Evaluating interventions in applied science is difficult. The more complex the area where the intervention is to be applied, the more important it is that evaluation studies achieve comparison groups that are balanced for important but unknown factors. This means randomised trials. We report three trials this month in a particularly complex area of medicine - educational interventions. All three studies evaluated seemingly sensible ideas: educational counselling for women who had suffered an unexpected complication at delivery, education to encourage breastfeeding, and an education programme to prevent postnatal depression among women at risk. All three were well designed but none showed any significant benefit from the intervention. Dr Wing Hung Tam and colleagues from Hong Kong (pages 853–859) were clearly disappointed to find no benefit from routine educational counselling for women who had suffered unexpected adverse outcomes during childbirth. They wondered whether this might have been because the control group also got some support, or because only women with relatively mild distress agreed to participate, but the most likely reason remains that there was indeed no benefit from routine counselling at times of stress. Other studies in the puerperium have shown that similar sorts of counselling may even be detrimental, and outside pregnancy there is increasing evidence that routine counselling after such stressful events as a major road traffic accident is harmful. We should still talk to patients who have had a bad pregnancy outcome, but there is no need to go out of our way to divert resources to give them any extra feedback. Dr Labarere's group from Grenoble in France (pages 847–852) studied the effect of an educational intervention to encourage breastfeeding. We already know from other trials that providing one-to-one support to breastfeeding mothers produces a modest increase in breastfeeding success. These authors evaluated whether a 30-minute midwife-delivered educational session had any extra benefit. The answer was no. The disappointed authors cast around for any possible design faults in their trial, but their results are probably correct. They certainly accord with a number of other similar studies of breastfeeding education. Another seemingly sensible innovation was evaluated in the IDEA trial from Australia (pages 842–846). IDEA stands for “Identify women at risk of postnatal depression, Educate her about the risk, and Alert carers to the danger”. What could be more sensible? The trial was well designed, it included a large sample and the result was clear; no significant reduction in the rate of high postnatal depression score at 12 weeks post-partum. These sorts of studies are hugely important. The health service is always short of resources and it is wasteful to move staff from traditional patient care to education and counselling unless there is good evidence that doing so reduces adverse outcomes. They will encourage clinical directors and managers to get doctors and midwives back to caring for patients so they can prevent birth trauma in the first place. When managers do persuade staff back to the labour ward, how can they ensure that the staff really do deal with emergencies properly? One obvious way is to make sure that they all undergo regular obstetric “fire drill” training sessions. It seems to me that this must be a good idea. After all, other emergency services, and the army and airline pilots regularly do emergency drills. Why should obstetrics be different? Black and Brocklehurst (pages 837–841) were more sceptical. They wondered if there was any evidence that such training actually prevented adverse outcomes, or even made staff do the right thing more often. It turns out that there have been no decent quality studies. Such courses may still be effective and most people probably believe this. It is fortunate that, partly as a result of Black and Brocklehurst's research, the Department of Health has recently commissioned a group in Bristol led by Dr Bryony Strachan to conduct a high quality evaluation of individual and team drill in obstetric emergencies.
Read full abstract