Lipid-based nutrient supplement at initiation of antiretroviral therapy does not substitute energy from habitual diet among HIV patients–a secondary analysis of data from a randomised controlled trial in Ethiopia

Introduction
Malnutrition is common among people with HIV in sub-Saharan Africa. Nutritional supplementation at initiation of antiretroviral treatment (ART) has shown beneficial effects, but it is not known if supplementation replaces or supplements the habitual energy intake in a context of food insecurity.
Methods
As part of a randomised controlled trial among people with HIV initiating ART in Ethiopia, we assessed whether the provision of a lipid-based nutrient supplement (LNS) affected energy intake from the habitual diet. People with HIV aged ≥18 years with a body mass index (BMI) >17 were randomly allocated 2:1 to receive either early (month 1-3 after ART initiation) or delayed (month 4-6 after ART initiation) supplementation with LNS (≈4,600 kJ/day). Participants with BMI 16-17 were all allocated to early supplementation. The daily energy intake from the habitual diet (besides the energy contribution from LNS) was assessed using a 24-h food recall interview at baseline and at monthly follow-up visits. Linear mixed models were used to compare habitual energy intake in (1) early versus delayed supplementation groups and (2) supplemented versus unsupplemented time periods within groups.
Results
Of 301 participants included, 67% of the participants were women, mean (±standard deviation [SD]) age was 32.9 (±8.9) years and 68% were living in moderately or severely food insecure households. Mean (±SD) reported habitual energy intake at baseline was 5,357 kJ/day (±2,246) for women and 7,977 kJ/day(±3,557) for men. Among all participants, there were no differences in mean habitual energy intake between supplemented and unsupplemented groups in neither the first 3 (P = 0.72) nor the following 3 months (P = 0.56). Furthermore, habitual energy intake did not differ within groups when comparing periods with or without supplementation (P = 0.15 and P = 0.20). The severity of food insecurity did not modify the effect of supplementation in habitual energy intake (P = 0.55). Findings were similar when participants with BMI 16-17 were excluded.
Conclusion
Our findings indicate that the LNS provided after ART initiation supplement, rather than substitute, habitual energy intake among people with HIV, even among those who are food insecure. This supports the feasibility of introducing nutritional supplementation as part of HIV treatment.