The Changing Epidemiology of Inflammatory Bowel Disease: What Goes Up May Come Down

Abstract

Recent epidemiologic studies have shown that although the incidence of inflammatory bowel disease (IBD) is rapidly increasing in newly industrialized countries, at the turn of the 21st century the incidence had stabilized in the Western world. In this issue of Inflammatory Bowel Diseases, Torabi and colleagues present their findings on the temporal trends and geographic variations in IBD incidence in Manitoba from 1990 to 2012 using the Manitoba Health population registry and the University of Manitoba IBD epidemiology database. Their results demonstrate an overall decrease in the incidence of IBD during the study period. They also found significant regional variations in disease incidence within Manitoba, with rates of new diagnosis of IBD remaining high in several regions. Lastly, the study found that a higher proportion of the indigenous population had a lower rate of IBD. These findings provide new insights on the changing epidemiology of IBD in the Western world. The overall declining incidence of IBD and identification of persistently low and high-risk populations in Manitoba, which traditionally has had some of the highest incidence rates of IBD, is intriguing and can provide new avenues of research for epidemiologists in the field.

With globalization of inflammatory bowel disease (IBD), there has been a resurgence of interest in identifying the environmental determinants of the disease. Epidemiologic studies have shown a rapid rise in the incidence of IBD in newly industrialized countries in the Middle East, Asia, and South America over the past 2 decades¹. Conversely, although IBD incidence increased dramatically in the Western world in the 20th century, the rates have stabilized in the 21st century. These patterns of disease incidence highlight the role of westernization in development of IBD. A number of studies have examined the relationship between environmental factors associated with westernization and risk of IBD; unfortunately, very few environmental influences have consistently been linked to IBD. Therefore, identification of environmental risk factors continues to be an area of great unmet need.

Aside from case–control and cohort studies that commonly examine the relationship between 1 or more environmental factors in relation to disease status, ecologic studies have long been used in epidemiology to identify patterns of disease occurrence across geographic regions. These studies can broadly provide cues on potential lifestyle and environmental risk factors. Additionally, temporal trends in disease occurrence can also provide clues on potential risk factors associated with disease through identification of high-risk populations. In this issue of Inflammatory Bowel Diseases, Torabi and colleagues examined the temporal trends and geographic variations in IBD incidence in Manitoba, a central province in Canada, from 1990 to 2012². Their results demonstrate an overall decrease in the incidence of IBD during the study period. Specifically, the incidence declined, at 1.89% and 0.96% per year in Crohn’s disease (CD) and ulcerative colitis (UC), respectively. They also found significant regional variations in disease incidence within Manitoba. Winnipeg and southern regions had the highest incidence of IBD, whereras the central and northern regions had the lowest incidence. Interestingly, the authors found that despite the overall decline in incidence during the study period, rates of new diagnosis of IBD remained high in several small regions within Manitoba. Lastly, the study found that a higher proportion of the indigenous population had a lower rate of IBD. In contrast, other sociodemographic factors such as average household income and proportion of immigrants and minorities were not associated with rates of IBD in multivariable analyses.

Torabi and colleagues used the Manitoba Health population registry, which contains records on demographic information (sex, date of birth), place of residence, migration, and death on all residents of the province. Cases of IBD were identified using the University of Manitoba IBD epidemiology database, a validated population-based registry. The authors used postal code and census subdivision code of residence to assign each IBD case to 1 of 296 small geographic areas (SGAs) in Manitoba. A spatial Poisson regression was used to model IBD incidence standardized to the censor population in 2006 according to SGAs. Temporal trends in IBD incidence were examined using average annual percentage changes (APCs).

There are a number of strengths that are worth noting. The large sample size, comprehensive IBD cohort, detailed information on SGAs and demographics, and long follow-up time allowed for finer characterization of the relationship between IBD incidence and geographic location. There are several limitations that are also worth acknowledging. Most importantly, the study design is subject to the ecologic fallacy, which occurs by inferring that associations at the aggregate level are true at the individual level. Additionally, compared with studies that examine data at the individual level, ecologic studies are particularly susceptible to confounding. For instance, lower incidence of IBD in the indigenous population may in fact be related to variations in genetic predisposition in this population and not environmental factors.

The findings in the study raise several important questions that deserve future research. First, studying SGAs with similar socioeconomic and ethnic makeup but with significant differences in IBD incidence can help eliminate the possibility that variations observed may be related to nonenvironmental factors such as genetic makeup and diagnostic biases related to regional differences in health care utilization. Second, it will be interesting to examine whether there are similar trends in the incidence of other immune-mediated disorders in Manitoba during the same time period. This could help guide research toward studying environmental risk factors that are shared across immune-mediated disorders. Third, the findings that the incidence of IBD has already peaked in the early part of the 21st century and is now declining in Manitoba may at first appear contradictory to the general belief that IBD incidence is rising. However, a recent systematic review by Ng and colleagues also demonstrated that the incidence of IBD is stabilizing in Western countries¹. Additionally, at least 1 other study has also demonstrated declining incidence in IBD in eastern provinces in Canada³. It is, however, unclear whether other regions in the west may also be experiencing declining IBD incidence. Lastly, identification of low- and high-risk populations can guide future research toward establishing cohorts aimed at identifying environmental and genetic drivers of the disease.

In conclusion, this study provides new insights on the changing epidemiology of IBD in the Western world. The overall declining incidence of IBD and identification of persistently low- and high-risk populations in Manitoba, which traditionally has had some of the highest incidence rates of IBD, are intriguing and can provide new avenues of research for epidemiologists in the field.

Supported by: Dr. Khalili is supported by the National Institute of Diabetes and Digestive and Kidney Diseases (NIDDK; R03 DK113337). Dr. Khalili is also supported by the American Gastroenterological Association (AGA) Pfizer Young IBD Investigator Grant, the American College of Gastroenterology (ACG) Pilot Grant, and the Helmsley Charitable Trust Foundation.

Conflicts of interest: Dr. Khalili has received consulting fees from Abbvie and receives grant funding from Pfizer and Takeda.