[States and predictors of postpartum smoking relapse: a population based longitudinal study].

Objectives　Although more than half of women who smoke attempt to stop smoking after conception, many relapse after delivery. We conducted a population-based longitudinal study to identify the predictors of postpartum smoking relapse.Methods　Participants were expectant mothers living in Nagoya city, Japan, who notified Health Centers of their pregnancy from April 2014 to March 2015. A self-administered questionnaire was given to the expectant mothers that includes individual factors in the pregnancy: expectant mother's smoking status, age, marital status, experience of parturition, (mother's and father's) occupations, infertility treatment reception, feelings when pregnancy was confirmed, plans to return to parents' house for delivery, accessibility to help with childcare, household member(s) smoking in the same room, alcohol consumption, and depressive symptoms continuing more than 2 weeks.　We followed their smoking status at their children's "3-month-old health check-up" (3 months) and "1-year-and-6-months-old health check-up" (18 months) held in Health Centers until March 2017.　The data were analyzed using a combination of Chi-square or Fisher's exact test and logistic regression modeling. The analyses were conducted separately in primiparas and multiparas in addition to all expectant mothers.Results　Participants were 24,413 mothers; 18,041 were followed up at 3 months and 14,163 at 18 months.　Of the 18,041 mothers at 3 months, 1,031 primiparas and 695 multiparas stopped smoking when they confirmed pregnancy; 89 (8.6%) primiparas and 107 (15.4%) multiparas relapsed at 3 months. Of the 14,163 mothers at 18 months, 789 primiparas and 568 multiparas stopped smoking when they confirmed pregnancy; 155 (19.6%) primiparas and 174 (30.6%) multiparas relapsed smoking at 18 months.　As a result of logistic regression modeling, "multiparas," "younger (<25 years old)," "not married (only in multiparas)," "no plan to return to mother's parent's house for delivery," "household member(s) smoking in the same room (only in primiparas)," and "depressive symptoms (only in all mothers and primiparas)" were the predictors of postpartum smoking relapse at 3 months. "Multiparas," "not married (only in all mothers)," "no help with childcare (only in all mothers)," and "household member(s) smoking in the same room" were the predictors of postpartum smoking relapse at 18 months.Conclusion　More mothers relapsed with smoking after 3 months than before 3 months. The predictors of postpartum smoking relapse differed between 3 and 18 months. Support to continue smoking cessation was needed for each mother at an appropriate time not only in pregnancy but also after delivery.

ROLES OF PUBLIC HEALTH CENTERS (HOKENJO) IN TUBERCULOSIS CONTROL IN JAPAN.

Public health centers (PHCs, hokenjo in Japanese) are local government authorities responsible for public health in Japan. PHCs have an important role in tuberculosis (TB) control. Typically, their responsibilities include 1) the recommendation to admit infectious TB patients to an isolation ward, 2) health checkups with chest X-ray of those in a close contact with infectious TB patients, and 3) public subsidy of medical expenses for TB treatments. Facing the emergence of multi-drug resistant tuberculosis (MDR-TB), the national TB control program was drastically changed; the Japanese version of the Directly Observed Treatment in Short-course (DOTS) strategy was started in 2005. New roles were added to PHCs' responsibilities; 1) active screening of latent TB infection by interferon gamma release assays for those in a close contact with infectious TB patients, 2) community DOTS to promote treatment adherence to outpatients, 3) cohort analysis of outcomes of TB treatment, and 4) national MDR-TB surveillance. These roles are important in preventing MDR-TB and eliminating TB in Japan.

A PCR method for VKORC1 G-1639A and CYP2C9 A1075C genotyping useful to warfarin therapy among Japanese.

Warfarin is widely prescribed for patients with the risk of thromboembolism around the world. The inter-individual and inter-racial differences in appropriate dosage depend highly on age, body weight, and genetic factors. A lot of studies including genome-wide association studies revealed that vitamin K epoxide reductase complex, subunit 1 (VKORC1) G-1639A and Cytochrome P450 (CYP) 2C9 A1075C are the most strong genetic factors for determining warfarin effects in Asians and Africans. Since we developed a quick and inexpensive genotyping method, polymerase chain reaction with confronting two-pair primers (PCR-CTPP), the method was applied for these genotypes to examine the possibility to clinical use. Subjects were 436 examinees (117 males and 319 females, aged 32 to 85 years) who attended a health checkup program in Japan. The PCR-CTPP for VKORC1 G-1639A and CYP2C9 A1075C was conducted for the subjects, as well as the samples genotyped by DigiTag2 method. The allele frequencies of VKORC1 G-1639A were 0.085 for G and 0.915 for A, and those of CYP2C9 A1075C were 0.979 for A and 0.021 for C, being in Hardy-Weinberg equilibrium (p = 0.658 and p = 0.514, respectively). These frequencies were similar to those reported in the HapMap project. Genotyping for both SNPs by PCR-CTPP was replicated by DigiTag2 method. Our results indicated that the PCR-CTPP could be one of the alternative methods for genotyping VKORC1 G-1639A and CYP2C9 A1075C for Asians and Africans with similar allele frequencies to Japanese.

A C to T polymorphism of urokinase plasminogen activator (P141L) is associated with Helicobacter pylori infection.

Urokinase plasminogen activator (uPA) plays an important role in tumor invasion and certain inflammatory diseases. However, few studies have paid attention to how the uPA is associated with Helicobacter pylori infection and gastric atrophy. This study investigated associations of a C-to-T polymorphism of uPA (P141L, rs2227564) in exon 6 in 454 Japanese health checkup examinees (126 males and 328 females) aged 35 to 85 without a history of cancer. The uPA was genotyped by polymerase chain reaction with two-pair primers. The genotype distribution was in Hardy-Weinberg equilibrium (p=0.52) and the frequency of the T allele was 0.239. The risk of H. pylori sero-positivity was significantly reduced with the T/T genotype; the odds ratio (OR) relative to the C/C genotype was 0.34 (95% confidence interval [CI]: 0.14 to 0.86). Of the sero-negative subjects, 21 with atrophy were infected with H. pylori but lost their sero-positivity. After reclassifying them together with the sero-positive subjects, the corresponding OR was 0.40 (95% CI: 0.16 to 1.00), confirming that the T/T genotype decreased the risk of H. pylori infection. This gene polymorphism was not associated with the risk of gastric atrophy. In conclusion, this study indicated a possibility that the uPA minor homozygous genotype was associated with a reduction of H. pylori infection risk. Further studies are required to confirm these findings.

Structure and roles of public health centers (hokenjo) in Japan.

Public health centers (PHCs, hokenjo in Japanese) are local government authorities responsible for public health in Japan. As of 2010, 494 centers were operating under the Ministry of Health, Labour and Welfare of Japan. While the general rule is that one PHC covers 300,000 inhabitants, several centers cover a population of more than 1 million. The roles of PHCs described in the Community Health Act include 1) propagation and improvement of information on community health, 2) vital statistics and other statistics on community health, 3) improvement of dietary conditions and food sanitation, 4) environmental sanitation including housing, water supply, sewage, waste disposal and public cleaning, 5) medical and pharmaceutical affairs, 6) matters involving public health nurses, 7) promotion and improvement of public medical services, 8) maternal, child, and elderly health, 9) dental health, 10) psychiatric health, 11) health of patients under long-term care due to incurable diseases, 12) prevention of infectious diseases, 13) laboratory tests on sanitation/environment, and 14) other functions needed to maintain/promote health in the community. Among those many roles, infectious disease controls are one of the most important. Concerning tuberculosis control, PHCs are responsible for the isolation of patients, health check-ups of those in close contact with infectious TB patients, and public subsidy of medical expenses for tuberculosis treatments. Food poisoning controls are also an important responsibility of PHCs, as are the conduct of surveys to trace suspicious foods and laboratory testing of samples from patients. To make these many measures effective, sufficient numbers of public health professionals are required.

No association between Helicobacter pylori seropositivity and ornithine decarboxylase (ODC) A317G polymorphism, and no modification by NAD(P)H:qinone oxidoreductase 1 (NQO1) C609T.

Our previous epidemiologic study reported that NAD(P)H:quinone oxidoreductase 1 (NQO1) 609C/C with full enzyme activity was a high risk genotype for Helicobacter pylori (H. pylori) seropositivity. Since NQO1 stabilizes ornithine decarboxylase (ODC), which attenuates the innate immune response through elevated polyamines, ODC functional polymorphisms may also influence H. pylori seropositivity. This study aimed to examine the association with ODC A317G polymorphism, as well as the modification by NQO1 C609T. The two polymorphisms were determined by polymerase chain reaction with confronting two-pair primers (PCR-CTPP) among 465 health checkup examinees in Nagoya. The ODC A317G genotype frequency was 35.9% for A/A, 49.3% for A/G, and 14.8% for G/G. The sex-age-adjusted odds ratio (OR) of the ODC gene for H. pylori seropositivity was not significant (OR = 1.09 for G/A and OR = 1.02 for G/G, relative to A/A). Among subjects with any NQO1 genotype, no association was observed between the ODC ploymorphism and H. pylori seropositivity. Results of the present study did not support the hypothesis that the different genetic traits in the ODC-polyamine pathway are associated with susceptibility to persistent H. pylori infection. The higher frequency of the ODC 317A allele in the Japanese population than that in the Caucasian population is firstly reported. The genetic traits through the ODC-polyamine pathway will be further investigated.

Plasma carotenoid, alpha-tocopherol and retinol concentrations and risk of colorectal adenomas: A case-control study in Japan.

To investigate associations between plasma carotenoids, alpha-tocopherol and retinol with colorectal adenomas risk, we measured concentrations in 224 asymptomatic colorectal adenoma cases and 230 population-based controls matched for age and sex. After adjustment for age, history of colorectal adenomas and cancers, BMI, smoking, drinking status, multivitamin consumption and plasma total cholesterol, the risk of colorectal adenomas in the highest quartile was approximately half of that of men in the lowest quartile for alpha-carotene (OR=0.38; 95% CI: 0.18-0.84; P(trend)=0.01), beta-carotene (OR=0.51; 95% CI: 0.24-1.07; P(trend)=0.03) and total carotenoids (OR=0.48; 95% CI: 0.22-1.03; P(trend)=0.04). In addition, a protective association for alpha-carotene in women was also indicated, but which did not reach statistical significance (OR=0.53; 95% CI: 0.19-1.52; P(trend)=0.35). Our findings suggest a protective effect of carotenoids against the development of colorectal adenomas.

Interleukin-2 gene polymorphisms associated with increased risk of gastric atrophy from Helicobacter pylori infection.

BACKGROUND: Gastric atrophy induced by Helicobacter pylori is thought to predispose patients to noncardiac gastric cancer development. However, the host genetic factors that influence the progression of gastric atrophy have not been elucidated. In this study, we examined the effects of cytokine polymorphisms on H. pylori-induced gastric atrophy. METHODS: Blood samples were taken from 454 Japanese subjects. The interleukin-2 (IL-2; T-330G), IL-4 (C-33T), and IL-13 (C-1111T) polymorphisms were genotyped by polymerase chain reaction with confronting two-pair primers (PCR-CTPP). Anti-H. pylori IgG antibody and pepsinogen I and II were measured to diagnose H. pylori infection and atrophic gastritis. RESULTS: The odds ratios (ORs) for the association between IL-2 polymorphism [OR = 2.78, 95% CI (confidence interval) = 1.26-6.17 (T/T to G/G)] or IL-4 polymorphism [OR = 2.22, 95% CI = 1.01-4.89 (T/C to C/C)] were increased significantly with gastric atrophy, whereas the corresponding OR of IL-13 polymorphism was decreased with gastric atrophy [OR = 0.61, 95% CI = 0.39-0.96 (C/T and T/T to C/C)]. There were no significant H. pylori seropositivity-related differences between these polymorphisms. We examined the relationship between these polymorphisms and gastric atrophy separately in H. pylori-seropositive and -seronegative groups. In the H. pylori-seropositive group, the IL-2 T/T (OR = 2.78, 95% CI = 1.12-6.93) had a significant association with gastric atrophy. CONCLUSIONS: These results reveal that the IL-2 gene polymorphism is associated with an increased risk of gastric atrophy induced by H. pylori infection and might predispose to gastric cancer.

Association of p53 codon Arg72Pro and p73 G4C14-to-A4T14 at exon 2 genetic polymorphisms with the risk of Japanese breast cancer.

BACKGROUND: The association between breast cancer risk and genetic polymorphisms of p53 at codon 72 (Arg72Pro) has been investigated by several studies, but the results are not consistent. The aim of this case-control study conducted in Nagoya, Japan, was to reconfirm the results of prior studies of polymorphisms of p53 Arg72Pro, and to test if polymorphisms of p73 G4C14-to-A4T14 at exon 2 (G4A) were also associated with breast cancer risk. METHODS: The cases were 200 breast cancer patients who visited Aichi Cancer Center Hospital. The controls were 282 local citizens who underwent a health check-up. All cases and controls were recruited from Chubu Japan. Genotyping was carried out by polymerase chain reaction with confronting two-pair primers. RESULTS: The p53 genotype distribution was 40.4% for Arg72 homozygous, 48.9% for heterozygous, and 10.7% for Pro72 homozygous in controls, and 32.0%, 50.0%, and 18.0% in cases, respectively. A comparison between cases and controls indicated a significantly increased risk for Pro72 homozygosity in cases (odds ratio=2.14; 95% confidence interval=1.21-3.79). The genotypic frequencies for p73 G4A were 54.3% for G/G, 39.7% for G/A, and 6.0% for A/A in controls; and 59.0%, 32.0%, and 9.0% in cases, respectively. There were no significant differences in p73 G4A frequency between cases and controls. CONCLUSIONS: This study implies an association of breast cancer risk with the p53 polymorphism Arg72Pro, but not with p73 G4A.

Duplex polymerase chain reaction with confronting two-pair primers (PCR-CTPP) for genotyping alcohol dehydrogenase beta subunit (ADH2) and aldehyde dehydrogenase 2 (ALDH2).

AIMS: Alcohol dehydrogenase beta subunit (ADH2) Arg47His and aldehyde dehydrogenase 2 (ALDH2) Glu487Lys were genotyped by a duplex polymerase chain reaction (PCR) with confronting two-pair primers (PCR-CTPP), which allows DNA amplification with one-tube PCR including eight primers, and subsequent electrophoresis. METHODS: Several PCR conditions were tested to establish the optimal conditions for distinguishing the allele-specific bands for the two polymorphisms. Under the optimal PCR conditions, 454 Japanese health check-up examinees were genotyped. RESULTS: The allele-specific bands were successfully amplified under the optimal conditions of the duplex PCR-CTPP. The genotype distributions were within the Hardy-Weinberg equilibrium. The bands produced by the duplex PCR-CTPP genotyping were clearer than those produced by PCR-CTPP, conducted solely for ADH2. CONCLUSIONS: ADH2 Arg47His and ALDH2 Glu487Lys were successfully genotyped by this newly developed duplex PCR-CTPP, an inexpensive and time-saving genotyping tool, which will be useful in epidemiological studies on alcoholism, as well as risk estimation of alcohol-related diseases.

Helicobacter pylori seropositivity and the myeloperoxidase G-463A polymorphism in combination with interleukin-1B C-31T in Japanese health checkup examinees.

OBJECTIVE: Genetic susceptibility appears to play an important role in Helicobacter pylori (HP) infection. The present study was conducted to re-examine the reported association between the myeloperoxidase (MPO) G-463A polymorphism and HP seropositivity in different subjects and to investigate interactions with smoking behavior and the interleukin-1B (IL-1B) C-31T polymorphism. METHODS: The subjects were 468 health checkup examinees in Nagoya, who consented to anonymous genotyping of residual blood samples. Genotyping was conducted by PCR-RFLP for MPO G-463A and PCR-CTPP for IL-1B C-31T. RESULTS: Among the successfully genotyped 437 participants without a cancer history, the HP seropositive rate was 56.2% for -463GG (n = 354), 49.4% for -463GA (n = 77) and 83.3% for -463AA (n = 6). The gender-age-adjusted odds ratio (aOR) for GA/AA relative to GG was 0.86 (95% confidence interval, 0.32-1.34) among males, 0.84 (0.47-1.49) among females, 0.83 (0.22-3.05) among current smokers and 0.87 (0.50-1.51) among never smokers. Analysis by IL-1B C-31T genotype revealed a significantly reduced OR of 0.41 (0.18-0.93) among the participants with IL-1B -31CT. The OR for IL-1B -31TT relative to -31CC/CT was 1.59 (1.00-2.55) among those with MPO -463GG and 3.36 (1.16-9.77) with MPO -463GA/AA. None of the gene-environment or gene-gene interactions proved to be statistically significant. CONCLUSIONS: The association between MPO G-463A and HP seropositivity was not reproduced in this study. The effect of IL-1B -31TT was more prominent among individuals with the low expression MPO -463A allele, but it remains to be confirmed for other datasets.

ABO blood type, Lewis and Secretor genotypes, and chronic atrophic gastritis: a cross-sectional study in Japan.

BACKGROUND: the H type I structure, synthesized by the secretor (Se) enzyme in gastric foveolar cells, and its metabolite, Lewis b (le(b)) antigen, mediate the adhesion of Helicobacter pylori ( H. Pylori) to the gastric epithelium, whereas H. Pylori does not bind to modified forms of Le(b) specific for blood types A and B. Such host factors as Le and Se genotypes and ABO blood type may affect the establishment of H. Pylori infection and, once infected, the risk of chronic atrophic gastritis. METHODS: we investigated the cross-sectional relation of abo blood type and Le and Le genotypes to gastric atrophy, assessed by serum pepsinogen levels, in japanese residents from two sources. RESULTS: among the 151 h. Pylori-positive participants of the h. Pylori eradication program, odds ratios (ors) for gastric atrophy, adjusted for age, sex, and smoking, were elevated for blood types a (or = 5.35; 95% confidence interval (ci), 2.11-13.58) and b (or = 4.79; 95% ci, 1.77-12.93) relative to type o. Ors for blood types a and b were also elevated in h. Pylori-negative subjects. These associations were not observed among 250 h. Pylori-positive health check-up examinees. The le genotype was not associated with gastric atrophy in either study population. The se/ se genotype was associated with statistically nonsignificant elevation of gastric atrophy risk in both populations. CONCLUSIONS: the present data showed a strong association of blood types a and b with gastric atrophy in one, but not the other, study population. Discrepant results between the two populations warrant further investigation. Background: the h type i structure, synthesized by the secretor (se) enzyme in gastric foveolar cells, and its metabolite, lewis b (le(b)) antigen, mediate the adhesion of helicobacter pylori ( h. Pylori) to the gastric epithelium, whereas h. Pylori does not bind to modified forms of le(b) specific for blood types a and b. Such host factors as le and se genotypes and abo blood type may affect the establishment of h. Pylori infection and, once infected, the risk of chronic atrophic gastritis. Methods: we investigated the cross-sectional relation of abo blood type and le and se genotypes to gastric atrophy, assessed by serum pepsinogen levels, in japanese residents from two sources. Results: among the 151 h. Pylori-positive participants of the h. Pylori eradication program, odds ratios (ors) for gastric atrophy, adjusted for age, sex, and smoking, were elevated for blood types a (or = 5.35; 95% confidence interval (ci), 2.11-13.58) and b (or = 4.79; 95% ci, 1.77-12.93) relative to type o. Ors for blood types a and b were also elevated in h. Pylori-negative subjects. These associations were not observed among 250 h. Pylori-positive health check-up examinees. The le genotype was not associated with gastric atrophy in either study population. The se/ se genotype was associated with statistically nonsignificant elevation of gastric atrophy risk in both populations. Conclusions: the present data showed a strong association of blood types a and b with gastric atrophy in one, but not the other, study population. Discrepant results between the two populations warrant further investigation.

High anti-Helicobacter pylori antibody seropositivity associated with the combination of IL-8-251TT and IL-10-819TT genotypes.

BACKGROUND: Helicobacter pylori induces inflammation of gastric mucosa regulated by several interleukins. This study examined associations between anti-Helicobacter pylori immunoglobulin G antibody seropositivity and functional polymorphisms of interleukin-8 T-251 A and interleukin-10 T-819C. MATERIALS AND METHODS: The subjects were 454 health check-up examinees (126 males and 328 females) without a history of cancer, aged 35-85 years, residing in Nagoya, Japan. After written informed consent was obtained individually, residual blood was anonymously applied for anti-Helicobacter pylori immunoglobulin G antibody testing and genotyping by the polymerase chain reaction with confronting two-pair primers. RESULTS: The genotype frequency of interleukin-8 T-251 A was 52.2% for TT, 39.5% for TA, and 8.3% for AA, and that of interleukin-10 T-819C was 49.5% for TT, 39.9% for TC and 10.6% for CC. Although the differences in the positive rates among the genotypes were not marked, 115 individuals with interleukin-8-251TT (low expression genotype) and interleukin-10-819TT (high expression genotype) had a higher rate (63.5%) than the others (52.0%). Relative to the combination of interleukin-8-251TT and interleukin-10-819TT, the sex-age-adjusted odds ratio for those with the other combinations was 0.62 (95% confidence interval, 0.39-0.98). The adjusted odds ratio among 65 current smokers was 0.13 (0.03-0.61). CONCLUSIONS: The observed association suggests that individuals with interleukin-8-251TT and interleukin-10-819TT, a combination presumably causing mild inflammation, have a higher probability of the continuing Helicobacter pylori infection, especially among current smokers.

Subjects with TNF-A-857TT and -1031TT genotypes showed the highest Helicobacter pylori seropositive rate compared with those with other genotypes.

BACKGROUND: A possible association between Helicobacter pylori seropositivity and tumor necrosis factor (TNF) a G-308A has been reported in Korea. The present study examined the associations of H. pylori with functional polymorphisms, TNF-A G-308A, C-857T, and T-1031C and TNF-B A252G in Japanese subjects. METHODS: The total of 1374 study subjects included 241 outpatients who participated in an H. pylori eradication program (HPE), 679 first-visit outpatients (FVO) at a regional cancer hospital, and 454 local residents who received a health checkup examination (HCE). RESULTS: The frequency of the TNF-A -308A allele was only 1.3% of 480 chromosomes in the HPE group, so the FVO and HCE groups were not genotyped for that polymorphism. The genotype frequency of TNF-A C-857T was 69.2% CC, 27.7% CT, and 3.1% TT; that of TNF-A T-1031C was 69.4% TT, 28.1% TC, and 2.5% CC; and that of TNF-B A252G was 36.8% AA, 48.2% AG, and 15.0% GG. TNF-A -857T was tightly linked to TNF-A -1031T and TNF-B 252A. No significant associations between H. pylori seropositivity and polymorphisms of TNF-A C-857T and TNF-B A252G were observed. However, a reduced odds ratio adjusted for sex, age, and recruitment source was observed for TNF-A -1031CC (0.43; 95% confidence interval, 0.20-0.91) relative to TNF-A -1031TT. Subjects with TNF-A -857CC and -1031CC showed the lowest seropositivity (38.2% of 34 participants), while those with TNF-A -857TT and -1031TT showed the highest (66.7% of 42 participants). CONCLUSIONS: This study suggests that the possibly high expression genotype of TNF-A may increase susceptibility to persistent H. pylori infection.A possible association between Helicobacter pylori seropositivity and tumor necrosis factor ( TNF) A G-308A has been reported in Korea. The present study examined the associations of H. pylori with functional polymorphisms, TNF-A G-308A, C-857T, and T-1031C, and TNF-B A252G in Japanese subjects.

Lack of consistency in the associations of Helicobacter pylori seropositivity with Se and Le polymorphisms among Japanese.

BACKGROUND: In our previous study, a marked association between Helicobacter pylori seropositivity and functional polymorphisms of the secretor and Lewis genes (odds ratio [OR], 0.32; 95% confidence interval [CI], 0.14-0.70 for se/se genotype relative to Se/Se genotype; and OR, 10.3; 95% CI, 3.16-33.8 for high-risk group defined by the combination of Se and Le relative to low-risk group) had been observed for 239 non-cancer Japanese outpatients of the gastroenterology clinic (OGC) undergoing gastroscopy at Aichi Cancer Center Hospital. METHODS: The present study was a confirmatory study to examine the association for 679 first-visit outpatients (FVO) of all clinics at the same cancer hospital and for 465 health checkup examinees (HCE) in the same city. RESULTS: The associations between H. pylori seropositivity and the Se and Le genotypes were nonsignificant or even in the opposite direction among the FVO (OR, 1.52; 95% CI, 1.00-2.32 for se/se genotype relative to Se/Se genotype; and OR, 0.77; 95% CI, 0.43-1.40 for the high-risk group defined similarly to the previous study), and among the HCE (OR, 1.25; 95% CI, 0.75-2.07; and OR, 1.07; 95% CI, 0.50-2.26, respectively). The discrepancy between the previous and present results was not explained by the difference in the distributions of age, sex, smoking, and H. pylori seroprevalence. CONCLUSION: Even in the same ethnic group, different sources of subjects may demonstrate inconsistent findings due to an unidentified effect modification. Inconsistent findings have rarely been reported by the same research group, but they are very important to understand the whole picture of the association under study.