intervention [ 116 ] and to report a comprehensive, systematic RCT recruitment and retention plan based on services marketing principles. Few studies have expanded the consideration of retention factors beyond sociodemographic characteristics and/or used theoretical underpinnings to guide the selection of retention factors when examining predictors of RCT completers [ 25 , 26 , 92 , 101 , 164 ].…”

Team members influence retention in a First Peoples’ community-based weight-loss program

“…Poor retention is a problem because longer time spent in weight-loss programs is associated with better weight outcomes ( Jiandani et al, 2016 ); poor retention may distort study findings, likely biasing toward program effectiveness, and require better retention strategies ( Rae et al, 2013 ). Retention in weight-loss programs is generally associated with older age ( Leahey et al, 2010 , Jiandani et al, 2016 , Jiang et al, 2016 , Babatunde et al, 2017 , Burgess et al, 2017 , Leung et al, 2017 , Stoutenberg et al, 2017 , Alexander et al, 2018 , Tomioka et al, 2019 ), though some studies have found no such association ( Moroshko et al, 2011 , Latner and Ciao, 2014 ). Retention rates have been found to be higher among females ( Jiang et al, 2016 ), but also among males ( Burgess et al, 2017 ) or not associated with gender ( Moroshko et al, 2011 , Stoutenberg et al, 2017 ).…”

“…Retention in weight-loss programs is generally associated with older age ( Leahey et al, 2010 , Jiandani et al, 2016 , Jiang et al, 2016 , Babatunde et al, 2017 , Burgess et al, 2017 , Leung et al, 2017 , Stoutenberg et al, 2017 , Alexander et al, 2018 , Tomioka et al, 2019 ), though some studies have found no such association ( Moroshko et al, 2011 , Latner and Ciao, 2014 ). Retention rates have been found to be higher among females ( Jiang et al, 2016 ), but also among males ( Burgess et al, 2017 ) or not associated with gender ( Moroshko et al, 2011 , Stoutenberg et al, 2017 ). Other common factors associated with retention are initial weight or BMI and initial weight-loss success.…”

“…To our knowledge only two studies have evaluated individual-level demographics associated with attrition in community lifestyle programs for Indigenous peoples. In a diabetes prevention program for Native American/Native Alaskans, age and gender predicted attrition in both program (4–6 month) and maintenance (12 months or longer) phases ( Jiang et al, 2016 ). Age and fitness level predicted attrition at 12 months from a culturally tailored community-setting fitness class for older Hawaiian adults ( Tomioka et al, 2019 ).…”