My grandfather used to say “Exercise? I get plenty of exercise – pushing myself back from the dinner table every night!” Of course, he walked almost daily on the Coney Island boardwalk in the middle of the winter and lived well into his 90s, so this was really just a way to get a laugh from us.
In this week’s New England Journal of Medicine (Jan 21, 2021)*, you can find the International “polypill” study known as “TIPS-3”, where half of 5,700 people with elevated risk factors for cardiovascular disease were randomized for 6 years, to either receive a “cocktail” pill that contained a statin, a beta-blocker, a thiazide diuretic, and another anti-hypertensive, or a placebo. They were also randomized to aspirin vs. no aspirin. They were all healthy but scored badly on a math algorithm (INTERHEART risk formula) based on their age, blood pressure, and other factors. The primary outcome for the study was defined as a composite of all cardiovascular events (including death from stroke, cardiovascular causes, heart attack, heart failure, and a few other events). Additionally, they tabulated death from any cause, cancer, or recurrence of any cardiovascular event. There is a mass of data in the publication, but at first glance, it looks good: the survival curves diverge at 1 year and remain separated. For the primary outcome – (death from cardiovascular causes, heart attack, stroke, heart failure, sudden death) there were 126 events in the polypill group (4.4%) and 157 (5.5%) in the placebo group… a very minor benefit. In the “death from any cause” column, the benefit shrinks further. For the polypill group vs. the placebo group, there were 5.2% vs. 5.7% deaths. The authors note that drug delivery was imperfect due to the COVID pandemic and supply line disruption. A half-percent survival advantage is not a home run, especially for a four-medication cocktail.
The polypill model was conceived in 2003 and mathematically predicted to have an 80% lower risk of stroke and heart attack, not the 0.5-1% that was seen. If all-cause mortality is the benchmark, and the benefit is 0.5%, for every 1,000 people taking the polypill for 5 years, we would expect to save 5 people. In epidemiology, we use the term “NNT” (number needed to treat) meaning the number of medicated patients needed to help 1. The NNT, in this case, is 200. That means that 199 out of 200 polypill takers do not benefit.
Despite the large problems of distribution, cost, inconvenience, and relatively disappointing results (after all, they used FOUR of our best medications for cardiovascular disease reduction) the editorial in the same issue concludes “These challenges deserve attention: the findings of TIPS-3 support a contention that polypills will be central to any comprehensive strategy to improve global cardiovascular health.”
So, let’s talk about walking, rather than pill-taking. To counter this splashy, undeserved enthusiasm, there are dozens of studies I could cite, but let’s start with Hakim et al, NEJM, Jan 8, 1998. This was a retrospective study of seven-hundred elderly non-smoking men who were enrolled in the Honolulu Heart Study. The men who walked over a mile a day had a 23.8% mortality over 12 years, vs. the men who walked less, who had a 40.5% mortality. The amount of walking was directly proportional to the improvement in mortality rates. This was retrospective… so maybe the “walkers” were already healthier? In another study (BMJ), 3,200 older adults 65 and up followed for 5 years, walking speed correlated with cardiovascular deaths, with the slower third walking speed having a 44% increased risk of death compared with the upper third. The last study to highlight (out of hundreds!) taken from the NEJM, was also prospective, looking at 72,000 nurses between 40 and 65 years-old with no cardiovascular disease. They were followed for 8 years for both total “energy output” into exercise and just walking. The “walkers” who walked 3 hours per week had a 35% lower risk of cardiovascular event.
This is mainstream medicine’s best offer to us, the aging U.S. population. They will argue that if we all take a polypill, all 70 million baby boomers, for 5 years, we will avert .35 million deaths. But if every baby-boomer could walk for 3 hours a week, the data suggests we could save 28 million deaths. I think I’ll walk, thanks.
Slow walking speed and cardiovascular death in well functioning older adults: prospective cohort study: https://www.bmj.com/content/339/bmj.b4460
A Prospective Study of Walking as Compared with Vigorous Exercise in the Prevention of Coronary Heart Disease in Women: https://www.nejm.org/doi/full/10.1056/NEJM199908263410904