Locals get travellers' diarrhoea too: risk factors for diarrhoeal illness and pathogenic Escherichia coli infection across an urban-rural gradient in Ecuador

Abstract

Objectives: Diarrhoea is a common and well-studied cause of illness afflicting international travellers. However, traveller's diarrhoea can also result from travel between high and low disease transmission regions within a country, which is the focus of this study.

Methods: We recruited participants for a case-control study of diarrhoea at four sites along an urban-rural gradient in Northern Ecuador: Quito, Esmeraldas, Borbón and rural communities outside of Borbón. At each of these sites, approximately 100 subjects with diarrhoea (cases) were recruited from Ministry of Health clinics and were age-matched with subjects visiting the same clinics for other complaints (controls).

Results: Travellers to urban destinations had higher risk of diarrhoea and diarrhoeagenic Escherichia coli (DEC) infections. Travel to Quito was associated with diarrhoea (aOR = 2.01, 95% CI = 1.10-3.68) and travel to Guayaquil (another urban centre in Ecuador) was associated with Diffuse Adherent E. coli infection (OR = 2.09, 95% CI = 1.01-4.33). Compared to those not travelling, urban origins were also associated with greater risk of diarrhoea in Esmeraldas (aOR = 2.28, 95% CI = 1.20-4.41), and with higher risk of diarrhoeagenic E. coli infections in Quito (aOR = 2.61, 95% CI = 1.16-5.86), with >50% of travel from Quito and Esmeraldas specified to another urban destination.

Conclusions: This study suggests that individuals travelling from lower-transmission regions (rural areas) to higher transmission regions (urban centres) within a single country are at a greater risk of acquiring a diarrhoea-related illness. Investments to improve water, sanitation and hygiene conditions in urban areas could have impacts on outlying rural areas within a given country.

Keywords: E. coli diarrhéique; E. coli à adhérence diffuse; Ecuador; Equateur; diarrhoea; diarrhoeagenic E. coli; diarrhée; diffusely adherent E. coli; rural; travel; urbain; urban; voyage.

Figure 1

Region of Study. This map highlights the region (red dashed circle) and four study sites (yellow stars) where the Eco-ZUR study was conducted. These four sites comprise an urban-rural gradient in northern Ecuador: including Quito (pop. ~1.62 M), Esmeraldas (pop. ~162 000), Borbón (pop. ~5000) and the rural villages (pop. ~50–500).

Figure 2

Prevalence of reported travel between study sites in the past year. Travel destination is denoted by the arrow, with the thickness of the arrow weighted by prevalence of reported travel to this destination. (a) Travel from urban/more populated cities to more rural/less populous communities. (b) Travel from rural/less populous communities to urban/more populated cities. It is important to note that not all travel is represented in these figures since data was only collected from participants at the four study locations (yellow stars). Individuals residing in Guayaquil, Santo Domingo and San Lorenzo may engage in more urban to rural travel or rural to urban travel that is not illustrated above.

Figure 3

Multivariate logistic regression model for characterizing Ecuadorian travelers. Demographic characteristics associated with domestic travel in the past year among study participants. All variables listed were included together in the model.

Figure 4

Risk by origin of travel. Multivariate logistic regression models for risk of domestic travel in the past year from locations across an urban-rural gradient on (a) diarrhoeal disease (b) diarrhoeagenic Escherichia coli infections and (c) diffuse adherent E. coli (DAEC). Odds ratios are adjusted based on sex, age, race, economic status, study site (random effect for “all sites” models only), diarrhoea case status (DEC models only), and known risk factors for diarrhoea disease transmission, i.e. household sanitation classification and reported household water treatment, with the exception of the DAEC subset, which is presented as unadjusted due to the relatively rare nature of these events.

Figure 5

Risk by destination of travel. Multivariate logistic regression models for risk of domestic travel in the past year to destinations across an urban-rural gradient on (a) diarrhoeal disease (b) diarrhoeagenic Escherichia coli infections and (c) diffuse adherent E. coli (DAEC). Odds ratios are adjusted based on sex, age, race, economic status, study site (random effect for “all sites” models only), diarrhoea case status (DEC models only), and known risk factors for diarrhoea disease transmission, i.e. household sanitation classification and reported household water treatment, with the exception of the DAEC subset, which is presented as unadjusted due to the relatively rare nature of these events.