Differences in levels of physical activity between White and South Asian populations within a healthcare setting: impact of measurement type in a cross-sectional study

Objective: We investigate differences between White and South Asian (SA) populations in levels of objectively measured and self-reported physical activity. Design: Cross-sectional study. Setting: Leicestershire, UK, 2010-2011. Participants: Baseline data were pooled from two diabetes prevention trials that recruited a total of 4282 participants from primary care with a high risk score for type 2 diabetes. For this study, 2843 White (age=64 +/- 8, female=37%) and 243 SA (age=58 +/- 9, female=34%) participants had complete physical activity data and were included in the analysis. Outcome measures: Moderate-intensity to vigorous-intensity physical activity (MVPA) and walking activity were measured using the International Physical Activity Questionnaire (IPAQ), and a combination of piezoelectric pedometer (NL-800) and accelerometer (Actigraph GT3X) were used to objectively measure physical activity. Results: Compared to White participants, SA participants self-reported less MVPA (30 vs 51 min/day; p<0.001) and walking activity (11 vs 17 min/day; P=0.001). However, there was no difference in objectively measured ambulatory activity (5992 steps/day vs 6157 steps/day; p=0.75) or in time spent in MVPA (18.0 vs 21.5 min/day; p=0.23). Results were largely unaffected when adjusted for age, sex and social deprivation. Compared to accelerometer data, White participants overestimated their time in MVPA by 51 min/day and SA participants by 21 min/day. Conclusions: SA and White groups undertook similar levels of physical activity when measured objectively despite self-reported estimates being around 40% lower in the SA group. This emphasises the limitations of comparing self-reported lifestyle measures across different populations and ethnic groups.