The association of peptic ulcer and schizophrenia: A population-based study

The association of peptic ulcer and schizophrenia:

A population-based study

Chun-Hui Liao

, Chen-Shu Chang

, Shih-Ni Chang

, Chih-Hsin Muo

,

Hsien-Yuan Lane

,

Fung-Chang Sung

⁎

, Chia-Hung Kao

⁎⁎

Introduction

Schizophrenia is a chronic and disabling psychiatric disorder with a worldwide prevalence of approximately 1% [1]. Patients with schizophrenia have been reported to have a higher mortality rate due to physical illness than the general population [2–4]. Although these patients have been reported to have a higher risk of several physical illnesses [5,6], there are fewreports about the association between schizophrenia and peptic ulcer, and these reports are not conclusive [7]. Peptic ulcer is a common digestive system disease with a prevalence of around 10–15% [8,9]. There are multiple possible risk factors of peptic ulcer reported, including Helicobacter pylori (H. pylori) infection, alcohol, smoking and nonsteroidal anti-infammatory drugs (NSAIDs) Schizophrenic patients were found to have a higher rate of H. pylori infection

[10], which is one of the important factors contributing to peptic ulcer. They were also found to have a high prevalence of dyspepsia [11]. Schizophrenia patients also have higher risk of alcohol abuse and smoking [12–14], which could attribute to peptic ulcer. On the other hand, schizophrenia is caused partially by a hyperactivity of dopaminergic system[15]. Dopamine systemwas documented having a protective effect toward peptic ulcer [16–20]. There was an observational study based on in-patients' database showed that schizophrenia patients had lower risk than the general population [21]. The resultswas limited with the source of patients,whichwere includedwith in-patients. There has been little fnding about the association between peptic ulcer and schizophrenia in the recent decades.

The incidence of peptic ulcer related to H. pylori infection has

declined in recent decades [22,23], but the risk of peptic ulcer among schizophrenic patients might be different. Besides, schizophrenia patients might take antipsychotic agents to control their psychotic

symptoms. Antipsychotics included dopamine receptor antagonist, which might decrease the protective effect of dopamine agonist [24].

The aim of this study is to investigate the risk of peptic ulcer among patients with schizophrenia compared with the general population. Medications were considered as covariates in this study. In addition, this is the frst population-based study to investigate peptic ulcer diagnosed by endoscopy among schizophrenic patients.

Method

Data sources and study subjects

The Taiwan National Health Insurance program, which covers over 23.03 million residents, or about 99.2% of the Taiwan population, was reformed onMarch 1, 1995, by the Bureau of National Health Insurance (BNHI). BNHI has released scrambled data to the National Health Research Institutes (NHRI) to establish 18 National Health Insurance Research Databases (NHIRDs) (http://w3.nhri.org.tw/nhird/date_01. html). The Longitudinal Health Insurance Database (LHID2000) we used in this study included all the medical information of 1 million insurants randomly selected among benefciaries during the period 1996–2000. There were no signifcant differences in the distribution of gender and age between the original claims data and the sampling data. The claims data in the LHID2000 was extended to December 31, 2009, and retrospectively collected until January 1, 1996. The diagnosis codes in the NHIRD were in accordance with those in the International Classifcation of Diseases, 9th Revision, Clinical Modifcation (ICD-9-CM). NHRI had scrambled patient identifcations with surrogate numbers to secure patient privacy. The present study was approved by the ethic committee at China Medical University and Hospital.

The dataset comprised 1874 patients with newly diagnosed schizophrenia (ICD-9-CM: 295) during the years 1998–2001; the date of

schizophrenia diagnosis was the index date. We excluded patients who had gastrointestinal ulcers (ICD-9-CM: 531–534) diagnosed before schizophrenia was diagnosed (n = 378); in all, 1496 patients were selected as the schizophrenia group. For comparison, we randomly selected people without a history of schizophrenia and peptic ulcer. These controls were frequency-matched by age, sex and index year with the cases in the corresponding schizophrenia cohort,with a control sample size four-fold of the corresponding schizophrenia cohort. We measured incidences of peptic ulcer (ICD-9-CM: 531–534) until the end of 2009 and the diagnosis was checked by the prescription of H2-blocker and prompt-pump inhibitors (PPI).

We also considered potentially confounding variables, including urbanization levels, monthly income, alcoholism, depression, bipolar, anxiety and medicine prescription history. The NHRI divided the urbanization of Taiwan into 7 levels, with level 1 refecting the most urbanized

and level 7 the least [25]. We collapsed levels 5–7 into level 5 because there were few schizophrenic patients in those levels. The monthly income level cut-off was based on insurance fee level.

Medicine prescription history included the prescription of antipsychotics, aspirin, lithium, valproate sodium, antidepressives, anxiolytics

and hypnotics, and NSAIDs, which were identifed within 180 days before the end-point, that is, the date of diagnosed gastrointestinal ulcer, death or failure to follow-up, or the end of 2009. The cumulative days of prescription were counted as the sum of days of use within 180 days before the end point.

Statistical analysis

The SAS 9.1 statistical package (SAS Institute Inc., Cary, NC, USA)was used for all statistical analyses in this study, and the signifcance level was set at 0.05. The chi-square test and t-test were used to determine any differences in sociodemographics and medical history between the two groups. We calculated the incidence rate ratio (IRR) and 95% confdence intervals (95% CI) for each variable of the two groups using Poisson regression. The disease-free rate and the differences in risk between the two groups were estimated with the Kaplan–Meier method and the log-rank test. The Cox proportional hazards model was used to assess the hazard ratios (HR) of gastrointestinal ulcers by covariates. Model 1 was adjusted sociodemographics (included sex, age, urbanization level, monthly income), alcoholism (ICD-9-CM: 305.0), depression

(ICD-9-CM: 296.2, 296.3, 300.4 and 311), bipolar (ICD-9-CM: 296.0– 296.1, 296.4–296.9) and anxiety (ICD-9-CM: 300.0, 300.2, 300.3,

308.3 and 309.81). Model 2 was adjusted sociodemographics, alcoholism, and medications, including aspirin, lithium, valproate sodium,

antidepressives, anxiolytics and hypnotics, and non-steroidal antiinfammatory drugs (NSAIDs). In further analysis, we also assessed the

association for peptic ulcer between schizophrenia and antipsychotics used within 180 days before the end point. Data analysis considered also the risk associated the length of medication for NSAIDs. The joint effects for ulcer among antipsychotics, lithium, anti-depression and anxiolytics between two cohorts were estimated.

Results

We sampled 7480 subjects in all, 1496 in the schizophrenia group and 5984 in the non-schizophrenia group. There were more male schizophrenic patients (males 55.0% vs. females 45.0%) and the mean age was 36.5 years (SD = 15.8). Compared with the non-schizophrenia group, schizophrenic patients lived in areas with lower urbanization, had lower income, and a higher prevalence of alcoholism, depression, bipolar, anxiety and received more medication (including antipsychotics, aspirin, lithium, valproate sodium,

antidepressives, anxiolytics and hypnotics, and NSAIDs) (Table 1). In the 12-year

study period, the incidence of gastrointestinal ulcers in the schizophrenia group was 27% higher than in the non-schizophrenia group (12.1 vs. 9.54per 1000 person-years) (Table 2). The rate of receiving treatment of H. pylori eradication after the diagnosis of

peptic ulcer in schizophrenia patients was lower than comparisons (7.64% vs. 14.4%, data not shown). The incidence increased with age in both groups, but was the lowest for

ages b20 years in the non-schizophrenia group. The log-rank test revealed that the disease-free rate was lower among patients with schizophrenia than controls (p b 0.0001) during the 12-year follow-up (Fig. 1).

Table 3 shows that schizophrenic patients had an adjusted HR of 1.17 and 0.86 (95%

CI = 0.97–1.41 and 0.69–1.07) to have gastrointestinal ulcers compared with those in the non-schizophrenia group inmodels 1 and 2. Stratifed by the antipsychotics usedwithin 180 days before the end point, those who received antipsychotics have the highest risk of developing ulcer (HR = 1.85 and 1.22; 95% CI= 1.24–2.76 and 0.86–1.75 in models 1 and 2) followed by schizophrenia patients who did receive antipsychotics and schizophrenia patients who did not receive antipsychotics compared to those who did not

receive antipsychotics. In alcoholism, or patient's use of antidepressives, anxiolytics and hypnotics, and NASIDs have a higher risk of gastrointestinal ulcers in model 2 (HR = 1.88, 1.42, 1.51 and 1.58, respectively).

The association between gastrointestinal ulcer and cumulative days of NSAID usewas

analyzed (Table 4). The cut-off day was based on the tertile of cumulative use days. The

risk increased from 1.34 to 1.88 with NSAID use compared to those that did not receive NSAIDs. In comparisons, they used only antipsychotics, anti-depression, or anxiolytics and hypnotics, had a higher risk of developing ulcer compared with those who did not receive antipsychotics, lithium, valproate sodium, anti-depression, and anxiolytics and

hypnotics (Table 5). But in schizophrenia patients, those who received more than two

medicines had a higher risk. Discussion

This population database cohort study demonstrated that there was a 1.27-fold increase in the incidence of peptic ulcer among schizophrenia patients than the non-schizophrenia patients in the matched control

group. We did not fnd a signifcant association between schizophrenia and peptic ulcers after controlling the confounding factors. This suggests that the association between schizophrenia and peptic ulcer was infuenced by multiple factors.

The incidence of peptic ulcer in schizophrenic patients has been reported with conficting results. Some reports showed that there was a

lower incidence of peptic ulcer among patients with schizophrenia [26,27], and others revealed that therewas an equal frequency of peptic ulcer between schizophrenic patients and the general population [28, 29]. There have been few studies on the association between peptic ulcer and schizophrenic patients, especially in the last 30 years. Our study was the frst population-based study to examine the association between peptic ulcer and schizophrenia, and the fndings were in contrast to those of previous reports. The diagnosis of peptic ulcer in our

study was dependent on endoscopic results because Taiwan's BNHI has set a limit on the prescription of H2-blockers and prompt-pump inhibitors (PPI), which are used for patients with positive ulcer diagnoses through endoscopy. Nevertheless, in previous reports, the diagnosis

of peptic ulcer depended on the results of an autopsy or a patient's subjective complaints, the use of which would complicate the design of a

specifc population-based study.

The mean age of the index cases is older than the most likely onset age in our study. The possible reason is the delay medical services for schizophrenia in our population. The association of delay medical engagement and discontinued therapy with the higher risk of peptic should be investigated further more.

H. pylori has been reported to be associatedwith the development of

peptic ulcer [22,23,30] and there is a greater prevalence of H. pylori infection in those of a lower socioeconomic status [31]. Patients

with schizophrenia have been found to have a higher prevalence of

H. pylori infection [10]. Although we did not have direct data on the incidence of H. pylori infection, the patientswith schizophrenia had lower

incomes than the control group, which might have contributed to the risk of H. pylori infection and the greater prevalence of peptic ulcer. But we found that there was lower rate of received eradication therapy of H. pylori among the patients with schizophrenia. The prevalence of H. pylori infection, as well as the prevalence of peptic ulcer, has been reported to be in decline in the last two decades [32,33]. This might be

due to improvements in quality of life and environmental hygiene [34]. Although there were some reports that showed the association of endocrine abnormalities, which are increased in schizophrenia, with H. pylori [35], our results could not conclude that relation of H. pylori infection and schizophrenia. Further evaluation should be furthermanaged. We also found that alcohol dependence had a signifcant, positive

associationwith peptic ulcer. This resultwas similar to those of previous reports [36–38].

Patients with schizophrenia were prescribed NSAIDsmore frequently

than controls in our study. We found a signifcant association between peptic ulcer and the use of NSAID use, which was compatible with the

results of previous reports [39,40]. Patients who used antidepressants, anxiolytics and hypnotic are also associated with elevated hazard of peptic ulcer. After controlling those risk factors of drugs, patients with schizophrenia had no signifcant higher risk of peptic ulcer. The association of peptic ulcer and anxiety had been revealed by Goodwin et al. [41], and we also found that patients with prescription of anxiolytics and hypnotics, or anti-depression and anxiolytics and hypnotics had a higher risk of developing peptic ulcer. Therefore, we have the hypothesis that schizophrenia patients might have higher risk of peptic ulcer due to higher rate of taking anxiolytics and hypnotics, or anti-depression and NSAID.

The strengths of our study included the large sample size from the

national representative database, the validated diagnoses of schizophrenia with the follow-up prescription of antipsychotics and of peptic

ulcer with endoscopy, and the follow-up prescriptions of H2-blockers and PPI.

Limitations and perspectives

There were some limitations in our study. First, our database could not provide information on smoking status, diet and lifestyle, and

body mass index, which could be risk factors for peptic ulcer. The replication is needed to include those risk factors to help to specify the possible causes of the higher risk of peptic ulcer among this group. Second, the prevalence of H. pylori infection was not revealed in our database;

therefore, we could not evaluate the direct association between schizophrenia, H. pylori, and peptic ulcer. Since H. pylori infection is the major

cause of peptic ulcer, the association of incidence of H. pylori infection and schizophrenia should be evaluated furthermore. Third, the severity

of peptic ulcer could not be differentiated in our database. The future study could be tried to compare the severity of schizophrenia patients with the control group and to follow up the treatment outcome among the schizophrenia patients. Furthermore, our database could not provide the information about the actual onset date of schizophrenia, and we could not evaluate the association of peptic ulcer and schizophrenia among them before they engaged to the medical service.

Implications

In conclusion, we found that schizophrenic patients are more likely to be diagnosed with peptic ulcer in the follow-up period. This suggests that the changing patterns of peptic ulcer incidence among the general population do not seemto be occurring in this group. This might be due to a higher rate of taking anti-depression, anxiolytic and hypnotics or NSAID and alcoholism among this group. The possibility of peptic

ulcer should be considered if the schizophrenic patients have gastrointestinal complaints especiallywhen they have anxiety or depressive disorders. The antipsychotic agents seem not increase the risk of peptic

ulcer. The patients with schizophrenia had higher opportunity to be prescribed NSAID, which is a risk factor for peptic ulcer. Clinicians should be careful in monitoring the possible risk of peptic ulcer among these patients.