Fig. 4

Treatment effect estimates at six months post-recruitment according to alternative missing not at random assumptions. The missing not at random assumptions are compared to the primary and complete case analyses. Each shaded rectangular strip shows the full posterior distribution. The darkness at a point is proportional to the probability density, such that the strip is darkest at the maximum density and fades into the background at the minimum density. The posterior mean and 95% credible interval (CrI) are marked. * interaction between treatment group and time period. For PSS-SR, negative differences favour the POPPI intervention, and for HrQoL positive differences favour the POPPI intervention. MAR = Missing At Random; MNAR = Missing Not At Random. A Primary treatment effect estimate. B Treatment effect on health-related quality of life score