Helicobacter pylori (H. pylori) infection is a common chronic infection, affecting more than 50% of the world’s population [1]. The prevalence has changed over the last 10 years, showing a decrease in developed countries but remaining high in most developing countries [1, 2]. Like other developing countries in Southeast Asia, the prevalence of H. pylori in Vietnam remains high. A recent study in Hanoi in the North of Vietnam showed that the prevalence of the infection was 76.8% [3]. There was no such community-based prevalence study on H. pylori infection in Ho Chi Minh City (HCMC), despite it is one of the most overcrowded cities globally with a population of 8.933.082 in total and a population density of 12,000 p/km². Its population density is five times higher than Hanoi (2,455 p/km²), three times higher than Shanghai city (3,800 p/km²), eighty times higher than the average of Asia (150 p/km²) [4, 5]. Furthermore, the link between H. pylori infection and crowded living conditions has been shown in several studies [6, 7]. Therefore, assessing the prevalence of H. pylori in HCMC is an essential public health issue in HCMC.

H. pylori infection is usually acquired during childhood and tends to persist if untreated [8–10]. H. pylori infection is a risk for duodenal-gastric ulcers, gastric cancer, and its treatment has been proved to decrease cancer risk in individuals with a family history of gastric cancer [11, 12]. Therefore, reducing H. pylori infection rate in the young population is crucial for decreasing the burden of gastric cancers in future.

We therefore conducted a study to estimate the prevalence of H. pylori infection and its geographical spread in school-aged children across the 24 districts of HCMC, Vietnam.

A total of 1854 pupils were invited to participate but 20.3% refused. This refusal rate was similar across districts. Eighteen pupils were excluded (4 children who did not perform the stool test, 5 children were using antibiotics in the last 4 weeks and 9 children had a missing age or gender). The remaining 1,460 pupils were included in the present analysis. Of these, 730 (50%) were boys, and the mean age was 10.1 ± 2.7 years with a range of 6–15 years (Table 1).

The overall prevalence of H. pylori infection was 87.7% (1,280/1,460). The prevalence was significantly higher in boys (90.0%, χ2 test with p = 0.003), and in children aged 9–11 years (90.7%, χ2 test with p = 0.007). The prevalence of H. pylori infection according to the age groups and gender is reported in Figure 2. In both sexes, the prevalence increased with age, peaked in the 9–11 age group, and then decreased when the child was 12 years or more.

Figure 3 illustrates a linear increasing trend in the prevalence of H. pylori infection across the four-points index of crowd in HCMC (80.5% in the rural area, 88.5% in the peri-urban area, 89.3% in the urban area, 90.9% in the super-urban area, Cochran χ2 test with p < 0.001).

There was no interactions between the four points index and age group or gender by using likelihood ratio χ2 test (p = 0.64). When splitting according to age subgroups and to gender, this increasing linear trend maintained (Figure 4). Prevalence of H. pylori infection increased significantly as PD or PE increased within both age and gender subgroups (p = 0.02).

Table 2 shows the results from the logistic regression performed to assess demographic factors associated with H. pylori infection. Multiple logistic regression analysis showed that age, gender, and living area were significantly related to H. pylori infection. The prevalence of H. pylori was significantly increased in children aged 9–11 years [OR = 1.91, 95% Cl: 1.29–2.85, p = 0.003] and in boys [OR = 1.5, 95% Cl: 1.08–2.06, p = 0.015]. A multivariate analysis also clearly showed that the prevalence of H. pylori infection significantly increased with the level of crowded: PD < 20,000 p/km² and ED < 35,000 p/km² [OR = 1.85; 95% Cl: 1.19–2.88, p < 0.001]; PD 20,000–35,000 p/km² and ED < 35,000 p/km² [OR = 2.05; 95% Cl: 1.10–2.09, p < 0.001]; and PD ≥ 35,000 p/km² or ED ≥ 35,000 p/km² [OR = 2.33; 95% Cl: 1.44 – 3.10, p < 0.001].

Our study is the first research that reports the current prevalence of H. pylori infection among school-aged children in HCMC. The overall prevalence of H. pylori in the representative sample of pupils in HCMC was 87.7%. Several studies on the prevalence of H. pylori infection in Vietnamese children ranged from 32.1% in the Mekong region [17] to 55.5% in Nghe An [18]. The prevalence observed in our study was higher compared with those reported in the previous studies, reflecting the increasing trend of H. pylori infection over the past decades. However, most of those previous studies were conducted in the countryside or mountainous areas, had a small sample size, and did not address all school-aged children. Moreover, all previous reports used sera diagnostic methods known to have low accuracy and could not differentiate a lifetime infection from an active infection. While in our study, all pupils at all grades in primary and secondary schools in 24 districts of HCMC were recruited. In addition, the Premier Platinum HpSA Plus stool test (Meridian Bioscience, USA) used in our study has proved a reliable tool to detect H. pylori infection with an accuracy of 93.4% [17, 18]. Nguyen TVH et al. conducted a validation study of the Premier Platinum HpSA PLUS test (Meridian bioscience, USA) in 232 Vietnamese children and reported a sensitivity of 97% and specificity of 95% [19]. Therefore, our results reflect an accurate infection prevalence of school-aged children in HCMC.

Several studies in Japan and Taiwan reported that H. pylori infection in children decreased in each age group [20–23]. In contrast, two studies from China showed that the infection increased with age [24, 25]. In our study, the prevalence of H. pylori infection increased with age up to 11 years, and then began to decline at 12 years of age in both boys and girls (Figure 2). These evidences suggest that the incidence rate of H. pylori across age groups varies greatly in different countries and may depend on different socio-economic statuses, living conditions, and lifestyle factors. Furthermore, the prevalence of H. pylori in HCMC was already high in the youngest ones, indicating that the acquisition of H. pylori occurs in very early childhood. Therefore, the building of programs to prevent and reduce the incidence of H. pylori has to be prioritized in public health policies.

Although HCMC takes up just 10% of the Southeast (SE) region’s land area (2,061.2/23,564.24 km²), it concentrates up to 50% population (8,993.082/17,828.907 inhabitants) of the SE region with PD of 4,292 p/km² [5, 13]. However, it is widely believed that the population has been seriously underestimated. Indeed, one study was conducted to estimate the actual population of HCMC by counting motorcycles, which is the main transport in HCMC and 90% of household-owned motorcycles, showed that the actual population was higher than three times the officially counted [4]. Therefore, the actual PD of HCMC could be up to 12,000 p/km², which means HCMC is an overcrowded city. Furthermore, several studies showed that the H. pylori infection is related to the crowded level. In order to control for all these qualities, we decided to build a 4-points index for the geographical division of HCMC mainly based on population density to have an insight on the level of crowds across the 24 districts of the city. And this index could aid us in explaining the H. pylori infection distribution in HCMC. During the process, we realized that some of the city center districts had relatively low PD compared to others, despite they are the busiest districts, where most offices and buildings are located, and almost all employees come there to work every day. Therefore, using both PD and PE to define the crowdy index woud be more reliable to find the possible link between H. pylori infection and the crowd level of HCMC.

By using the 4-points index, our study showed that there was a linear increasing trend in the prevalence of the infection across four areas in HCMC (Figure 3), and its trend was also found in both age or genders subgroup (Figure 4). Furthermore, the crowded index did not correlate with age group and gender by using likelihood ratio chi-square test (p = 0.64). That means the prevalence significantly increased as population density or employees density increased and is the same way whatever the age or gender subgroup. Indeed, a study performed in the North of Vietnam found that the higher crowded living area was a positive risk factor for H. pylori infection [3]. Additionally, reports from China [24], Japan [21], Jordan [26], and Nepal [27] also observed that the infection rate significantly increased in people living in crowded areas. However, these studies stratified the living area only according to population density whereas our study developed the 4-point index based on both population density and employees density as previously described and explained. That crowdy index can create a more comprehensive understanding of the relation between H. pylori infection and the level of populated concentration in HCMC. Therefore, designing the programs to prevent and reduce the H. pylori infection in HCMC should focus on the highly concentrated population areas.

The differences in the target population, the diagnostic tests, and the cut-off values of the test cause some difficulties when comparing our results to the findings in other countries. Using the same stool antigen test, a cross-sectional study in Portugal published in 2011 reported that the prevalence of H. pylori infection in children (0–15 years old) was 32% [28]. The survey conducted in China, among all children aged (0–15 years) published in 2020, reported the prevalence was 32.6% [24]. Another cross-section study conducted in Thailand in 2009 reported that the prevalence in children (5–7 years old) was 44.8% [29].

At the early stage of the study, the required sample size was 1854 pupils after applying the appropriate formula, as demonstrated in the study design part. In fact, we had only 1,478 pupils were enrolled in the study, which was smaller than the initial number. However, we were assumed that the loss rate of 20% at the beginning of the study development. That means 1,483 children were required to enroll in our study. Therefore, our sample size was large enough to find a difference.

The study has several strengths included this was an all school-ages community-based study that was the first to be conducted in HCMC. The sample size was large with 1854 pupils could represent the sample of pupils of HCMC. Moreover, the stool test used to detect the infection status is highly accurate and also have validated for Vietnamese children with high sensitivity and high specificity [19].