Measuring the violence prevention climate: Development and evaluation of the VPC-14

Background: Violence and aggression are common in inpatient mental health hospital settings and cause problems for staff, patients and organisations. An important factor in treatment efficacy is ward atmosphere, and one element of this is the violence prevention climate. Objectives: To develop and test the psychometric properties of a new scale to measure perceptions of the violence prevention climate among staff and patients in mental health inpatient settings. Design: Scale development and cross-sectional validation study. Setting and participants: Three hospital sites within an independent sector provider of secure mental health care. Participants were patients and staff residing in/working on wards in the adult male and female mental health care pathways. Methods: The study was conducted in three stages: scale development, pilot testing and psychometric evaluation. The scale items were developed from systematic literature review, informant interviews (staff) and focus groups (patients) and expert review. The resulting scale was subject to pilot testing with staff and patients (n = 58 and n = 25). The reliability and validity of the scale was examined by administering it to 326 staff and 95 patients. Exploratory factor analysis was used to establish construct validity, and this was further assessed with Rasch modelling. Internal consistency was assessed by calculation of Cronbach's alpha coefficients. Convergent and discriminant validity were measured by comparing results with existing validated instruments. Temporal stability of the items was assessed using test-retest reliability coefficients. Results: The VPC-14 is a 14-item scale demonstrating good psychometric properties. Exploratory factor analysis revealed two subscales, staff actions and patient actions, each demonstrating good internal consistency (Cronbach's alpha.89 and.76). All items demonstrated good temporal stability. Rasch modelling confirmed the unidimensionality of the two subscales, and items demonstrated high construct validity. Moderate correlations were found between subscales of the VPC-14 and the EssenCES, whilst no correlations were found with items in the ACMQ, thus demonstrating good convergent and discriminant validity. Conclusion: The VPC-14 is currently the most robust available measure of the inpatient violence prevention climate. It is quick and easy to administer, considers views of both staff and patients and thus can be introduced as standard practice in a ward setting. Potential uses include tracking the violence prevention climate longitudinally and in evaluation of new policy and procedural interventions.