Health benefits of religious involvement? The flawed evidence quoted by Dr. Peter Saunders and the Christian Medical Fellowship.

Dr. Peter Saunders, advocating that doctors should proselytise to patients in consultations, trotted out on Radio 4 last week the rather exhausted canard that there is a huge evidence base to suggest a positive relationship between religion and health. Dr. Saunders states: “We are talking about in the vicinity of 1,200 research studies and 400 reviews in peer-reviewed medical journals. 81% show a positive correlation between faith and health; 4% a negative correlation and 15% sit on the fence.” This astonishing number of papers was mentioned by Dr. Harold Koenig in his “Handbook of Religion and Health”, and trotted out again in a (no doubt impartial) report by the Christian Medical Fellowship (CMF).

Over a thousand studies? Really? Others don’t see the statistics in quite this way. The researcher Richard Sloan (Professor of Behavioral Medicine at Columbia University Medical Center) analysed 266 papers identified as relevant to religion and health by the healthcare literature database Medline that appeared over the course of one year. However, only 42 of 266 (17%) were relevant to claims about the effects of religious involvement on health.

Sloan continues: “So proponents of a connection between religion and health are technically correct when they write that there are a great many studies on the topic. But they are wrong when they report that the vast majority demonstrate positive relationships between religious involvement and better health. In fact, the vast majority of these studies have nothing whatsoever to do with the health benefits of religious involvement”.

Let’s have a look at some of these papers supposedly demonstrating the effects of religious faith on health. Of 89 studies on cardiovascular health cited by Koenig and the Christian Medical Fellowship report, 33 were studies on denominational differences in health or compared groups whose religiosity was unknown e.g. comparing cholesterol levels in Seventh day Adventists to age-matched New York city residents. Such studies show us nothing about the effect of religious involvement and health. Eleven studies were reviews of other studies, case reports or “mere descriptions of projects”. Three were published only in abstract form and cannot therefore be critically appraised. This leaves 34 of the original 89 papers that Koenig and the CMF claim to be the basis of the relationship between religious involvement and healthcare. Sloan reviewed these and found 30 had serious methodological flaws or were “so misinterpreted or misrepresented that they cannot possibly be used as evidence for an effect of religion on health”. 

That leaves 4 studies out of the original 89. Extrapolating this to the 1200 cited by Dr. Saunders, only in the region of 50 are likely to actually be about the effect of religious involvement in health, and we don’t know whether they show positive or negative effects.

The quality of many of the papers cited is also dubious in the least. In one of the 81% of 'positive' studies cited by Koenig, the CMF, and Dr. Saunders, eight ‘remote healers’ prayed for patients with high blood pressure, who did indeed have a reduction in systolic blood pressure greater than the control group. No patient details are given. The eight remote healers were four “Science of Mind Practitioners”, one was a Presbyterian Institute Minister, another was a Church of Christ Minister, one was “Director of the Seventh Sense Institute” and another was “a gifted individual whose healing abilities have been verified by both doctors and scientists”. Data was presented only for the patients of four healers who “had the highest number of returned patients”. It is a wonder that the authors managed to publish a study that drops patients from its analysis who are likely to have a negative outcome.

Whilst the Christian Medical Fellowship and Dr. Peter Saunders might consider this a study that supports the discussion of religion with patients, I would hope most doctors would look at such studies and consider that proselytising to patients on the basis of such nonsense would be a serious breach of their duty to treat patients on the basis of evidence (not to mention a breach of General Medical Council guidelines).

References

Sloan R. Blind Faith. New York: St. Martin’s Griffin.

Koenig H, McCullough M, Larson D. Handbook of Religion and Health. New York: OUP.

Bunn A & Randall D. cmf file 44 (2011) - health benefits of Christian faith. Available at http://admin.cmf.org.uk/pdf/cmffiles/44_faith_benefits.pdf