Characterization of central adiposity in women 50 years and older in Kenya

Abstract

University of Washington Abstract Characterization of central adiposity among women 50 years and older living with HIV in Kenya Kuya Lokuruka Chair of Supervisory Committee: Carey Farquhar, Department of Global Health Background: As life expectancy for people living and aging with HIV in sub-Saharan Africa is increasing, non-communicable diseases (NCDs) such as obesity are emerging as major contributors to morbidity and mortality. Effective assessment and management of obesity in routine clinical settings is critical to reducing cardio-metabolic risk.Methods: This was a secondary data analysis of a 6-month observational study that enrolled women aged ≥ 50 years at the Kenyatta National Hospital (KNH) Comprehensive Care Clinic (CCC). Participants consented to an interview, examination, blood draw, and DEXA scan. Data were extracted on visceral adipose tissue (VAT) mass, total body fat percentage, BMI, and anthropometric measurements, including waist circumference (WC), waist-to-hip ratio (WHR), and waist-to-height ratio (WHtR). Socio-demographic and clinical characteristics collected included ART regimen, viral load, date of HIV diagnosis, HbA1c, blood pressure, depression status, and cognitive function. Univariate and multivariable analyses using logistic regression were used to determine correlates of elevated VAT and waist-to-height ratio. Agreement between anthropometric measurements and VAT was assessed using receiver operating characteristic (ROC) analysis. Results: Overall, 179 women were enrolled, with a median age of 58 years (IQR: 53-62). Most of the participants had at least a primary or secondary education (59.2%) and were widowed or currently married (62.5%). About half of the participants (50.3%) were employed, and most (68.7%) earned less than 10,000 Kenya shillings (about 77 USD) per month. The majority of participants had been living with HIV for more than ten years (81%), were on a DTG-based regimen (84.4%), and were virally suppressed (92.2%). Many participants had mild cognitive impairment (66.5%) and did not have depression (81.6%). The majority of participants were overweight/obese according to body fat percentage (97.2%) and BMI (79.5%). Many had central obesity according to WC (66.3%) and by WHtR (86.0%). Cardiometabolic risk stratification by VAT had the majority in the low risk category (88.3%). In univariate analysis, those with mild or moderate depression were more likely to have elevated VAT. Odds ratio [OR] = 3.26; 95% Confidence Interval [CI]: 1.15 – 9.18, p = 0.026. This association remained after adjusting for age; adjusted OR = 3.24; 95% CI: 1.14 - 9.23; p-value= 0.028. In univariate analysis, those participants who had been diagnosed with HIV more than ten years before the study were three times more likely to have an elevated waist-to-height ratio (OR = 2.88; 95% CI: 1.13 – 7.33, p = 0.027). This remained significant in multivariable analysis adjusting for age; aOR was 2.82 (CI: 1.10 - 7.25) with a p-value of 0.028. On ROC analysis, WHtR had an Area Under the Curve (AUC) of 0.852 (95% CI: 0.758 – 0.946), with an optimal cut-off of 0.60, the best balance of sensitivity (81%) and specificity (74.5%) for predicting VAT. Conclusion: A high obesity prevalence was seen using anthropometric measurements, with the majority having low cardiometabolic risk. Women who had longer duration of HIV and those with mild to moderate depression had increased odds of obesity. The waist-to-height ratio measure could be useful for cardiometabolic assessment in the clinic setting.