December 16, 2017

Is your blood pressure too high?

I don't intend to make this set of essays a medical one but I do at times mention findings from medical research that are of particular interest -- and blood pressure is very widely attended to.  Many doctors measure it every time you visit. So it is clearly of some importance.  In particular, high blood pressure is often a precursor to heart attacks and stroke -- which are no fun at all.

So it attracted a lot of controversy recently when the American medical authorities (The American College of Cardiology/American Heart Association (ACC/AHA)) increased the level at which blood pressure becomes a problem.  Previously the maximum "safe" level was 140/90, which happens to be about where my blood pressure levels are. The new "safe" level is 130/80.  Above that level you should receive treatment to bring your pressure down.  That has of course thrown tens of millions into the "diseased" basket -- which produced howls of outrage from many sources -- including me.

A recent article In JAMA by a Dr. Philip Greenland has however hit back and said that the problem is the high level of unhealthy lifestyles among Americans -- including unhealthy diets.  He said that the critics are shooting the messenger -- and his article is overall a good defence of the new guidelines

An article by the ever-skeptical Prof. John Ionnadis is more cautious, however.  He doesn't altogether disagree with the new guidelines but points out that the reseasrch on whih it is based has some rather large flaws if used to guide policy.  It is good reseearch but as a basis for public policy guidelins, extraordinary rigor is required in the research.  The big flaw in the existing research being that it is based on an unrepresentative sample of people who already had heart symptoms.  How far can we generalize from them?  Possibly not at all and probably only weakly,

I would like to add some further criticisms:  Some of the benefits of therapy were tiny.  Adding one extra drug to a conventional regime, for instance, gave an improvement in health outcomes of just one half of one percent (0.54). That could well be illusory.  The authors appear to rely on the finding being statistically significant but, given the large sample size (9361) practically everything is guaranteed to be significant. Statistical significance in that case means nothing.

One also has to be pretty suspicious about the proportion of the population who have ideal cardiovascular health -- from 0.5% in a population of African American individuals to 12% in workers in a South Florida health care organization. One understands that Africans do tend to die younger but saying that 99.5% of that population has some degree of risk seems extreme -- perhaps extreme enough for the finding to be ignored.

A final difficulty I see lies principally with Dr Greenland's article.  He stresses the importance of a "healthy" diet in getting heart attacks down.  That's a very conventional view  but is it right?  And if it is right, how do we know our diet is healthy?  Up until a couple of years ago fat was regarded as bad and sugar as safe, but that has now been stood on its head. The opposite is now the accepted wisdom.  So color me skeptical.  If there is such a thing as a healthy diet, I doubt if anyone knows what it is.  So what do I eat to avoid a heart attack.  I don't think Dr Greenland or anyone else knows.

So I come down to the conclusion by Prof. Ioannidis, who states, "The ability to generalize these gains across diverse settings in clinical practice and to use limited resources wisely remains an open challenge." In other words, we don't know when someone would be helped by the new guidelines. They are a long way from gospel.




Go to John Ray's Main academic menu
Go to Menu of longer writings
Go to John Ray's basic home page
Go to John Ray's pictorial Home Page
Go to Selected pictures from John Ray's blogs