Irritable bowel syndrome (IBS) affects an estimated 10%–15% of the population with substantial impacts on quality of life and work productivity (1,2). The diagnosis relies on the symptom-based criteria: Today, it is defined as abdominal pain associated with altered bowel habits in the absence of other explanatory conditions (3–5) but with slight variations in definition during the past 30 years (6,7).
Fear about the potential serious nature of bowel symptoms may underlie much of the explanation for seeking care in patients with IBS vs reference individuals (8), and community estimates suggest that more than half of patients with IBS fear that their illness will shorten their life span (9). Although the existing literature demonstrates no increased risk of mortality from IBS, most studies have not been population based (10–13), and the largest studies so far followed <400 patients with IBS (14,15).
Despite clear diagnostic criteria, most clinicians consider IBS a diagnosis of exclusion (16). Indeed, more than 50% of patients with IBS will undergo colonoscopy at some point in their diagnostic workup, given patient and provider fears over missing organic disease (17). However, the role of colonoscopy for predicting disease outcome has not been defined.
We used a nationwide histopathology register to examine the overall risk of death in a large cohort of individuals with IBS undergoing a colorectal biopsy compared with matched reference individuals. In a secondary cohort, we examined mortality in patients with IBS not undergoing biopsy.
Patients with IBS were selected from the Epidemiology Strengthened by histoPathology Reports in Sweden (ESPRESSO) study, a histopathology-based cohort consisting of 6.1 million gastrointestinal (GI) pathology reports collected across Sweden's 28 pathology departments from 1965 to 2017 (18). Between 2015 and 2017, the cohort was assembled by collecting all Swedish histopathology data from gastrointestinal tract sites accompanied by information on date of biopsy, topography (location of biopsy), and morphology. Each biopsy sample was categorized according to an individual's person identity number (PIN) and included information on date of birth and sex. Through linkage with individual PINs, it was possible to follow individuals over time across available Swedish registers (see below) and link data to information on healthcare diagnoses according to International Classification of Diseases (ICD) codes. The cohort contained 2.1 million unique individuals corresponding to 6.1 million histopathology data entries (accounting for multiple biopsies in one individual). Individuals with histopathology data were matched with up to 5 controls from the general population (without GI histopathology) and all first-degree relatives and first spouses to encompass a total study population of 13.0 million individuals.
Definition of IBS
We first identified all individuals undergoing a colorectal biopsy (see Table S1, Supplementary Digital Content 1, http://links.lww.com/AJG/B438) in the ESPRESSO cohort. We used ICD codes to identify individuals within this cohort with a first-ever diagnosis of IBS (ICD-9 564B; ICD-10 K58) in inpatient or nonprimary outpatient care from 2002 to 2016 (see Table S1, Supplementary Digital Content 1, http://links.lww.com/AJG/B438) before or after a colorectal biopsy. Both patients with IBS and their reference individuals were excluded from all analyses if at baseline they had an earlier diagnosis of inflammatory bowel disease, celiac disease, or colorectal cancer (see Table S1, Supplementary Digital Content 1, http://links.lww.com/AJG/B438) because these disorders may lead to false-positive IBS diagnoses (Figure 1).
Patients with IBS were divided into those with diarrhea predominance (IBS-D) and those without diarrhea predominance (IBS non-D) (see ICD and SNOMED codes in Table S1, Supplementary Digital Content 1, http://links.lww.com/AJG/B438). In secondary analyses, we did not restrict patients with IBS to individuals who had undergone colonoscopy but identified patients with IBS from the whole ESPRESSO study base (biopsied individuals, reference individuals with no record of biopsy at date of matching, and first-degree relatives [regardless of biopsy status]).
Registers and covariates
Demographic data (dates of birth and death, immigration/emigration, sex, age, county of residence, and education) from all study participants were retrieved from the Total Population Register (19) and the longitudinal integrated database for health insurance and labor market studies (LISA) using an individual's PIN (20). Data were linked to the Swedish Patient Register (21) to obtain data on inpatient and outpatient medical encounters. The diagnostic validity of the National Patient Register is generally 85–95% (21), with high validity seen for IBS diagnoses (22).
Index individuals with IBS were matched on age, sex, calendar year, and county with up to 5 reference individuals from the Swedish general population (19). Reference individuals had no GI biopsy before date of matching but could undergo biopsy thereafter. Because controls were also matched for calendar year, they were subject to the same health care and access to medication as patients with IBS.
Patients with IBS were also compared with their siblings. Siblings were identified through the Total Population Register. Sibling comparisons allowed us to examine the influence of intrafamilial confounding associated with shared genetic and early environmental factors on mortality in IBS.
Data on date of death were retrieved from the Total Population Register, which records almost all deaths within 30 days of occurrence (19). Combined with comprehensive information on migration to and from Sweden, nearly all the country's population can be accounted for in medical research endeavors. Data on cause-specific mortality were obtained from the Swedish Cause of Death Registry (23) (covering >99% of all deaths) and cross-referenced with the Swedish Cancer Register (24) to evaluate deaths related to malignancies.
Within the study population, we determined medical comorbidities related to mortality in the past 5 years before study entry using the ICD-10 codes prospectively recorded in routine clinical practice to determine the presence of diabetes, cardiovascular disease (CVD), neurologic disease, and cancer (see Table S2, Supplementary Digital Content 1, http://links.lww.com/AJG/B438).
Follow-up time started on latest of either a first-ever IBS diagnosis or first colorectal biopsy (only individuals with both medical IBS diagnosis and colorectal biopsy were eligible for case inclusion). The same date was used as index date for matched reference individuals so as not to introduce immortal time bias. Follow-up ended with death, emigration, or December 31, 2017 (for cause-specific mortality, follow-up ended on December 31, 2016). We calculated the mortality rate (death per 1,000 person-years of follow-up) and used the Cox regression to estimate unadjusted and multivariable-adjusted hazard ratios (HRs) and 95% confidence intervals (CIs) for total and cause-specific mortalities.
For our primary definition of IBS, we required a colorectal biopsy independent of its mucosal appearance. Using an alternative definition of IBS, we restricted our data to individuals with a normal mucosa (see Table S1, Supplementary Digital Content 1, http://links.lww.com/AJG/B438) to see what role a medical IBS diagnosis may play for the risk of death. In another analysis, we examined patients with IBS with a normal mucosa and those with nonspecific colorectal inflammation separately (see Table S1, Supplementary Digital Content 1, http://links.lww.com/AJG/B438) to see whether the underlying mucosal appearance could predict future mortality in IBS. Data on histopathology were obtained through the ESPRESSO (18). Inflammation was classified according to the recommendations from the Swedish quality and standardization committee for GI histopathology: KVAST (25). To evaluate the influence of intrafamilial confounding, we compared patients with IBS and their siblings. We also performed an analysis according to disease subtypes (IBS-D and IBS non-D). We also performed an analysis using more extensive exclusion criteria of IBS mimics, excluding those with a previous diagnosis of giardiasis, exocrine pancreatic insufficiency, porphyria, eosinophilic gastroenteritis, sucrose–isomaltase deficiency, pelvic floor dysfunction, or microscopic colitis (see Table S1, Supplementary Digital Content 1, http://links.lww.com/AJG/B438). Finally, because patients with IBS undergoing colonoscopy or sigmoidoscopy with biopsy may represent a more severe subset of disease, we used an alternative definition of IBS in a separate analysis examining individuals with IBS who did not undergo a colorectal biopsy compared with matched general population reference individuals.
All analyses were conditioned on matching factors (sex, age, county, and year of biopsy), except in siblings where we conditioned on matching set within family. We further adjusted for baseline medical comorbidities (cancer, diabetes, CVD, and neurologic disease) and education because earlier research suggests an inverse relationship between education and seeking medical care (26) (and ascertainment of IBS) and mortality (27). Because psychiatric disease could function as both a confounder and a mediator for IBS and mortality risk, we evaluated its presence and role in cause-specific mortality in all patient groups but did not adjust for it in our regression models. Statistics were performed using SAS statistical software v9.4. A P-value <0.05 was considered statistically significant. CIs for mortality rates were computed by assuming a Poisson distribution. The study was approved by the Stockholm Ethics Review Board. Informed consent was waived by the board because the study was strictly register based (28).
We identified 51,401 incident cases of IBS from 2002 to 2016 undergoing colonoscopy (Figure 1). After exclusions, there remained 45,254 patients with IBS who were matched to 217,316 reference individuals (Figure 1).
On average, the time interval between the first IBS diagnosis and biopsy was 3.0 (SD 4.9) years. In 58% of patients (n = 26,432), the biopsy was performed <1 year before/after an IBS diagnosis.
Biopsies could be performed both before and after an IBS diagnosis. Table 1 shows the characteristics of study participants according to IBS status. Both patients with IBS and matched reference individuals were predominantly women, with a mean age in the mid-40s. Compared with matched reference individuals without IBS, patients with IBS were more likely to be born in the Nordic countries and suffer from comorbidities. The median follow-up of study participants was 7.3 years (Table 2).
The crude incidence for death was higher in IBS (9.4 per 1,000 person-years) than in general population reference individuals who had not undergone biopsy at the time of matching (7.9 per 1,000 person-years) (Table 2), as was the cumulative incidence over time presented using the Kaplan-Meier estimators (Figure 2).
This translated into a 1.10-fold increased risk of death (95% CI = 1.05–1.14) (Table 2). This positive association was completely attenuated after adjustment for potential confounders, including education and baseline medical comorbidities (HR = 0.96; 95% CI = 0.92–1.00). All-cause mortality did not vary with follow-up time or comorbidity, but we found a small decreased risk of mortality in women with IBS (multivariable HR = 0.92; 95% CI = 0.99–0.97) (Table 3). Compared with age-matched reference individuals, individuals with younger IBS onset had somewhat increased mortality, although mortality was decreased in groups of older IBS onset.
Patients with IBS were at a lower risk of CVD (HR = 0.92; 95% CI = 0.87–0.98), cancer (HR = 0.93; 95% CI = 0.86–1.00), and psychiatric disease (multivariable HR = 0.81; 95% CI = 0.71–0.92) (Table 2).
In a subgroup analysis of patients with IBS with <1 year between biopsy and IBS diagnosis (IBS = 26,432; general population reference individuals = 130,073), the adjusted HR was 0.94 (95% CI = 0.88–1.00). We examined mortality in 116,802 unique patients with normal mucosal biopsies (i.e., no inflammation of any type) according to the IBS status (see flowchart of the identification of study participants in Figure S3, Supplementary Digital Content 1, http://links.lww.com/AJG/B438, and patient characteristics in Table S6, Supplementary Digital Content 1, http://links.lww.com/AJG/B438). In this comparison, IBS was associated with a lower risk of death (HR = 0.80; 95% CI = 0.74–0.85, see Table S7 and Figure S4, Supplementary Digital Content 1, http://links.lww.com/AJG/B438).
When comparing 29,456 patients with IBS and their siblings (n = 53,228) (see Figure S5 and Table S8, Supplementary Digital Content 1, http://links.lww.com/AJG/B438), IBS was associated with a 9% increased risk of death (95% CI = 1.00–1.18, see Figure S6 and Table S9, Supplementary Digital Content 1, http://links.lww.com/AJG/B438). Because a diagnosis of IBS with diarrhea (IBS-D) can potentially represent several other non-IBS diagnoses, we examined the role of IBS subtypes (IBS-D and IBS non-D) in 45,184 patients with IBS (see Figure S7 and Table S10, Supplementary Digital Content 1, http://links.lww.com/AJG/B438). We found no association with death in IBS-D but a lower risk of death in IBS non-D (see Figure S8 and Tables S11 and S12, Supplementary Digital Content 1, http://links.lww.com/AJG/B438). Excluding patients with additional IBS mimickers (giardiasis, exocrine pancreatic insufficiency, porphyria, eosinophilic gastroenteritis, sucrose–isomaltase deficiency, pelvic floor dysfunction, or microscopic colitis) left an IBS population of 42,777 compared with 202,972 controls. In this population, the unadjusted HR for mortality was 1.09 (95% CI = 1.05–1.14), with an adjusted HR of 0.96 (95% CI = 0.92–1.00).
Finally, we examined the role of colorectal inflammation in mortality (see Figure S9 and Table S13, Supplementary Digital Content 1, http://links.lww.com/AJG/B438). Compared with matched reference individuals, patients with IBS without inflammation had a lower mortality (HR = 0.88; 95% CI = 0.83–0.94, see Table S14 and Figure S10, Supplementary Digital Content 1, http://links.lww.com/AJG/B438), whereas patients with IBS with nonspecific colorectal inflammation had an increased mortality (HR 1.16; 95% CI = 1.04–1.30, see Table S15 and Figure S10, Supplementary Digital Content 1, http://links.lww.com/AJG/B438).
Alternative IBS cohort
In a cohort of 41,427 patients with IBS not undergoing a colorectal biopsy (alternative definition), mortality was not increased (HR = 1.02; 95% CI = 0.99–1.06) compared with 204,890 nonbiopsied reference individuals (see Table S3 and Figure S1; Tables S4 and S5 and Figure S2, Supplementary Digital Content 1, http://links.lww.com/AJG/B438).
In this nationwide population-based study of >45,000 patients with IBS, we found no increased risk of all-cause or cause-specific death. Extensive sensitivity analyses consistently found HRs between 0.8 and 1.2, with a slightly increased mortality in those with colorectal inflammation and a lower mortality in those with a normal mucosa. Compared with siblings without IBS, mortality in IBS-affected siblings was not significantly increased.
To date, there have been limited population-based data on mortality in IBS. The existing literature has been largely based on small numbers of patients with IBS seeking care at single tertiary care centers and followed longitudinally without a comparator group (10–13). To the best of our knowledge, the largest population-based survey study so far included less than 400 patients with IBS and was limited to one US county (14). However, that study similarly found no increased risk of death in IBS (HR = 1.06; 95% CI = 0.86–1.32) (14).
The lack of excess mortality risk in IBS in the current and other studies seemingly runs counter to the established impacts on quality of life, work productivity, and healthcare resource use of the disease. Although patients with IBS are greatly affected by their symptoms, tension arises when healthcare professionals rarely find a non-IBS diagnosis despite frequently exhaustive evaluations (29). More than 50% of patients with IBS worry that IBS will shorten their life span (9). Importantly, many people with IBS will never seek care for their symptoms (between 30% and 90%) (30), and thus, our formally diagnosed population might represent a more severe subset of individuals with IBS. In this context, it is important that we analyzed both patients with IBS who had undergone and not undergone colorectal biopsies; in neither of the 2 groups did we detect any excess mortality.
Patients with IBS were slightly less likely to die from cancer (HR = 0.93) and had a 19% reduced risk of death from psychiatric disease. In surveys of patients with IBS, fear of potential malignancy is common (>20%) (9) and associated with increased health care seeking (8). Thus, our findings provide reassurance to patients and their providers that a diagnosis of IBS is unlikely to be a harbinger of undetected cancer. The reduced risk of death from psychiatric disease is surprising given the well-established connection between IBS, decreased quality of life, and psychiatric comorbidities (1). Therefore, we urge caution when interpreting these results. It should be noted that these data do not suggest that the patients with IBS in this cohort have lower incident rates of psychiatric disease but rather lower rates of death directly attributed to psychiatric disease—an important distinction. Because we were reliant on registry information on death from psychiatric illness, there may be some misclassification bias in that psychiatric disease could lead to excess mortality by other means (i.e., failure to monitor one's chronic medical disease). To that point, death from CVD was the most common cause for excess mortality in Swedish patients with unipolar depression, a common comorbidity in patients with IBS (31).
In our sensitivity analyses, the only group with IBS and an increased risk of mortality was that with histopathology suggestive of mucosal inflammation (16% increased mortality). Biomarkers of inflammation have consistently been linked to risk of all-cause, cancer, and CVD mortality (32). Some clinicians would argue that this subpopulation with known colonic inflammation would inherently represent an alternative diagnosis to IBS, which is not thought to be associated with colonic inflammation. However, the true incidence of minor mucosal abnormalities in IBS is unknown, with some evidence suggesting a variety of subtle morphologic changes among the colonic mucosa of patients with IBS (33) and characterizing this subpopulation was beyond the scope of the current study. Although one might expect diarrhea to be a similar marker of organic disease, we found no increased mortality among those with IBS-D. Meanwhile, patients with IBS and a normal mucosa had an even lower mortality than their age- and sex-matched reference individuals who also had a normal mucosa.
Our study benefited from several notable strengths. It was 100 times larger than the largest previous population-based study of IBS and mortality (14). The great statistical power allowed us to calculate the precise risk estimates and rule out more than marginal risk increases even in subgroups of patients. Of note, the upper limit of our adjusted risk estimate for all-cause mortality was 1.00.
Our linkage to histopathology data allowed us to examine the role of a normal mucosa vs inflammation. The well-validated Swedish registers with linkage through personal identity numbers (34) ensure virtually no loss to follow-up. We were able to conduct extensive sensitivity analyses to confirm our main findings, specifically among patients with IBS in the general population without histopathology, siblings, and varying degrees of histopathologic inflammation. We also accounted for relevant comorbidities using the Patient Register with a positive predictive value of 85%–95% (19).
We should acknowledge several limitations, including the risk of residual confounding (as in all observational studies). Most importantly, there is a potential for selection bias in a cohort of individuals undergoing biopsy for a disease that does not require biopsy for diagnosis. Therefore, our population may represent a subset of patients with IBS in whom suspicion for organic disease is high enough to warrant endoscopic investigation. Nevertheless, a sensitivity analysis of >40,000 individuals and not undergoing a colorectal biopsy showed no increase in mortality compared with nonbiopsied reference individuals. Because most of those with IBS will never seek care for their symptoms (30), and all patients with IBS in our main analysis underwent lower endoscopy with biopsy, our analysis may represent a more severe subset of health care–seeking individuals. Despite this, we still did not find any increased mortality. To counter the influence of health care–seeking patterns in patients with IBS, we were able to adjust for education through the LISA database (20), which is an important predictor of care seeking among those with gastrointestinal symptoms (35). Furthermore, we relied on a single diagnostic code for the identification of patients with IBS. However, the ICD system is believed to have strong validity in IBS (22), with an IBS diagnosis stable over time (10,36). Conditioning the IBS diagnosis on having an associated gastrointestinal biopsy also likely increases data validity and reduces the risk of misclassification error, which can occur when transcription error occurs in the ICD coding. Limiting IBS diagnoses to a single code also minimizes the potential for immortal time bias in a study examining the risk of death. Because Sweden is a single, relatively homogenous country, there is a possibility that our findings may have limited generalizability to other parts of the world. However, Sweden has seen significant immigration in recent years such that approximately 9% of our IBS cohort and approximately 14% of our control cohort were born outside of a Nordic country. Finally, we did not have access to other laboratory values that may indicate an alternative diagnosis or severity of disease. We would argue, however, that histopathology data are likely a more reliable indicator of an alternative diagnosis, and the yield of additional investigations in patients presenting with symptoms suggestive of IBS is quite low (29).
In conclusion, we show that IBS is not associated with increased mortality. Our findings should provide reassurance to patients with IBS that their disease is unlikely to be life shortening while simultaneously allowing clinicians to spend more time on patient education and effective treatment approaches for a disease with significant economic and societal burdens.
CONFLICTS OF INTEREST
Guarantor of the article: Jonas F. Ludvigsson, MD, PhD.
Specific author contributions: K.S. and J.F.L. wrote the first draft of the paper. K.S., J.S., and J.F.L. conceived and designed the study with input from the other authors. J.F.L. supervised the project. J.F.L. and K.S. funded the study. J.S. performed the statistics. All authors interpreted the data and contributed to the writing of the paper. All authors revised and approved the final version.
Financial support: K.S. is supported by an American Gastroenterological Association (AGA) career development award. H.K. is supported by NIDDK K23DK099681. A.T.C. is supported by NIH DK098311.
Potential competing interests: K.S. has received research support from AstraZeneca, Takeda, and Gelesis, served as a speaker for Shire, and served as a consultant to Bayer Ag, Synergy, and Shire. J.F.L. coordinates a study on behalf of the Swedish IBD quality register (SWIBREG). This study has received funding from Janssen. O.O. has been the principal investigator for projects (unrelated to the current paper) at KI partly financed by investigator-initiated grants from Janssen and Pfizer.
WHAT IS KNOWN
- ✓ Mortality concern is a driver of care seeking in patients with IBS.
- ✓ Data on mortality in IBS are limited to small studies and may suffer from selection bias.
WHAT IS NEW HERE
- ✓ In this nationwide cohort study of >45,000 individuals with IBS, we found no association with mortality.
- ✓ Despite having significant impacts on quality of life, IBS does not seem to affect the risk of death.
1. Chey WD, Kurlander J, Eswaran S. Irritable bowel syndrome: A clinical review. JAMA 2015;313:949–58.
2. Lovell RM, Ford AC. Global prevalence of and risk factors for irritable bowel syndrome: A meta-analysis. Clin Gastroenterol Hepatol 2012;10:712–21.e714.
3. Mearin F, Lacy BE, Chang L, et al. Bowel disorders. Gastroenterology 2016;150(6):1393–407.e5.
4. Hyams JS, Di Lorenzo C, Saps M, et al. Functional disorders: Children and adolescents. Gastroenterology 2016;150(6):1456–68.e2.
5. Adriani A, Ribaldone DG, Astegiano M, et al. Irritable bowel syndrome: The clinical approach. Panminerva Med 2018;60:213–22.
6. Longstreth GF, Thompson WG, Chey WD, et al. Functional bowel disorders. Gastroenterology 2006;130:1480–91.
7. Thompson WG, Longstreth GF, Drossman DA, et al. Functional bowel disorders and functional abdominal pain. Gut 1999;45(Suppl 2):II43–47.
8. Kettell J, Jones R, Lydeard S. Reasons for consultation in irritable bowel syndrome: Symptoms and patient characteristics. Br J Gen Pract 1992;42:459–61.
9. Halpert A, Dalton CB, Palsson O, et al. What patients know about irritable bowel syndrome (IBS) and what they would like to know. National Survey on Patient Educational Needs in IBS and development and validation of the Patient Educational Needs Questionnaire (PEQ). Am J Gastroenterol 2007;102:1972–82.
10. Owens DM, Nelson DK, Talley NJ. The irritable bowel syndrome: Long-term prognosis and the physician-patient interaction. Ann Intern Med 1995;122:107–12.
11. Tang YR, Wang P, Yin R, et al. Five-year follow-up of 263 cases of functional bowel disorder. World J Gastroenterol 2013;19:1466–71.
12. Holmes KM, Salter RH. Irritable bowel syndrome—a safe diagnosis? Br Med J (Clin Res Ed) 1982;285:1533–4.
13. Harvey RF, Mauad EC, Brown AM. Prognosis in the irritable bowel syndrome: A 5-year prospective study. Lancet 1987;1:963–5.
14. Chang JY, Locke GR III, McNally MA, et al. Impact of functional gastrointestinal disorders on survival in the community. Am J Gastroenterol 2010;105:822–32.
15. Ford AC, Forman D, Bailey AG, et al. Effect of dyspepsia on survival: A longitudinal 10-year follow-up study. Am J Gastroenterol 2012;107:912–21.
16. Spiegel BM. Do physicians follow evidence-based guidelines in the diagnostic work-up of IBS? Nat Clin Pract Gastroenterol Hepatol 2007;4:296–7.
17. Talley NJ, Gabriel SE, Harmsen WS, et al. Medical costs in community subjects with irritable bowel syndrome. Gastroenterology 1995;109:1736–41.
18. Ludvigsson JF, Lashkariani M. Cohort profile: ESPRESSO (Epidemiology Strengthened by histoPathology Reports in Sweden). Clin Epidemiol 2019;11:101–14.
19. Ludvigsson JF, Almqvist C, Bonamy AK, et al. Registers of the Swedish total population and their use in medical research. Eur J Epidemiol 2016;31:125–36.
20. Ludvigsson JF, Svedberg P, Olen O, et al. The longitudinal integrated database for health insurance and labour market studies (LISA) and its use in medical research. Eur J Epidemiol 2019;34:423–37.
21. Ludvigsson JF, Andersson E, Ekbom A, et al. External review and validation of the Swedish national inpatient register. BMC Public Health 2011;11:450.
22. Navkiran J, Backman AS, Linder M, et al. Validation of the use of the ICD-10 diagnostic code for irritable bowel syndrome in the Swedish National Patient Register. Gastroenterology 2014;146:S543.
23. Brooke HL, Talback M, Hornblad J, et al. The Swedish cause of death register. Eur J Epidemiol 2017;32:765–73.
24. Barlow L, Westergren K, Holmberg L, et al. The completeness of the Swedish cancer register: A sample survey for year 1998. Acta Oncol 2009;48:27–33.
25. Swedish Society of Pathology. KVAST-dokument. Lund, Sweden: Swedish Society of Pathology, 2018.
26. Westin M, Ahs A, Brand Persson K, et al. A large proportion of Swedish citizens refrain from seeking medical care—lack of confidence in the medical services a plausible explanation? Health Policy 2004;68:333–44.
27. Pappas G, Queen S, Hadden W, et al. The increasing disparity in mortality between socioeconomic groups in the United States, 1960 and 1986. N Engl J Med 1993;329:103–9.
28. Ludvigsson JF, Haberg SE, Knudsen GP, et al. Ethical aspects of registry-based research in the Nordic countries. Clin Epidemiol 2015;7:491–508.
29. Ford AC, Lacy BE, Talley NJ. Irritable bowel syndrome. N Engl J Med 2017;376:2566–78.
30. Canavan C, West J, Card T. The epidemiology of irritable bowel syndrome. Clin Epidemiol 2014;6:71–80.
31. Osby U, Brandt L, Correia N, et al. Excess mortality in bipolar and unipolar disorder in Sweden. Arch Gen Psychiatry 2001;58:844–50.
32. Zuo H, Ueland PM, Ulvik A, et al. Plasma biomarkers of inflammation, the Kynurenine pathway, and risks of all-cause, cancer, and cardiovascular disease mortality: The Hordaland Health Study. Am J Epidemiol 2016;183:249–58.
33. Kirsch R, Riddell RH. Histopathological alterations in irritable bowel syndrome. Mod Pathol 2006;19:1638–45.
34. Ludvigsson JF, Otterblad-Olausson P, Pettersson BU, et al. The Swedish personal identity number: Possibilities and pitfalls in healthcare and medical research. Eur J Epidemiol 2009;24:659–67.
35. Almario CV, Ballal ML, Chey WD, et al. Burden of gastrointestinal symptoms in the United States: Results of a nationally Representative survey of over 71,000 Americans. Am J Gastroenterol 2018;113:1701–10.
36. Canavan C, Card T, West J. The incidence of other gastroenterological disease following diagnosis of irritable bowel syndrome in the UK: A cohort study. PLoS One 2014;9:e106478.