American Journal of Gastroenterology ISSN 0002-9270 C  2007 by Am. Coll. of Gastroenterology doi: 10.1111/j.1572-0241.2007.01232.x Published by Blackwell Publishing Utilization of Health-Care Resources by Patients With IBD in Manitoba: A Profile of Time Since Diagnosis Teresa Longobardi, Ph.D., and Charles N. Bernstein, M.D. University of Manitoba Inflammatory Bowel Disease Clinical and Research Centre and Department of Medicine, Winnipeg, Manitoba, Canada OBJECTIVES: We tested the hypothesis of nonlinear longitudinal trends in health-care utilization by individuals with Crohn’s disease (CD) and ulcerative colitis (UC) in Manitoba. METHODS: Administrative databases were used to report resource use in 2000/1. A total of 5,485 cases of CD and UC and 45,279 matched controls were separated into incident cases (0–2 yr of disease), cases with longstanding disease (3–10 yr), and cases with very longstanding disease (>10 yr). Relative risk ratios (RRR) indicating the likelihood of resource use, given disease duration, were computed using multinomial logistic regression analysis. Sensitivity analysis was conducted to test the robustness of results to altering the disease duration cutoffs. RESULTS: Independent of disease duration, in general, outpatient utilization was over twice as likely among IBD cases compared with controls whether or not the contact was made for IBD-specific reasons. The likelihood of utilization was greatest among incident cases for outpatient visits with an internist (RRR 6.16, 95% CI 5.11–7.43) and surgical visits (RRR 3.78, 95% CI 3.14–4.55). Inpatient stays for IBD-specific reasons in general were considered dependent on disease duration; in particular, there was a fourfold higher likelihood for the incident cases relative to their controls. For non-IBD-specific reasons, IBD cases were 1.5 times as likely to have inpatient stays, regardless of disease duration. CONCLUSIONS: Our results suggest that within the first 2 yr from disease diagnosis the most costly resources were employed. We can likely measure the greatest proportion of treatment effects on resource use within a relatively short period. (Am J Gastroenterol 2007;102:1683–1691) INTRODUCTION Crohn’s disease (CD) and ulcerative colitis (UC) are chronic inflammatory bowel diseases (IBD). These diseases are char- acterized by periods of relapse and remission following initial onset. Although there are episodic flares, 75–90% of patients can be found in remission at any one time (1). This makes dis- tinguishing or stratifying the analysis between symptomatic and asymptomatic patients important in any analysis of re- source utilization; a state which is dependent on disease du- ration. Although mortality rates are slightly higher than that of the general population (2–5), particularly for Crohn’s dis- ease (6), patients typically live many years to nearly a full life span with these diseases. Disease flares impact quality of life, including the ability to work, and may lead to costly hospital- izations that include surgery in the most severe exacerbations. As peak IBD incidence is typically in early adulthood (7), the lifetime burden of disease is estimated to be substantial (8, 9). However, few population-based IBD datasets have been used to describe how the history of disease impacts lifetime health- care utilization patterns and its resulting economic burden on society. With the advent of novel expensive biological therapeutics to treat IBD, to adequately evaluate the cost-effectiveness of these therapies, we believe it is necessary to establish base- line estimates in order to be able to quantify the direct and indirect cost savings that may result from reduced hospital visits, length of stays, surgeries, and reductions in work loss. In this current study we estimated the profile of time since diagnosis of the IBD population in Manitoba, Canada. We adapted the University of Manitoba IBD Epidemiology Database (UMIBDED) (7) to test the hypothesis of a constant trend in utilization attributable to IBD following initial inci- dence. The methodology applied in our earlier work with the U.S. National Health Information Survey (NHIS) data was applied to this study (10, 11). METHODS The University of Manitoba IBD Epidemiology Database Manitoba is a central Canadian province with a rela- tively stable population of approximately 1.1 million in- habitants (http://www.citypopulation.de/index.html). It grew about 2.5% over the 10 yr from April 1991 to April 2001. 1683