Prevalence of Giardia and Cryptosporidium
A total of 488 fecal samples were collected and transported in ice boxes to the parasitology laboratory of the Centro de Investigación en Alimentación y Desarrollo and were stored between 5 °C and 7 °C for 24–72 h until analysis. Eighty-seven percent (n = 150), 8% (n = 15) and 5% (n = 8) of the participant children gave 3, 2 and 1 sample(s) per child, respectively.
The overall prevalence of pathogenic and commensal protozoa found in the participant children (n = 173) is shown in Table 1. More than half (n = 103) of the children had protozoan infections (60%), and 29% (n = 50) had infections with two or more protozoan genera. No difference was found in the prevalence of protozoan infections between females and males (p = 0.290; data not shown). Cryptosporidium spp. showed the highest prevalence (n = 47, 27%), which was followed by G. intestinalis. (n = 40, 23%). In addition, E. histolytica/dispar/moshkovskii were also detected, albeit at a lower prevalence. However, the species of the E. histolytica/dispar/moshkovskii complex were not identified. On the other hand, E. nana had the highest prevalence (n = 57, 33%) of all pathogenic and commensal protozoa detected in this study. I. bütschlii had a lower prevalence (n = 1, 0.6%). Finally, the helminth H. nana was only found in 4 children (2%) (not shown in Table 1). E. histolytica is well recognized as a pathogenic amoeba, but E. dispar, E. nana, E. coli and I. bütschlii are considered a non-pathogenic amoebae . However, pathogenicity of E. moschkovskii remains unknown .
Infection and nutritional status
At baseline, the average age of the study children (n = 173) was 8.8 (±2.8) years old. Fifty-one percent (n = 88) and 49% (n = 85) were female and male, respectively. No difference was found between the proportions of female and male participants (p = 0.747). Our study children consisted of 26 preschools and 147 primary school children with average ages of 3.6 (±0.98) and 9.7 (±1.93), respectively. In relation to the anthropometric measurements, the proportions below −2 SD in W/A, H/A (stunting) and BMI/A were 1.5% (n = 3), 3.3% (n = 6) and 3% (n = 5), respectively. The means (SD) for the W/A, H/A and BMI/A Z scores were 0.135 (±1.12), −0.53 (±1.41) and 0.42 (±1.40), respectively.
No difference was found in the age, weight, height, W/A Z scores between the children with and without giardiasis or cryptosporidiosis (Table 2). However, children with giardiasis or cryptosporidiosis had significantly lower H/A and BMI/A Z scores than uninfected children (p = 0.001 and p = 0.028 and p = 002 and p = 0.030 respectively) (Table 2).
Distribution of risk factors of the families of participating children
One hundred and seventy-three mothers and 45 fathers of the children responded to the questionnaires. Most of the participating children came from homes with unmarried mothers and fathers (Table 3). At the time of the interview, most mothers were housewives and most fathers were employed. Most parents had not completed their secondary education, and more than half of the families were living with 1 or less valid minimum wages. Twenty-four percent of the families of the participating children were living in crowded conditions, and 90% were living in households made of materials that were appropriate for the weather conditions of the study area. Also, most of the children’s families drank water directly from the tap (59%). In addition, 43% of the families had domestic animals (Table 3).
Factors associated with G. intestinalis and Cryptosporidium spp.
G. intestinalis and Cryptosporidium spp. were separately analyzed as dependent variables and risk factors as independent variables. Univariate analysis revealed that the drinking water type, presence of domestic animals in the household, symptomatology and seasonality for sample collection for giardiasis; and drinking water type and presence of domestic animals in the household for cryptosporidiosis fulfilled the acceptance criterion for examination in the multiple regression analysis (Table 4). The civil status, education and economic activity of the parents; family income; crowding and household conditions did not meet the criteria (data not shown). The nutritional status, sex and age are recognized to be factors that influence the prevalence of intestinal parasitic infections [24, 25]. They were used as adjustment variables in the multiple regression logistic analysis (stepwise), and preliminary models for giardiasis and cryptosporidiosis were produced. Interaction (p > 0.1) and collinearity (p < 0.7) were not found for those models, and the final models were defined (Table 5). The Giardia-infected children had higher risk to present abdominal pain (OR = 4.0, 95%CI = 1.11–13.02, p = 0.030) adjusted by sex, age, W/A, H/A, and BMI/A Z-scores. On the other hand, children drinking tap water were at a higher risk (OR = 5.0, 95%CI =1.41–17.20, p = 0.012) of cryptosporidiosis with the same adjustment variables (Table 5).