Background The last decades several individualised patient reported outcome measures have been developed, which allow the patients to choose the items or domains they consider most important. In rehabilitation, treatment is tailored to the patient’s individual goals. It has therefore been suggested that patient specific instruments may be better suited to capture the effects of rehabilitation. The Patient Generated Index (PGI) is an individualised instrument designed to measure health related quality of life (HRQoL), which has been used since the early nineties.

Objectives To evaluate the psychometric properties of the Norwegian version of the Patient Generated Index (PGI) for patients with rheumatic diseases participating in rehabilitation or self-management programmes.

Methods Patients attending three rehabilitation institutions completed self-report questionnaires (including the PGI) at arrival, 1-2 days after arrival (test-retest), and after 5 and 52 weeks. The PGI, scored from 0 to 100, was assessed for data quality, reliability, validity and responsiveness. Test-retest reliability was assessed by the intraclass correlation coefficient (ICC) and standard error of measurement (SEM)/smallest detectable change (SDC). Construct validity was assessed by testing pre-defined hypotheses of relationships between PGI and the other outcome measures by Pearson correlation. Responsiveness was assessed by the standardised response mean (SRM).

Results 145 patients aged 29 to 75 years with different inflammatory rheumatic diseases agreed to participate in the study. 118 (81%) filled in the PGI correctly at baseline. Test-retest reliability showed an ICC (95% CI) of 0.86 (0.74, 0.93), a SEM of 7.2 and a SDC of 20.0. Low to moderate correlation was found between the PGI and the SF-36, QoLS, MHAQ and the NRS for pain, fatigue and disease activity. Highest correlations were found between the PGI and the NRS disease activity and the SF-36 role physical (0.47, p>0.001). At 5 weeks the PGI had an SRM of 0.23 and at12 months it was 0.43, which was the highest among the study instruments (figure 1).

Conclusions The PGI showed good psychometric properties, in particular it showed higher responsiveness at long term than any of the other standardised instruments. A limitation is that relatively many patients found it difficult to fill in the PGI. Further research is needed to increase the feasibility of the PGI. The presence of a facilitator who can guide patients who have problems with completing the PGI or an electronic version with feedback might increase the completion rates. Despite these challenges we recommend the PGI for evaluation of rehabilitation programmes for patients with rheumatic diseases.

Disclosure of Interest None Declared