Evaluating Psychometric Properties of Parent- or Caregiver-Report Instruments on Child Maltreatment: Systematic Reviews Using the COSMIN Methodology

Child maltreatment (CM) is a devastating public health problem that affects more than half of all children worldwide. The United Nations (UN) launched an initiative to eliminate CM as part of their 2030 Agenda for Sustainable Development Goals. To monitor progress towards achieving the goal of eradicating CM, all UN member states should annually report their national CM prevalence and progress in reducing CM. However, no consensus on which instruments are best for investigating CM has been achieved. This thesis aimed to evaluate the psychometric properties of all currently available parent- or caregiver-report instruments on any type of CM and recommend those with the best psychometric quality. A systematic search of six databases (CINAHL, Embase, ERIC, PsycINFO, PubMed and Sociological Abstracts) was conducted by following the Preferred Reporting Items for Systematic Reviews and Meta-Analyses (PRISMA) statement. The assessment of psychometric properties was performed using the Consensus-based Standards for the Selection of Health Measurement Instruments (COSMIN) methodology for assessing the psychometric properties of patient-report outcome measurement instruments in a systematic review. In total, 109 development and validation studies reporting on the psychometric properties of 15 selected instruments were included: 15 studies reported on the content validity; 25 reported on the criterion validity, construct validity, and reliability; and 69 reported on the responsiveness. The methodological quality of the studies was generally adequate; however, the quality of the studies reporting on content validity was poor overall. The psychometric quality of the instruments’ content validity was generally sufficient, but sufficient quality was determined based on reviewers’ subjective opinions of the content of the instrument itself (items, response options, and instructions) due to the lack of direct evidence from the studies. The psychometric quality of the construct validity, criterion validity, and reliability were overall either indeterminate or not reported because of incomplete or missing data on the psychometric properties. The quality of the responsiveness was also overall either insufficient or not reported. High-quality evidence on all psychometric properties was limited. None of the included instruments can be recommended as the most suitable for use in clinical practice and research. Nine instruments are promising based on the available psychometric evidence, but need additional psychometric evidence before being recommended