To assess the feasibility of randomising a sufficiently large number of pregnant women who screen positive for intimate partner violence (IPV) to perform a full-scale effectiveness trial. This pilot randomised trial was nested within a cohort of IPV-positive pregnant women who accepted an e-health package in Spain and Denmark. This study was co-designed with patient input using a modified Zelen’s design and a qualitative evaluation. Eligible women were randomised to an intervention (received the e-health package) or to a control (received it with a delay). The primary outcome was the proportion of women screened positive for IPV who consented to be randomised to receive an e-health package with a delay. Twenty-nine of the 51 women in the cohort were randomised. Of these, 12 were allocated to the control group, and 7 (58.3%) consented to receive the e-health package with a delay, i.e., 24.1% of the total randomised and 13.7% of the total cohort. The proportion of randomized women who adhered to treatment was 31.03% (9/29). Complete outcome data were obtained in 7/17 (41.17%) and 2/12 (16.66%) women in the intervention and control groups, respectively. Qualitative interviews indicated that the women perceived the intervention as beneficial and that delaying it was not acceptable. This pilot randomised trial indicated that a future full-scale randomised study would likely be unfeasible based on our a priori criteria to stop the trial. If a future large trial is launched, it should include an internal pilot with strict go/no-go criteria monitored by an independent committee. NCT04978064. E-health Psychological Intervention in Pregnant Women Exposed to Intimate Partner Violence (eIPV). Registered 15 July 2021, https://clinicaltrials.gov/study/NCT04978064 • In the evaluation of e-health interventions for intimate partner violence during pregnancy, it was unclear if those at risk who met the eligibility criteria would consent to randomisation if allocated to the control group and if it would be feasible to collect outcome data. • Of the total randomised and of those allocated to the control group, only around a quarter and a half, respectively, consented to take part in the trial. Complete outcome data collection was feasible in only a third of those randomised. • This pilot randomised trial indicated that a future full-scale randomised trial would likely be unfeasible. If a future large trial is launched, it should include an internal pilot phase with strict go/no-go criteria monitored by an independent trial oversight committee. Observational design should be considered as an alternative to randomised evaluation.
Zapata-Calvente et al. (Sun,) studied this question.