ABSTRACT
BACKGROUND: Tuberculosis is a leading cause of death worldwide, and food insecurity is known to negatively influence health outcomes through multiple pathways. Few studies have interrogated the relationship between food insecurity and tuberculosis outcomes, particularly independent of nutrition. METHODS: We conducted a prospective cohort study of adults initiating first-line treatment for clinically suspected or microbiologically confirmed drug-sensitive tuberculosis at a rural referral center in Haiti. We administered a baseline questionnaire, collected clinical data, and analyzed laboratory samples. We used logistic regression models to estimate the relationship between household food insecurity (Household Hunger Scale) and treatment failure or death. We accounted for exclusion of patients lost to follow-up using inverse probability of censoring weighting and adjusted for measured confounders and nutritional status using inverse probability of treatment weighting. RESULTS: We enrolled 257 participants (37% female) between May 2020 and March 2023 with a median age (interquartile range) of 35 (25-45) years. Of these, 105 (41%) had no hunger in the household, 104 (40%) had moderate hunger in the household, and 48 (19%) had severe hunger in the household. Eleven participants (4%) died, and 6 (3%) had treatment failure. After adjustment, food insecurity was significantly associated with subsequent treatment failure or death (odds ratio 5.78 [95% confidence interval, 1.20-27.8]; P = .03). CONCLUSIONS: Household food insecurity at tuberculosis treatment initiation was significantly associated with death or treatment failure after accounting for loss to follow-up, measured confounders, and nutritional status. In addition to the known importance of undernutrition, our findings indicate that food insecurity independently affects tuberculosis treatment outcomes in Haiti.
ABSTRACT
BACKGROUND: Childhood iron deficiency anemia (IDA) is an important contributor to under-five mortality in the developing world. There is evidence that Community Health Worker (CHW) delivered programs to increase maternal knowledge of child health practices may decrease childhood IDA. This study reports findings on the association between a long standing CHW intervention and childhood anemia status in rural Haiti. METHODS: Using structural equations and mediation analyses on data from a household-based survey of 621 mother/child dyads, we tested the hypothesis that CHW would have a direct positive effect on maternal knowledge and an indirect effect on childhood anemia in rural Haiti. RESULTS: CHW contact was significantly associated with maternal knowledge of key child health practices (ß = 0.193, SE = 0.058, p = 0.001). However, knowledge was not associated with childhood anemia (ß = -0.008, SE = 0.009, p = 0.382). Maternal knowledge categories significantly affected by CHW contact included diarrheal prevention knowledge (ß = 0.111, SE = 0.045, p = 0.013) and signs of malnutrition (ß = 0.217, SE = 0.071, p = 0.002). There was no significant association with knowledge of vitamin A and iron rich foods (ß = 0.057, SE = 0.032, p = 0.074), which is the intervention most likely to impact childhood anemia. In all path models tested, we identified the control variables low household socio-economic status, mothers' anemia status, and child's age less than 24 months as significant predictors of childhood anemia. CONCLUSIONS: CHWs delivered interventions are associated with improved maternal knowledge of child health practices in rural Haiti; however, this knowledge is not associated with improved childhood anemia. Concurrently with CHW-delivered programs, interventions household poverty are implied to impact childhood health outcomes in resource poor settings.
ABSTRACT
The HIV/AIDS pandemic disproportionately afflicts regions of the world that have minimal access to formal schooling and low literacy rates. Health educational interventions are difficult to evaluate efficiently in these settings because standard approaches such as written questionnaires cannot easily be employed. Here, we describe a novel method of rapidly assessing health interventions among large groups that does not require the ability to read or write. We tested this evaluation tool within the context of a community-based HIV/AIDS drama education program in a low-literate region of rural Haiti. The evaluation was simple, easy to use, and confirmed substantial improvements in knowledge after the intervention. The tool also provided information that helped alter the intervention to improve educational outcomes in subsequent productions of the drama. This evaluation method can be utilized for very little cost, and may be replicated in resource-poor, non-literate settings throughout the developing world.
