Conflicts of interest rife in those setting diabetes and cholesterol guidelines
Posted Oct 20 2011 7:39am
One of my aims with this site is to expose health-related misinformation. There’s usually no shortage of material here. Many of the things we’re encouraged to do, like eat margarine, cut back on saturated fat, eat ‘healthy’ wholegrains, have no evidence base for them. So, how come we have this sort of ‘education’ rammed down our throats. Well, these ideas mean big business.
Last Sunday night I was asked by a national radio station here in the UK (Radio 5 live) to comment on a milk that is being touted as healthier than ordinary milk on the basis that it is low in saturated fat. During the interview (available to listen to below) I felt compelled to make two basic points:
Just because something is low in saturated fat does not make it inherently healthy. Methylated spirits is contains no saturated fat at all, but it’s still not good to drink.
There really is no good evidence linking saturated fat to heart disease, and neither has eating less of it been shown to benefit health.
The interviewer suggested many doctors would disagree with my second point. It’s a fair point, but then it would be up to those who disagree to produce the relevant science to back up their claims. The thing is, when you ask for the science, it’s never forthcoming. As an example, take a look at this exchange I had with the heart surgeon (with links to Unilever) who went on a seeming one-man crusade against butter.
What is it that can cause there to be a yawning abyss between official recommendations and what the science shows? Sometimes, the answer lies in conflicts of interest. It’s not uncommon for doctors, scientists and ‘key opinion leaders’ to have financial or other interests in industry. Imagine, say, you were sitting on a panel considering what the upper limit of ‘normal’ cholesterol levels should be. How imagine you were paid consultancy fees by one or more company that makes cholesterol-reducing drugs. Or perhaps drug companies fund your department or professorial chair. Heck, maybe you even own shares (stock) in these companies. Do you think that might influence your judgement at all?
The issue of conflicts of interests was a subject of a recent piece of research published in the British Medical Journal . Researchers in the US looked at the potential conflicts of interest in panel members setting guidelines regarding the management of diabetes or cholesterol.
Of 14 sets of guidelines, five had no declaration of conflicts of interest at all. Taken as a whole, about half (48 per cent) of the 288 panel members reported conflicts of interest at the time the guidelines were published. 73 panel members reported no conflicts of interest, but it turns out this was not true for 8 (11 per cent) of them. Overall, more than half (52 per cent) of panel members had conflicts of interest. In short, conflicts of interest are very common in guideline panels, and these can go undeclared.
This piece of research reminded me of an episode some years ago concerning a group known as the National Cholesterol Education Program (NCEP) – an ‘expert panel’ in the US responsible for setting cholesterol policy. In 2004, the NCEP made recommendations for cholesterol norms to be lowered to the levels recommended by many doctors today. It subsequently transpired that 8 out of 9 members of the panel had financial links with drug companies making statin drugs. By the way, an independent review of the NCEPs guidelines concluded that: ‘we found no high-quality clinical evidence to support current treatment goals for [LDL] cholesterol’ .
If you want to understand how health policy can be out-of-step with the science, it pays to follow the money.
1. Neuman J, et al. Prevalence of financial conflicts of interest among panel members producing clinical practice guidelines in Canada and United States: cross sectional study. BMJ 2011;343:doi:10.1136/bmj.d5621
2. Hayward RA, et al. Narrative review: lack of evidence for recommended low-density lipoprotein treatment targets: a solvable problem. Ann Int Med 2006;145:520-530