15 May 2020
Our Evidence Manager, Luke Price, explains how we can take a critical view of new studies and make sure we have all the facts. When it comes to helping people to live longer, there are other factors at play than how optimistic we are.
When new research hits the media, it can be all too easy to take a ‘glass half full’ approach and assume that what’s being reported is a true and straight forward representation of the evidence.
A classic example of this appeared in the news last month, with media outlets reporting on new research suggesting that those with a more positive outlook are likely to live longer than those who are less optimistic. This builds on a growing body of evidence attempting to link optimism with health, wellbeing and longevity outcomes.
With a few notable exceptions (for example in The Guardian and on the NHS website), reporting on this study has been wholly uncritical, leading with headline findings such as optimistic people having between 11-15% longer lifespans and also being more likely to live until 85.
However, the findings of this study need to be taken with a pinch of salt for a whole host of reasons.
We must always take a cautious approach to new research.
First off, it’s important to note that the research only shows an association between optimism and a longer life. This is not the same as saying that being more optimistic directly causes a longer life – correlation does not equal causation.
Secondly, the research only focuses on a narrow set of people who are disproportionately white, better off and from two specific occupations (nursing and ex-army) compared to the general population of the USA. This raises an important question: how likely is it that these findings will apply to the much more diverse and less well-off broader population in the USA, let alone in the UK?
Thirdly, with studies of this kind you have to ‘control’ for other factors in order to isolate the impact of one variable (in this case optimism) on the final outcome (longer life). There are a huge range of things beyond optimism that might enable someone to live longer, including socio-economic factors such as gender, race and income, health conditions such as type 2 diabetes or having had a heart attack, and health behaviours such as smoking, diet and physical activity.
The headline findings reported by news outlets do include controls for some socio-economic factors and health conditions, but not for health behaviours. When those are considered the associated 11%-15% increase in lifespan drops to about 9%. Comparing this to the impact that never having been diagnosed with type 2 diabetes and never having had a heart attack has on how much longer we can expect someone to live (17% and 18% respectively) really puts things into perspective.
Finally, the findings reported by many news outlets ignore that when you include health behaviours in the analysis, the increased chance of living to 85 due to being optimistic becomes statistically insignificant (meaning that the difference in numbers could be entirely due to random chance).
As I’ve hopefully shown, taking the ‘glass half full’ approach to evidence is risky. At best, it leads to the sharing of ideas that are firmly in the ‘more research needed’ category. At worst, it can lead us to pursue policies and practice that might achieve very little – or even worsen the problems they’re meant to solve.
That’s why we should always take a cautious approach to new research.
It’s not that such research is ‘bad’, or that all reporting of such research is uncritical. Rather, we must scrutinise it to understand the potential limitations, debate its relevance and explore the implications that it may or may not have for policy and practice.