UKCO2019 Oral Communications (1) (2) (5 abstracts)
1Bradford Institute for Health Research, Bradford, UK; 2University of North Carolina Gillings School of Global Public Health, Chapel Hill, USA; 3University of York, York, UK; 4Leeds Institute of Clinical Trials Research, University of Leeds, Leeds, UK.
Background: Weight perception may differ by ethnicity but has not been examined among migrants to the United Kingdom (UK). Similarly, studies have not assessed whether a figure rating scale (FRS) or perceived weight question (PWQ) are more accurate in evaluating body size perception.
Methods: At 24 months postpartum, women with singleton pregnancies (n=1201) within the Born in Bradford cohort (Bradford, UK) completed the nine-item Stunkard FRS and a 7-category PWQ ranging from very underweight to very overweight. Data from both scales were condensed to four weight categories representing the World Health Organisation cut-offs. Weighted kappa statistics were applied to assess agreement between measured weight categories with the FRS and PWQ. χ2 tests and Fishers exact tests were used to assess differences in misperception by ethnicity and migration and logistic regression was used to assess odds of underestimation.
Results: Thirty percent of white British and 23% of Pakistani-origin women had obesity, with no differences by migration status. Compared to white British women, underestimation was more frequent for Pakistani-origin women, particularly those foreign-born, with higher prevalence of underestimation with the PWQ (18% vs 10%; P<0.001) than FRS (14% vs 6%; P<0.001). Agreement between measured and perceived weight categories were similar when using the FRS or PWQ for white British women (FRS: κ (95%CI): 0.612 (0.56, 0.65); PWQ: 0.614 (0.55, 0.68)). Overall agreement was lower for Pakistani-origin women with higher accuracy for the FRS (0.583 (0.52, 0.63) vs PWQ: 0.472 (0.40, 0.54)). Compared to white British women, odds of underestimation were 155% higher among Pakistani-origin women when using the FRS compared to 114% with the PWQ.
Conclusions: Ethnic differences in weight misperception may influence how women regard the applicability of public health messaging on weight. The measures used to assess misperception can influence the findings, suggesting visual scales rather than questions should be used to assess weight perception.
Keywords: Ethnicity, migration, obesity, misperception
Disclosures: None.