I just spent a fascinating couple of days in Cambridge at the ISBNPA satellite meeting More than the sum of the parts? Integration of individual and environmental approaches to changing population-level physical activity behaviour. It was a good opportunity to think about the kinds of evidence we need to be generating in order to inform policy and practice in this area.

As Liz Richardson of CRESH articulated clearly in a recent blog post, the danger of evidence-free (or at least, evidence-lite) policy is that if the benefits of ‘greenspace’ for public health are overstated, or oversimplified, then if they don’t eventually transpire we run the risk of the baby being thrown out with the bathwater. On the existing evidence, it seems pretty likely that there are some benefits of exposure to/access to natural spaces for public health and wellbeing, but that the situation is too complex to simply say ‘greenspaces are good for people’ without further specification. There’s a fine balance to be struck between the old non-committal ‘more research is needed’, and just getting on with it because the evidence is ‘good enough’.

So, the evidence we generate needs to be robust, sensitive to the complexities of these relationships, and ideally not packed with caveats that mean informing policy and practice become a massive headache. Evaluation of complex interventions (MRC guidance here) is prominent in public health research these days, and rightly so; we need to figure out how to generate rigorous evidence on interventions that don’t fit the typical Randomised Controlled Trial (RCT) mould. A great example of an evaluation of an environmental intervention that I learned about this week is that of the Connswater Community Greenway in Belfast. But as Rich Mitchell pointed out in another CRESH blog post a while ago, intervention research is not the be all and end all of generating the evidence we need. One critical issue that’s hard to deal with in intervention research, is that we may be looking for (or specifically interested in) long-term impacts of environmental change that are just not detectable in a typical intervention evaluation follow up of one or two years.

Here at the European Centre, we’ve been thinking a lot about how we do environment and health research in a truly interdisciplinary fashion (paper here). It seems fundamental to this research issue that we use a mixture of methods and approaches, and we do our best to integrate these. Amongst others we do lab studies, detailed survey research, large scale population-based studies, in-depth qualitative/mixed-method studies,  and importantly figuring out how to synthesise the complex evidence base in systematic reviews and how all this evidence gets communicated. Beyond Greenspace will hopefully turn out to be a useful part of this puzzle.