Background
The increasing prevalence of type 2 diabetes mellitus (T2DM) in the Middle East in general, and the UAE in particular, has been attributed to several lifestyle factors, including lack of physical activity.  However, the contribution of physical inactivity to weight gain and diabetes in the UAE has been poorly investigated. 

Aim
Investigate the association of physical inactivity with T2DM in a cohort of Emiratis, and determine whether it is independent of body mass index (BMI). 

Methods
Emirati participants, with and without T2DM, were randomly recruited from Imperial College London Diabetes Centre (ICLDC, Abu Dhabi), in an age and gender stratified manner. T2DM was defined according to American Diabetes Association Criteria. We collected physical activity data of participants using the Global Physical Activity Questionnaire (GPAQ). By following GPAQ guidelines, we calculated participants’ activity levels and categorized them into; Low Physical Activity (LPA); Moderate Physical Activity (MPA); and High Physical Activity (HPA). We used regression analysis to test the relationship between physical inactivity (LPA) with T2DM risk among Emiratis. All statistical analyses were performed in SPSS (Version 19.0). Data are presented as mean ± SD.   

Results
Physical activity data were collected from 966 Emiratis (479 males and 487 females; age = 37.4 ± 6.4 years; BMI = 32 ± 6.3 Kg/m2); 361 with normoglycaemia, 199 with prediabetes, and 406 with T2DM. Participants had a mean age of. The prevalence of LPA in the cohort was 67.6%, with similar rates across the three groups; 67.0%, 66.3%, and 68.7%, among participants with normoglycaemia, prediabetes, and T2DM, respectively. Weighed prevalence of T2DM was higher among participants with LPA (18.3%) compared to those with MPA or HPA combined (16.5%). In addition, weighed prevalence of T2DM in the LPA group was highest among participants with morbid obesity, followed by those with obesity, and with normal BMI (Table 1).  Furthermore, when controlling for BMI group, weighed prevalence of T2DM was similar among participants with LPA compared to those with either MPA or HPA (Table 1). Regression analyses showed no significant association of LPA with T2DM risk, with odds ratio (OR) of 1.07 (95% CI 0.81-1.40, P=0.648). After adjustment for BMI, the OR of LPA for T2DM risk decreases to 0.97 (95% CI 0.733-1.278, P=0.817). 

Discussion
Our data show high rates of physical inactivity irrespective of diabetes or obesity status among Emiratis. Moreover, there was no additive influence of physical inactivity on T2DM risk after correction for adiposity. Whether other risk factors, including adiposity and genetic predisposition can better explain the epidemic of T2DM in this high-risk population remain to be investigated.

Physical activity in an Emirati cohort: relationship with diabetes and obesity