Following on from the recent observational study reporting benefits for coffee drinkers, a further research report from Harvard Medical School has found its way into the collective media consciousness.

In "Physical Activity after Diagnosis and Risk of Prostate Cancer Progression: Data from the Cancer of the Prostate Strategic Urologic Research Endeavor" (Cancer Research, 24 May 2011) Richman et al report "Men who walked briskly for 3 hours per week or more had a 57% lower rate of progression than men who walked at an easy pace for less than 3 hours per week."

This is a significant claim.

The many limitations of this research

The basic problem with this type of research is that the more analyses you do, the more significant results you will turn up. 

There is no way of telling whether these associations occurred by chance, because of a causal link between vigorous walking and prostate cancer progression, or because of some other unknown factor that is connected both to walking and to disease progression.

Sample size

There is a further problem in that they only studied around 1,000 men, and very few of them experienced disease progression.  Rare outcomes are inherently far more susceptible to statistical chance and require larger sample sizes to compensate. 

We can clearly see that this affected the findings of this study from the very wide confidence intervals around the estimates of supposed benefit.

To put this in context, the main estimate of benefit was a hazard ratio of 0.43.  This represents the "57%" lower chance of disease progression in the vigorous walking group relative to the risk of progression in less vigorous walkers.

However, the confidence interval (0.21-0.91) tells us that the true value probably lies somewhere between 9% and 79%.  This is a very wide range, and one that should make us cautious about the results.

Reverse causation

This is compounded by the fact that, in this instance, it is entirely plausible that progression of prostate disease - along with treatment - would lead to less walking activity.  This is known as "reverse causation". 

The researchers tried to account for this by looking at disease progression instead of mortality, but it still seems possible to be a problem.

So what can we take from this study?

In conclusion, we can't say for certain that vigorous walking will reduce a man's risk of prostate cancer progression.  However, there is some encouragement here to suggest there may be specific benefits on top of the general health benefits of taking exercise.

Liz Woolf, UKPL Steering Group member, commented:  "Just to be safe, it is important that people with cancer check with their doctor before taking up any new form of exercise."

Find out more

• NHS Choices Behind the Headlines report
• Get the abstract of the original research
• Read the analysis by Cancer Research UK