You are here
A model of an inflammatory bowel disease population-based registry: The Forlì experience (1993–2013)
Digestive and Liver Disease, Available online 21 September 2017
Background and aims
The article presents a population-based registry designed to estimate incidence and prevalence of inflammatory bowel disease (IBD) in the area of Forlì (north-eastern Italy).
The registry included all patients with IBD ulcerative colitis (UC) and Crohn’s disease (CD) in the Forlì area from 1993 to 2013. A data manager matched records from various sources.
Seven hundred ninety-one patients were registered during the study period, 564 (71.3%) with UC and 227 (28.7%) with CD. The standardized annual incidence rate for UC was 12.8 per 100,000 females (95% CI 11.1–14.4) and 15.7 per 100,000 males (95% CI 13.9–17.5). That of CD was 7.0 per 100,000 for females (95% CI 5.7–8.3) and 5.4 per 100,000 males (95% CI 4.3–6.4). The prevalence of CD and UC on 1 January 2014 was 109.2 per 100,000 inhabitants (95% CI 94.3–124.2) and 266.4 per 100,000 inhabitants (95% CI 243.4–289.7), respectively.
Although further studies are needed, the data suggest that incidence and prevalence of IBD in Italy are underestimated.
Keywords: Crohn’s disease, Incidence, Population-based registry, Prevalence, Ulcerative colitis.
Inflammatory bowel diseases (IBD), consisting of ulcerative colitis (UC) and Crohn’s disease (CD), are characterized by chronic inflammation of the gastrointestinal tract in genetically susceptible individuals exposed to environmental risk factors 1 2 3 . The etiology of IBD has been extensively studied in the past few decades  , but pathogenesis is not fully understood 1 4 .
Considerable variation in the incidence of IBD has been observed around the world, with a wide range of estimates both within and between geographic regions. IBD is believed to be associated with industrialization, with the highest incidence and prevalence rates of IBD being observed in North America and Europe  . The incidence of IBD may be rising in developing nations as they have become industrialized 6 7 , although population-based data to confirm this trend are sparse. Several studies have reported that the incidence of IBD has increased markedly over the last part of the 20th century, while other studies have suggested a plateau or even a decline in certain geographic regions 8 7 . Earlier diagnosis, more effective treatments and better clinical management have also had an impact on the epidemiology of IBD, since they have led to an increasing prevalence 9 10 .
In order to establish the burden of IBD, to improve our understanding of the causal mechanisms, and to estimate the economic and social consequences of IBD, epidemiological studies may play a key role. Population-based registries are considered the most accurate approach for collecting epidemiological data.
With this rationale, we created a population-based IBD registry in the area of Forlì, northeast Italy. In the present article, we report a study that was primarily aimed at evaluating IBD incidence and prevalence in for the years 1993–2013. The secondary objective was to perform a preliminary evaluation of incidence trends.
2.1 The IBD registry
The population-based IBD registry, the first of its kind in Italy, was established in 2009 as a joint project between the Department of Gastroenterology and Digestive Endoscopy of the Morgagni-Pierantoni Hospital and the Romagna Cancer Registry. On 31 December 2016, the population of the Health Care District area of Forlì was 186,330, of whom 90,137 were men and 96,193 women. The area is served by a single referral centre for IBD, where the registry is based. Gastroenterology departments of comparable level of care are situated in most of the other health districts (total n = 11) of the Emilia-Romagna Region.
The registry was created as a specialised registry, i.e., an epidemiological structure capable of ensuring continuous and exhaustive recording of a given disease in a given geographic region. The objectives of the registry are: (1) to establish the descriptive epidemiology of the disease; (2) to identify the risk factors and the changes in their prevalence over time; and (3) to evaluate the impact of health policy on the control of the disease.
The data collection began retrospectively for the years 1993–2009, and continued prospectively. The registry is based on the following sources of information: the hospital discharge records (for patients treated both in the area of Forlì and elsewhere in Italy); the diagnostic services, in particular the pathology departments; the general practitioner databases; the Department of Gastroenterology and Digestive Endoscopy database; the health administrative services; and the death certificate registries.
An adapted version of the information software system developed by the Romagna Cancer Registry of the Istituto Scientifico Romagnolo per lo Studio e la Cura dei Tumori (IRST) IRCCS, Meldola, was used. A personal record was created for each patient comprising personal data (surname, name, date and place of birth, place of residence and date of migration) obtained from the municipal population registries (n = 15).
Incident cases are coded according to the International Classification of Diseases, 9th revision, Clinical Modification (ICD-9-CM). All available information with the following ICD-O codes are reviewed by a gastroenterologist: 555.0, 555.1, 555.9, 556, 556.0, 556.1, 556.2, 556.3, 556.4, 556.5, 556.6, 556.8, and 556.9. Confirmed cases of CD and UC are registered with the code 555 and 556, respectively.
The accuracy of data is checked using the standard method of cancer registration  . In particular, the place of residence at the time of first diagnosis of IBD – a prerequisite for inclusion in the incidence archive of the registry – is checked through the municipal population registries or through the archives of the district health Authority  .
The vital status of all incident cases was updated on 13 December 2013. Information was obtained through record linkage with the regional mortality registry and the local municipal population registries. Incident cases of IBD between 1993 and 2011 were record-linked to the Romagna Cancer Registry to identify the patients who developed cancer after IBD (non melanoma skin cancers were excluded).
The present study was approved by an Institutional Ethics Committee (ID: IRST 100.37).
2.2 Statistical analysis
The absolute frequency distribution and percentage by disease and gender were calculated. The patient age was calculated using the date of birth and the date of diagnosis of IBD. The median age and 25th and 75th age percentiles were calculated for each condition.
Crude and age-standardized incidence rates per 100,000 residents were calculated. Annual number of residents by sex and 5-year age group were obtained from the municipal population registries. The standardized incidence rates were calculated with the direct method using the European population as a standard. Ninety-five confidence intervals (95% CI) around the standardized rates were calculated assuming a Poisson distribution.
The 21-year prevalence rates were calculated as point rates on 1st January 2014. The cases who died with the disease and those diagnosed with IBD before 1993 were excluded.
During 1993–2013, the IBD Registry recorded information on 791 patients ( Fig. 1 ). There were 564 (71.3%) cases of UC and 227 (28.7%) of CD. The standardized annual incidence rate for UC was 12.8 per 100,000 females (95% CI 11.1–14.4) and 15.7 per 100,000 males (95% CI 13.9–17.5) ( Table 1 ). Among females, two incidence peaks were observed at age 20–24 years and 45–49 years ( Fig. 2 ). The standardized annual incidence rate for CD was 7.0 per 100,000 female inhabitants (95% CI 5.7–8.3) and 5.4 per 100,000 male inhabitants (95% CI 4.3–6.4) ( Table 1 ). Among females, an incidence peak was found at age 20–24 years, followed by a decrease in older age groups. A decrease after the age of 50 was observed for both genders ( Fig. 3 ).
|CD||No.||Crude rate (x100,000 inhabitants)||Age-standardized (Europe) rate (x100,000 inhabitants)||95% CI|
|UC||No.||Crude rate (×100,000 inhabitants)||Age-standardized (Europe) rate (×100,000 inhabitants)||95% CI|
UC incidence rates did not change substantially over time (1993–1995: UC incidence rate 13.3; 2011–2013: UC incidence rate 13.8) ( Fig. 4 ), whereas CD incidence rates decreased from 2002 onwards and reached the level of 3.9 in 2011–2013 ( Fig. 5 ).
Twenty-four (3.0%) patients were lost to follow-up. On 1st December 2014, 705 (89%) patients were alive and 62 (8.1%) had died. The prevalence of CD and UC on 1st January 2014 – not reported elsewhere – was 109.2 per 100,000 inhabitants (95% CI 94.3–124.2) and 266.4 per 100,000 inhabitants (95% CI 243.4–289.7), respectively.
The record linkage with the Romagna Cancer Registry, updated in 2011, showed that 42 of the 721 patients with an IBD diagnosis during the period 1993–2011 developed a cancer. Five of them were diagnosed with colorectal cancer (data not shown).
Available information about time trends in incidence, prevalence and public health impact of IBD in Italy is not fully reliable. As demonstrated by a recent Italian review about CD, there are many data but coming from heterogeneous sources  . Many studies, the most important being the EpiCom experience, have shown the importance of conducting large epidemiological studies and meta-analyses to better understand IBD trends, risk factors, and behaviour  . In Italy, this research effort is still in its early stages. In part, this reflects the idea that IBDs is a condition affecting mainly North American and North European populations. Growing incidence rates in Mediterranean populations indicate that this is no longer the case.
A review of descriptive epidemiological studies from North America by Loftus et al., published in 2004, reported UC and CD incidence rates varying between 2.2 and 14.3 per 100,000 inhabitants for UC and between 3.1 and 14.6 cases for CD  . The range of prevalence rates per 100,000 inhabitants was 37–246 and 26–199, respectively.
In Europe, according to the EpiCom study, annual incidence rates are 8.2 for UC, 5.4 for CD, and 1.7 for IBD-Unclassified. The study found that the combined annual incidence rates of UC and CD in Western Europe is twice as high as the rates in Eastern Europe. This gradient was smaller than originally expected based on the North–South gradient previously observed, probably due to bias in inception cohorts.
A high degree of variability has been a constant limitation of Italian data. In the 1990s, two population-based studies were carried out. The work based in Florence, by Trallori et al., yielded UC and CD incidence rates of 9.6 and 3.4 per 100,000 inhabitants/year, respectively, and prevalence rates of 121 and 40 per 100,000 inhabitants, respectively  . The multicentre study conducted by Tragnone et al., a collaborative project between our institution and other seven centres located in north and central Italy, reported UC and CD incidence rates of 5.2 and 2.3 per 100,000 inhabitants/year, respectively  . Two studies covering the administrative regions of Lombardia and Lazio reported UC rates of 9.7 and 12.2/14.5, and CD rates of 5.9 and 7.0 17 18 . According to a joint report by three Italian Gastroenterology and Endoscopy scientific associations, the national incidence in adults can be estimated at 7.2 for UC and 3.9 for CD  .
Overall, Italy seems to occupy an intermediate position but is closer to Northern Europe 13 20 . A very recent epidemiological review of CD studies has confirmed a high variability of the reported rates. Factors accounting for this include the lack of a national registry (except for paediatric patients) and differences in data sources and study methods. In several geographic areas, the reported incidence rates range between 2.3 and 15.4 per 100,000  .
Thus, our data on CD are comparable to those from the literature, especially the ECCO EpiCom study and the most recent Italian studies. Conversely, the observed incidence of UC was higher than expected, the interpretation of which is not straightforward. In the future, detailed analysis of incidence data will clarify this finding. In line of principle, we cannot exclude the hypothesis of an over-registration bias, that is, the inclusion of cases of ischaemic colitis and segmental colitis-associated diverticulosis with clinical and histological features overlapping those of UC.
As regards time trends, the preliminary analysis presented here suggests that UC incidence has been stable over time. CD rates, conversely, have reached a peak around the turn of the century followed by a decrease in the subsequent decade. Considering that IBD incidence is increasing worldwide, we suggest that this trend may reflect changes in clinicians’ awareness. Expanding the registration in time and space will allow a better understanding of the temporal pattern of the disease.
The population-based registry described here is a simple tool to use and, in our opinion, could be adapted to cover a wider geographical area. In fact, for the past two years we have been working towards extending the registry to cover the entire region of Romagna (1,100,000 inhabitants). The most significant obstacles to overcome will probably be the heterogeneity in data sources (low level of data digitalization, numerous healthcare institutes with different clinical records systems) and the need for closer collaboration between physicians and data managers.
Comments on the incidence of cancer are premature, given the limited duration of follow-up and – more important – the young age of IBD patients at registration. In the future, a record-linkage with the Romagna Cancer Registry to evaluate the cancer incidence will be among our primary objectives.
Carrying out epidemiological studies on IBD is a complex task due to the insidious nature of the diseases, the variability of clinical manifestations, and the difference in the ways in which diagnostic procedures are performed in each referral center. An active collaboration between the specialized personnel of an IBD Unit and epidemiologists (especially for retrospective studies) and the creation an efficient registration system are essential for the accuracy of epidemiological studies. An IBD population-based registry has an enormous potential. The IBD registry of Forlì is the first of its kind in Italy and will eventually cover the entire region of Romagna (more than 1,100,000 inhabitants). The comprehensive nature of the registry will guarantee accurate IBD incidence and prevalence rates and an improved organization of patient care. In addition, a record-linkage with the Romagna Cancer Registry will allow to perform cohort studies of cancer incidence in IBD patients.
Conflicts of interest
The authors wish to thank Dr. Stefania Maltoni for her help in data collection.
-  A.-M. Lowe, P.-O. Roy, M. B-Poulin, P. Michel, A. Bitton, L. St-Onge, P. Brassard, et al. Epidemiology of Crohn’s disease in Quebec, Canada. Inflamm Bowel Dis. 2009;15:429-435
-  C.N. Bernstein, J.F. Blanchard, P. Rawsthorne, A. Wajda, et al. Epidemiology of Crohn’s disease and ulcerative colitis in a central Canadian province: a population-based study. Am J Epidemiol. 1999;149:916-924
-  C.N. Bernstein, A. Wajda, L.W. Svenson, A. MacKenzie, M. Koehoorn, M. Jackson, et al. The epidemiology of inflammatory bowel disease in Canada: a population-based study. Am J Gastroenterol. 2006;101:1559-1568
-  C.G. Loftus, E.V. Loftus, W.S. Harmsen, A.R. Zinsmeister, W.J. Tremaine, J.L. Melton, et al. Update on the incidence and prevalence of Crohn’s disease and ulcerative colitis in Olmsted County, Minnesota, 1940–2000. Inflamm Bowel Dis. 2007;13:254-261
-  J.H. Gollop, S.F. Phillips, L.J. Melton, A.R. Zinsmeister, et al. Epidemiologic aspects of Crohn’s disease: a population based study in Olmsted County, Minnesota, 1943–1982. Gut. 1988;29:49-56
-  E.V. Loftus, M.D. Silverstein, W.J. Sandborn, W.J. Tremaine, W.S. Harmsen, A.R. Zinsmeister, et al. Ulcerative colitis in Olmsted County, Minnesota, 1940–1993: incidence, prevalence, and survival. Gut. 2000;336-343
-  E.V. Loftus, M.D. Silverstein, W.J. Sandborn, W.J. Tremaine, W.S. Harmsen, A.R. Zinsmaster, et al. Crohn’s disease in Olmsted County, Minnesota, 1940–1993: incidence, prevalence, and survival [erratum appears in Gastroenterology 1999;116:1507]. Gastroenterology. 1998;114:1161-1168
-  R.E. Sedlack, J. Whisnant, L.R. Elveback, L.T. Kurland, et al. Incidence of Crohn’s disease in Olmsted County, Minnesota, 1935–1975. Am J Epidemiol. 1980;112:759-763
-  D.K. Podolsky. Inflammatory bowel disease. N Engl J Med. 2002;347:417-429
-  A. Ekbom. The epidemiology of IBD: a lot of data but little knowledge. How shall we proceed?. Inflamm Bowel Dis. 2004;10(Suppl. 1):S32-S34
-  O.M. Jensen, D.M. Parkin, R. Maclennan, C.S. Muir, R.G. Skeet. Cancer registration: principles and methods. (IARC-International Agency for Research on Cancer and International Association of Cancer Registries, 1991) Scientific publications N.95
-  C. Galeone, C. Pelucchi, G. Barbera, C. Citterio, C. La Vecchia, A. Franchi, et al. Crohn’s disease in Italy: a critical review of the literature using different data sources. Dig Liver Dis. 2017;49:459-466
-  J. Burisch, N. Pedersen, S. Čuković-Čavka, M. Brinar, I. Kaimakliotis, D. Duricova, O. Shonová, I. Vind, S. Avnstrøm, et al. East-West gradient in the incidence of inflammatory bowel disease in Europe — the ECCO-EpiCom inception cohort. Gut. 2014;63:588-592
-  E. Loftus. Clinical epidemiology of inflammatory bowel disease: incidence, prevalence, and environmental influences. Gastroenterology. 2004;126:1504-1517
-  G. Trallori, D. Palli, C. Saieva, G. Bardazzi, A.G. Bonanomi, G. d’Albasio, et al. A population-based study of inflammatory bowel disease in Florence over 15 years (1978–92). Scand J Gastroenterol. 1996;31:892-899
-  A. Tragnone, G. Corrao, F. Miglio, R. Caprilli, G.A. Lanfranchi. Incidence of inflammatory bowel disease in Italy: a nationwide population-based study. Int J Epidemiol. 1996;25:1044-1052
-  R. Di Domenicantonio, G. Cappai, M. Arca, N. Agabiti, A. Kohn, P. Vernia, et al. Occurrence of Inflammatory bowel disease in central Itay: a study based on health information systems. Dig Liver Dis. 2014;46:777-782
-  C. Fornari, F. Madotto, G. Fiorino, S. Ardizzone, A. Bortoli, F. Caprioli, et al. Inflammatory bowel disease in Italy: incidence trends and patients’ characteristics. Value Health. 2013;16:A501
-  AIGO, SIED, SIGE. Libro Bianco della Gastroenterologia Italiana. (2011)
-  S. Shivananda, J. Lennard-Jones, R. Logan, N. Fear, A. Price, L. Carpenter, et al. Incidence of inflammatory bowel disease across Europe: is there a difference between north and south? Results of the European Collaborative Study on Inflammatory Bowel Disease (EC-IBD). Gut. 1996;39:690-697