Activities and Outlook Churnalism Part 1

Our first churnalism posts were about diet and exercise because journalism (churnalism?), and our society, are obsessed with the impact of these daily activities on our health. That obsession makes some sense. We know that activity and diet do affect your health. People who exercise do better than those who do not and if you begin gorging yourself on heaps of refined sugars (such as those in Krispie Kream donuts, H/T Leana Wen) your health will suffer. Journalism goes astray not by covering diet and exercise but by trying to make the relationship between diet, exercise, and health more sensational than it is. In the next two posts we acknowledge that we all do quite a bit more than exercise and eat and, thus, move on to the sometimes truly absurd relationships that have been proposed between things some of us choose to do and our health. 

What else do we do? You might go to the movies once a week or the shooting range twice a month. You might go on a date, or a long hike with a friend, or spend an afternoon reading outside. You do these things simply because you enjoy them (or maybe to prove that humans do, in fact, have free will). We bring up these types of activities because sometimes a researcher begins to wonder if any of these things have a measurable impact on your health. These studies are often quirky and seem to be catnip for journalists. Not surprisingly, when journalists cover these research reports they often commit the seven deadly sins of churnalism. 

Want to avoid Alzheimer’s disease? Stay in the sauna.

For thousands of years humans have found that brief periods of heat and sweat could be pleasurable, relaxing, and restorative. These practices independently originated in Native American, Asian, African and European cultures. Like dating, hiking or reading, nobody needs justification to go to a sauna -- it feels good. A sauna leaves one drenched in a good sweat and you come out with your skin feeling clean and your cheeks radiating warmth. Many people feel like it relaxes muscles and soothes joints.

But, like so many activities, the intrinsic pleasure is not enough for some scientists (and journalists). A New York Times article in 2016 makes the provocative claim that people who regularly use a sauna are 66% less likely to develop Alzheimer’s disease than those who seldom use a sauna. It is hard to express the utter absurdity of this claim. First recognize that we presently have not found anything that prevents Alzheimer’s. Every intervention that you have heard about preventing this terrible disease (exercise or crossword puzzles) is ineffective when rigorously studied – teaser for an upcoming sensible medicine post. There may be some benefit associated with the things we do to improve cardiovascular health but even these are uncertain and when we prescribe them it is usually with the thought, “well, it will help the heart if not the brain.” The most powerful risk factor for Alzheimer's disease is a genetic factor, the apolipoprotein E genotype. The effect of this risk factor varies from study to study and population to population but, in a representative study, having this genetic predisposition raised the risk of Alzheimer's disease in 75-year-olds from about 3% to 16%. This means that regular use of a sauna is about equivalent to getting rid of the most profound genetic risk factor we have identified. Before going further, it is pretty clear that our churnalist is committing the fourth sin, neglecting plausibility.

The news story is based upon a journal article that found that “compared with men who used a sauna once a week, those who used a sauna four to seven times a week had a 66 percent lower risk for dementia and a 65 percent lower risk for Alzheimer’s disease.” This study used the database of behaviors and health outcomes from the Kuopio Ischaemic Heart Disease study which looks at residents of Eastern Finland. At this point, we are sure you can guess some of the problems, but first, an aside about saunas and Finns.

People from Finland love saunas. Warming in a sauna is a part of the culture. Finland has about 5.3 million people and about 2 million saunas. It is possible for every Finn to be in a sauna at the same time. The same thing can’t be said for Chicago or San Francisco. In fact, when the great Finnish F1 driver Valtteri Bottas was shown relaxing with his trainer in the epic Netiflix documentary Drive to Survive, he was shown in a sauna.

Now to the study. The authors compared people who used the sauna once per week to those who used it nearly daily. If you have been paying attention to this series, you already recognized confounding. You have to suspect that there are some differences between people who sauna daily and those who sauna only weekly. Daily users might have more free time, more wealth, more access to a sauna, and maybe more health conscious. The study attempts to adjust for some measures of health (alcohol intake, smoking, blood pressure, diabetes and other health and behavioral factors) but does not adjust for socioeconomic status. Moreover, the journal article does not provide baseline characteristics of the groups, so we do not know how different regular sauna users and semi-regular users are.

In addition to the issues of confounding, there are issues of generalizability or “external validity.” This is a study done entirely in Finland. What do we make of these results in America where the news story appears? In the US, very few people have any access, let along regular access, to a sauna.

Finally, we get back to the issue of plausibility and the massive size of the effect reported. According to this report, if Finns, who already take saunas, increase their sauna use to 4-7 times a week the majority of Alzeihmer’s would be prevented. Can you imagine what would happen if Americans, who never use a sauna, took up the habit? We might expect the beneficial effect to be even greater!

Finnish people have been using the sauna since 7000 BC. They never needed a study to show sauna use is associated with implausibly large reductions in dementia. Different Finns use saunas differently. Different Finns have different rates of Alzheimer’s disease. The types of Finns who use saunas the most have the lowest rate of the disease. Whether sauna use has anything to do with it remains unknown. Americans, meanwhile, almost never use saunas and have nearly no access to them (and still have enormous carbon footprints), yet we are being led to believe they have extraordinary benefits. The seventh sin of churnalism rears its head. Instead of accepting the easy story, the more interesting question would be exploring who are the types of Finns who live for the sauna.

Nothing Takes the Smile off Your Face Like Dying

The NPR story begins simply, “Good news for the cheery.”3 The story reports a medical journal article that suggests that optimists live longer: “Researchers found that the most optimistic men and women demonstrated, on average, an 11-15% longer lifespan, and had far greater odds of reaching 85 years old, compared to the least optimistic group.” The article’s take home message is clear. If you want to improve your chances of living a long life, think positive thoughts. Now, there’s a feel-good message.

The article does a commendable job of summarizing the underlying research. The authors studied a group of women and a group of men who filled out a survey that measures optimism. They followed them for decades. People who scored as more optimistic on the questionnaire lived longer.

The article continues:

Now, researchers say they can't tell from this study how optimism might affect longevity. Optimistic people might be more motivated to try to maintain good health — such as maintaining a decent diet, engaging in regular exercise and not smoking. They may also be better at regulating stress, Lee [the study’s author] says. The burden of unrelieved stress is well known to have negative effects on health, including an increase in heart disease, liver disease and gastrointestinal problems.

Here the journalist has committed the sin of incuriousness. She has accepted the conclusion offered by the authors and immediately launched into making up an explanation. In doing so she misses what is so interesting about the research and its flaws. When you dig into the research, all of the conclusions begin to fall into doubt.

One key detail is omitted from the news coverage. After filling out the survey, women who died in the first 2 years were excluded from the study, and men who died in the first year were excluded. Researchers explain this choice: these people “were excluded from analyses to mitigate the possibility that imminent health decline affected their reports of optimism.”

That is a big sentence. In other words, the researchers concede something important. How can we know if people were pessimistic and then had a decline in health, or they were pessimistic because their health was in decline? This is reverse causality or the protopathic bias. Although one term sounds cooler, they both mean that there might be an association, and the association might even be cause and effect, but the cause and effect is in the opposite direction of your hypothesis. By removing people who died 2 and 1 year after the survey, the researchers admit that reverse causality is a consideration, and they have proposed a way of controlling for it. However, it is also clear their attempts might be inadequate.  

People with chronic medical conditions do not first begin to deal with those conditions 1-2 years before their deaths, they often deal with them for decades. Because you are dealing with medical problems, like being sick or overweight—even if you do not yet carry formal medical diagnoses—you might be less optimistic than someone lucky enough to be in exceptional health. It might be that good health drives optimism, and not that optimism drives good health. The authors’ adjustment might work perfectly well for that person who is diagnosed with a rapidly fatal disease that affects their outlook on life, but it does not work for that person with the chronic disease that affects them over decades. Realizing this, the speculation about the reasons why optimism translates into longevity seems at least premature if not irrelevant.  

This example demonstrates a twist on our first sin, mistaking association with causation. First the journalist has missed the possibility of reverse causality. This, in itself, is an unforced error since the researchers essentially pointed it out. Then the journalist assumes causation, suggesting that adopting an optimistic outlook could lead to a longer life. 

Of course, the most rigorous way to test the question would be a randomized trial. But just imagining such a trial reveals the absurdity of these conclusions. You would need to randomly assign volunteers to be optimistic or pessimistic. The intervention -- the thing you are prescribing -- is instructing people to try to think positively. Of course, that advice would be summarily dismissed by most people. Even if the subjects accepted their assignments (maybe they are being paid particularly well), could someone actually will themselves to be more optimistic? What is the point of the research itself? Even if we could eliminate reverse causality, and optimistic people really do live longer, the finding is only useful if we believe we can get people to change the hue of their lenses (from blue to rose colored or vice versa). But if you, like us, believe that telling someone to be happy is only slightly less dangerous than telling your adolescent child to smile more, then you will agree that this finding is worthless. Even thinking about this paper drives us towards our inner curmudgeon.

Let’s stop here. Our next and final introductory post about churnalism will continue to explore errors in interpreting, and reporting, articles about our habits and our health.