Baseline cohort data from HARMONY, a cluster randomised controlled trial of culturally safe domestic violence management in general practice.

Abstract

Background: Domestic violence and abuse (DVA) is a globally prevalent, health damaging problem. In high income countries, migrant /refugee populations from low/middle income countries often consist of young families. DVA is more prevalent, and barriers to disclosure greater among migrant/refugee than native-born families. Consequently, general practice (GP) patient populations are increasingly diverse, but evidence for culturally safe and effective GP management is minimal. The HARMONY study tested a culturally safe DVA intervention to improve GP identification and referral among South Asian families.

Methods: HARMONY was a pragmatic cluster RCT in 19 GP clinics in two regions of Melbourne, Australia. Eligible practices required ≥ 1 South Asian GP; used one of two common software programs; and agreed to have GrHanite™ software on practice computers. This analysis investigated baseline DVA and ethnicity identification in routine electronic GP data. Deidentified aggregated data for female patients aged ≥ 18 with DVA identification, referral, and South Asian ethnicity data were extracted from medical records. Chi Square for comparison of proportions.

Results: Twenty-four clinics were recruited and randomised, but five dropped out due to Covid. Fifty-two percent (50/96) of 19 clinics' staff were South Asian. While 46.7% of female patients (21,220/45,438) were aged 26-45 years, 17.6% (7,874/45,438) were South Asian. There were more South Asian patients in Comparison 20.6% (4,193/20,312) than Intervention 14.7% (3,681/25,126) clinics. South Asian women had less access to Medicare (83% vs 97%) and pensions (13% vs 22%). At baseline, clinicians recorded 0.58% (265/45,438) DVA-affected women. Notably, they identified fewer South Asian (0.38%) (28/7,874) than non-South Asian women experiencing DVA (0.63%) (237/37,564),-0.28% (0.12%-0.43%), p = 0.004. No referrals were identified.

Conclusion: DVA was notably under-identified in these patient populations, but worse among South-Asian female patients. While almost one in six of HARMONY's female population were South Asian, fewer than four in 1000 South Asian women were identified as experiencing DVA. Greater effort to regularly train and to support GP clinics to better identify DVA is vital but even more important in diverse communities to provide culturally safe DVA identification, care and documentation.

Trial registration: ACTRN12618001845224 on 13/11/2018.