RESUMO
Data aggregation in mental health is complicated by using different questionnaires, and little is known about the impact of item harmonization strategies on measurement precision. Therefore, we aimed to assess the impact of various item harmonization strategies for a target and proxy questionnaire using correlated and bifactor models. Data were obtained from the Brazilian High-Risk Study for Mental Conditions (BHRCS) and the Healthy Brain Network (HBN; N = 6,140, ages 5-22 years, 39.6% females). We tested six item-wise harmonization strategies and compared them based on several indices. The one-by-one (1:1) expert-based semantic item harmonization presented the best strategy as it was the only that resulted in scalar-invariant models for both samples and factor models. The between-questionnaires factor correlation, reliability, and factor score difference in using a proxy instead of a target measure improved little when all other harmonization strategies were compared with a completely at-random strategy. However, for bifactor models, between-questionnaire specific factor correlation increased from 0.05-0.19 (random item harmonization) to 0.43-0.60 (expert-based 1:1 semantic harmonization) in BHRCS and HBN samples, respectively. Therefore, item harmonization strategies are relevant for specific factors from bifactor models and had little impact on p-factors and first-order correlated factors when the child behavior checklist (CBCL) and strengths and difficulties questionnaire (SDQ) were harmonized.
Assuntos
Transtornos Mentais , Psicopatologia , Criança , Feminino , Humanos , Adolescente , Masculino , Reprodutibilidade dos Testes , Psicometria , Saúde Mental , Inquéritos e Questionários , Transtornos Mentais/diagnóstico , Transtornos Mentais/psicologiaRESUMO
BACKGROUND: Semi-structured diagnostic interviews and symptom checklists present similar internal reliability. We aim to investigate whether they differ in predicting poor life outcomes in the transition from childhood to young adulthood. METHODS: For this longitudinal study, we used data from the Brazilian High Risk Cohort Study for Childhood Mental Health Conditions. Eligible participants were aged 6-14 years on the day of study enrolment (January to February, 2010) and were enrolled in public schools by a biological parent in Porto Alegre and São Paulo, Brazil. 2511 young people and their caregivers were assessed at baseline in 2010-11, and 1917 were assessed 8 years later (2018-19; 76·3% retention). Clinical thresholds were derived using semi-structured parent-report interview based on the Diagnostic and Statistical Manual of Mental Disorders, according to the Developmental and Well-being Assessment (DAWBA), and clinical scores as defined by the Child Behavior Checklist (CBCL; T-score ≥70 considered positive caseness). At 8 years, participants were assessed for a composite life-threatening outcome (a composite of death, suicide attempts, severe self-harm, psychiatric inpatient admission, or emergency department visits) and a composite poor life chances outcome (a composite of any criminal conviction, substance misuse, or school dropout). We evaluated the accuracy of DAWBA and CBCL to predict these outcomes. Logistic regression models were adjusted for age, sex, race or ethnicity, study site, and socioeconomic class. FINDINGS: DAWBA and CBCL had similar sensitivity, specificity, predictive values, and test accuracy for both composite outcomes and their components. Any mental health problem, as classified by DAWBA and CBCL, was independently associated with the composite life-threatening outcome (DAWBA adjusted odds ratio 1·62, 95% CI 1·20-2·18; CBCL 1·66, 1·19-2·30), but only CBCL independently predicted poor life chances (1·56, 1·19-2·04). Participants classified by both approaches did not have higher odds of the life-threatening outcome when compared with participants classified by DAWBA or CBCL alone, nor for the poor life chances outcome when compared with those classified by CBCL alone. INTERPRETATION: Classifying children and adolescents based on a semi-structured diagnostic interview was not statistically different to symptom checklist in terms of test accuracy and predictive validity for relevant life outcomes. Classification based on symptom checklist might be a valid alternative to costly and time-consuming methods to identify young people at risk for poor life outcomes. FUNDING: Conselho Nacional de Desenvolvimento Científico e Tecnológico; Fundação de Amparo à Pesquisa do Estado de São Paulo; and Medical Research Council, European Research Council. TRANSLATION: For the Portuguese translation of the abstract see Supplementary Materials section.
Assuntos
Lista de Checagem , Saúde Mental , Adolescente , Humanos , Criança , Adulto Jovem , Adulto , Estudos de Coortes , Brasil , Estudos Longitudinais , Reprodutibilidade dos TestesRESUMO
OBJECTIVES: Model configuration is important for mental health data harmonization. We provide a method to investigate the performance of different bifactor model configurations to harmonize different instruments. METHODS: We used data from six samples from the Reproducible Brain Charts initiative (N = 8,606, ages 5-22 years, 41.0% females). We harmonized items from two psychopathology instruments, Child Behavior Checklist (CBCL) and GOASSESS, based on semantic content. We estimated bifactor models using confirmatory factor analysis, and calculated their model fit, factor reliability, between-instrument invariance, and authenticity (i.e., the correlation and factor score difference between the harmonized and original models). RESULTS: Five out of 12 model configurations presented acceptable fit and were instrument-invariant. Correlations between the harmonized factor scores and the original full-item models were high for the p-factor (>0.89) and small to moderate (0.12-0.81) for the specific factors. 6.3%-50.9% of participants presented factor score differences between harmonized and original models higher than 0.5 z-score. CONCLUSIONS: The CBCL-GOASSESS harmonization indicates that few models provide reliable specific factors and are instrument-invariant. Moreover, authenticity was high for the p-factor and moderate for specific factors. Future studies can use this framework to examine the impact of harmonizing instruments in psychiatric research.