Prevalence and risk factors of functional gastrointestinal disorders in Vietnamese infants and young children

We performed a cross-sectional study to assess the community prevalence of FGIDs in infants and toddlers, aged 0–48 months in Vietnam. Recruitment was conducted in two study sites in Hanoi, including a governmental hospital and a governmental kindergarten. Subjects aged 0 – 6 months were recruited from the vaccination centre in the National children’s Hospital while subjects aged 6.1 months – 48 months were recruited from both the vaccination centre as well as the kindergarten. Subjects were screened by paediatricians and researchers for eligibility and then recruited on the same day. Healthy infants and young children were recruited with the exclusion for subjects with physician-identified organic disorders, physical and/or mental deficiency, history of gastrointestinal surgery, and the use of long-term medication.

Data was collected from October 2018 to March 2019. Written informed consent was obtained from the parent/legal guardian after providing details of the survey and all concerns answered, before the questionnaire was administered. A previously validated questionnaire to diagnose FGIDs in infants and young children was used for data collection [18‐20]. The questionnaire was translated into Vietnamese language by a professional translation company and tested for clarity and applicability using a smaller sample of mothers of infants and young children. The final version of the questionnaire (in English language) is provided as a supplementary file (see Supplementary file 1). The questionnaire for FGIDs was tailored for the group of infants aged 0 – 6 months and young children aged 6.1 months – 48 months, which consisted of 3 sections: Sect. 1 contained questions on demographics characteristics including age, gestational age, gender, birth order, mode of delivery, prenatal complications, anthropometry, growth pattern, feeding practices, as well as medical history; Sect. 2 covered subject’s FGIDs history with explicit questions on gastrointestinal symptoms including stool pattern (using the validated Amsterdam and Bristol stool chart [21, 22]), spitting and crying habits for infants and young children; Sect. 3 collected information on the family social and economic details such as income, educational level of both parents and the occurrence of domestic violence in the household. The last section also included questions on adverse life events faced by the child and family, such as exposure of child or mother to physical or verbal abuse. Section 1 and 2 were collected via an interview from the research team while Sect. 3 was self-administered. After completing the questionnaire, the subjects were diagnosed using the ROME IV criteria and grouped according to the case definitions for infantile colic, regurgitation, dyschezia, functional diarrhoea and functional constipation [1, 3].

The sample size was calculated using an estimated 10% prevalence for infantile colic and childhood constipation using Epi Info Stat Calc [4]. Assuming a confidence level of 95%, power of 80% and a precision found to the nearest 2%, a sample size of 750 infants in ages 0–6 months (group 1) and 750 young children in ages 6.1 months – 48 months (group 2) were considered adequate. Group 2 was further stratified for age according to the following categories: 6.1 – 12 months, 12.1 – 24 months, 24.1 – 36 months and 36.1 – 48 months (Table 1).

The data was entered into a Microsoft Excel (Microsoft Office 365 MSO,16.0.012527.21230). The SPSS statistical analysis software (windows version 22.0) was used for the data analysis. Descriptive statistics were used to present the demographic characteristics such as age and weight of the sample and the prevalence of FGIDs in infants and young children. Descriptive statistics used includes means, medians, standard deviations, ranges, and percentages, as required by the distribution characteristics of each variable. Qualitative variables were expressed in terms of frequency and percentage. Quantitative variables were presented as means ± standard deviation (standard distribution) and median or quartile (non-standard distribution). The difference between quantitative variables was investigated by Student’s t-test. The difference between qualitative variables was compared by chi-square test (or McNemar). Multivariate logistic regression was used to evaluate the potential risk factors (such as demographic data, exposure to adverse events, social and economic factors) associated with the development of FGIDs. Odds ratio (OR) estimates and 95% confidence interval of OR, as well as the p-values of the chi-square test for each risk factor were provided. All statistical tests used a significance level of 0.05. Multiple logistic regression was performed on variables that were found to have significant associations.

The protocol, information letter to the parents/legal guardian and written informed consent form were approved by the Research Ethics Committee of the Vietnam National Children’s Hospital. The survey was conducted in accordance with the guidelines of the Declaration of Helsinki, the International Conference on Harmonization (ICH) guidelines on Good Clinical Practice (GCP), and applicable regulatory requirements of the Scientific and Medical Council National Children’s Hospital, Hanoi Medical University and Authority of Quynh Loi Kindergarten. The study was registered in the Netherlands Trial Registry (identifier: NL7286/NTR7495).

Mothers of 1662 infants and young children between the age of 0 – 48 months old were invited to participate in the study. Of these, 1511 mothers (752 in group 1 aged 0–6 months and 759 in group 2 aged 6.1 – 48 months) completed the questionnaires for the final analysis (Table 1). Data was excluded from subjects who did not complete the screening procedure (n = 151).

All subjects recruited were Vietnamese, with a mean age of 3.17 ± 1.73 months for group 1 and 24.40 ± 12.99 months for group 2. A total of 829 subjects (54.9%) were males. Most subjects were living in the urban areas (78.2%). Around half of the subjects (51.2%) provided information on the range of income and most of the families (39.0%) had an annual household income below 273,000,000 VND (estimate around US$11,800). High perception of financial sufficiency was reported in 88.5% of the subjects stating that income met their needs. In terms of parental educational background, the majority of mothers (39.6%) had a Bachelor’s degree whereas the majority of the fathers (41.3%) had a high school education.

In total, 151 subjects (10.0%) were diagnosed with at least one FGID, according to the Rome IV criteria. Only nine infants (0.6%) had more than one FGID of which the most common combination was infant colic and regurgitation. The prevalence of FGIDs among infants and young children according to residence area and age group is presented in Table 2. Infant regurgitation was the most prevalent FGID in early infancy (9.3%). The peak onset for regurgitation was observed between 2—3 months of age. Functional constipation was the most common disorder in the 7 months – 48 months old (5.6%). The prevalence of functional constipation was relatively high between the period of 5—6 months and 24—30 months of age. The overall prevalence for male subjects (3.9%) with reported functional constipation was significantly higher (p = 0.003) compared to female subjects (2.1%). All subjects at 0—6 months of age with functional constipation had reported formed stool consistency as evaluated by the Amsterdam Stool Chart. For the constipated subjects 6 months and above, stool consistencies of very hard with separate hard lumps, hard, and dry and lumpy were reported according to the Bristol stool chart. Colic prevalence of 2.5% was observed in early infancy, which peaked between the period of 1—2 months. Infant dyschezia was only reported in early infancy (0.9%), with peak onset between 2—3 months. Functional diarrhoea was not reported in both infants and young children.

Supplementary table S1 shows early life factors and infant feeding practices against FGIDs outcome [see Supplementary file 2]. No association was found for infantile colic and dyschezia against gender, gestational age, birth weight, growth curve, mode of delivery and feeding. The risk of having infant regurgitation was significantly lower among infants with exclusive breastfeeding at 2 – 3 months (OR = 0.137, 95%CI = 0.028 – 0.675, p = 0.015), exclusive breastfeeding at 3 – 4 months (OR = 0.121, 95%CI = 0.028 – 0.516, p = 0.004) and formula initiation at 0 – 1 months (OR = 0.060, 95%CI = 0.010 – 0.378, p = 0.003). Being a male (OR = 3.665, 95%CI = 1.563 – 8.593, p = 0.003) and formula feeding initiation at 1 – 2 months (OR = 18.558, 95%CI = 1.569 – 219.444, p = 0.020) increased the risk of functional constipation.

With regards to feeding practices among infants 0 – 6 months old, a high percentage (95.2%) of breastfeeding initiation at birth was observed. The average duration of exclusive breastfeeding lasted up to 1.9 months, with the average starting age for formula feeding at 0.5 months. 71.8% of subjects between the age of 7 – 48 months reported the use of follow-on formula. In addition, 9.4% of the young children consumed additional vitamin and/or mineral supplements i.e. vitamin D, multi-vitamin, calcium and docosahexaenoic acid (DHA).

Supplementary table S2 shows the association between FGIDs and socio-demographic factors including residence area, household income, family size, birth order, and parental education level [see Supplementary file 3]. No risk factor was identified for infantile colic and dyschezia. Higher paternal education level (Master’s degree) was associated significantly with a reduced risk of infant regurgitation (OR = 0.022, 95%CI = 0.002 – 0.327, p = 0.006). The risk of functional constipation was significantly higher for children in family with annual household income between 273,000,000 – 546,999,999 VND or estimate around 11,800 – 23,800 USD (OR = 5.887, 95%CI = 1.794 – 19.321, p = 0.003) and family with one child only (OR =  > 999.999, 95%CI =  > 999.999—> 999.999, p < 0.001).

Supplementary table S3 shows the association between FGIDs and stressful events as faced by the main caregiver and the child [see Supplementary file 4]. Most of the subjects reported that neither they nor their children were exposed to any of the listed stressful events.