Many aspects of the physical, social and economic environments influence our health behaviours, including what we eat and how active we are. In turn these behaviours influence our risk on non-communicable diseases (NCDs) such as diabetes.

Cross-sectional studies

There are several study designs that we use to investigate these factors – for example, cross-sectional studies of populations  which may reveal correlations between aspects of the environment and behaviours. Taken on their own, such studies cannot provide strong evidence of cause and effect, but they do often reveal possible relationships that are worth further investigation. A simple example is the difference in behaviours, such as physical activity and diet, between urban and rural settings. The work on physical activity in Cameroon demonstrated marked urban rural differences, and has pointed the way for further research.

A simple example is the difference in behaviours, such as physical activity and diet, between urban and rural settings. The work on physical activity in Cameroon demonstrated marked urban rural differences, and has pointed the way for further research.

Cohort studies

A stronger design to capture cause and effect is a cohort study, where a population can be followed over time. Measurements at baseline and at regular follow up points means that we can assess changes in the determinants we have hypothesized to be important, the behaviours we think are associated with them.

Good examples of this approach being used within a UK setting come from the EPIC Norfolk Study and the Fenland study. The intention is to establish similar cohorts with collaborators in low and middle income settings.

Natural experiments

Given the rate of change in many low and middle income countries, a potentially strong study design to identify causal relationships is the natural experiment, where we can study the effects of changes on the environment on particular populations, often comparing them to other populations where the change being studied isn’t happening.

One such ‘natural experiment’ is the introduction of a tax on sugar sweetened beverages. Just such as tax came into force in Barbados on September 1st 2015 and members of CEDAR are assisting the University of the West Indies in evaluating the impact of this measure.

The Unit has considerable expertise across all these approaches in the UK and across Europe. It is working with collaborators to extend the application of this expertise into low and middle income countries.

The Interconnect Project-Global Data for Diabetes and Obesity Research

As well as behavioural influences on our risk of NCDs, our genes have a role to play.

The global variation in risk of type 1 and type 2 diabetes between populations is well described and the pattern suggests a complex but distinct interaction between genetic and environmental risk factors across the life course. Considerable progress has been made in understanding the relationship between lifestyle behavioural factors and the incidence of type 2 diabetes and also in the identification of genetic factors for type 1, type 2 and obesity.

This has made the study of gene-environment interactions possible but most studies of interaction have sought to explain variation within populations. Although studying variation in risk within populations is important, the next critical step is to attempt to move beyond within-population examination of gene-environment interaction to the examination of it between-populations.

InterConnect, a new international research initiative on gene-environment interaction in diabetes and obesity, is enabling this work. Supported by the European Union FP7 grant, the project is changing the way that data are used in population research into the causes of diabetes and obesity. It is supporting decentralized surveillance and data analysis of shared global results that is secure, scalable and sustainable.

• More at www.interconnect-diabetes.eu