Obesity and its association with sociodemographic factors, health behaviours and health status among Aboriginal and non-Aboriginal adults in New South Wales, Australia

Abstract

Background: High body mass index (BMI) is the second leading contributor to Australia's burden of disease and is particularly prevalent among Aboriginal peoples. This paper aims to provide insight into factors relating to obesity among Aboriginal adults and Aboriginal-non-Aboriginal differences.

Methods: Cross-sectional analysis of data from the 45 and Up Study, comparing obesity (BMI ≥30 kg/m²) prevalence and risk factors among 1515 Aboriginal and 213 301 non-Aboriginal adults in New South Wales. Age-sex-adjusted prevalence ratios (PRs) for obesity by sociodemographic factors, health behaviours and health status were estimated (multivariable log-binomial regression) for Aboriginal and non-Aboriginal participants separately. We quantified the extent to which key factors (physical activity, screen time, education, remoteness, area-level disadvantage) accounted for any excess Aboriginal obesity prevalence.

Results: Obesity prevalence was 39% among Aboriginal and 22% among non-Aboriginal participants (PR=1.65, 95% CI 1.55 to 1.76). Risk factors for obesity were generally similar for Aboriginal and non-Aboriginal participants and included individual-level and area-level disadvantage, physical inactivity, and poor physical and mental health, with steeper gradients observed among non-Aboriginal participants for some factors (P_interaction <0.05). Many risk factors were more common among Aboriginal versus non-Aboriginal participants; key factors accounted for >40% of the excess Aboriginal obesity prevalence.

Conclusion: A substantial proportion of the excess obesity prevalence among Aboriginal versus non-Aboriginal participants was explained by physical activity, screen time, education, remoteness and area-level disadvantage. Socioeconomic and health behaviour factors are potential targets for promoting healthy BMI, but these must be considered within the context of upstream social and cultural factors. Adults with health needs and disability require particular attention.

Keywords: epidemiology; health Inequalities; obesity.

Figure 1

(A) Distribution of body mass index among 45 and Up study participants by Aboriginal status and (B) proportion (and 95% CI) obese by age group, sex and Aboriginal status.

Figure 2

Obesity prevalence and prevalence ratios among Aboriginal and non-Aboriginal participants according to sociodemographic factors. Total n=1515 Aboriginal participants and 213 301 non-Aboriginal participants across all models; a category for missing data was included for each exposure variable but these results are not presented. Prevalence ratios are adjusted for age (in 5-year increments up to age <80 and ≥80 years) and sex, where appropriate. Horizontal lines represent 95% CIs around point estimates; the CIs in the non-Aboriginal sample are very narrow and are thus obscured by the marker for the point estimate. *Significant trend across categories among Aboriginal participants (only tested for ordinal variables). † Significant trend across categories among non-Aboriginal participants (only tested for ordinal variables). SEIFA, Socio-Economic Index for Areas.

Figure 3

Obesity prevalence and prevalence ratios among Aboriginal and non-Aboriginal participants according to health behaviour factors. Total n=1515 Aboriginal participants and 213 301 non-Aboriginal participants across all models; a category for missing data was included for each exposure variable but these results are not presented. Prevalence ratios are adjusted for age (in 5-year increments up to age <80 and ≥80 years) and sex, where appropriate. Horizontal lines represent 95% CIs around point estimates; the CIs in the non-Aboriginal sample are very narrow and are thus obscured by the marker for the point estimate. *Significant trend across categories among Aboriginal participants (only tested for ordinal variables). † Significant trend across categories among non-Aboriginal participants (only tested for ordinal variables).

Figure 4

Obesity prevalence and prevalence ratios among Aboriginal and non-Aboriginal participants according to health status factors. Total n=1515 Aboriginal participants and 213 301 non-Aboriginal participants across all models; a category for missing data was included for each exposure variable but these results are not presented. Prevalence ratios are adjusted for age ((in 5-year increments up to age <80 and ≥80 years) and sex, where appropriate. Horizontal lines represent 95% CIs around point estimates; the CIs in the non-Aboriginal sample are very narrow and are thus obscured by the marker for the point estimate. *Significant trend across categories among Aboriginal participants (only tested for ordinal variables). † Significant trend across categories among non-Aboriginal participants (only tested for ordinal variables).