An assessment of kilocalories and protein in the diets of HIV-infected adults in Kenya



Journal Title

Journal ISSN

Volume Title



INTRODUCTION: The detrimental effects of HIV are well known throughout the world. This public health burden is particularly evident in Kenya, Africa, where 6.7% of adults are infected with HIV. One problem of those living with HIV in this resource-poor country is the concurrent existence of malnutrition. Previous investigators undertook many studies to bridge the link between HIV and malnutrition. However, they were unable to gather individual diet information to allow accurate estimation of kilocalorie and protein needs. These deficits hinder accurate dietary interventions for these patients. \r\nMETHODS: Over the period of April 2009 to August 2009, we designed and implemented a dietary instrument (3-day recall survey) to assess the kilocalorie and protein consumption for HIV-infected adults in Kijabe, Kenya. We used this data to compare subject consumption to predicted nutrient needs. We used Harris-Benedict (HB) and Mifflin formulas for kilocalorie need predictions and World Health Organization (WHO) equation for protein need predictions. We then characterized the relationship between dietary intake, BMI and CD4 levels.\r\nRESULTS: A total of 201 patients were surveyed, 122 (60.7% [27% males, 73% females]) met inclusion criteria. There was no statistical difference between HB and Mifflin equations for kilocalories. Males averaged 68.8% (SD 23.3) of estimated kilocalorie requirements (HB) and 100.5% (SD 45.3) of protein. Women averaged 74.4% (SD 24.4) of kilocalorie needs (HB) and 100.5% (SD 42.5) of protein. Differences between genders were not statistically significant (p=0.247 [kilocalorie], p=0.936 [protein]). There was a significant correlation between protein intake and CD4 for males (r=0.7035, p=0.0004) but not for females (r=-0.1911, p=0.1546). There was no statistical significance found between kilocalories (male r=0.104, p=0.654; female r=0.0420, p=0.765) and CD4. No relationship was found between nutrient intake and BMI (male p=0.690; female p=0.477).\r\nCONCLUSIONS: A 3-day recall dietary assessment appears to be an adequate method to obtain dietary information for adult HIV patients in East Africa. Further, we conclude that current predictive formulas for protein underestimate needs in the male HIV population. \r\n