Factors differentiating dropouts from completers in a longitudinal, multicenter clinical trial

Background: Subject dropout from a prospective, longitudinal trial can produce biases in the remaining sample that affect study findings and their interpretation, yet little is known about factors contributing to dropout. Objective: To determine characteristics differentiating those who complete from those who drop out of a longitudinal multicenter clinical trial. Methods: In this study, 578 parents and other caretakers of infants at risk for cardiopulmonary arrest enrolled in a longitudinal trial investigating the psychosocial impact of cardiopulmonary resuscitation training. In this secondary analysis, the baseline sociodemographic, emotional, psychosocial, and infant characteristics of those who dropped from the trial were compared with those who completed the study. Results:The study was completed by 60% (n = 347) of the participants. Those most likely to drop out were fathers or other caretakers (vs. mothers) employed outside the home who spoke English (vs. Spanish); were assigned to an experimental group (vs. a control group); had higher levels of depression, hostility, and overall psychosocial distress; and held negative views about health care. Conclusion: Although a few sociodemographic characteristics differentiated dropouts from completers, personal, emotional and psychosocial factors were the predominant predictors of dropouts. Other reputed sources of retention difficulties (e.g., income, education, minority status, lack of social support, or problems with family functioning) did not predict dropout.