O6.8. THE EXTERNAL VALIDITY OF EARLY PSYCHOSIS RESEARCH: IMPLICATIONS FOR UNDERSTANDING CLINICAL POPULATIONS AND MEASUREMENT-BASED CARE

Research findings over the past two decades have improved our understanding of schizophrenia and related disorders, including their onset and early course. However, this growing uptake also necessitates attention to the representativeness of research samples. In order to assess the implicit assumption of generalizability, we examined characteristics of research participants from within a large, catchment-based early intervention program for first episode psychosis (EIP) and compared them across a broad range of demographic and clinical factors to EIP patients from the same program who did not participate in research. Within a well-established EIP clinical research infrastructure operating in Montreal, Canada since 2003, patients (ages 14–35) who consented to participate in one of two major services-oriented projects funded by a national health research agency (n = 300) were compared with patients who elected not to participate during the same time periods of recruitment (n = 214). All subjects were drawn from a large, geographically defined catchment area of approximately 300,000 individuals with no competing public or private services in the same region. Data was systematically collected from all patients (with approval from the local research ethics board) based on a desire to engage in ongoing program evaluation. Group representativeness was assessed in dimensions of sociodemographic measures, pathways to care, psychiatric symptoms (positive psychotic, negative psychotic, depression, anxiety), and functioning (global functioning, social and occupational functioning) at entry to the EIP program. Between-group differences were assessed using basic descriptive statistics including t-tests, chi-squared tests, and Mann-Whitney U tests, as appropriate. Patients who participated in research studies were more likely to be diagnosed with affective psychosis than non-participants (35% vs. 21%, respectively; p<0.001), to have proportionally longer median durations of untreated illness (9.11 months vs. 5.67 months, respectively; p<0.003, and to have higher baseline total symptom scores on the Scale for the Assessment of Positive Symptoms (35.58 vs. 30.80, respectively; p<0.001) and the Brief Psychiatric Rating Scale (67.21 vs. 63.67, respectively; p<0.001). Participants also trended towards being more engaged in post-secondary education than non-participants (50.67% vs. 42.99%, respectively; p=0.086) and came from environments of lower rather than higher socio-economic status (71.3% to 63.31%, respectively; p=0.084). Even in longstanding catchment-based EIP settings with a history of community outreach, research samples may be capturing subgroups that are not representative of the presenting clinical population in important yet potentially divergent ways. Given the recent ascendance of population-based approaches in mental health, researchers should be aware of the possibility of similar discrepancies in their own studies and careful to interpret study findings in light of questions regarding generalizability. Finally, these findings suggest strategies for ensuring representativeness in study recruitment, highlight the need to contextualize the reporting of recruited samples with comparators, and have implications for what is prioritized in measurement-based care efforts.

Récupéré en direct depuis OpenAlex et désinversé. Les résumés ne sont pas conservés dans cette base de données : les index inversés représentent 8,6 Go des 9,3 Go de texte de la base, et le serveur dispose de 13 Go libres.