Elsevier

The Lancet

Volume 398, Issue 10296, 17–23 July 2021, Pages 238-248
The Lancet

Articles
Body-mass index and diabetes risk in 57 low-income and middle-income countries: a cross-sectional study of nationally representative, individual-level data in 685 616 adults

https://doi.org/10.1016/S0140-6736(21)00844-8Get rights and content

Summary

Background

The prevalence of overweight, obesity, and diabetes is rising rapidly in low-income and middle-income countries (LMICs), but there are scant empirical data on the association between body-mass index (BMI) and diabetes in these settings.

Methods

In this cross-sectional study, we pooled individual-level data from nationally representative surveys across 57 LMICs. We identified all countries in which a WHO Stepwise Approach to Surveillance (STEPS) survey had been done during a year in which the country fell into an eligible World Bank income group category. For LMICs that did not have a STEPS survey, did not have valid contact information, or declined our request for data, we did a systematic search for survey datasets. Eligible surveys were done during or after 2008; had individual-level data; were done in a low-income, lower-middle-income, or upper-middle-income country; were nationally representative; had a response rate of 50% or higher; contained a diabetes biomarker (either a blood glucose measurement or glycated haemoglobin [HbA1c]); and contained data on height and weight. Diabetes was defined biologically as a fasting plasma glucose concentration of 7·0 mmol/L (126·0 mg/dL) or higher; a random plasma glucose concentration of 11·1 mmol/L (200·0 mg/dL) or higher; or a HbA1c of 6·5% (48·0 mmol/mol) or higher, or by self-reported use of diabetes medication. We included individuals aged 25 years or older with complete data on diabetes status, BMI (defined as normal [18·5–22·9 kg/m2], upper-normal [23·0–24·9 kg/m2], overweight [25·0–29·9 kg/m2], or obese [≥30·0 kg/m2]), sex, and age. Countries were categorised into six geographical regions: Latin America and the Caribbean, Europe and central Asia, east, south, and southeast Asia, sub-Saharan Africa, Middle East and north Africa, and Oceania. We estimated the association between BMI and diabetes risk by multivariable Poisson regression and receiver operating curve analyses, stratified by sex and geographical region.

Findings

Our pooled dataset from 58 nationally representative surveys in 57 LMICs included 685 616 individuals. The overall prevalence of overweight was 27·2% (95% CI 26·6–27·8), of obesity was 21·0% (19·6–22·5), and of diabetes was 9·3% (8·4–10·2). In the pooled analysis, a higher risk of diabetes was observed at a BMI of 23 kg/m2 or higher, with a 43% greater risk of diabetes for men and a 41% greater risk for women compared with a BMI of 18·5–22·9 kg/m2. Diabetes risk also increased steeply in individuals aged 35–44 years and in men aged 25–34 years in sub-Saharan Africa. In the stratified analyses, there was considerable regional variability in this association. Optimal BMI thresholds for diabetes screening ranged from 23·8 kg/m2 among men in east, south, and southeast Asia to 28·3 kg/m2 among women in the Middle East and north Africa and in Latin America and the Caribbean.

Interpretation

The association between BMI and diabetes risk in LMICs is subject to substantial regional variability. Diabetes risk is greater at lower BMI thresholds and at younger ages than reflected in currently used BMI cutoffs for assessing diabetes risk. These findings offer an important insight to inform context-specific diabetes screening guidelines.

Funding

Harvard T H Chan School of Public Health McLennan Fund: Dean's Challenge Grant Program.

Introduction

The global prevalence of overweight and obesity has doubled over the past four decades, with 1·9 billion (39%) adults living with overweight, and an additional 650 million (13%) adults living with obesity in 2016.1, 2 Although studies published in the past 5 years suggest that the rate of increase in overweight and obesity in high-income countries might be slowing,2, 3 there is growing evidence that this epidemic has accelerated in low-income and middle-income countries (LMICs), where approximately 67% of people with obesity now reside.4, 5, 6 The unprecedented increase in overweight and obesity in LMICs has paralleled the alarming rise in the prevalence of diabetes and other cardiovascular risk factors in these countries, such that 79% of the estimated 463 million people with diabetes reside in LMICs.7 However, data on how overweight and obesity, measured with standard metrics such as body-mass index (BMI), relate to diabetes risk across LMICs, and whether the variation observed in country-level studies is also observed at larger geographical scales are scarce.

Research in context

Evidence before this study

We searched PubMed (with the medical subject heading search tool) on April 15, 2020, using the terms “body mass index” OR “anthropometry” AND “diabetes mellitus” AND “low- and middle-income countries” NOT “comment” NOT “case reports”. We searched for manuscripts published in any language from database inception to April 15, 2020. We found two pooled studies on the association between body-mass index (BMI) and diabetes. One study pooled nationally representative surveys from six low-income and middle-income countries (LMICs) and evaluated the association between BMI categories and non-communicable disease multimorbidity (including nine chronic conditions, one of which was diabetes). The second study pooled data on 900 000 individuals recruited from 18 cohorts across seven Asian countries and did not include nationally representative data. Several large studies on the prevalence and projected trends of overweight, obesity, and diabetes across LMICs have been published, but none of these studies have evaluated the association between BMI and diabetes risk in these settings and how this association varies by geographical region and sex.

Added value of this study

To our knowledge, this study uses the largest harmonised dataset collected to date of nationally representative, individual-level data on BMI and a biological measure of diabetes in 685 616 adults across 57 LMICs spanning six world regions. We did robust analyses, stratified by sex and geographical region, to assess the association between BMI (as a continuous and categorical exposure) and diabetes, defined biologically as a fasting plasma glucose concentration of 7·0 mmol/L (126·0 mg/dL) or higher; a random plasma glucose concentration of 11·1 mmol/L (200·0 mg/dL) or higher; or a glycated haemoglobin of 6·5% (48·0 mmol/mol) or higher, or by self-reported use of diabetes medication. We also present receiver operating curve analyses of optimal BMI cutoffs when assessing diabetes risk. The results show substantial variability in the association between BMI and diabetes risk by region and sex, and they add to our current understanding of the association between BMI and diabetes risk in countries poorly represented in previous literature.

Implications of all the available evidence

Given the rapidly growing burden of overweight, obesity, and diabetes in LMICs, urgent population-level strategies are needed to reverse current and projected trends. Additionally, our findings highlight that interventions and the BMI thresholds at which clinicians and policy makers consider metabolic risk to be increased vary across LMICs. Finally, in specific regions, screening might also need to include younger adults than is currently recommended by most guidelines.

Although the association between high BMI and metabolic risk is well established,8, 9 the current understanding of BMI and its association with key clinical outcomes has been shaped by a vast number of studies that have, to date, almost exclusively been done in high-income countries.8, 10, 11 The exception has been the increasing number of studies done in Asian and south Asian countries,12, 13, 14 which have directly informed clinical guidelines recommending the lowering of BMI thresholds that define overweight to better characterise metabolic risk in these populations.14 Importantly, single-country studies in LMICs have also indicated variability in the association between BMI and diabetes risk when standard BMI thresholds are used,15, 16 but differences in this association across LMICs, which are highly heterogeneous, remain largely unexplored.

In this study, we aimed to characterise the association between BMI and diabetes risk in LMICs at the country level, and stratified by geographical region and sex. To achieve this aim, we used the largest harmonised dataset of individual-level survey data compiled to date, including biologically measured diabetes status, to characterise the risk of diabetes across the full range of BMIs in LMICs.

Section snippets

Data sources and study population

In this cross-sectional study, we did a pooled analysis of individual-level data from 58 nationally representative population-based surveys across 57 LMICs. The requirements for inclusion of a national survey and the search methods used have been described previously.17, 18 Further details specific to this analysis are provided in the appendix (pp 3–4). Briefly, eligible surveys were done during or after 2008; had individual-level data; were done in a low-income, lower-middle-income, or

Results

We identified 58 nationally representative surveys done in 57 LMICs, of which 49 were STEPS surveys and nine were non-STEPS surveys (table 1). Data on a diabetes biomarker were available in all 58 surveys. The diabetes biomarker used in 47 (81%) surveys was point-of-care fasting capillary glucose (appendix pp 33–34). Plasma equivalents were provided in all but eight of these surveys. For these eight surveys, we converted capillary glucose measurements to plasma glucose. No differences were

Discussion

In this cross-sectional study of 685 616 individuals across 57 LMICs, we found that a greater risk of diabetes was observed at a BMI of 23 kg/m2 or higher, with a corresponding increase in diabetes risk of 43% in men and 41% in women when compared with those who have a BMI of 18·5–22·9 kg/m2. ROC analyses showed variability across sex and geographical regions in the BMI cutoffs at which sensitivity and specificity were optimised for diabetes screening, ranging from a BMI cutoff of 23·8 kg/m2 in

Data sharing

Deidentified participant-level data used in this study are publicly available and can be obtained on request or via the weblinks provided in the appendix (p 70). The STROBE checklist can also be accessed on request. In addition, data dictionaries will be shared by the corresponding author on request.

Declaration of interests

MKA reports receiving a grant from Merck and Co awarded to Emory University, outside the submitted work. DJW reports serving on a data monitoring committee for Novo Nordisk SOUL and FLOW trials. JBM reports serving as an academic associate for the American clinical laboratory, Quest Diagnostics. All other authors declare no competing interests.

References (32)

  • KF Adams et al.

    Overweight, obesity, and mortality in a large prospective cohort of persons 50 to 71 years old

    N Engl J Med

    (2006)
  • SS Khan et al.

    Association of body mass index with lifetime risk of cardiovascular disease and compression of morbidity

    JAMA Cardiol

    (2018)
  • SA Patel et al.

    Is the “south Asian phenotype” unique to south Asians? Comparing cardiometabolic risk factors in the CARRS and NHANES studies

    Glob Heart

    (2016)
  • Appropriate body-mass index for Asian populations and its implications for policy and intervention strategies

    Lancet

    (2004)
  • AM Almajwal et al.

    Performance of body mass index in predicting diabetes and hypertension in the Eastern Province of Saudi Arabia

    Ann Saudi Med

    (2009)
  • GAV Murphy et al.

    The use of anthropometric measures for cardiometabolic risk identification in a rural African population

    Diabetes Care

    (2014)
  • Cited by (0)

    *

    Contributed equally

    View full text