Mariken Stegmann

are reported. If scores showed a ceiling effect with censoring from above, that is 100, 33  a Tobit model  was used. 34,35  In all models, adjustments weremade where necessary for baseline differences  between the groups. A P‐value ≤0.05 was considered statistically significant for all analyses. Finally,  the experiences of GPs and patients were reported  descriptively.  Results  Participants  During the study period, the contact details of 268 patients were sent to the research team  by oncologists and were screened for eligibility. Of these, 223 were eligible and 145 (65%)  agreed to participate. Finally, 53 and 61 patients completed the assessment in the OPT and  control groups, respectively (Figure 3). The mean age of participants was 74.0 years  (standard deviation [SD] = 6.4 years), 70.2% (n = 80) were male, and 71.9% (n = 82) had lung  cancer (Table 1). In the intervention group, both the OPT scores and/or a GP evaluation were  available for 47 patients (88.7%). There were no apparent differences in the baseline  characteristics between groups (Table 1).  Figure 3. CONSORT diagram of the study. CONSORT = Consolidated Standards of Reporting Trials.  OPT = Outcome Prioritisation Tool.  Patient empowerment  Continuous scores. For both groups, high DSE scores were observed (ceiling effect). 33 Correcting for this, the adjusted mean scores were 86.8 (SD = 18.2) for the OPT group and  44 Chapter 4