Ecological niche modelling for predicting the risk of cutaneous leishmaniasis in the Neotropical moist forest biome

Abstract

A major challenge of eco-epidemiology is to determine which factors promote the transmission of infectious diseases and to establish risk maps that can be used by public health authorities. The geographic predictions resulting from ecological niche modelling have been widely used for modelling the future dispersion of vectors based on the occurrence records and the potential prevalence of the disease. The establishment of risk maps for disease systems with complex cycles such as cutaneous leishmaniasis (CL) can be very challenging due to the many inference networks between large sets of host and vector species, with considerable heterogeneity in disease patterns in space and time. One novelty in the present study is the use of human CL cases to predict the risk of leishmaniasis occurrence in response to anthropogenic, climatic and environmental factors at two different scales, in the Neotropical moist forest biome (Amazonian basin and surrounding forest ecosystems) and in the surrounding region of French Guiana. With a consistent data set never used before and a conceptual and methodological framework for interpreting data cases, we obtained risk maps with high statistical support. The predominantly identified human CL risk areas are those where the human impact on the environment is significant, associated with less contributory climatic and ecological factors. For both models this study highlights the importance of considering the anthropogenic drivers for disease risk assessment in human, although CL is mainly linked to the sylvatic and peri-urban cycle in Meso and South America.

An investigation of the effects of environmental and ecologic factors on cutaneous leishmaniasis in the old world: a systematic review study

“…Moreover, it was observed that whereas temperature had direct effects on CL incidence, humidity exerted inverse effects [18]. Similar results were also reported by Ghatee et al [21] and Chavy et al [26]. In another study, Sharafi et al reported that the amount of evaporation, the number of sunny days, the rate of wind velocity and temperature had direct effects on the incidence of CL [29].…”

“…The effect of the time lag of temperature is related to the average incubation period of CL 17. Other studies have also ended up with the same results observing a significant positive correlation between the prevalence of CL and the temperature, such as Ali-Akbarpour et al's study in 2012 in different regions of Iran [32] and other countries [26,31,[33][34][35] as well as Roger et al' study in 2013 in Cayenne, French Guiana, 15 with lags of 4-8 months due to the incubation period of CL. Other studies have reported that an increase in temperature can rise the trend of CL cases [15,19,27,[36][37][38][39], and the morbidity of the disease starts in March and reaches its peak in July.…”

“…After evaluation, 45 full-text articles were removed, and 23 full-text articles met the eligibility criteria for the systematic review. Finally, after removing nine articles, 14 full-text articles were included in the systematic review (shown in Table 1) [1,6,9,11,12,[15][16][17][18][19][20][21][22][23][24][25][26][27][28][29][30]. Based on the findings, the effects of environmental factors on the incidence of CL in the Old World are discussed below.…”

Spatial epidemiology of yellow fever: Identification of determinants of the 2016-2018 epidemics and at-risk areas in Brazil

“…In that way, an alternative may be to focus on the occurrence of human cases [25], considering that the disease records show favourable conditions for the circulation of the pathogen, whatever the hosts and vectors, including the secondary ones [24,26]. Choice of relevant explicative variables is crucial for such purposes.…”

“…We resampled occurrence points to consider the more likely place of infection better, as proposed recently for cutaneous leishmaniasis modelling and to reply on the BAM framework [25]. To increase the likelihood to sample the area of infection, we randomly distributed the occurrence points in a buffer area expected to represent the most likely area of infection.…”