Determinants of Helicobacter pylori infection and eradication failure among healthcare workers in a high-exposure clinical setting: a single-centre retrospective cohort study in China

Abstract

Objective

Healthcare workers (HCWs) are at potential risk of Helicobacter pylori (H. pylori) infection due to occupational exposure, yet epidemiological data in this population remain scarce. This study aimed to investigate the prevalence, risk factors, eradication efficacy and antibiotic resistance patterns of H. pylori infection among HCWs.

Design

Retrospective cohort study.

Setting

A tertiary hospital in Beijing, China.

Participants

725 HCWs who underwent H. pylori testing via urea breath test between January 2020 and December 2023.

Primary and secondary outcome measures

Demographic characteristics, lifestyle factors, symptomatic presentations, treatment regimens and antibiotic resistance profiles were extracted from electronic medical records. Univariate analyses using χ² tests identified associations between variables, followed by multivariate logistic regression to adjust for confounders and determine independent predictors of H. pylori eradication failure.

Results

The overall H. pylori positivity rate was 22.9% (166/725). Male gender (OR=1.76, 95% CI 1.20 to 2.59) and alcohol consumption (OR=2.40, 95% CI 1.26 to 4.58) were identified as independent risk factors. Notably, H. pylori-positive individuals had a lower incidence of acid reflux symptoms (9.1% vs 17.3%, p=0.011). The eradication rate was 92.0% (95% CI 84.3% to 96.7%), but clarithromycin resistance reached 47.1% (95% CI 36.4% to 58.0%), with 12.8% (95% CI 6.6% to 21.5%) showing dual resistance to clarithromycin and quinolones.

Conclusions

Targeted screening for male HCWs and alcohol users, along with bismuth-containing quadruple therapy, is recommended. High clarithromycin resistance underscores the need for local antibiotic stewardship.

STRENGTHS AND LIMITATIONS OF THIS STUDY.

• Data integration from three distinct, validated sources provided a multidimensional assessment of exposures and outcomes.

• The use of objective measures and standardised checklists minimised measurement bias and ensured consistency.

• Strict and prespecified criteria enhanced internal validity by defining a homogeneous cohort of healthcare workers.

• The single-centre retrospective design may introduce selection bias and limit the generalisability of the findings.

• The reliance on self-reported questionnaires for lifestyle exposures introduces the potential for information bias.

Introduction

Helicobacter pylori (H. pylori), a gram-negative bacterium colonising the human gastric mucosa, remains one of the most prevalent chronic bacterial infections worldwide.¹ Current estimates indicate that over 50% of the global population is infected with H. pylori, with significant geographical disparities.² H. pylori infection is a major aetiological factor for chronic gastritis, peptic ulcer disease, gastric mucosa-associated lymphoid tissue lymphoma and gastric adenocarcinoma.³ The WHO classifies H. pylori as a group I carcinogen,⁴ underscoring its role in gastric carcinogenesis. Beyond its direct health impacts, H. pylori infection imposes substantial economic burdens, including costs related to diagnosis, treatment and management of complications.

Healthcare workers (HCWs) represent a unique population at heightened risk for infectious diseases due to occupational exposure. While previous research has extensively explored bloodborne pathogens⁵ and respiratory infections,⁶ the risk of H. pylori transmission in healthcare settings remains under-recognised. HCWs face recurrent exposure to H. pylori primarily through two mechanisms: direct patient contact during endoscopic procedures, oral care or handling of gastric secretions, which may facilitate bacterial transmission⁷; and environmental contamination, as H. pylori retains viability on surfaces for extended periods, posing transmission risks in high-touch clinical environments.^{8 9} Emerging evidence suggests that HCWs exhibit higher H. pylori levels compared with the general population.¹⁰

Despite the clinical and occupational significance of H. pylori infection, its epidemiology among HCWs remains inadequately characterised due to pervasive methodological limitations in existing studies. Many studies are constrained by small-scale cross-sectional designs that overlook correlations with lifestyle variables, occupational exposures or antibiotic resistance trends.11,13 Geographical bias further obscures the true burden, as data from low-resource regions, where H. pylori prevalence and occupational hazards are highest, remain scarce. Additionally, asymptomatic infections, which account for up to 70% of H. pylori carriers, are systematically neglected in current screening paradigms, potentially perpetuating silent transmission reservoirs within high-risk HCW populations.^{14 15}

Given the elevated occupational exposure risks, diverse clinical manifestations of H. pylori infection and insufficient robust data among HCWs, this study systematically investigates H. pylori epidemiology, risk determinants and therapeutic outcomes, which aims to inform evidence-based occupational health policies, enhance targeted screening strategies and optimise antimicrobial stewardship protocols in high-risk clinical environments.

Methods

Study design

This single-centre retrospective cohort study adhered to the applicable 2007 Strengthening the Reporting of Observational Studies in Epidemiology guidelines¹⁶ and was conducted from January 2020 to December 2023.

Data collection

Data were systematically collected from three validated sources: (1) institutional occupational health databases, providing demographic and occupational variables (age, gender, body mass index (BMI), marital status and departmental affiliation); (2) self-administered questionnaires completed during annual health evaluations, which captured lifestyle parameters, including smoking status, alcohol consumption and frequent dining out and (3) clinical data were extracted from the electronic medical record (EMR) system of the Civil Aviation General Hospital (Beijing, China). Symptom profiles for all participants were prospectively assessed during their annual health examinations using the Gastrointestinal Symptom Rating Scale.¹⁷ In addition, detailed histories of prior H. pylori eradication regimens were retrieved from the same EMR system. Self-reported race or ethnicity data were not collected as the study population was homogenous in this regard, being exclusively recruited from a single centre in Beijing, China.

Participants

The inclusion criteria were: (1) full-time or part-time HCWs; (2) H. pylori infection was evaluated using the ¹³C/¹⁴C-urea breath test (UBT), which was administered universally to all HCWs during health checks. For the ¹³C-UBT, eradication was defined as a Delta Over Baseline of less than 4, while for ¹⁴C-UBT, which was defined as disintegrations per minute of less than 100¹⁸; (3) complete records for demographics, lifestyle factors and clinical symptoms. Patients were excluded if they met any of the following criteria: (1) under 18 or over 70 years of age and (2) incomplete records. A total of 1185 HCWs at our institution underwent the ¹³C/¹⁴C-UBT. Based on predefined inclusion/exclusion criteria, 725 HCWs (61.2%) were stratified into two cohorts: H. pylori-negative (n=559, 77.1%) and H. pylori-positive (n=166, 22.9%). H. pylori-positive HCWs will receive a 14-day bismuth-containing quadruple therapy (comprising a proton pump inhibitor, bismuth and two antibiotics) as the primary empirical treatment for H. pylori eradication, in accordance with the standard antibiotic regimens and dosages specified in the Fifth Chinese National Consensus Report on the management of H. pylori infection¹⁹; H. pylori status will be evaluated via ¹³C/¹⁴C-UBT performed 4–8 weeks after therapy completion.

Statistical analysis

All analyses were performed using SPSS V.25.0 (IBM, SPSS). Continuous variables were reported as mean±SD and compared using Student’s t-test, categorical variables were reported as frequencies (%) and analysed with χ² or Fisher’s exact tests. Variables with significant univariable associations were subsequently entered into a backward stepwise logistic regression to build the final multivariable model. The final model retained age, gender, alcohol use and department. Prior to regression, multicollinearity was assessed via the variance inflation factor, with all variables exhibiting a value below 5, indicating absence of significant multicollinearity. Model fit was rigorously evaluated: the Omnibus test confirmed the model’s statistical significance, the Hosmer-Lemeshow test demonstrated good calibration and the Nagelkerke R² quantified the proportion of variance explained. A two-sided p<0.05 defined statistical significance.

Patient and public involvement

Patients and/or the public were not involved in the design, conduct, reporting or dissemination plans of this research.

Results

Prevalence and demographic characteristics of H. pylori infection

As shown in table 1, the analysis of demographic factors revealed significant gender-based differences in H. pylori infection rates, with males showing a higher prevalence compared with females (34.3% vs 22.0%, p=0.001). This association was further supported by multivariate logistic regression (OR=1.76, 95% CI 1.20 to 2.59, p=0.004), indicating that male sex is an independent risk factor. No significant differences were observed in age (mean: 39.44±12.32 vs 39.16±12.06, p=0.801) or BMI (23.32±3.70 vs 23.59±3.43, p=0.400) between the H. pylori-negative and H. pylori-positive groups. Marital status (p=0.785) and departmental distribution (p=0.411) also showed no statistically significant association with infection status. These findings suggest that gender is the primary demographic factor influencing H. pylori infection risk in this cohort, while age, BMI, marital status and occupational department do not play significant roles.

Table 1. H. pylori prevalence and demographic characteristics.

Variable	H. pylori-negative (n=559)	H. pylori-positive (n=166)	χ2	P value
Gender			10.432	0.001
Male	123 (22.0)	57 (34.3)
Female	436 (78.0)	109 (65.7)
Age (years)	39.44±12.32	39.16±12.06	0.253	0.801
BMI (kg/m²)	23.32±3.70	23.59±3.43	−0.841	0.400
Marital status			0.074	0.785
Married	422 (76.0)	123 (75.0)
Unmarried	137 (24.5)	43 (25.9)
Departments			7.175	0.411
Logistics	18 (3.2)	11 (6.6)
Emergency	90 (16.1)	25 (15.1)
Internal medicine	113 (20.2)	33 (19.9)
Surgery	147 (26.3)	42 (25.3)
Medical	66 (11.8)	12 (7.2)
Administrative	63 (11.3)	23 (13.9)
Others	62 (11.1)	20 (12.0)

Categorical variables are expressed as the number of individuals, accompanied by percentages in parentheses, while continuous variables are presented as the mean±SD.

BMI, body mass index; H. pylori, Helicobacter pylori.

Lifestyle factors and clinical symptoms

The analysis of lifestyle factors is shown in table 2, which revealed a significant association between alcohol use and H. pylori infection, with a higher prevalence in the positive group (10.4% vs 5.0%, p=0.013), further supported by multivariate analysis (OR=2.40, 95% CI 1.26 to 4.58, p=0.008). Smoking (p=0.528) and irregular meal patterns (p=0.912) showed no significant association. Among clinical symptoms, the absence of acid reflux was linked to reduced infection risk (9.4% vs 17.2%, p=0.011), aligning with multivariate results (OR=0.49, 95% CI 0.27 to 0.88, p=0.018). Other symptoms, including epigastric pain (p=0.091), heartburn (p=0.121) and constipation (p=0.305), demonstrated no statistically significant differences between groups. These findings highlight alcohol consumption as a key modifiable risk factor and suggest a potential inverse relationship between H. pylori infection and acid reflux symptoms.

Table 2. Lifestyle and clinical symptoms between two groups.

Variable	H. pylori-negative (n=559)	H. pylori-positive (n=166)	P value
Lifestyle
Alcohol use	28 (5.0)	17 (10.4)	0.013
Smoking	30 (5.4)	11 (6.7)	0.528
Irregular meals	90 (16.2)	26 (15.7)	0.912
Symptoms
Acid reflux	96 (17.2)	15 (9.4)	0.011
Epigastric pain	54 (9.7)	9 (5.4)	0.091
Heartburn	88 (15.7)	18 (10.8)	0.121
Constipation	80 (14.3)	29 (17.5)	0.305

Categorical variables are expressed as the number of individuals, accompanied by percentages in parentheses.

H. pylori, Helicobacter pylori.

Multivariate analysis of risk factors

The multivariate analysis identified three independent factors significantly associated with H. pylori infection. As shown in table 3, male sex exhibited a 1.76-fold increased risk of infection compared with females (95% CI 1.20 to 2.59, p=0.004). Alcohol use (an average of more than two alcoholic drinks per day for men or more than one drink per day for women)²⁰ was strongly linked to infection, with drinkers having 2.4 times higher odds than non-drinkers (95% CI 1.26 to 4.58, p=0.008). Conversely, the absence of acid reflux emerged as a protective factor, reducing infection risk by 51% (OR=0.49, 95% CI 0.27 to 0.88, p=0.018). These results confirm that male gender and alcohol consumption are key modifiable risk factors, while the lack of acid reflux may reflect altered gastric physiology in H. pylori-positive individuals.

Table 3. Multivariate logistic regression analysis.

Variable	SE	Wald	OR	95% CI	P value
Male	0.197	8.263	1.76	1.20 to 2.59	0.004
Alcohol use	0.330	7.006	2.40	1.26 to 4.58	0.008
No acid reflux	0.299	5.637	0.49	0.27 to 0.88	0.018

Eradication outcomes and antibiotic resistance

The overall H. pylori eradication efficacy was 92.0% (80/87). Antibiotic resistance analysis revealed clarithromycin resistance in 47.1% (41/87) of cases, quinolone resistance in 22.1% (19/86) and dual resistances to both clarithromycin and quinolone in 12.8% (11/86).

Discussion

The findings of this retrospective study provide critical insights into the epidemiology, risk factors, clinical manifestations and therapeutic challenges of H. pylori infection among HCWs. The overall H. pylori prevalence in our HCW cohort was 22.9%, significantly lower than the estimated 50% in the general population of the same region.^{21 22} The discrepancy may reflect differences in occupational hygiene practices.

HCWs are rigorously trained in infection control measures, including hand hygiene, use of personal protective equipment and disinfection protocols, which may reduce H. pylori transmission. In developed nations like the USA, H. pylori prevalence ranges from 20% to 30%, while rates in developing countries often exceed 70%.²³ Previous studies have shown that improved sanitation and socioeconomic status correlate with declining H. pylori prevalence,^{24 25} aligning with our observation. However, the lower prevalence in our cohort might also stem from underdiagnosis due to asymptomatic infections, a hypothesis supported by our finding of reduced acid reflux symptoms in H. pylori-positive individuals.²⁶ Our HCW prevalence mirrors that of high-income populations, suggesting that occupational hygiene practices may mitigate risks even in endemic regions. This underscores the importance of institutional infection control programmes as a modifiable factor in H. pylori transmission.

Male HCWs exhibited a 34.3% infection rate, significantly higher than females (22.0%, p=0.001). A previous study indicated that behavioural factors, such as alcohol consumption, smoking and social dining habits, partially explain this disparity.²⁷ In our study, male alcohol users had a higher infection risk, consistent with evidence that alcohol disrupts gastric mucosal integrity and facilitates bacterial colonisation. Emerging research suggests hormonal influences on H. pylori susceptibility. Oestrogen has been shown to enhance mucosal defence mechanisms, potentially explaining the lower infection rates in females.²⁸ A study by Kang et al²⁹ demonstrated that oestrogen receptor-beta activation reduces gastric inflammation and H. pylori colonisation in animal models. This biological advantage, combined with healthier lifestyle choices among female HCWs (eg, lower alcohol consumption), may synergistically protect against infection. Male HCWs in surgical or emergency departments often engage in prolonged, high-stress shifts, which may compromise immune function and increase susceptibility. Conversely, female HCWs in administrative roles may have fewer direct patient contacts, reducing exposure risks. These occupational dynamics warrant further investigation.

Alcohol consumption emerged as a strong independent risk factor (OR=2.40, p=0.008), corroborating studies linking alcohol to H. pylori persistence and virulence.³⁰ Ethanol metabolites, such as acetaldehyde, directly damage gastric epithelial cells and upregulate H. pylori adhesion molecules (eg, BabA), enhancing bacterial attachment.^{31 32} A meta-analysis³³ clearly showed that heavy drinkers (>40 g alcohol/day) had a higher risk of H. pylori positivity, mirroring our results. Targeted interventions, such as alcohol reduction programmes and workplace policies limiting alcohol-related social events, could mitigate infection risks. In our previous retrospective analysis of 3340 patients with H. pylori infection,²² alcohol use was documented in 429 subjects (12.5%), and the findings indicated that alcohol use was significantly associated with a reduced risk of eradication failure (Adjusted odds ratio (AOR) =0.513, p<0.01). In the current cohort of HCWs, the prevalence of alcohol use among H. pylori-positive subjects was 10.4% (17/164). The consistent direction and effect size of this association across both studies suggest that the relationship is not specific to occupational context but rather reflects a broader epidemiological trend. Educational campaigns emphasising the H. pylori-alcohol link may empower HCWs to adopt healthier behaviours. H. pylori-positive individuals reported significantly fewer acid reflux symptoms (9.1% vs 17.3%, p=0.011), a paradoxical finding given H. pylori association with hyperacidity in peptic ulcer disease.³⁴ Chronic H. pylori infection suppresses acid secretion through urease hydrolysing urea to ammonia and carbon dioxide, neutralising gastric acid and facilitating H. pylori colonisation.³⁵ Thus, symptom-driven screening protocols are insufficient for HCWs. Instead, routine non-invasive testing should be integrated into occupational health programmes, especially for high-risk subgroups.

Our overall H. pylori eradication rate of 92.0% highlights the efficacy of tailored regimens. However, the high clarithromycin resistance rate poses a significant challenge.³⁶ Regional variations in H. pylori clarithromycin resistance rates are evident, with Italy and Japan reporting resistance rates of 30%, compared with 40% in Turkey and 50% in China.³⁷ European data reveal an upward trend in clarithromycin resistance, reaching an aggregate rate of 17%.³⁸ In our previous study,³⁹ clarithromycin and levofloxacin resistance rates were 44.4% (75/169) and 26.6% (45/169), respectively. Among HCWs in the current study, corresponding rates were 47.1% (41/87) for clarithromycin and 22.1% (19/86) for quinolones. The remarkable consistency in resistance profiles between these cohorts indicates that the high prevalence of antibiotic resistance is not occupation-specific but reflects a regional epidemiological pattern. The potassium-competitive acid blocker vonoprazan enhances acid suppression, improving antibiotic efficacy.⁴⁰ In Japan, the most frequently administered treatment for anti-H. pylori is the vonoprazan-amoxicillin-clarithromycin (VAC) triple therapy.⁴¹ Prior research² demonstrated that VAC triple therapy achieved H. pylori eradication rates of 89.2% (intention-to-treat analysis) and 90.2% (per-protocol analysis). However, efficacy declined significantly in clarithromycin-resistant strains, with an eradication rate of 76.2%. These findings underscore the imperative to address antimicrobial resistance as a general public health priority, transcending occupational boundaries.

This study has several limitations, including its single-centre design, which may restrict the generalisability of findings to HCWs in other regions or settings. Specifically, the generalisability of our findings is limited by the demographic and geographic characteristics of our cohort, as the participants were all HCWs from a single tertiary hospital in Beijing, China, and were predominantly of Han Chinese ethnicity. As we did not collect self-reported race or ethnicity data, the applicability of our results to other racial or ethnic populations, or to healthcare settings in other regions, may be constrained. Additionally, retrospective data collection methods that led to missing variables, limiting causal interpretations, and the small eradication cohort hindered subgroup analysis of treatment efficacy. Future research should prioritise multicentre studies to validate risk factors and resistance patterns across diverse populations, incorporate molecular epidemiology and conduct longitudinal follow-up to evaluate recurrence rates and long-term clinical outcomes.

Conclusions

This study illuminates the unique epidemiology of H. pylori infection among HCWs, characterised by male predominance, alcohol-related risks, attenuated symptoms and alarming antibiotic resistance. By bridging clinical, occupational and microbiological perspectives, our findings advocate for proactive screening, resistance-guided therapy and institutional policies to safeguard HCW health.

Data availability statement

Data are available on reasonable request.