Contrasting patterns in the small-scale heterogeneity of human helminth infections in urban and rural environments in Brazil

Abstract

Marked heterogeneity exists in the patterns of parasitic infection between individuals, households and communities. Analysis of parasite distributions within populations is complicated by the fact that parasite distributions are highly aggregated and few studies have explicitly incorporated this distribution when investigating small-scale spatial heterogeneities. This study aimed to quantify the small-scale (within- and between-household) heterogeneity of helminth infection in an area of Minas Gerais State, Brazil, with rural and urban sectors. Parasitological data from a cross-sectional survey of 1,249 individuals aged 0-86 years from 242 households were analysed. Within-household clustering of infection was assessed using random effect logistic regression models and between-household spatial heterogeneity was assessed using a Bayesian negative binomial spatial model. The overall prevalence of hookworm (Necator americanus) was 66.9%, the prevalence of Schistosoma mansoni was 44.9% and the prevalence of Ascaris lumbricoides was 48.8%. Statistical analysis indicated significant (within) household and (between household) spatial clustering of hookworm in both rural and urban areas and of S. mansoni in rural areas. There was no evidence of either household or spatial clustering of S. mansoni in urban areas. The spatial correlation of S. mansoni was estimated to reduce by half over a distance of 700 m in the rural area. Rural hookworm had a much smaller half-distance (28 m) and urban hookworm showed an even smaller half-distance (12 m). We suggest that such species-specific differences in patterns of infection by environment are primarily due to variation in exposure and parasite life cycle, although host genetic factors cannot be ruled out.

Fig. 1

Age-intensity profiles of (a) hookworm and (b) S. mansoni in rural (solid lines) and urban (dashed lines) areas of Americaninhas, southeastern Brazil. Although different levels, a similar age profile was observed in rural and urban areas, with hookworm intensity steadily rising with age and S. mansoni intensity convex in form.

Fig. 2

Frequency distribution of egg counts (eggs/gram faeces) per person in urban and rural areas of Americaninhas, southeastern Brazil. (A) hookworm and (B) S. mansoni. Eggs counts in both areas were highly aggregated.

Fig. 3

Maps of small-scale spatial heterogeneity of the intensity of infection in rural and urban areas of Americaninhas, southeastern Brazil. Each dot represents one household, with the shading showing the quartile of the standardized parasite density ratio as derived by the Bayesian spatial model. North is the top of the page. (A) Hookworm in urban areas, where the quartiles are 1, 1.1 and 1.3; (B) Hookworm in rural areas where the quartiles are 0.9, 1.6 and 3.1; (C) S. mansoni in urban areas where the quartiles are 1, 1.1 and 1.2; and (D) S. mansoni in rural areas where the quartiles are 0.4, 1.2 and 5. For each species, marked spatial heterogeneity was evident in both areas. High infection intensities of both hookworm and, to a lesser extent, S. mansoni are evident in the northeast corner of the urban area. In rural areas, high infection intensities of S. mansoni occur in the northeast corner.