Validity and Reliability of the Spanish Version of the "Abuse Assessment Screen" among Pregnant Women

Objective: To assess the reliability, accuracy, and construct validity of the Spanish Abuse Assessment Screen (AAS) among pregnant women using the Spanish version of Index of Spouse Abuse (ISA) as a reference standard. Design and Sample: Cross-sectional survey. A total of 1,329 pregnant women were selected in nine primary care centers during 2008-2009. Measures: The Spanish ISA was self-administered first, followed by the AAS, administered by the midwife. Sensitivity, specificity, and predictive values of the Spanish AAS were compared with the Spanish version of the ISA as a reference standard. Results: Percentage of agreement between initial and retest administration of the Spanish AAS was high, from 96.4% to 100%. Specificity was for all types of abuse above 97%, but sensitivity values were much lower (33.3%, 22.9%, 6.9%, for severe physical abuse, minor psychological abuse, and minor physical abuse, respectively). The sensitivity of severe psychological abuse was perfect. Construct validity was good. Conclusion: The Spanish AAS has good test-retest reliability, specificity, and construct validity. The sensitivity was good for severe psychological abuse and moderate for severe physical abuse. Further formal psychometric evaluations, in other languages from countries with low prevalence of abuse, remains a priority for clinical and research efforts in pregnancy domestic violence screening.