Adaptive work in the primary health care response to domestic violence in occupied Palestinian territory: a qualitative evaluation using Extended Normalisation Process Theory.

BACKGROUND A health system response to domestic violence against women is a global priority. However, little is known about how these health system interventions work in low-and-middle-income countries where there are greater structural barriers. Studies have failed to explore how context-intervention interactions affect implementation processes. Healthcare Responding to Violence and Abuse aimed to strengthen the primary healthcare response to domestic violence in occupied Palestinian territory. We explored the adaptive work that participants engaged in to negotiate contextual constraints. METHODS The qualitative study involved 18 participants at two primary health care clinics and included five women patients, seven primary health care providers, two clinic case managers, two Ministry of Health based gender-based violence focal points and two domestic violence trainers. Semi-structured interviews were used to elicit participants' experiences of engaging with HERA, challenges encountered and how these were negotiated. Data were analysed using thematic analysis drawing on Extended Normalisation Process Theory. We collected clinic data on identification and referral of domestic violence cases and training attendance. RESULTS HERA interacted with political, sociocultural and economic aspects of the context in Palestine. The political occupation restricted women's movement and access to support services, whilst the concomitant lack of police protection left providers and women feeling exposed to acts of family retaliation. This was interwoven with cultural values that influenced participants' choices as they negotiated normative structures that reinforce violence against women. Participants engaged in adaptive work to negotiate these challenges and ensure that implementation was safe and workable. Narratives highlight the use of subterfuge, hidden forms of agency, governing behaviours, controls over knowledge and discretionary actions. The care pathway did not work as anticipated, as most women chose not to access external support. An emergent feature of the intervention was the ability of the clinic case managers to improvise their role. CONCLUSIONS Flexible use of ENPT helped to surface practices the providers and women patients engaged in to make HERA workable. The findings have implications for the transferability of evidenced based interventions on health system response to violence against women in diverse contexts, and how HERA can be sustained in the long-term.