Helicobacter pylori infection is acquired during early childhood (1,2). The prevalence of the infection is higher in developing countries than developed countries (1,2). Low socioeconomic status (SES), low maternal education, and H pylori–infected family members are common risk factors for the infection (2–7). Differences in H pylori prevalence among ethnic groups within the same country or population have been reported (2,8–10). For example, in the United States, H pylori seroprevalence was 13% and 4% among 1- to 3-year-old black and white children, respectively, and at ages 18 to 23 years, it was 43% and 8% in blacks and whites, respectively (9). In Germany, the prevalence of H pylori infection was 4.8% in children with German nationality compared with 66.7% in the children with Turkish nationality (10). Many of these differences were attributed to disparities in SES.
Ethnicity is a complex variable that stands for both socioeconomic variables and social and cultural practices that may affect the transmission of H pylori infection. The Israeli population comprises 2 major ethnic groups, the Jewish and the Arab populations, which have distinguished cultural and socioeconomic features. In the present study, we examined the prevalence of H pylori infection in Israeli children and assessed the existence of interactions between ethnicity and SES in relation to H pylori seroprevalence during childhood and adolescence.
PATIENTS AND METHODS
The Israeli population comprises 2 major ethnic groups: the Jewish population (80%), of which 11% to 13.9% is estimated to be ultraorthodox religious, and the Israeli Arab population (20%) (11–13). The Arab population in Israel resides mainly in rural areas in towns or neighborhoods separated from the Jewish population. The population groups are distinguished by socioeconomic, demographic, and social features (11–13). Some of these characteristics are presented in Table 1.
Collection of Data and Sera
Sera from the serum bank of the Israel Center for Disease Control were used in the present study. The sera of the Arab children (N = 575) and of the Jewish children from the general population (N = 584) were collected from 2000 to 2001, and those of children belonging to an ultraorthodox religious Jewish community (N = 464) were collected from 1997 to 2007. The serum bank consists of samples obtained at 3 collection sites for children ages 0 to 20 years in community and hospital laboratories, spread across the country. The sera of the participants were from patients seeking medical attention, but not specifically for gastrointestinal symptoms. The samples consist of residual sera from diagnostic tests remaining after the required biochemical tests have been performed and designated to be discarded. The sera were obtained from inpatient and outpatient subjects. Collection of samples is ongoing, and the sampling covers all of the months of the year. The sampling does not include specimens taken from cases suspected of immunological disorders. Sera are kept frozen (−70°C) until tested. The samples do not carry the personal identification of the subjects. They have a unique identifier plus details of the sex, age, ethnicity (Arab or Jewish), and place of residence (at the level of town). For Jewish participants, family origin was determined by the father's country of birth (second-generation Israeli, Eastern-origin countries [Africa/Asia/South America], Western-origin countries [Europe/Russia/North America/Australia]).
The socioeconomic rank of the place of residence as defined by the Israel Central Bureau of Statistics (14) was used as an indicator for SES. The ranks were on a scale of 1 to 10; the lower the rank, the lower the SES. This SES composite index reflects a combination of basic characteristics such as financial resources, housing, density, motorization, education, employment profile, and so on (14).
Localities, towns, and cities at socioeconomic ranks of 1 to 3 were classified as low SES, 4 to 6 socioeconomic ranks were classified as intermediate SES, and ranks of 7 or higher were classified as high SES. The study was approved by the institutional review board of Tel Aviv University.
Detection of Anti-H pylori IgG Antibodies
H pylori–specific immunoglobulin G (IgG) antibodies were measured in the serum samples using Dade Behring's (Miami, FL) Enzygnost anti-Helicobacter pylori II/IgG commercial kit according to the manufacturer's instructions. The kit's sensitivity and specificity in children are 92.7% and 95.7%, respectively. In children younger than 6 years, the respective values are 91.6% and 97.1% (15). Sera with results ≥0.250 were considered seropositive to H pylori IgG antibodies.
Detection of Anti-CagA Antibodies
A random subsample of seropositive sera to anti–H pylori IgG antibodies (N = 156) was tested for the presence of IgG antibodies to CagA-H pylori strains. The test was performed using the CagA IgG enzyme-linked immunosorbent assay kit (Genesis Diagnostics, Lansing, MI) following the manufacturer's instructions.
The differences in H pylori seropositivity according to age, sex, population group, and SES were examined using the χ2 test. H pylori seroprevalence and 95% confidence intervals (CIs) were calculated. Multivariate analyses using logistic regression models were conducted to determine the correlates of H pylori sero-positivity and to assess the presence of interactions between ethnicity and SES with H pylori positivity. Adjusted odds ratio (OR) and 95% confidence intervals (CIs) were derived from these models. Two-tailed P < 0.05 was considered statistically significant. The data were analyzed with SPSS version 17 (SPSS Inc, Chicago, IL).
A total of 1623 serum samples were included in the study. The samples belonged to children with a mean age of 8.2 years (standard deviation [SD] 5.4, range 0–20 years); 847 (52.2%) samples were taken from boys.
There were 513 (31.6%, 95% CI 29.4–33.9) H pylori sero-positive samples. H pylori seropositivity was similar among boys and girls: 30.0% (95% CI 27.0–33.2) and 33.4% (95% CI 30.2–36.8), respectively (P = 0.14). H pylori seropositivity was 22.9% (95% CI 19.7–26.5) among Jewish children from the general population, 25.2% (95% CI 21.5–29.4) among ultraorthodox Jewish children compared with 45.6% (95% CI 41.5–49.7) among Arab children (P < 0.001). H pylori seropositivity was inversely associated with SES: 37.1%, 31.0%, and 16.9% among participants of low, intermediate, and high SES, respectively (P for trend < 0.001). The prevalence of H pylori increased with age in the 3 study groups, and it was consistently higher among Arab children (Fig. 1). Data on family origin were available for 268 (45.9%) Jewish children. H pylori seropositivity was higher among children of Eastern countries origin: 29.7% compared with 21.7% in children of Western countries origin, and 13.5% in second-generation Israeli children (P = 0.03).
In a pooled multiple logistic regression analysis, Arab participants had >2-fold increased likelihood of H pylori seroprevalence compared with Jewish ones (adjusted OR 2.16, 95% CI 1.67–2.79). In children who belonged to low SES and intermediate SES, the likelihood of H pylori seropositivity increased by 2.5- and 1.8-fold, respectively, as compared with children of high SES. H pylori seropositivity increased significantly with age (model 1, Table 2). A significant interaction between SES and ethnicity was found while controlling for age and sex; when compared with Jewish participants from high SES, the likelihood of H pylori seropositivity significantly increased by 2.03-fold among Jewish children from intermediate SES, whereas in Arab children from intermediate SES, it was increased by 2.42-fold (model 2, Table 2). Similarly, compared with Jewish participants from high SES, the OR for H pylori seroprevalence in Jewish children from low SES was 2.26, whereas among Arab children from low SES, it was 5.72 (model 2, Table 2).
CagA seropositivity was determined in a random sample of 156 (30%) H pylori–positive children. Overall CagA seropositivity was 42.9% (95% CI 35.4–50.8). CagA seropositivity was 45.0% and 40.8% in boys and girls, respectively (P = 0.59), and it was 40.8% and 45.0% in Jewish and Arab children, respectively (P = 0.59). The mean age of CagA seropositive children and CagA seronegative children was similar: 9.6 (SD 6.5) and 9.9 (SD 5.3), respectively (P = 0.74). CagA seropositivity was 50.0% in children from low SES, 41.2% in children from intermediate SES, and 30.3% in children from high SES (P for trend 0.05). The age-adjusted OR for CagA seropositivity was 2.31 (95% CI 0.95–5.65) (P = 0.06) in children from low SES and 1.60 (95% CI 0.63–4.07) (P = 0.31) in children from intermediate SES, as compared with children from high SES.
We examined the seroprevalence and correlates of anti-H pylori IgG antibodies in a large sample of Israeli children and adolescents. The overall H pylori seroprevalence was 32%; it was higher in Arab children (46%) than in Jewish children (23%–25%). H pylori seropositivity among Jewish children is similar to that reported among children from developed countries, whereas in Arab children, the prevalence is comparable with that observed in developing countries (1,2). In common with previous reports (6–9), the prevalence of H pylori increased with age among both Arab and Jewish children, but it was consistently higher among Arab children, suggesting higher transmission rates in the former group. The great difference in H pylori seroprevalence among adolescents and young children may in part be explained by birth cohort effect rather than by late acquisition of the infection; however, given the nature of prevalence data, we cannot draw firm conclusions. The likelihood of the infection was inversely related to SES, regardless of ethnicity and age. These findings are in agreement with previous studies (5,7). Ethnic differences in the prevalence of H pylori have been reported (7–10), and they were attributed to the differences in SES. In the present study, the ethnic differences in H pylori seroprevalence persisted even after controlling for SES and age. Therefore, socioeconomic disparities between ethnic groups may only partially explain such findings. Furthermore, our findings showed a significant interaction between ethnicity and SES and H pylori seroprevalence; that is, in the same SES strata, while controlling for age and sex, the likelihood of the infection was still much greater among Arab than Jewish participants, indicating that there are variables other than SES that account for the ethnic differences in H pylori prevalence. It is possible that the proportion of H pylori–infected family members, who are believed to be the main reservoir of H pylori (3,4,16), is higher in the Arab population than in the Jewish population. In a previous study that we carried out in Arab villages, we found that the risk of H pylori infection among preschool-age and school-age children was significantly increased if they had a sibling infected with H pylori(4,16). In the study mentioned above, we found 82% H pylori seropositivity among mothers (mean age 32.5, SD 5.1 years) of the participants (4). Among Jewish women ages 20 to 39 years, H pylori seropositivity was 49% (unpublished data). In the present study, H pylori seropositivity in infants, which is assumed to be the result of maternal antibodies, was 3-fold higher in Arab than in Jewish infants, a finding that may mirror differences in maternal H pylori infection.
Family size and household density in the Arab population are greater than in the Jewish population and this also may increase the chances of H pylori transmission among the Arab population. This may be true only to a certain extent, however, because the prevalence of H pylori in Arab participants was higher than that in children who belonged to an ultrareligious Jewish community, which is characterized by high fertility rates, extended families, and low SES (12). It may be possible that the religious hand-washing before touching food rules in the Jewish ultraorthodox community contribute to the difference in the risk of H pylori transmission in these 2 crowded populations.
CagA is one of the main virulence factors involved in the pathogenesis of H pylori infection (1,17). In a subsample of H pylori seropositive children, we found high CagA seropositivity of 43%. A previous study from Israel reported lower CagA prevalence of 25% among H pylori isolates of symptomatic children who were investigated for gastrointestinal symptoms (18). The serologic examination that we used reflects cumulative exposure to CagA-positive strains rather than current infection with CagA strains, and this may in part explain the higher prevalence in our study. CagA seropositivity was greater in children from low SES as compared with children from high SES communities. It is possible that the chances to be infected with CagA strains grow in low SES communities because of higher transmission rates of H pylori infections in these settings.
The present study has several limitations. The sampling frame consisted of residual sera taken from clinical samples. This may overestimate the prevalence of H pylori infection, because some of these children may have been referred for investigation for conditions related to H pylori infection. This is not expected to affect the observed associations between ethnicity and SES with H pylori seroprevalence because all of the population groups in Israel have similar access to health services based on the National Health Law. The sera from Arab and Jewish children from the general population were collected 10 years ago. The samples that belonged to children from the ultraorthodox Jewish community were collected from 1997 to 2007; SES and the prevalence of H pylori infection can change in a 10-year period. This may not be the case, however, because, in essence, this community does not appear to change its behavior. The age-adjusted prevalence of H pylori infection in sera that were collected from 1997 to 1999, 2000 to 2001, and 2002 to 2007 was 25.6%, 20.6%, and 24.1%, respectively, suggesting neither significant difference in H pylori seroprevalence over time in this group nor a trend of decline in seroprevalence. In addition, the SES rank of residence place was used as a marker of the individual-level SES. This may induce nondifferential misclassification of SES, which leads to underestimation of the association measurement.
In conclusion, H pylori seropositivity among Israeli children and adolescents is determined by age, ethnicity, socioeconomic variables, and the interaction between them. Socioeconomic disparities may not totally explain the ethnic differences in H pylori prevalence. CagA seropositivity is especially high in low SES settings, and this may have public health and clinical implications.
1. Suerbaum S, Michetti P. Helicobacter pylori infection. N Engl J Med 2002; 347:1175–1186.
2. Torres J, Perez-Perez G, Goodman KJ, et al. A comprehensive review of the natural history of Helicobacter pylori infection in children. Arch Med Res 2000; 31:431–469.
3. Weyermann M, Rothenbacher D, Brenner H. Acquisition of Helicobacter pylori infection in early childhood: independent contributions of infected mothers, fathers, and siblings. Am J Gastroenterol 2009; 104:182–189.
4. Muhsen K, Athamna A, Athamna M, et al. Prevalence and risk factors of Helicobacter pylori infection among healthy 3- to 5-year-old Israeli Arab children. Epidemiol Infect 2006; 134:990–996.
5. Ozen A, Ertem D, Pehlivanoglu E. Natural history and symptomatology of Helicobacter pylori in childhood and factors determining the epidemiology of infection. J Pediatr Gastroenterol Nutr 2006; 42:398–404.
6. Malaty HM, Logan ND, Graham DY, et al. Helicobacter pylori infection in preschool and school-aged minority children: effect of socioeconomic indicators and breast-feeding practices. Clin Infect Dis 2001; 32:1387–1392.
7. Opekun AR, Gilger MA, Denyes SM, et al. Helicobacter pylori infection in children of Texas. J Pediatr Gastroenterol Nutr 2000; 31:405–410.
8. Malaty HM, Graham DY, Wattigney WA, et al. Natural history of Helicobacter pylori infection in childhood: 12-year follow-up cohort study in a biracial community. Clin Infect Dis 1999; 28:279–282.
9. Malaty HM, El-Kasabany A, Graham DY, et al. Age at acquisition of Helicobacter pylori infection: a follow-up study from infancy to adulthood. Lancet 2002; 359:931–935.
10. Rothenbacher D, Bode G, Berg G, et al. Prevalence and determinants of Helicobacter pylori infection in preschool children: a population-based study from Germany. Int J Epidemiol 1998; 27:135–141.
11. Central Bureau of Statistics. Statistical Abstracts of Israel 2010. Publication number 61. http://http://www.cbs.gov.il/reader/shnatonhnew_site.htm
. Accessed April 6, 2010.
12. Gurovich N, Cohen-Kastro E. Ultra-orthodox Jews Geographic Distribution and Demographic, Social and Economic Characteristics of the Ultra-orthodox Jewish population in Israel 1996–2001. Jerusalem: Central Bureau of Statistics; 2004.
13. Measurement and Estimates of the Population of Ultra-orthodox Jews. Jerusalem: Central Bureau of Statistics; 2011.
14. Characterization and Classification of Local Authorities by the Socioeconomic Level of the Population 2001. Jerusalem: Central Bureau of Statistics; 2004.
15. Kindermann A, Konstantopoulos N, Lehn N, et al. Evaluation of two commercial enzyme immunoassays, testing immunoglobulin G (IgG) and IgA responses, for diagnosis of Helicobacter pylori infection in children. J Clin Microbiol 2001; 39:3591–3596.
16. Muhsen K, Athamna A, Bialik A, et al. Presence of Helicobacter pylori in a sibling is associated with a long-term increased risk of H. pylori infection in Israeli Arab children. Helicobacter 2010; 15:108–113.
17. Graham DY, Yamaoka Y. Disease-specific Helicobacter pylori virulence factors: the unfulfilled promise. Helicobacter 2000; 5 (suppl 1):S3–S9.
18. Benenson S, Halle D, Rudensky B, et al. Helicobacter pylori genotypes in Israeli children: the significance of geography. J Pediatr Gastroenterol Nutr 2002; 35:680–684.