BACKGROUND
Missing cases present a challenge to our ability to evaluate the effects of web-based psychotherapy trials. As missing cases are often lost to follow up, less is known about their characteristics, their likely clinical outcomes, or the likely effect of the treatment being trialled.
OBJECTIVE
To explore the characteristics of missing cases, their likely treatment outcomes, and the ability of different statistical models to approximate missing post-treatment data.
METHODS
A sample of internet-delivered cognitive behavioural therapy participants, in routine care (n = 6701 with 36% missing cases at post-treatment), was used to identify predictors of dropping out of treatment and predictors that moderated clinical outcomes, such as psychological distress, anxiety and depressive symptoms. These variables were then incorporated into a range of statistical models that approximated replacement outcomes for missing cases, with the results compared using sensitivity and cross-validation analyses.
RESULTS
Treatment adherence, as measured by the rate of an individual’s progress through the treatment modules, and higher symptom scores at pre-treatment, were identified as the dominant predictors of missing cases probability (Nagelkerke R2 = 60.8%), as well as the rate of symptom change. Low treatment adherence, in particular, was associated with increased odds for presenting as missing cases during post-treatment assessment (eg, OR = 161.1:1) and at the same time, attenuate the rate of symptom change across anxiety (up to 28% of the total symptom 48% reduction effect), depression (up to 41% of the total 48% symptom reduction effect) and psychological distress symptom outcomes (up to 52% of the total 37% symptom reduction effect) at the end of an eight week window. Reflecting this pattern of results, statistical replacement methods that overlooked the features of treatment adherence, and baseline severity, underestimated missing case symptom outcomes by as much as 40% at post-treatment.
CONCLUSIONS
The treatment outcomes of the cases that were missing at follow up were distinct from the remaining observed sample. Thus, overlooking the features of missing cases is likely to result in an inaccurate estimate of the effect of treatment.
CLINICALTRIAL
Association of diabetes and tuberculosis: impact on treatment and post-treatment outcomes
