Last updated May 16, 2017 at 10:13 am
Cycle commute cuts cardiac conniptions! Research on regular riding brings good tidings!
These are the sorts of by-lines that appear in the media when a health study is published, and often further scrutiny of the original research points to a bit of journalistic licence inflating the link between activity and benefit. So last week I once again put on my sceptical hat when the University of Glasgow published a study on methods of commuting alongside reductions in cancer and cardiovascular disease (CVD) which made headlines.
Now, just to make one thing clear: I desperately want cycling to prevent me from getting CVD or cancer. If it can, I would be indestructible, because cycling is heavily ingrained in my life. It has been a significant mode of transport since I was at primary school, and even now I use my bike to get to work when I don’t have to ferry children to their various institutions.
The coincidence of this news breaking last week was too much for me to ignore. One of my 2-year-old girls learnt how to pedal a bike which had me daydreaming about trading prams for family bike rides. The vision was like a rainbow shining through the tears (a tears-bow?) brought on by my bike having recently suffered a terminal injury. While rebuilding my new bike, I have been soaking up said tears with the pages of “Bike Nation: How cycling can save the world”, which includes an opening chapter entitled “The miracle pill: bikes make everyone more healthy”. While I am quietly confident that the author was suggesting that health benefits are experienced by the people actually cycling, and not literally ‘everyone’, the essence mirrored the findings of this recent study – more cycling results in a population with lower rates of CVD and of certain cancers.
Where I start to get a bit edgy with health studies is how they are interpreted and reported in the media. I understand media outlets need to generate hype to bring in readers. But by the time correlations found in studies make it to the headlines, they imply causal relationships. Just because the cyclists had lower incidences of certain diseases, doesn’t necessarily mean that cycling was the cause. It may be that people that are more likely to cycle are also more health conscious and have much better diets. So we say that exercise and positive health outcomes are correlated or associated (or whatever term you would like to use), and not that exercise directly caused reductions in cancer or CVD mortality.
For a very over-the-top lesson in correlation vs causation, I draw your attention to the link between the number of letters in the Scripps National Spelling Bee and death by venomous spiders. Yes, these two things appear to correlate, but there is no causal link. We don’t go to high spider alert when spelling bees use longer words.
Image credit: Tyler Vigen, http://www.tylervigen.com/spurious-correlations
Delving into health studies can be quite tricky, but there are a few tips and tricks for quickly finding a reason to doubt them (many of these are summarised here).
The first thing I looked at was the sample size because, in general, the bigger the sample the more likely you are of representing the entire population. Albeit overly simplified to prove my point, it is like tossing a coin three times and getting heads every single time, then concluding that this coin only gives heads. This study used a quarter of a million people which is great, but if I told you the result of 250,000 coin tosses was 80% heads, your first assumption would probably be that some of the coins used had heads on both sides, right? And similarly, the nature of the subjects in this study, how they were selected, and how their conditions were altered and measured should be checked.
There are always confounding factors in these types of study. The authors do not have control over the group, so they cannot make sure their subjects’ exercise is regimented, they eat the same food or drink the same amount of alcohol. They can take this into account when they are looking at their data, but there are always related factors that can’t be removed or ignored. In the original paper, the authors show how many subjects from each group had family histories of disease, their depravity index (how rich or poor they are), body mass index, ethnicity, smoking status, and many other factors that made me happy enough that they had tried to eliminate, or at least identify, confounding factors.
The authors of this study should be praised because as well as openly highlighting the confounding factors, they also downplay the link between cycling and the health outcomes as an association, and not necessarily causal, and that it isn’t specifically cycling that produces benefits. Along with cyclists, the group that walked long distances (more than 6 miles per week) also had fewer incidences of, or deaths from, cancer or CVD than the inactive control group. The authors suggest that cycling is most likely performed at a greater intensity and for a longer period, essentially a dose dependent response of activity to the risk of cancer and CVD.
The take-home message from this study is that that any level of activity is linked with health benefits and this increases as the level of activity goes up. I don’t think this is going to be news to anyone, especially when it comes to heart disease. The surprise for me was the reductions in cancer rates, but then again sedentary lifestyles have been linked to higher rates of certain cancers. So, while the outcomes might not be surprising, they are important and required. Often changes in policy, such as government setting aside money for separated pedestrian and cycling paths, needs to have quantifiable benefits so the infrastructure spending can be compared with, and offset by, a larger health saving.
I think the author of ‘Bike Nation’ has a point when advocating specifically for cycling, that often this exercise is performed as a mode of transport and can easily be incorporated into everyday life. Whereas activities like going to the gym or swimming, we have to go out of our way to do; they are additional drains on our time. So, while cycling may not make me indestructible, by swapping inactivity for activity I am giving myself the best chances of avoiding premature death, a goal I try to achieve every day.
Images courtesy of the author, and public domain.