Beyond greenspace: an ecological study of population general health and indicators of natural environment type and quality

This study was published in the International Journal of Health Geographics and is available open access here.

For small areas across Britain (each containing about 1500 people), we calculated indicators of:

• The % of land covered by various different environment types, such as broadleaf woodland, grassland, built environment and freshwater (data from the UK Land Cover Map)
• The diversity of different environment types within each area
• The relative density of protected/designated areas (such as nature reserves and Sites of Special Scientific Interest)
• Bird Species Richness (using data kindly provided by the British Trust for Ornithology)
• Fresh water quality (based on Water Framework Directive data)

For each area, we also calculated the percentage of the population that said their health was generally ‘good or very good’ or ‘bad or very bad’, based on a question in the 2011 Census. This is quite a simple question, but it relates strongly to other markers of health and wellbeing, and the advantage is that we have the data for the entire population.For this analysis we just focused on data for England, due to differences in the way the census is conducted and data are released across the UK .

After accounting for the fact that the age distribution of the population is different in different places, and also for the level of socio-economic deprivation in each area, the analyses suggested that the different environmental indicators did indeed seem to be differently associated with population health. For example, we found that associations between good health and land cover type were particularly apparent for broadleaf woodland, coastal areas and improved grassland (which is the classification of much of our managed urban park space). The results also suggested that rates of good population heath were higher in areas with a greater diversity of land cover types, with greater bird species richness and in areas with higher density of protected/designated areas. Interestingly, the results suggested that population health was less good in areas with better freshwater quality, contrary to our hypotheses; we speculate on why this might be in the paper, but it’s hard to explain.

In accord with some previous research (e.g. this paper on greenspace, and our paper on coastal proximity) we also found that the associations between different environmental indicators and health varied by area socio-economic status. In general, the strongest associations were indicated in the most deprived areas of England, adding to suggestions that proximity – and access – to good quality natural environments may help ameliorate socio-ecomomic health inequalities. This process has been termed ‘equigenesis‘ – do good environments promote equality?

The study is subject to limitations of the ‘ecological’ study design – for example, we only have data on populations, not individuals, and we only measure everything at a single point in time (the study is cross-sectional). However, it has the advantage of very large numbers, and considers the environment and population across the whole of England. The companion study, investigating mental health in rural areas, overcomes some of the Census study limitations, and comes up with some comparable findings.

We reflect on what these studies mean in a post here.