One of the big problems with science is that you do the experiment and it doesn’t support your thesis. What to do? Slice and dice the data to make it look better, of course.
That’s just what happened in a study that purports to show that acupuncture alleviates depression in pregnancy. This would be an important finding if it were true, since some pregnant women suffer from depression and are wary of the risks of taking antidepressants in pregnancy.
In Acupuncture for Depression During Pregnancy, the authors set out to investigate whether acupuncture could alleviate symptoms of depression:
Qualified participants were randomized to acupuncture specific for depression, acupuncture not specific for depression, or prenatal massage…
Acupuncture specific for depression was tailored individually to address each participant’s depression-related patterns of disharmony according to the principles of traditional Chinese medicine and following a published standardized treatment manual.26 Acupuncture not specific for depression was also standardized and needles were inserted in real acupuncture points that did not address depression-relevant patterns of disharmony according to traditional Chinese medicine…
Swedish massage was provided in a standardized fashion and included effleurage and pétrissage strokes…
There were 141 participants in the study, randomized as follows: 46 received acupuncture specific for depression, 48 received acupuncture not specific for depression and 40 received massage. Response to treatment was defined as a more than 50% reduction on the Hamilton Rating Scale for Depression, as well as a HRDS score between 7-14. In other words, the patients were still depressed, but their symptoms had decreased markedly. Remission was defined as achieving an HRDS score of less than 7.
Did acupuncture specific for depression result in remission of depression. The authors acknowledge that it did not.
Remission rates were not significantly different between the group receiving acupuncture specific for depression (16 of 46, 34.8%) and the combined control groups (26 of 88, 29.5%; number needed to treat 19.1); the remission rates for the groups receiving acupuncture not specific for depression and prenatal massage did not differ (11 of 40, 27.5%; and 15 of 48, 31.2%, respectively; number needed to treat 26.7). Acupuncture specific for depression remission rates did not differ from either the group receiving acupuncture not specific for depression (P=.47; number needed to treat 13.7) or prenatal massage (P=.72; number needed to treat 28.3).
But did acupuncture for depression produce an improvement in symptoms (a response)? The authors claim that it did, stating:
Response rates were significantly higher for the group receiving acupuncture specific for depression (29 of 46, 63.0%) than for the combined control groups (39 of 88, 44.3%; P<.04)... The control interventions did not differ from each other, 15 of 40 (37.5%) for the group receiving acupuncture not specific for depression and 24 of 48 (50%) for the group receiving prenatal massage; P=.24)...
But on closer inspection, the data does not support the authors’ claims. Yes, the data show that acupuncture specific for depression produced a statistically significant higher rate of response than non-specific acupuncture (63% vs. 37%). And the data also show that acupuncture for depression produced a statistically significant higher rate of response than found in a group created by combining non-specific acupuncture + massage (63% vs, 44%). But, the difference between the response rate for acupuncture specific for depression and massage was not statistically significant (63% vs. 50%; P=0.20).
In other words, acupuncture specific for depression produced an effect that was not different from that of one of the controls. But by creating a group that combined non-specific acupuncture (ineffective) with massage (effective), they were able to create a composite that was ineffective, and then they compared the experimental group with the composite. But that doesn’t change the fact that acupuncture for depression was no better than massage.
Have the authors shown that acupuncture specific for depression is effective for treating depression in pregnant women? I don’t think so. If it offers no advantage over massage, we can forget about acupuncture and simply offer depressed pregnant women a day at the spa. It is no less effective and probably far more enjoyable.
