I was recently trying to make sense of Patrick Holford’s support for homocysteine testing and his remarkable confidence in his own guidance: How to Eliminate Your Risk of Ever Having a Heart Attack. One of the irritations about reading Holford’s material is that it’s like being pecked by a hummingbird, darting from one assertion to another so rapidly that it’s difficult to get a “Huh?” in edgewise. It is usually very difficult to review Holford’s claims because he doesn’t provide references in his web articles although his text is dense with implied footnotes, scientific papers and reference texts. For this piece, I shall concentrate on Holford’s inappropriate interpretation of official statistics for cardiovascular disease and some simple mathematical errors.
Patrick Holford doesn’t seem to write in a rhetorical style other than one modelled after Cassandra and her prophecies of doom. Fortunately, Patrick Holford has a solution for most modern ills that he warns us about, and, usefully, it is typically available from his associated site, Health Products for Life. A recent example of this is, Are you being electrified? in which Holford guides us through the horrors of being immersed in electromagnetic radiation and advises us on various protective measures including the purchase of an Electrosmog detector that he sells.
Holford assures us that he does his research. Not only did he run an electrosmog detector around his own home:
[n]ext, I did additional research. I read well-respected medical journals. I looked into safety regulations in a wide variety of countries, including the UK, the United States, Russia, Sweden, Switzerland and New Zealand. I also delved into the technological and biotech publications.
After all that impressive array of research into electromagnetic radiation, it is a little disappointing that Holford only provided 10 references but that is better than his usual practice.
To return to the issue of homocysteine testing. I should probably state upfront that I don’t believe that it is currently possible to eliminate your risk of having a heart attack; it may be possible and even practical to reduce it but not to eliminate it. Socio-economic factors are widely reported to have a significant role in cardiovascular disease; as do race, gender, age, smoking, co-morbid diseases such as diabetes and family history (the British Heart Foundation statistics report gives a good overview of some of these factors). Research scientists and epidemiologists have worked very hard to discover whether there are dietary or pharmacological interventions that can overcome the impact of all of these contributory factors, but so far, there are not.
Holford has a hummingbird approach to proving his alarming claim that “[h]eart attacks and strokes have reached epidemic proportions”. He darts here and there, throwing in assertions from ancient history, the late nineteenth century and modern cultural comparisons, all without citing any references.
Britons, for example, have approximately nine times as much heart disease by middle age as the Japanese, although this difference is starting to narrow as the Japanese adopt more Western diets and lifestyles. But it wasn’t always this way.
Autopsies performed on mummified Egyptians who died in 3000 BC show signs of deposits in the arteries but no actual blockages that would result in a stroke or heart attack. Fast forward a few thousand years to 1890, where according to American health records, the incidence per 100,000 people of heart attacks was near zero. By 1970, however, that incidence had risen to 340 deaths per 100,000.
It’s a tremendous amount of work to track down all of these vague references (some of which I suspected to be wrong) so I concentrated on the assertions that I could readily check. E.g.:
[a]lmost one in two men and women die from heart attacks or strokes…In Britain, a quarter of a million people die prematurely every year from heart attack or stroke”.
No, they don’t; both of these statements are wrong. I know this because I read the same British Heart Foundation statistics report on which these assertions are probably based. Chapter 1 (pdf) of the British Heart Foundation (BHF) report summarises the mortality statistics.
Diseases of the heart and circulatory system (cardiovascular disease or CVD) are the main cause of death in the UK and account for just over 216,000 deaths in 2004. More than 1 in 3 people (37%) die of CVD…
CVD is one of the main cause of premature death in the UK (death before the age of 75). 32% of premature deaths in men and 24% of premature deaths in women are from CVD…CVD caused just under 60,000 premature deaths in the UK in 2004.
It’s a considerable stretch to interpret 37% as “[a]lmost one in two”. The total number of deaths from CVD is a “quarter of a million”; the number of premature deaths that involve CVD is “just under 60,000”. There is no room for complacency, but the figures are lower than the alarming ones that Holford stated.
It’s bad enough when Holford misreports figures that are clearly stated in the BHF text; it becomes more embarrassing/awkward when he seems to interpret some of the data tables. E.g., Holford writes, “a quarter of all deaths from heart attack occur in people under the age of 65”. No, they don’t; not according to my calculations. At least, not according to Table 1.2 of the BHF report and some basic arithmetic which should surely be well within the attention grasp and number-handling skills of someone awash with the fish oil capsules that Holford recommends on a regular basis. Heart attacks are typically classified as coronary heart disease or CHD: I have reproduced these data from Table 1.2 of the report.
Deaths from CHD by sex and age, 2004, United Kingdom
||75 & over
I don’t take fish oil supplements but it seems to me you can add the total number of deaths in the final two columns and subtract this new figure from the number of total deaths to calculate the number of deaths from CHD in the under-65s. From there, it should be a straightforward calculation to estimate whether it is accurate to say that “a quarter of all deaths from heart attack occur in people under the age of 65”.
The number of CHD deaths in people over the age of 65:
We need to subtract this number from the total number of deaths from CHD to obtain the number of people who died under the age of 65:
It doesn’t even need arithmetic skills, just estimation, to know that if 14,244 deaths from CHD occur in the under-65s, this is not even close to “a quarter” of 105,842. The percentage is 13.5.
Holford is wrong when he claims that, “a quarter of all deaths from heart attack occur in people under the age of 65”: according to the BHF figures, an accurate estimate of the number is 13.5%. I did Holford the courtesy of running various permutations of the numbers such as men only (19.5%), or calculating the numbers for all deaths from diseases of the circulatory system rather than just CHD (11%) but none of the numbers come out at anything approaching “a quarter”.
I shall spare you the reading around that I did on the issue of autopsies in egyptian mummies but I will say that only the very wealthy could afford the elaborate funeral rites of mummification, so this might skew the results. Further, it seems as if life expectancy was somewhere between the ages of 30-40 and it is possible that tuberculosis and various other endemic diseases killed many of these people. This post is not the appropriate place to counter the implication that it is only people with ‘blockages’ who experience fatal heart attacks or strokes: consider coronary artery spasm which can occur in arteries that appear normal; or sudden cardiac arrest that arises from an unsuspected abnormality.
I scouted about for information about heart disease and late 19th century America. Last year, the New York Times published an overview of some fascinating research into the history of some chronic diseases: So Big and Healthy Grandpa Wouldn’t Even Know You.
The biggest surprise emerging from the new studies is that many chronic ailments like heart disease, lung disease and arthritis are occurring an average of 10 to 25 years later than they used to. There is also less disability among older people today, according to a federal study that directly measures it. And that is not just because medical treatments like cataract surgery keep people functioning. Human bodies are simply not breaking down the way they did before…
Scientists used to say that the reason people are living so long these days is that medicine is keeping them alive, though debilitated. But studies like one Dr. Fogel directs, of Union Army veterans, have led many to rethink that notion.
The study involves a random sample of about 50,000 Union Army veterans. Dr. Fogel compared those men, the first generation to reach age 65 in the 20th century, with people born more recently…
Instead of inferring health from causes of death on death certificates, Dr. Fogel and his colleagues looked at health throughout life. They used the daily military history of each regiment in which each veteran served, which showed who was sick and for how long; census manuscripts; public health records; pension records; doctors’ certificates showing the results of periodic examinations of the pensioners; and death certificates.
They discovered that almost everyone of the Civil War generation was plagued by life-sapping illnesses, suffering for decades. And these were not some unusual subset of American men — 65 percent of the male population ages 18 to 25 signed up to serve in the Union Army. “They presumably thought they were fit enough to serve,” Dr. Fogel said…
After the war ended, as the veterans entered middle age, they were rarely spared chronic ailments…
Eighty percent had heart disease by the time they were 60, compared with less than 50 percent today.
The studies by Fogel and others do not directly contradict Holford’s assertions about heart disease in the late nineteenth century but they do highlight some relevant facts that might question the evidence base for his assertions.
In another post, I shall address the issue of whether your H-score and homocysteine tests are proxy measures for CVD risk or established causal factors in disease. I’ll look at the intervention trials that have experimented with lowering homocysteine levels through supplementation etc. and whether they have succeeded in lowering those levels, but not the incidence of clinical disease. Ultimately, it is more relevant to consider whether you have decreased the incidence of disease or reduced its severity rather than concentrating on a particular measurement. When you alter a particular measurement such as a cholesterol or homocysteine level, but you don’t reduce the incidence of (say) heart attacks or strokes, then it raises the suspicion that you have identified a false surrogate endpoint.
Holford might “read well-respected journals”, but when one considers the mangling of the straightforward BHF report, he doesn’t always seem to manage to distil or communicate the facts in an accurate and transparent way. This might lead a reader to consider whether Holford’s summary of the research evidence in support of homocysteine tests and supplementation is an accurate reflection of the literature and sufficiently strong to support his recommendations or his claims that you can “eliminate your risk of ever having a heart attack”.