Brain Injury in Domestic Abuse Related Death Reviews: Missed opportunities: a retrospective analysis of DARDR reviews

Domestic abuse related death reviews record events capable of injuring the brain for most of the people who died. Agencies almost never considered brain injury as a result, and what the reviews record is associated with abuse towards a person and not with abuse by them. A coded field in the national index would make that record usable across the whole library. A stratified sample of 131 published reviews was read in full and coded to a person-level register of 681 people. Of these, 236 have at least one recorded occasion before the fatal incident on which an event capable of injuring the brain is described. Among the 154 people who died, the proportion is 80 per cent. Brain injury was considered by an agency or by the review for 12 of the 681. For 134 of the 236, the review states that no agency considered it. Among the 336 people given a role in the abuse, two or more recorded occasions is associated with every category of abuse recorded towards the person. The odds ratios run from 8.60 (95% CI 4.95 to 14.93) for four or more abuse entries to 2.81 (95% CI 1.40 to 5.64) for economic abuse. It shows no association with abuse recorded by the person towards someone else (1.07, 95% CI 0.66 to 1.73). Stratifying by the number of timeline events in each review moves the largest estimate from 8.60 to 8.28 and leaves the null at 1.06. The Home Office library has no facet for brain injury or learning disability, and codes disability only for the person who died. None of the 535 works in the Threshold evidence corpus examines death reviews as a source. The findings are machine-coded and evidence-linked, not human-validated. They count what reviews wrote down and should be read as a floor. Published by Oxon Advisory Press, Didcot, as a technical note. Edition 1.0, September 2026. ISBN 978-1-0669550-1-5 (PDF).