Thirty-three studies (reported in 35 publications) of observational design were included (the number of participants was not reported).
Educational programmes for students entering health care professions (6 articles).
Two studies reported skills, in addition to knowledge and attitudes, as an outcome measure. In one of these, two groups received training (of 2 days and 3.5 days) and were compared to historical controls who had not received training. Participants on the longer course had sustained improvements in knowledge and attitudes at 6 months, and both trained groups had higher scores than the control group on a simulated performance measure at 9 months. In the second study, a pre- and post-test comparison after a half-day workshop found significant improvement in knowledge and attitudes at 6 months, but no change in detection rates of domestic violence. Two studies reported positive participant attitudes toward training. The remaining two studies did not report on outcome measures. The included studies were of short training programmes with little evaluation of clinical performance measures as outcomes.
Educational programmes for practitioners in health care settings (4 articles referring to 3 training programmes).
One study reported improvements in professional knowledge and comfort levels. There were increased rates of screening for domestic violence and in safety assessments, but the majority of women remained unscreened and there was no statistically-significant increase in the proportion of domestic violence cases identified. One study reported positive participant attitudes toward training. The remaining study reported low attendance at training and no improvement in domestic violence documentation rates. The included studies were of short training programmes with limited follow-up.
Education of health care professionals coupled with health care system changes (9 articles referring to 8 training programmes).
One of the studies carried out a 1-year and 8-year follow-up of a domestic violence screening protocol in an emergency department in conjunction with health care provider training. The detection rates increased from a baseline of 5.6% to 30% at the 1-year follow-up and decreased to 7.7% at the 8-year follow-up. There was an increase in reported screening, increased knowledge reported by health care providers, and an increase in the number of abused women citing the media or health care providers as referral sources in one study (data not reported). One study reported that a triage screening protocol was not successful, with poor protocol compliance by emergency department staff. One study reported an increase in domestic violence detection rates at 2 and 3 months (data not reported) compared to baseline, with the health system changes being more effective than staff training. One study of an intervention involving three counselling sessions for pregnant abused women, compared with postpartum women without counselling, found no between-group differences in resource use of contact with the police. One study of the implementation of a domestic violence protocol three times during pregnancy, in conjunction with maternity care coordinator training, reported an increase in detection rates from 5.4 to 16.2% and improved documentation on the perpetrator. One study of an emergency department-based advocacy programme reported an increase in domestic violence detection rates from 1.2 to 1.7%. The remaining study reported no increase in detection rates or referrals. The included studies were of limited educational programmes with no consistent reinforcement.
Evaluation of screening protocols in studies giving no information on training of providers (16 studies).
The findings of this group of studies were inconsistent. While some of the studies reported improvements in domestic violence detection rates, none of the studies reported on the long-term effectiveness of the screening protocols or on the outcome for abused women. No information was available in relation to health care professional training for these studies. The long-term effectiveness of screening protocols and the impact of screening and counselling on abused women were not evaluated.