Seroprevalence of Helicobacter pylori in Hispanics living in Puerto Rico: A population‐based study

Abstract Background Helicobacter pylori is an important etiologic factor for peptic ulcers and gastric cancer, one of the top ten leading causes of cancer death in Puerto Rico. However, the prevalence of H. pylori infections in this population was previously unknown. The aim of this study was to examine the seroprevalence of H. pylori and its associated risk factors in Puerto Rico. Materials and Methods A cross‐sectional study was designed using an existing population‐based biorepository. Seropositivity was determined using the Premier™ H. pylori immunoassay. Helicobacter pylori seroprevalence was estimated with 95% confidence using marginal standardization following logistic regression. To assess the risk factors associated with H. pylori seropositivity, a multivariable log‐binomial model was fitted to estimate the prevalence ratio (PR) and its 95% confidence interval (95% CI). Results A total of 528 population‐based serum samples were analyzed. The mean age of the study population was 41 ± 12 years, of whom 55.3% were females. The overall seroprevalence of H. pylori was 33.0% (95% CI = 28.3%‐38.1%). Increasing age and having <12 years of education were significantly (P < .05) associated with H. pylori seropositivity in the multivariable model; however, residing in counties with low population density reached marginal significance (P = .085). Conclusions We report that H. pylori infection is common among Hispanics living in Puerto Rico. The H. pylori seroprevalence observed in Puerto Rico is similar to the seroprevalence reported in the overall population of the United States. The association between H. pylori seroprevalence and the risk factors analyzed offers insight into the epidemiology of gastric cancer in Puerto Rico and warrants further investigation.

which led to the classification of H. pylori as a class I human carcinogen by the International Agency for Research on Cancer (IARC). 5 In the United States (US), there are marked racial/ethnic disparities in gastric cancer, with higher incidence and mortality rates among Hispanics and other racial/ethnic groups compared to non-Hispanic Whites. 6 Substantial variability in H. pylori prevalence is observed globally according to geographic location, age, and socioeconomic status.
A birth cohort effect has been observed in H. pylori prevalence in most of the world. 7,8 The Centers for Disease Prevention and Control (CDC) estimate that the prevalence of H. pylori infection in developing countries is 70% and is 30%-40% in most industrialized countries.
In the US, the prevalence of H. pylori infection is approximately 31%, with variable prevalence among diverse racial/ethnic groups: 21% in Whites, 52% in African Americans, and 64% in Mexican Americans. 6,9 Marked differences in H. pylori infection have also been reported among Hispanic individuals from different regions of Central and South America. [10][11][12] Individuals with low levels of education and low socioeconomic status have reported to have more than 5.5-and 6.6-times higher odds of being infected with H. pylori. 13 Gastric cancer is the fifth most common incident cancer and the third leading cause of cancer death worldwide. 14 Gastric cancer incidence rates vary dramatically across regions and countries and are higher in less developed countries. In general, gastric cancer is more common in East Asia, Eastern Europe, and the mountainous regions of Latin America. 15,16 In Puerto Rico during 2008-2012, gastric cancer was the fifth leading cause of cancer death for men and the eighth cause for women. 17 The age-standardized incidence and mortality rates (per 100 000 population) were 9.0 and 6.1 for males, and 5.2 and 3.5 for females, respectively. 16 The specific risk factors associated with gastric cancer, including the prevalence of H. pylori infection among Hispanics living in Puerto Rico, are currently unknown. The aim of this study was to examine the seroprevalence of H. pylori and its associated risk factors in Hispanics living in Puerto Rico using samples from a population-based biorepository.

| Study design and population
We used a representative sample from an existing population-based biorepository of archived, frozen serum samples (n = 1645) from a seroepidemiologic survey of viral hepatitis and other infections in Puerto Rico. The study design has been previously described by Pérez et al. 18 In this study, a secondary analysis was performed using an estimated sample size of 528 subjects, which assumed a conservative H. pylori seroprevalence. In brief, the parent seroepidemiologic study "Epidemiology of Hepatitis C in the Adult Population of Puerto Rico" aimed to recruit noninstitutionalized individuals aged 21-64 years old residing in Puerto Rico at the time of the survey (2005)(2006)(2007)(2008). This study used a stratified, multistage, and probability cluster design of all households in Puerto Rico. The data collection instrument designed for this study was modeled after questionnaires previously used in household studies in Puerto Rico. The questions were based on the questionnaire developed by the ARIBBA (Alliance for Research in El Barrio, New York, and Bayamón) Project, a study that evaluated factors that influenced risk behaviors among Puerto Rican injection drug users. 19 Two field managers visited every selected occupied household, where subjects that agreed to participate in the study completed the following procedures: 1 informed consent and pretest counseling; 2 a personal interview to obtain information about sociodemographic characteristics, medical history, tattooing and body piercing practices, knowledge on viral infections, and self-report of hepatitis A and hepatitis B vaccination; 3 an audio computer-assisted self-interview to ascertain cigarette and alcohol use, drug use, sex-related risk behaviors, and history of incarceration; and 4 collection of blood and urine samples. Of the selected residents, 1654 (77.9%) participated in the study. Although the age distribution of the sample was similar to that of the adult population of Puerto Rico according to the Census 2000, females were slightly overrepresented as they made up 56.4% of the study sample.
The secondary analysis in this study was performed using an estimated sample size of 528 subjects, which assumed a conservative H. pylori seroprevalence of 20%, a precision of 3.5%, 95% confidence, and a sample loss of 5%. Serum samples were analyzed using the Premier™ H. pylori enzyme immunoassay (Meridian Bioscience, Inc. Cincinnati, OH, USA) according to the manufacturer's specifications.
This immunoassay was specifically designed for the in vitro qualitative detection of H. pylori IgG antibodies in human serum and plasma samples with a relative sensitivity of 99.2% and a relative specificity of 96.0%.

| Statistical analysis
Prevalence of H. pylori, overall and by sociodemographic characteristics, was estimated using marginal standardization following logistic regression. 22 Due to the complex sampling design, the prevalence estimation, and the simple and multivariate log-binomial models were weighted according to the probability of participation in each household block and the inverse of the probability of selection according to the geographic strata, household blocks, and sex distribution according to postcensal estimates in Puerto Rico. Simple log-binomial models were fitted to estimate the unadjusted prevalence ratio (PR) and its 95% confidence interval (CI) associated with each independent variable. The PR was used in this analysis as a measure of association because it is considered more conservative, consistent, and appropriate for cross-sectional studies compared to the prevalence odds ratio. 23,24 Variables were entered into the multivariate log-binomial model if the variables were known or hypothesized risk factors for H. pylori, and the P values associated with their regression coefficients were <.05. 25 All data were analyzed using Stata for Windows release 14.0 (Stata Corporation, College Station, TX, USA).

| Prevalence of H. pylori, overall and by sociodemographic characteristics
The mean age of the study participants from which the selected serum samples were collected in the parent study was 41 ± 12 years, of whom 55.3% were females ( Table 1). The majority of participants were married, had at least 12 years of education, and lived in lowdensity populated counties in Puerto Rico. The overall seroprevalence of H. pylori in this cohort was 33.0% (95% CI: 28.3%-38.1%).
Helicobacter pylori seroprevalence increased significantly with age; the highest seroprevalence (44.3%) was observed among participants 50-64 years old. A comparable seroprevalence was observed among men and women (35.7% and 30.9%, respectively). Higher H. pylori seroprevalence was detected among participants with <12 years of education (46.2%) and those that lived in low-density populated counties (41.0%).
The prevalence of H. pylori seropositivity was also significantly higher among those with <12 years of education and low population density in their residential county. After multivariable adjustment, these prevalence ratios were attenuated, but remained significant (PR = 1.4, 95% CI = 1.0-1.8) and marginally significant (PR = 1.3 95% CI = 1.0-1.6), respectively. Although H. pylori seroprevalence was significantly (P < .05) associated with marital status in the simple regression model, the prevalence ratio was attenuated and was not significant in the multivariable model.

| DISCUSSION
Helicobacter pylori is a major risk factor for the development of gastric cancer, the third leading cause of cancer death worldwide. 14 In Puerto Rico, gastric cancer is one of the top ten leading causes of cancer death 17  In conclusion, the seroprevalence reported in this study provides baseline information regarding H. pylori infection among Hispanics living in Puerto Rico and will serve as a basis for future studies of H. pylori-associated diseases, such as peptic ulcer disease and gastric adenocarcinoma. The H. pylori prevalence observed among this cohort may have an impact in the incidence of gastric neoplasia and may contribute to a disproportionate cancer burden among individuals with low socioeconomic status. Additional epidemiological studies have even more relevance in the aftermath of the hurricane Maria in Puerto Rico 43 and the possibility of an increase of H. pylori infections after this natural disaster. Infectious disease outbreaks are a major concern after a natural disaster given the limited access to food and safe drinkable water, overcrowded shelters, poor hygiene practices, exposition to wastewater, and inadequate access to medical care. 44 Studies have described an increase in H. pylori infections among individuals who survived natural disasters. 45 Therefore, a better understanding of the risk factors associated with H. pylori infection in this Hispanic population will be of the utmost importance given the established association between H. pylori and gastric cancer, and the high burden of this malignancy in Puerto Rico.