SEARCH

SEARCH BY CITATION

Abstract

  1. Top of page
  2. Abstract
  3. Introduction
  4. Methods
  5. Results
  6. Discussion
  7. Acknowledgements
  8. Ethics
  9. Contributors
  10. References

Summary

Background

Recently, interest has been revived in whether people with coeliac disease, in contrast to other inflammatory gastrointestinal diseases, have an increased risk of schizophrenia.

Aim

To compare the risk of schizophrenia in people diagnosed with coeliac disease, ulcerative colitis and Crohn's disease with the general population.

Methods

We used data from the UK General Practice Research Database. People with coeliac disease, Crohn's disease and ulcerative colitis were matched individually with five age-, sex- and general practice-matched controls. The prevalence of schizophrenia was calculated and compared between disease groups and their respective controls. We calculated odds ratios for schizophrenia using conditional logistic regression adjusting for smoking status.

Results

In people with coeliac disease, Crohn's disease and ulcerative colitis the prevalence of schizophrenia was 0.25%, 0.27% and 0.24%, respectively, compared with a general population prevalence of 0.37%. The adjusted odds ratios showed no association between schizophrenia and gastrointestinal disease (coeliac disease vs. controls 0.76, 95% CI: 0.41–1.4; Crohn's disease vs. controls 0.74, 95% CI: 0.44–1.3; ulcerative colitis 0.71, 95% CI: 0.44–1.1).

Conclusions Contrary to recent findings we found no evidence of an increased risk of schizophrenia in people with coeliac disease compared with the general population.


Introduction

  1. Top of page
  2. Abstract
  3. Introduction
  4. Methods
  5. Results
  6. Discussion
  7. Acknowledgements
  8. Ethics
  9. Contributors
  10. References

Recently, interest has been revived in whether people with coeliac disease, in contrast to other inflammatory gastrointestinal diseases, have an increased risk of schizophrenia.1 Eaton et al.1 carried out a population-based case–control study where they used people attending hospital with schizophrenia and identified those (or their parents) who had had a diagnosis of coeliac disease (or inflammatory bowel disease). They compared the prevalence of each gastrointestinal disease with a general population control group. While they claimed that ascertainment bias was not present, as they only counted a diagnosis of gastrointestinal disease made prior to the psychiatric one, it is possible that coeliac disease was ascertained more readily as new serological tests have made it easier to make the diagnosis, particularly in ‘clinically silent’ cases. This is particularly so in comparison with inflammatory bowel disease where a clinical presentation almost always precedes the diagnosis.

Although others have also described the association between coeliac disease and schizophrenia2 some have questioned whether it is causal3 and this is clearly a controversial area. We have therefore examined the risk of diagnosed schizophrenia in three population-based cohorts of people diagnosed with coeliac disease; Crohn's disease and ulcerative colitis and compared it with the risk in the general population.

Methods

  1. Top of page
  2. Abstract
  3. Introduction
  4. Methods
  5. Results
  6. Discussion
  7. Acknowledgements
  8. Ethics
  9. Contributors
  10. References

We obtained data from the General Practice Research Database (GPRD), a source of longitudinal records of routine primary care visits for more than 8 million people registered at general practices across the United Kingdom. We identified all people with a recorded diagnosis of coeliac disease, Crohn's disease or ulcerative colitis in their GPRD record between June 1987 and April 2002 (for coeliac disease) or December 2001 (for Crohn's disease or ulcerative colitis). We then obtained up to five comparison subjects per person, who did not have each respective disease, matched to each subject in our disease groups by age, gender, general practice and follow-up time of their general practice record. Further details of our cohorts are published elsewhere.4, 5 Each subject was assigned a date of diagnosis corresponding to the date of their first record of coeliac disease, Crohn's disease, ulcerative colitis or schizophrenia. We calculated the prevalence of schizophrenia by dividing the number of diagnoses by the respective total number of people in each group. We then estimated odds ratios for the risk of having schizophrenia ever diagnosed using conditional logistic regression adjusting additionally for smoking status.

Results

  1. Top of page
  2. Abstract
  3. Introduction
  4. Methods
  5. Results
  6. Discussion
  7. Acknowledgements
  8. Ethics
  9. Contributors
  10. References

Approximately 4732 people with coeliac disease, 5961 with Crohn's disease and 8301 with ulcerative colitis were included in our study. There were 23 620, 29 843 and 41 589 controls respectively. The prevalence of schizophrenia was 0.25% (n = 12) in coeliac disease, 0.27% (n = 16) in Crohn's disease and 0.24% (n = 20) in ulcerative colitis. The overall prevalence of schizophrenia in the general population controls was 0.37%. In people with inflammatory bowel diseases the risk of schizophrenia appeared slightly lower than their general population controls although this was not statistically significant at the 5% level (Table 1).

Table 1.  Risk of schizophrenia in people with coeliac disease, Crohn's disease and ulcerative colitis
 Schizophrenia, n (%)
NoYesOdds ratio (95% CI )Adjusted odds ratio* (95% CI)
  1. * Adjusted for smoking status.

Coeliac disease
 Control23 539 (99.7)81 (0.3)11
 Case4720 (99.8)12 (0.2)0.74 (0.40–1.4)0.76 (0.41–1.4)
Crohn's disease
 Control29 734 (99.6)109 (0.4)11
 Case5945 (99.7)16 (0.3)0.74 (0.44–1.3)0.74 (0.44–1.3)
Ulcerative colitis
 Control41 428 (99.6)161 (0.4)11
 Case8281 (99.8)20 (0.2)0.62 (0.39–0.99)0.71 (0.44–1.1)

Of the people with coeliac disease and schizophrenia nine (75%) were diagnosed with the latter diagnosis prior to the diagnosis of their gastrointestinal disease. The equivalent proportions for Crohn's disease and ulcerative colitis were 62.5% and 75% respectively.

Discussion

  1. Top of page
  2. Abstract
  3. Introduction
  4. Methods
  5. Results
  6. Discussion
  7. Acknowledgements
  8. Ethics
  9. Contributors
  10. References

Contrary to recent findings we found no evidence of an increased risk of schizophrenia in people with coeliac disease when compared with the general population. The overall prevalence of schizophrenia in our general population control groups was 0.37% and although there is substantial geographical variation in the ‘lifetime’ prevalence of schizophrenia around the world, this figure is similar to that reported from other studies based in European populations.6 Although misclassification of both intestinal disease and schizophrenia is a possible bias we think it likely inconsequential as the recorded diagnoses of many diseases including schizophrenia in the GPRD have been demonstrated to be accurate.7 Although there may be residual confounding by socioeconomic status when we controlled for smoking habit, which shows a strong socioeconomic association in the UK at present, no substantial changes in the effect estimates occurred. While our study was large, because schizophrenia is rare, it is possible that we have missed a positive association through lack of statistical power, i.e. chance. However, our confidence intervals for the coeliac disease did not overlap the point estimate from the previous study from Eaton et al.1 suggesting that the findings do differ.

Explanations for the divergent results from the study by Eaton et al.1 may have been that we have somehow underestimated the prevalence of schizophrenia in our cohorts of gastrointestinal disease. This is plausible, as it has been suggested that people with schizophrenia might receive less adequate health care than the general population and the majority of the people with both gastrointestinal disease and schizophrenia in our study had the latter diagnosis made first. However, this bias if present should not be differential among the gastrointestinal diseases in our study. By this we mean that if we have underestimated the associations (between schizophrenia and gastrointestinal disease) this would not be specific to just coeliac disease, i.e. it should affect inflammatory bowel disease also. Our study has only addressed the association between clinically diagnosed coeliac disease and not ‘silent’ or ‘undetected’ disease and it is possible that an association exists between this entity and schizophrenia. However, people with psychiatric disease could well be more likely to have a series of tests around the time of diagnosis for other conditions compared with the general population, one of which might be serology for coeliac disease. This would, in turn, likely lead to the finding of an association that is misleadingly. Clearly there are differences in both the study design and findings of our paper and studies by others, so more work is needed to clarify the relationship between coeliac disease and schizophrenia.

Overall we believe that there is little evidence from our study that there is a causal association between coeliac disease and schizophrenia.

Acknowledgements

  1. Top of page
  2. Abstract
  3. Introduction
  4. Methods
  5. Results
  6. Discussion
  7. Acknowledgements
  8. Ethics
  9. Contributors
  10. References

Funding for this study was provided by The Wellcome Trust. The authors have no competing interests to declare. Authors would like to thank the staff at the Epidemiological and Pharmacy Information Core and Chris Smith for their help with formatting the data.

Ethics

  1. Top of page
  2. Abstract
  3. Introduction
  4. Methods
  5. Results
  6. Discussion
  7. Acknowledgements
  8. Ethics
  9. Contributors
  10. References

Granted by the Scientific Ethical and Advisory Group to the GPRD.

Contributors

  1. Top of page
  2. Abstract
  3. Introduction
  4. Methods
  5. Results
  6. Discussion
  7. Acknowledgements
  8. Ethics
  9. Contributors
  10. References

JW and TC had the original idea for the study and carried out the analyses. All authors contributed to the interpretation and writing of the manuscript.

References

  1. Top of page
  2. Abstract
  3. Introduction
  4. Methods
  5. Results
  6. Discussion
  7. Acknowledgements
  8. Ethics
  9. Contributors
  10. References
  • 1
    Eaton W, Mortensen PB, Agerbo E, Byrne M, Mors O, Ewald H. Coeliac disease and schizophrenia: population based case control study with linkage of Danish national registers. BMJ 2004; 328: 4389.
  • 2
    Dohan FC. More on celiac disease as a model for schizophrenia. Biol Psychiatry 1983; 18: 5614.
  • 3
    Ross-Smith P, Jenner FA. Diet (gluten) and schizophrenia. J Hum Nutr 1980; 34: 10712.
  • 4
    Card T, Hubbard R, Logan RF. Mortality in inflammatory bowel disease: a population-based cohort study. Gastroenterology 2003; 125: 158390.
  • 5
    West J, Logan RF, Smith CJ, Hubbard RB, Card TR. Malignancy and mortality in people with coeliac disease: population based cohort study. BMJ 2004; 329: 7169.
  • 6
    Goldner EM, Hsu L, Waraich P, Somers JM. Prevalence and incidence studies of schizophrenic disorders: a systematic review of the literature. Can J Psychiatry 2002; 47: 83343.
  • 7
    Howard LM, Kumar C, Leese M, Thornicroft G. The general fertility rate in women with psychotic disorders. Am J Psychiatry 2002; 159: 9917.