The COOP/WONCA charts in an acute psychiatric ward. Validity and reliability of patients' self-report of functioning

We wanted to examine whether persons needing acute psychiatric admittance can give us reliable information about their functional status with the COOP/WONCA charts (Dartmouth Primary Care Cooperative Information Project-World Organization of National Colleges, Academies and Academic Associations of General Practitioners/Family Physicians), and if this information parallel that of therapists and ward staff. To examine internal consistency, external reliability and test-retest reliability, all patients consecutively admitted to an acute psychiatric department were asked to fill in the COOP/WONCA, Short-Form-36 Health Survey (SF-36) and Symptom Checklist 90--Revised (SCL-90-R) 1-2 weeks after admittance. Their therapists scored a special version of the GAF, and the ward staff scored an observer version of the COOP/WONCA. The first 59 patients repeated the COOP/WONCA scoring after a few days to estimate of test-retest reliability. Of a total of 267 persons admitted in the study period, 102 were included. Non-inclusion was largely due to early discharge. The internal consistency of the COOP/WONCA corresponded to a Cronbach's alpha=0.85. External reliability versus the SF-36 resulted in a correlation of r=-0.82 (P<0.001), and versus the observer version of COOP/WONCA we found r=0.46 (P<0.001). Correlation with GAF was low and not significant. The COOP/WONCA correlated significantly with the SCL-90-R scores (r=0.59, P<0.001). The test-retest correlation was r=0.85 (P<0.001). Correlations varied between sub-groups of patients, but all had consistently high correlations between the various self-scored measures. The COOP/WONCA gives a consistent and reliable report of acutely admitted psychiatric inpatients' own views of their functional capacity. The therapists' views and to some degree the ward staff's views diverge from the patients' opinions, especially for the more seriously disturbed patients.