Advertisement

Rate of Adverse Events and Associated Health Care Costs for the Management of Inflammatory Bowel Disease in Germany

Open AccessPublished:December 26, 2019DOI:https://doi.org/10.1016/j.clinthera.2019.11.012

      ABSTRACT

      Purpose

      Therapeutic management of inflammatory bowel disease (IBD) is challenging, and available therapies are associated with adverse events (AEs) that may lead to treatment discontinuation. This study evaluated the rate of drug-related AEs of special interest (AESIs) associated with IBD therapies and compare health care costs among patients with IBD who did and did not experience AESIs.

      Methods

      A retrospective cohort analysis was conducted using claims data from a German Sickness Fund (Allgemeine Ortskrankenkasse PLUS). Patients were diagnosed with ulcerative colitis (UC) or Crohn's disease (CD) and newly initiating treatment with immunosuppressant, anti–tumor necrosis factor α, or anti-integrin therapies from January 1, 2011, to December 31, 2015. Patients were required to have continuous insurance coverage and no evidence of use of these IBD therapies for 12 months before the date of newly initiating therapy (index date). Rates of AESIs were based on 28 different events or chronic conditions associated with IBD treatment. Direct health care costs were reported separately for patients who did or did not experience AESIs. Only treatment periods lasting ≥60 days were considered. AESI rates related to all possible treatment patterns were calculated and reported as the number of events per 10,000 patient-years. Health care costs were calculated based on IBD-related health care resource use.

      Findings

      A total of 1126 (CD, n = 676; UC, n = 450) patients met the inclusion criteria. Mean age was 36.5 years for patients with CD and 42.5 years for patients with UC; 60.5% and 47.6% were female, respectively. Median observed time since the index date was 1460 and 1552 days, for patients with CD and UC. The overall rate for any AESI was 1392.4 and 1917.9 events per 10,000 patient-years in patients with CD and those with UC. Severe infections and diabetes mellitus were the most common AESIs. Significant differences in mean total direct health care costs were found for CD patients with AESIs versus those without (€8920.08 and €6004.86; P < 0.001). A similar trend was observed with mean drug costs and mean medical costs. In UC, total direct health care costs, although generally higher in patients with AESIs, were not significantly different; however, medical costs were (€1946.93 vs €971.28; P < 0.001).

      Implications

      AEs are common in patients with IBD treated with current therapies and associated with substantial health care costs. An urgent need exists for development of IBD treatments that are associated with lower rates of AEs.

      Keywords

      Introduction

      Crohn's disease (CD) and ulcerative colitis (UC) are the main types of chronic inflammatory bowel disease (IBD); both are characterized by a relapsing course but with different manifestations.
      • Hein R.
      • Koster I.
      • Bollschweiler E.
      • Schubert I.
      Prevalence of inflammatory bowel disease: estimates for 2010 and trends in Germany from a large insurance-based regional cohort.
      • Baumgart D.C.
      The diagnosis and treatment of Crohn's disease and ulcerative colitis.
      • Baumgart D.C.
      • Sandborn W.J.
      Inflammatory bowel disease: clinical aspects and established and evolving therapies.
      • Prenzler A.
      • Mittendorf T.
      • Conrad S.
      • von der Schulenburg J.M.
      • Bokemeyer B.
      [Costs of Crohn's disease in Germany from the perspective of the statutory health insurance].
      • Floyd D.N.
      • Langham S.
      • Severac H.C.
      • Levesque B.G.
      The economic and quality-of-life burden of Crohn's disease in Europe and the United States, 2000 to 2013: a systematic review.
      • Burisch J.
      • Jess T.
      • Martinato M.
      • Lakatos P.L.
      The burden of inflammatory bowel disease in Europe.
      • Kaplan G.G.
      The global burden of IBD: from 2015 to 2025.
      IBD can present at a young age,
      • Kaplan G.G.
      The global burden of IBD: from 2015 to 2025.
      ,
      • Ananthakrishnan A.N.
      Epidemiology and risk factors for IBD.
      and the disease requires long-term, often costly treatment.
      • Floyd D.N.
      • Langham S.
      • Severac H.C.
      • Levesque B.G.
      The economic and quality-of-life burden of Crohn's disease in Europe and the United States, 2000 to 2013: a systematic review.
      ,
      • Limsrivilai J.
      • Stidham R.W.
      • Govani S.M.
      • Waljee A.K.
      • Huang W.
      • Higgis P.D.R.
      Factors that predict high health care utilization and costs for patients with inflammatory bowel diseases.
      The traditional treatment goal in IBD is to control inflammation and alleviate symptoms, and to reach corticosteroid-free remission of the disease.
      • Gomollon F.
      • Dignass A.
      • Annese V.
      • et al.
      European evidence-based consensus on the diagnosis and management of Crohn's disease 2016: part 1: diagnosis and medical management.
      • Harbord M.
      • Eliakim R.
      • Bettenworth D.
      • et al.
      Third European evidence-based consensus on diagnosis and management of ulcerative colitis. Part 2: current management.
      • Peyrin-Biroulet L.
      • Lemann M.
      Review article: remission rates achievable by current therapies for inflammatory bowel disease.
      Current IBD treatment guidelines recommend aminosalicylates (ASA), conventional immunosuppressants, or biologics for long-term disease management.
      • Dignass A.
      • Preiss J.C.
      • Aust D.E.
      • et al.
      Updated German guideline on diagnosis and treatment of ulcerative colitis, 2011.
      ,
      • Preiss J.C.
      • Bokemeyer B.
      • Buhr H.J.
      • et al.
      Updated German clinical practice guideline on "Diagnosis and treatment of Crohn's disease" 2014.
      For patients who fail to improve with conventional therapy, the choice of a specific agent for treatment escalation is driven by disease course and patient-based criteria such as comorbid conditions and previous therapies.
      • Gomollon F.
      • Dignass A.
      • Annese V.
      • et al.
      European evidence-based consensus on the diagnosis and management of Crohn's disease 2016: part 1: diagnosis and medical management.
      ,
      • Harbord M.
      • Eliakim R.
      • Bettenworth D.
      • et al.
      Third European evidence-based consensus on diagnosis and management of ulcerative colitis. Part 2: current management.
      ,
      • Dignass A.
      • Preiss J.C.
      • Aust D.E.
      • et al.
      Updated German guideline on diagnosis and treatment of ulcerative colitis, 2011.
      ,
      • Preiss J.C.
      • Bokemeyer B.
      • Buhr H.J.
      • et al.
      Updated German clinical practice guideline on "Diagnosis and treatment of Crohn's disease" 2014.
      If disease remission cannot be achieved with the first-line treatment and/or there are safety concerns associated with it, guidelines recommend a second or subsequent line of therapy with an agent previously listed. In Europe, the anti–tumor necrosis factor α (anti-TNFα) agents infliximab and adalimumab and the anti-integrin antibody vedolizumab are among the biologic therapies indicated for patients with moderately to severely active CD and UC who have failed to improve with conventional treatment, or an anti-TNFα agent in the case of vedolizumab, or who have contraindications for such therapies.
      The anti-TNFα agent golimumab is indicated for UC only.
      Few IBD treatments result in patients achieving complete clinical remission. In addition, patients often experience low initial effectiveness and loss of response to IBD treatment, and a substantial proportion of patients experience drug-related adverse events that can lead to early discontinuation of therapy.
      • Preiss J.C.
      • Bokemeyer B.
      • Buhr H.J.
      • et al.
      Updated German clinical practice guideline on "Diagnosis and treatment of Crohn's disease" 2014.
      ,
      • Loftus Jr., E.V.
      • Davis K.L.
      • Wang C.C.
      • Dastani H.
      • Luo A.
      Treatment patterns, complications, and disease relapse in a real-world population of patients with moderate-to-severe ulcerative colitis initiating immunomodulator therapy.
      • Narula N.
      • Kainz S.
      • Petritsch W.
      • et al.
      The efficacy and safety of either infliximab or adalimumab in 362 patients with anti-TNF-alpha naive Crohn's disease.
      Updated S3-guideline colitis ulcerosa. German Society for Digestive and Metabolic Diseases (DGVS)—AWMF registry 021/009.
      • Marehbian J.
      • Arrighi H.M.
      • Hass S.
      • Tian H.
      • Sandborn W.J.
      Adverse events associated with common therapy regimens for moderate-to-severe Crohn's disease.
      Furthermore, health care resource use (HCRU), as well as direct and indirect health care costs (the latter of which are related to lost working days), associated with IBD treatment are high, thus imposing a significant burden to patients, payers, and society.
      • Prenzler A.
      • Mittendorf T.
      • Conrad S.
      • von der Schulenburg J.M.
      • Bokemeyer B.
      [Costs of Crohn's disease in Germany from the perspective of the statutory health insurance].
      ,
      • Floyd D.N.
      • Langham S.
      • Severac H.C.
      • Levesque B.G.
      The economic and quality-of-life burden of Crohn's disease in Europe and the United States, 2000 to 2013: a systematic review.
      ,
      • Karve S.
      • Candrilli S.
      • Kappelman M.D.
      • Tolleson-Rinehart S.
      • Tennis P.
      • Andrews E.
      Healthcare utilization and comorbidity burden among children and young adults in the United States with systemic lupus erythematosus or inflammatory bowel disease.
      ,
      • Prenzler A.
      • Bokemeyer B.
      • von der Schulenburg J.M.
      • Mittendorf T.
      Health care costs and their predictors of inflammatory bowel diseases in Germany.
      It is currently unclear to what extent adverse events are associated with IBD treatment and how they affect the overall economic burden of IBD. Addressing these knowledge gaps may provide an understanding of real-world treatment patterns and associated HCRU and costs and highlight the need for additional treatment options for patients with IBD. The aim of the current study was: (1) to evaluate the rate of drug-related adverse events of special interest (AESIs) associated with IBD treatment among patients with IBD, and (2) to compare direct health care costs between patients with IBD who experienced an AESI versus those who did not.

      Patients and methods

      Study Design

      This retrospective cohort study used claims data from a regional German Sickness Fund (Allgemeine Ortskrankenkasse PLUS). In accordance with German law and the policy of the institutions, ethical approval and informed consent were not required because anonymized data were used. The dataset included, on a patient level, the following: data regarding all outpatient physician visits and diagnoses; inpatient hospitalizations, including length of stay, primary diagnoses, and performed procedures; outpatient prescription medications for IBD dispensed; and main sociodemographic characteristics of patients. The dataset covered 2 regions of Germany (Saxony and Thuringia), which, because of uniform reimbursement rules across Germany, have frequently been used in the past to describe the real-world treatment of German patients.
      • Wilke T.
      • Groth A.
      • Fuchs A.
      • et al.
      Real life treatment of diabetes mellitus type 2 patients: an analysis based on a large sample of 394,828 German patients.
      ,
      • Groth A.
      • Wilke T.
      • Borghs S.
      • Gille P.
      • Joeres L.
      Real life pharmaceutical treatment patterns for adult patients with focal epilepsy in Germany: a longitudinal and cross-sectional analysis of recently approved anti-epileptic drugs.

      Patients

      For inclusion in the study, patients were required to have received 2 confirmed outpatient International Classification of Diseases, Tenth Revision diagnoses of CD or UC (K50.-/K51.-) in 2 different quarters of the year and/or 1 respective inpatient diagnosis between January 1, 2011, and December 31, 2015, and be continuously insured by the sickness fund from the first observed IBD diagnosis until December 31, 2016. Death of a patient after the first IBD diagnosis was the only exception to this rule. If a patient received >1 diagnosis of both CD and UC, he or she was excluded from the sample. In addition, patients were required to have newly initiated treatment with an immunosuppressant (IS) (azathioprine, cyclosporine, methotrexate, or 6-mercaptopurine), an anti-TNFα (infliximab, adalimumab, golimumab, or certolizumab), or an anti-integrin (vedolizumab) during the identification period spanning January 1, 2013, to December 31, 2016.
      Because these therapies are indicated for the treatment of moderate to severe IBD, the initiation of these therapies was used as a proxy for at least moderate disease severity. The prescription date of the first treatment served as the study index date. Each patient was observed from the index date until end of data availability (December 31, 2016) or until the death of the patient or until occurrence of a medication gap related to any of the aforementioned treatments of >60 days, whichever occurred first.

      Adverse events of special interest

      AESIs included 28 events or chronic conditions, identified from the literature
      • Marehbian J.
      • Arrighi H.M.
      • Hass S.
      • Tian H.
      • Sandborn W.J.
      Adverse events associated with common therapy regimens for moderate-to-severe Crohn's disease.
      ,
      • McLean L.P.
      • Cross R.K.
      Adverse events in IBD: to stop or continue immune suppressant and biologic treatment.
      ,
      • McAuliffe M.E.
      • Lanes S.
      • Leach T.
      • et al.
      Occurrence of adverse events among patients with inflammatory bowel disease in the HealthCore Integrated Research Database.
      and validated by clinical experts to be potential AESIs of IBD treatment. AESIs were identified by means of the International Classification of Diseases, Tenth Revision and procedure codes according to German claims data. Only AESIs associated with inpatient encounters were considered (see the Supplemental Table I in the online version at https://doi.org/10.1016/j.clinthera.2019.11.012).
      Each observed patient-day was assigned to a specific treatment pattern (combinations of agents were defined as unique treatment patterns). Thus, each AESI could be assigned to a specific treatment pattern, and different subsequent treatment patterns per patient were possible.
      Treatment patterns of interest included ASA (mesalazine and sulfasalazine), oral corticosteroids (OCS) (prednisone, prednisolone, budesonide, hydrocortisone, methylprednisolone, and betamethasone), IS, anti-TNFα, and anti-integrin agents, as well as combinations of these agents. A sensitivity analysis was conducted to include only treatment patterns if they lasted ≥60 days, because AESIs observed during short treatment periods (<60 days) may not be related to the assigned treatment pattern but to previous ones.
      AESI rates related to treatment patterns were calculated and reported as the number of events per 10,000 patient-years (TTPY), considering the observed time for each treatment pattern. Results were reported for the most common 4 treatment patterns only.
      To account for any confounding factors, as available in the database, ANCOVA was conducted to assess the relation between AESI rates and treatment pattern, with the latter as a dependent variable. Treatment pattern, age, sex, comorbidities, and IBD-related symptoms in the baseline period (12 months before the index date) were used as independent variables and the observational time per treatment pattern per patient as weighting factor.

      HCRU and costs

      HCRU and costs were calculated from the perspective of a German sickness fund. Cost calculation was based on several resource use items. First, medical costs (inpatient and outpatient services) were calculated. Costs based on documented Diagnosis Related Group reimbursement, outpatient general practitioner and specialist visits (€0.10 per documented treatment point; the number of treatment points reflected the complexity of an outpatient general practitioner or specialist visit

      Vergütung steigt um 620 Millionen Euro [press release]. Published August 23, 2018. Accessed April 18, 2019. https://www.kbv.de/html/1150_36611.php.

      ), and rehabilitation stays (cost calculation according to reimbursed amounts). The second factor calculated was drug costs based on the list prices in respective years. The third factor was days absent from work because of IBD. Costs were based on average loss of productivity per day in Germany (in 2014, €165; in 2015, €193; and in 2016, €197).
      Direct health care costs represent all health care costs associated with IBD medical and drug services.

      Statistical analysis

      HCRU per observed patient-year and costs were calculated and reported separately for patients with CD and patients with UC by using descriptive statistics (mean, SD, median, and ranges). Depending on the type and distribution of the variables, Pearson's χ2 test, t test, or Mann–Whitney U test were used to calculate P values for comparisons. To account for the skewed distribution of costs, the Mann–Whitney U test was used to compare costs between patients who experienced an AESI versus those who did not. Generalized linear models were used to assess the relation between direct health care costs and specific treatment patterns, adjusting for available baseline characteristics.
      All data handling was executed in Microsoft SQL Server 2008 and Microsoft Excel 2010 (Microsoft Corporation, Redmond, Washington). All statistical analyses were performed in STATA 13.1 (StataCorp, College Station, Texas).

      Results

      Patients

      A total of 1126 (676 with CD; 450 with UC) patients who were newly initiating treatment with IS, anti-TNFα, or anti-integrin therapy were identified from the German sickness fund (Figure 1, Table I). Mean (SD) age for patients with CD was 36.5 years (14.5), and for patients with UC it was 42.5 years (17.5); 60.5% and 47.6% were female, respectively; and median observed time since the index date was 1460 and 1552 days. Patients with CD had a mean (SD) Charlson Comorbidity Index of 0.39 (0.92), and patients with UC had a mean Charlson Comorbidity Index of 0.78 (1.57). More patients with CD had an IBD-related surgery (CD, 6.8%; UC, 1.6%) and hospitalization (CD, 5.0%; UC, 3.6%) than patients with UC during the baseline period.
      Figure 1
      Figure 1Sample attrition diagram. Anti-TNFα = anti–tumor necrosis factor α; IBD = inflammatory bowel disease; IS = immunosuppressant; UC = ulcerative colitis
      Table ICharacteristics of study cohorts of immunosuppressant, anti–tumor necrosis factor α, and anti-integrin naive patients at the index date (baseline) or 12 months before the index date.
      CharacteristicPatients With CDPatients With UC
      Sample size676450
      Follow-up, d
       Mean1356.31399.6
       Median14601552
       SD453.0463.5
       95% CI1322.2–1390.51356.8–1442.5
      Age, years at index date
       Mean36.542.5
       Median3442
       SD14.517.5
       95% CI35.4–37.640.9–44.1
      Female sex60.5%47.6%
      OCS38.9%32.4%
      Aminosalicylate50.0%74.2%
      CD = Crohn's disease; OCS = oral corticosteroid; UC = ulcerative colitis.
      Note: All patients were naive to immunosuppressant, anti–tumor necrosis factor α, or anti-integrin treatment 12 months before the index date.
      Among patients with CD, IS was the most common index treatment, with 42.0% of observed days assigned to IS treatment, followed by anti-TNFα (24.9%), OCS + IS (9.0%), and ASA (6.0%) (Table II). Among patients with UC, IS was the most common index treatment, with 48.1% of observed days assigned, followed by anti-TNFα (16.4%), ASA (16.2%), and OCS + IS (5.8%).
      Table IIObserved treatment patterns among previously immunosuppressant (IS), anti–tumor necrosis factor α (anti-TNFα), and anti-integrin treatment-naive patients. Values are given as no. (%) of observed days.
      TreatmentPatients With CD (n = 676)Patients With UC (n = 450)Patients With CD (n = 676)Patients With UC (n = 450)
      Main AnalysisSensitivity Analysis
      No treatment
      None of the defined focus treatments.
      16,784 (4.3)7212 (2.5)
      OCS monotherapy21,133 (5.4)10,676 (3.6)18,230 (5.3)9236 (3.5)
      IS monotherapy162,964 (42.0)141,094 (48.1)155,028 (45.3)134,204 (51.2)
      Anti-TNFα monotherapy96,559 (24.9)47,931 (16.4)93,394 (27.3)46,389 (17.7)
      Anti-integrin monotherapy4580 (1.2)4479 (1.5)4381 (1.3)4305 (1.6)
      ASA monotherapy23,382 (6.0)47,551 (16.2)17,951 (5.2)39,366 (15.0)
      OCS + IS34,913 (9.0)17,008 (5.8)30,697 (9.0)13,550 (5.2)
      OCS + anti-TNFα12,876 (3.3)5572 (1.9)11,144 (3.3)4860 (1.9)
      OCS + anti-integrin1012 (0.3)1197 (0.4)774 (0.2)1105 (0.4)
      IS + anti-TNFα11,331 (2.9)7968 (2.7)9686 (2.8)7213 (2.8)
      IS + anti-integrin293 (0.1)849 (0.3)188 (0.1)797 (0.3)
      Anti-TNFα + anti-integrin230 (0.1)638 (0.2)146 (0.1)530 (0.2)
      OCS + IS + anti-TNFα1760 (0.5)689 (0.2)613 (0.2)466 (0.2)
      OCS + IS + anti-integrin44 (0.0)37 (0.0)
      OCS + anti-TNFα + anti-integrin0185 (0.1)108 (0.1)
      IS + anti-TNFα + anti-integrin116 (0.0)064 (0.0)
      ASA = aminosalicylate; CD = Crohn's disease; OCS = oral corticosteroid; UC = ulcerative colitis.
      Note: All patients were naive to IS, anti-TNFα, or anti-integrin 12 months before the index date.
      None of the defined focus treatments.

      Adverse events of special interest

      AESIs were identified in 15.5% of patients with CD and in 17.6% of patients with UC. The rate of any AESI (events per TTPY) was 1392.4 in patients with CD and 1917.9 in patients with UC. The most common AESIs for patients with CD (events per TTPY) were perianal disease (460.8), severe infections (249.1), diabetes mellitus (211.7), bone-related conditions (161.9), and congestive heart failure (124.5). The most common AESIs for patients with UC were diabetes mellitus (338.7), severe infections (254.0), congestive heart failure (216.4), bone-related conditions (197.6), and perianal disease (103.5) (Table III).
      Table IIIFrequency of adverse events of special interest (AESIs) among previously immunosuppressant, anti–tumor necrosis factor α, and anti-integrin treatment-naive patients.
      AESIPatients With CD (n = 676)Patients With UC (n = 450)
      AESI, No. (%)AESI per 10,000 Observed Patient-yearsAESI, No. (%)AESI per 10,000 Observed Patient-years
      Main AnalysisMain AnalysisSensitivity AnalysisMain AnalysisMain AnalysisSensitivity Analysis
      All AESIs (N; % patients)148 (105; 15.5)1392.41247.6154 (79; 17.6)1917.91935.5
      Acute hepatic failure
      Hepatic transplant
      Jaundice, idiopathic1 (0.7)12.55.6
      Hepatitis, drug-/toxin- induced, noninfectious1 (0.7)9.413.9
      Cholestasis
      Hepatic infarction
      Hepatomegaly3 (2.0)37.411.32 (1.3)18.827.9
      Autoimmune chronic active hepatitis
      Cirrhosis2 (1.4)24.911.32 (1.3)18.813.9
      Portal vein thrombosis1 (0.7)12.55.62 (1.3)18.827.9
      Hepatic vein thrombosis
      Liver abscess
      Malignancy (excluding neoplasm of rectum, anus, and anal canal)4 (2.7)49.822.57 (4.6)65.997.5
      Severe opportunistic infections1 (0.7)9.413.9
      Severe infections (excluding opportunistic infections, and herpes viral infection of perianal skin and rectum)20 (13.5)249.190.127 (17.5)254.0320.3
      Anaphylactic reactions
      Hypersensitivity reactions
      Angioedema
      Bone-related conditions13 (8.8)161.950.721 (13.6)197.6292.4
      Diabetes mellitus17 (11.5)211.773.236 (23.4)338.7459.5
      Death4 (2.7)49.816.97 (4.6)65.955.7
      Venous thromboembolism (excluding hepatic vein thrombosis)2 (1.4)24.95.6
      Pancreatic events6 (4.1)74.711.34 (2.6)37.641.8
      Congestive heart failure10 (6.8)124.550.723 (14.9)216.4278.5
      Cataracts and glaucoma2 (1.4)24.911.36 (3.9)56.583.6
      Skin lesions and/or arthralgia5 (3.4)62.328.24 (2.6)37.655.7
      Perianal disease37 (25.0)460.8169.011 (7.1)103.5153.2
      Small bowel resection21 (14.2)261.595.7
      CD = Crohn's disease; UC = ulcerative colitis.
      Note: For the sensitivity analysis, only treatment periods lasting ≥60 days were observed. All patients were naive to immunosuppressant, anti–tumor necrosis factor α, or anti-integrin treatment 12 months before the index date.
      IS monotherapy, anti-TNFα monotherapy, ASA monotherapy, and OCS + IS combination therapy were the treatment regimens (including combinations) associated with the highest AESI rates (Table IV). In patients with CD, AESI rates (events per TTPY) were 209.1 for IS + OCS combination therapy, 156.1 for ASA monotherapy, 128.5 for anti-TNFα monotherapy, and 109.7 for IS monotherapy. In patients with UC, AESI rates (events per TTPY) were 307.0 for ASA monotherapy, 181.1 for IS monotherapy, 171.1 for IS + OCS combination therapy, and 129.5 for anti-TNFα monotherapy.
      Table IVRates of adverse events of special interest per 10,000 patient-years for top 4 treatment patterns.
      TreatmentPatients With CD (n = 676)Patients With UC (n = 450)
      IS monotherapy109.7181.1
      Anti-TNFα monotherapy128.5129.5
      Aminosalicylate monotherapy156.1307.0
      OCS + IS209.1171.7
      Anti-TNFα = anti–tumor necrosis factor α; CD = Crohn's disease; IS = immunosuppressant; OCS = oral corticosteroid; UC = ulcerative colitis.
      Note: All patients were naive to IS, anti-TNFα, or anti-integrin treatment 12 months before the index date.
      In the ANCOVA, which assessed the relation between AESI rates and treatment pattern, all models presented only insignificant differences with regard to AESI rates for specific treatment patterns. Results were similar in the sensitivity analysis that considered treatment periods lasting ≥60 days. Overall reported AESI rates were similar to the main analysis (patients with CD, 1247.6 events per TTPY; patients with UC, 1935.5 events per TTPY) (Table III). This also applied to the rate of specific AESIs. The results of the ANCOVA did not identify any treatment pattern associated with a statistically significantly higher AESI rate.

      HCRU and costs

      Health care costs along with resource items are presented in Table V. Mean drug, medical, and total direct health care costs were compared between patients with AESIs versus those without AESIs. For CD, mean total direct health care costs for patients with AESIs was €8920.08 versus €6004.86 for those without AESIs (P < 0.001). In line with that, both mean drug costs (€6550.59 vs €4559.74; P = 0.001) and mean medical costs (€2302.19 vs €1400.25; P < 0.001) differed between CD patients with and without AESIs, respectively (Table VI). Results were confirmed in the sensitivity analysis when observing treatment periods ≥60 days only. (Supplemental Table II). For UC, mean total costs of patients with AESIs were generally higher than those without AESIs (€6437.87 vs €5707.94) but not statistically significant (P = 0.088). Mean medical costs were different between UC patients with AESIs versus those without (€1946.93 vs €971.28, respectively; P < 0.001).
      Table VMean inflammatory bowel disease (IBD)-related health care resource use (HCRU), loss of productivity, and costs per observed patient-years.
      HCRU and CostsPatients With CD (n = 676)Patients With UC (n = 450)
      Affected,

      No. (%)
      Mean (SD)Affected,

      No. (%)
      Mean (SD)
      Days absent from work associated with an IBD diagnosis416 (61.5)20.0 (39.7)248 (55.1)16.1 (34.0)
      Mean cost for loss of productivity, based on days absent from work, €3639.59 (7328.42)2935.95 (6266.90)
      Mean no. of IBD-related hospitalizations
      All hospitalizations with a diagnosis of CD or UC (International Classification of Diseases, Tenth Revision K50.-/K51.-).
      342 (50.6)0.3 (0.4)190 (42.2)0.2 (0.4)
      Mean no. of IBD-related surgeries
      All surgeries with OPS (procedure codes) 5–43 to 5–49.
      200 (29.6)0.2 (0.5)57 (12.7)0.1 (0.2)
      Mean inpatient costs, €1072.45 (1921.88)678.19 (1430.12)
      Mean no. of IBD-related GP visits634 (93.8)2.9 (1.5)426 (94.7)2.7 (1.4)
      Mean no. of IBD-related specialist visits532 (78.7)1.7 (1.4)345 (76.7)1.7 (1.4)
      Mean costs for outpatient GP/specialist visits, €467.90 (220.64)464.37 (227.12)
      Mean no. of IBD-related rehabilitations20 (3.0)0.0 (0.1)17 (3.8)0.0 (0.1)
      Mean rehabilitation costs, €48.35 (335.87)72.83 (426.36)
      Mean drug costs for IBD-related treatments,
      All focus treatments: OCS, aminosalicylate, immunosuppressant, anti–tumor necrosis factor α, and anti-integrin.
      4868.97 (6830.77)4620.69 (7321.22)
      Mean total costs, €10,097.25 (11,371.28)8772.03 (10,573.14)
      CD = Crohn's disease; GP = general practitioner; OCS = oral corticosteroid; OPS, Operations and Procedures; UC = ulcerative colitis.
      Note: All patients were naive to immunosuppressant, anti–tumor necrosis factor α, or anti-integrin treatment 12 months before the index date.
      All hospitalizations with a diagnosis of CD or UC (International Classification of Diseases, Tenth Revision K50.-/K51.-).
      All surgeries with OPS (procedure codes) 5–43 to 5–49.
      All focus treatments: OCS, aminosalicylate, immunosuppressant, anti–tumor necrosis factor α, and anti-integrin.
      Table VIComparison of direct health care costs between patients with and without adverse events of special interest (AESIs).
      SampleNo. (%)Drug Costs,
      Direct health care costs are presented as per patient-year.
      Medical Costs,
      Direct health care costs are presented as per patient-year.
      Total Costs,
      Direct health care costs are presented as per patient-year.
      MeanSDMeanSDMeanSD
      Patients with CDWithout AESIs571 (84.5)4559.746609.921400.251717.816004.866982.04
      With AESIs105 (15.5)6550.597774.662302.192797.028920.088415.12
      P0.001< 0.001< 0.001
      Patients with UCWithout AESIs371 (82.4)4662.247483.19971.281175.595707.947647.35
      With AESIs79 (17.6)4425.536599.491946.932192.296437.877374.33
      P0.650< 0.0010.088
      CD = Crohn's disease; UC = ulcerative colitis.
      Note: All patients were naive to immunosuppressant, anti–tumor necrosis factor α, or anti-integrin treatment 12 months before the index date.
      Direct health care costs are presented as per patient-year.
      Multivariable generalized linear modeling showed that in patients with CD, occurrence of an AESI during the follow-up period resulted in 24.9% higher costs (P = 0.012) (Table VII). Older age and IBD-related symptoms at index decreased the cost, whereas extraintestinal manifestations at baseline increased total direct health care costs. None of the included variables was associated with total direct health care costs in patients with UC.
      Table VIIMultivariable generalized linear model regression estimates of total direct health care costs.
      Direct health care costs are presented as per patient-year.
      ,
      Generalized linear model – outcome = costs per patient-year.
      VariableExp (coefficient)SEP95% CI
      Patients with CD, n = 675
       Any AENo

      Yes
      Reference

      1.249
      0.1110.0121.049–1.487
       Age at index0.9890.003< 0.0010.983–0.996
       IBD-related symptoms
      Within 12 months of the preindex period.
      No

      Yes
      Reference

      0.788
      0.0720.0100.659–0.944
       Extraintestinal manifestations
      Within 12 months of the preindex period.
      No

      Yes
      Reference

      1.309
      0.1290.0061.079–1.588
      Patients with UC, n = 450
       Any AENo

      Yes
      Reference

      1.060
      0.1290.6540.822–1.365
       Age at index0.9940.0040.1170.986–1.002
       IBD-related symptoms
      Within 12 months of the preindex period.
      No

      Yes
      Reference

      0.825
      0.1240.1220.657–1.052
       Extraintestinal manifestations
      Within 12 months of the preindex period.
      No

      Yes
      Reference

      0.992
      0.1510.9570.738–1.334
       SexMale

      Female
      Reference

      0.976
      0.1240.8470.766–1.245
       Charlson Comorbidity Index
      Within 12 months of the preindex period.
      0.9990.0450.9880.915–1.090
      AE = adverse event; CD = Crohn's disease; IBD = inflammatory bowel disease; UC = ulcerative colitis.
      Note: All patients were naive to immunosuppressant, anti–tumor necrosis factor α, or anti-integrin treatment 12 months before the index date.
      Direct health care costs are presented as per patient-year.
      Generalized linear model – outcome = costs per patient-year.
      Within 12 months of the preindex period.

      Discussion

      Several AESIs are commonly associated with IBD treatment in Germany. During a mean observational period of 1356 days for CD and 1340 days for UC, 15.5% of CD patients and 17.6% of UC patients who had newly initiated treatment with IS, anti-TNFα, or anti-integrin therapy experienced at least 1 AESI, which led to an inpatient hospitalization. In terms of HCRU and costs, AESIs were associated with a higher HCRU and higher direct health care costs in patients with CD but not in patients with UC.
      The main strengths of this study are the use of a large unselected sample of patients with IBD, the inclusion of a substantial number of AESIs derived from a literature review and discussions with clinical experts, and the assignment of each observed treatment day to a specific treatment pattern. To our knowledge, no study thus far has assessed the AESI risk and costs associated with most common IBD treatments based on German claims data.
      Because all reported events were associated with an inpatient hospitalization in this analysis, we considered these as serious AESIs, without any knowledge about severity of the condition or event that led to the hospitalization. This makes comparison of the AESI rates reported here versus rates reported in previous clinical trials challenging. In a previous systematic literature review of adverse events associated with IS and biologic treatment, infectious complications and congestive heart failure were the most frequent adverse events associated with these treatments.
      • McLean L.P.
      • Cross R.K.
      Adverse events in IBD: to stop or continue immune suppressant and biologic treatment.
      A higher risk of malignancies was also observed in association with the combination treatment of anti-TNFα and IS. Furthermore, data from different trials confirmed a favorable safety profile of vedolizumab.
      • Scribano M.L.
      Vedolizumab for inflammatory bowel disease: from randomized controlled trials to real-life evidence.
      A recent network meta-analysis of randomized clinical trials compared safety profiles of biologics for IBD and found that the most frequently reported adverse events were infections, serious infections, upper respiratory tract infections, nasopharyngitis, abdominal pain, headache, nausea, and pharyngitis.
      • Mocko P.
      • Kawalec P.
      • Pilc A.
      Safety profile of biologic drugs in the treatment of inflammatory bowel diseases: a systematic review and network meta-analysis of randomized controlled trials.
      In this study, we did not collect data on injection–site reactions but, in line with the aforementioned studies, could confirm that infections are one of the most frequently observed adverse events associated with IBD treatments. This study also confirms that malignancy as an AESI occurs in this population, even if not thoroughly analyzed in clinical trials. The frequent AESIs observed in this study but not reported in previous clinical trials included diabetes mellitus, bone-related conditions, small-bowel resections, and perianal disease. However, we included these adverse events based on both the scientific literature and discussion with clinical experts.
      In this study, there was no significant relation detected between specific treatment pattern and AESI rate. Different explanations might apply here. First, we assigned every observed day to a specific treatment pattern. However, overlapping effects of previous medications with current medications may exist, which make it difficult to assign a specific AESI to a specific treatment pattern. Second, unobserved patient characteristics might have led to bias in these results. Third, we only had access to prescription data and assumed that medications were taken as prescribed, which if not accurate, could have influenced the results.
      An evaluation of the reported HCRU and overall health care costs of €10,097 per patient-year for CD and €8772 per patient-year for UC by clinicians and payers deemed them to be plausible and in line with earlier studies.
      • Wilke T.
      • Groth A.
      • Long G.H.
      • Tatro A.R.
      • Sun D.
      Adverse events associated with the treatment of inflammatory bowel disease in Germany.
      A direct comparison with previous studies is challenging because we addressed an IBD population treated in Germany who were naive to previous IS, anti-TNFα, and anti-integrin treatment only, and because previous publications partly reported different cost categories and/or were based on older health care price structures. Nevertheless, our data confirm the findings of Stark et al,
      • Stark R.
      • Konig H.H.
      • Leidl R.
      Costs of inflammatory bowel disease in Germany.
      who concluded that the health care items that most influenced IBD-related health care costs were productivity losses and drug cost. Prenzler et al
      • Prenzler A.
      • Yen L.
      • Mittendorf T.
      • von der Schulenburg J.M.
      Cost effectiveness of ulcerative colitis treatment in Germany: a comparison of two oral formulations of mesalazine.
      examined the annual direct health care costs for patients with CD and patients with UC in Germany, based on the year 2007. They reported costs of €3767 for patients with CD (68.5% medication; 20.5% inpatient treatment) and €2478 for patients with UC (74% medication; 10% inpatient treatment). Because these data refer to a period 7–9 years before our observation was initiated, and these numbers do not include indirect costs, we interpret the cost numbers reported here to be generally in line with these results. We are not aware of any more recent HCRU or cost studies addressing a general IBD population in Germany.
      Our data confirm, at least among patients with CD, that AESIs are associated with a higher HCRU and higher direct health care costs. The difference between health care costs in UC patients with and without AESIs was not significant, mostly because of the low sample size in the non-AESI group. We are not aware of any previous analysis that addresses this research question.
      We acknowledge some limitations. First, our definition of AESI is not based on known clinical trials in patients with IBD, both in terms of selection of events or conditions of interest and in terms of characterization of their severity. Instead, we developed a broad list of 28 events or chronic diseases that were derived from the literature and discussions with IBD clinical experts who interpreted these to be potential adverse events associated with current IBD therapies. For chronic diseases, only incident inpatient diagnoses were counted as AESIs. We acknowledge that some of these conditions might not be AESI associated and may develop in an aging population without any IBD treatment. However, because of the comparatively young age of the population, we believe that at least a substantial proportion of newly diagnosed chronic conditions such as type 2 diabetes mellitus or congestive heart failure are associated with the prescribed treatments.
      Second, our assessment of AESI was based on inpatient diagnoses and procedural codes. We could not assess any direct relationship between an AESI and a specific treatment, nor could we confirm whether a specific prescribed treatment was taken by the patient at the specific date of an AESI event. Third, only AESIs associated with an inpatient encounter were captured because we wanted to focus on severe AESIs. Consequently, because of our more conservative approach, our reported rate of adverse events may be an underestimate. Fourth, because retrospective claims data were used, the availability of specific disease parameters was limited (eg, laboratory results or patient-reported outcomes). In addition, because cases were not randomized to each cohort but followed a sequential treatment pathway, systematic baseline differences between the treatment cohorts are likely. Furthermore, treatments may have been dictated by an individual patient profile, which is characterized by variables that cannot be observed in claims data. Characteristics of treating physicians such as their qualifications, knowledge, and perceptions toward therapies were not available, but it is not known whether and how these characteristics may influence treatment, AESI rates, HCRU, or costs.
      Finally, one of the main limitations of this analysis was small sample size as we addressed treatment-naive patients only. We chose this approach because previous treatments may have influenced reported AESI rates; however, the small sample size may have led to insignificance of results in several outlined multivariable analyses. We therefore caution against generalizing results from this study to the general IBD population because patients whose treatment is initiated with an IS, anti-TNFα, or anti-integrin may be clinically different from those who receive these treatments as second- or third-line therapy.

      Conclusions

      The incidence of IBD is increasing worldwide, and adverse events are common in patients with IBD treated with an IS, an anti-TNFα, an anti-integrin, OCS, and/or ASA and are associated with substantial HCRU and costs. During a follow-up period of ∼1350 days, 14% to 19% of patients were affected by an AESI. Our results therefore show an urgent need for development of IBD treatments that are associated with lower adverse event rates.

      Disclosures

      This research and its publication were sponsored by Genentech, Inc, a member of the Roche Group. The sponsor was involved in the study design, interpretation of data, in the writing of the manuscript, and in the decision to submit the article for publication. Dr. Wilke is an owner of Ingress-Health, a health economics and real-world evidence research consultancy hired by Genentech, Inc., and Roche Pharmaceuticals Ltd. to conduct this research. He has received honoraria from pharmaceutical/consultancy companies, including Novo Nordisk, AbbVie, Merck, GlaxoSmithKline, BMS, LEO Pharma, AstraZeneca, Bayer, Boehringer Ingelheim, and Pharmerit. As Professor at the Hochschule Wismar, Germany, Dr. Wilke is an employee of the state Mecklenburg-Vorpommern. Ms. Groth is an employee of IPAM e.V., a research institute of the Wismar University of Applied Sciences hired by Genentech, Inc, and Roche Pharmaceuticals Ltd to conduct this research. Drs. Tatro and Sun are employees of Genentech, Inc, a member of the Roche group, and own shares of Roche stock. The authors have indicated that they have no other conflicts of interest regarding the content of this article.

      Acknowledgments

      This analysis was funded by Genentech, Inc , a member of the Roche Group. Third-party medical writing and editorial support (formatting, proofreading, and copyediting) was provided by ApotheCom and was funded by F. Hoffmann–La Roche, Ltd .
      Dr. Wilke designed the study protocol and wrote and reviewed the manuscript. Ms. Groth performed statistical analyses and reviewed the manuscript. Dr. Long conceived the study question, designed the protocol, interpreted the data, and revised the manuscript. Dr. Tatro assisted in development of the study concept, reviewed the protocol and data, and revised the manuscript. Dr. Sun contributed to acquisition, analysis, or interpretation of the data, and wrote and reviewed the manuscript.

      Appendix.

      Supplemental Table IDefinition of adverse events (AEs) that were analyzed.
      VariableDefinitionIdentification of a new AE of interest
      Acute hepatic failure (liver necrosis, non–alcohol related liver cirrhosis, liver coma)ICD-10 code: K72.-No respective diagnosis in previous 180 days
      Hepatic transplantationOPS code: 5-504Every new code is assumed to indicate a new event
      Jaundice, idiopathicICD-10 code: R17No respective diagnosis in previous 180 days
      Hepatitis, drug-/toxin-induced, noninfectiousICD-10 code: K71.2–K71.6No respective diagnosis in previous 180 days
      CholestasisICD-10 code: K 71.0No respective diagnosis in previous 180 days
      Hepatic infarctionICD-10 code: K76.3No respective diagnosis in previous 180 days
      HepatomegalyICD-10 code: R16.0 and R16.2No respective diagnosis in previous 180 days
      Autoimmune chronic active hepatitisICD-10 code: K75.4No respective diagnosis in previous 180 days
      CirrhosisICD-10 code: K74.3–K74.6No respective diagnosis in previous 180 days
      Portal vein thrombosisICD-10 code: I81.-No respective diagnosis in previous 180 days
      Hepatic vein thrombosisICD-10 code: I82.0No respective diagnosis in previous 180 days
      Liver abscessICD-10 code: K75.0No respective diagnosis in previous 180 days
      MalignancyICD-10 codes: C00–C75/C81–C97/ D00–D09No respective diagnosis in previous 180 days
      Severe opportunistic infectionsICD-10 codes: A15–A19 (tuberculosis including mycobacterium)/B44.- (aspergillosis)/B45.- (cryptococcosis)No respective diagnosis in previous 180 days
      Severe infectionsICD-10 codes: A00–A99, B15–B19, B35–B49.-No respective diagnosis in previous 180 days
      Anaphylactic reactionsICD-10 code: T78.0/T78.2/T80.5/T88.6No respective diagnosis in previous 180 days
      Hypersensitivity reactionsICD-10 code: T78.1/J39.3/M36.4No respective diagnosis in previous 180 days
      AngioedemaICD-10 code: T78.3No respective diagnosis in previous 180 days
      Bone-related conditionsICD-10 code: M80.-/M81.- (osteoporosis), S12.-/S22.-/S32.-/S42.-/S52.-/S62.-/S72.-/S82.-/S92.-/T02.-/T12.- (fractures), M87.- (osteonecrosis), OPS 5-820 (hip replacement)No respective diagnosis in previous 180 days
      Diabetes mellitusICD-10 code: E10.-/E11.-/E12.-/E13.-/E14.-No respective diagnosis in previous 180 days
      DeathAs documented in the datasetNA
      Venous thromboembolismICD-10 code: I82.-No respective diagnosis in previous 180 days
      Pancreatic eventsICD-10 code: K85.-/K86.-No respective diagnosis in previous 180 days
      Congestive heart failureICD-10 code: I50.-No respective diagnosis in previous 180 days
      Cataracts and glaucomaICD-10 code: H25.-/H26.-/H40.-No respective diagnosis in previous 180 days
      Skin lesions and/or arthralgiaICD-10 code: F45.4-/M25.5-No respective diagnosis in previous 180 days
      Perianal diseaseICD-10 code: K60.-/K61.-/K62.-/K64.5/A60.1/C20.-/C21.-No respective diagnosis in previous 180 days
      Small bowel resectionOPS 5-454Every new code is assumed to indicate a new event
      ICD-10 = International Classification of Diseases, Tenth; OPS = Operations and Procedures.
      Supplemental Table IIDirect cost comparisons between patients who experienced an AESI versus those who did not, based on sensitivity analysis.
      OCSISanti-TNFαOCS + ISASA
      Patients with an AESIN24511518
      Cost, mean (SD), €12,151.17 (7645.66)6065.98 (8565.57)32,190.78 (8021.07)6070.52 (5164.23)4307.48 (8494.06)
      Patients without an AESIN5366612092141
      Cost, mean (SD), €2789.02 (3146.10)2351.95 (2793.67)27,706.03 (12,553.00)3467.34 (1472.94)1308.58 (797.78)
      AESI, adverse events of special interest; anti-TNFα = anti–tumor necrosis factor α; ASA = aminosalicylates; IS = immunosuppressant; OCS = oral corticosteroids; UC = ulcerative colitis.
      Only most frequently observed treatment types are presented.

      References

        • Hein R.
        • Koster I.
        • Bollschweiler E.
        • Schubert I.
        Prevalence of inflammatory bowel disease: estimates for 2010 and trends in Germany from a large insurance-based regional cohort.
        Scand J Gastroenterol. 2014; 49: 1325-1335https://doi.org/10.3109/00365521.2014.962605
        • Baumgart D.C.
        The diagnosis and treatment of Crohn's disease and ulcerative colitis.
        Dtsch Arztebl Int. 2009; 106: 123-133https://doi.org/10.3238/arztebl.2009.0123
        • Baumgart D.C.
        • Sandborn W.J.
        Inflammatory bowel disease: clinical aspects and established and evolving therapies.
        Lancet. 2007; 369: 1641-1657https://doi.org/10.1016/s0140-6736(07)60751-x
        • Prenzler A.
        • Mittendorf T.
        • Conrad S.
        • von der Schulenburg J.M.
        • Bokemeyer B.
        [Costs of Crohn's disease in Germany from the perspective of the statutory health insurance].
        Z Gastroenterol. 2009; 47: 659-666https://doi.org/10.1055/s-0028-1109059
        • Floyd D.N.
        • Langham S.
        • Severac H.C.
        • Levesque B.G.
        The economic and quality-of-life burden of Crohn's disease in Europe and the United States, 2000 to 2013: a systematic review.
        Dig Dis Sci. 2015; 60: 299-312https://doi.org/10.1007/s10620-014-3368-z
        • Burisch J.
        • Jess T.
        • Martinato M.
        • Lakatos P.L.
        The burden of inflammatory bowel disease in Europe.
        J Crohns Colitis. 2013; 7: 322-337https://doi.org/10.1016/j.crohns.2013.01.010
        • Kaplan G.G.
        The global burden of IBD: from 2015 to 2025.
        Nat Rev Gastroenterol Hepatol. 2015; 12: 720-727https://doi.org/10.1038/nrgastro.2015.150
        • Ananthakrishnan A.N.
        Epidemiology and risk factors for IBD.
        Nat Rev Gastroenterol Hepatol. 2015; 12: 205-217https://doi.org/10.1038/nrgastro.2015.34
        • Limsrivilai J.
        • Stidham R.W.
        • Govani S.M.
        • Waljee A.K.
        • Huang W.
        • Higgis P.D.R.
        Factors that predict high health care utilization and costs for patients with inflammatory bowel diseases.
        Clin Gastroenterol Hepatol. 2017; 15: 385-392https://doi.org/10.1016/j.cgh.2016.09.012
        • Gomollon F.
        • Dignass A.
        • Annese V.
        • et al.
        European evidence-based consensus on the diagnosis and management of Crohn's disease 2016: part 1: diagnosis and medical management.
        J Crohns Colitis. 2017; 11: 3-25https://doi.org/10.1093/ecco-jcc/jjw16
        • Harbord M.
        • Eliakim R.
        • Bettenworth D.
        • et al.
        Third European evidence-based consensus on diagnosis and management of ulcerative colitis. Part 2: current management.
        J Crohns Colitis. 2017; 11: 769-784https://doi.org/10.1093/ecco-jcc/jjx009
        • Peyrin-Biroulet L.
        • Lemann M.
        Review article: remission rates achievable by current therapies for inflammatory bowel disease.
        Aliment Pharmacol Ther. 2011; 33: 870-879https://doi.org/10.1111/j.1365-2036.2011.04599.x
        • Dignass A.
        • Preiss J.C.
        • Aust D.E.
        • et al.
        Updated German guideline on diagnosis and treatment of ulcerative colitis, 2011.
        Z Gastroenterol. 2011; 49: 1276-1341https://doi.org/10.1055/s-0031-1281666
        • Preiss J.C.
        • Bokemeyer B.
        • Buhr H.J.
        • et al.
        Updated German clinical practice guideline on "Diagnosis and treatment of Crohn's disease" 2014.
        Z Gastroenterol. 2014; 52: 1431-1484https://doi.org/10.1055/s-0034-1385199
      1. Entyvio [summary of Product Characteristics]. Takeda Austria GmbH, Linz, Austria2019
      2. Humira [summary of Product Characteristics]. AbbVie Deutschland GmbH & Co, Ludwigshafen, Germany2019
      3. Remicade [summary of Product Characteristics]. Janssen Biologics BV, Leiden, the Netherlands2019
      4. Simponi [summary of Product Characteristics]. Janssen Biologics BV, Leiden, the Netherlands2019
        • Loftus Jr., E.V.
        • Davis K.L.
        • Wang C.C.
        • Dastani H.
        • Luo A.
        Treatment patterns, complications, and disease relapse in a real-world population of patients with moderate-to-severe ulcerative colitis initiating immunomodulator therapy.
        Inflamm Bowel Dis. 2014; 20: 1361-1367https://doi.org/10.1097/mib.0000000000000089
        • Narula N.
        • Kainz S.
        • Petritsch W.
        • et al.
        The efficacy and safety of either infliximab or adalimumab in 362 patients with anti-TNF-alpha naive Crohn's disease.
        Aliment Pharmacol Ther. 2016; 44: 170-180https://doi.org/10.1111/apt.13671
      5. Updated S3-guideline colitis ulcerosa. German Society for Digestive and Metabolic Diseases (DGVS)—AWMF registry 021/009.
        Z Gastroenterol. 2018; 56: 1087-1169https://doi.org/10.1055/a-0651-8174
        • Marehbian J.
        • Arrighi H.M.
        • Hass S.
        • Tian H.
        • Sandborn W.J.
        Adverse events associated with common therapy regimens for moderate-to-severe Crohn's disease.
        Am J Gastroenterol. 2009; 104: 2524-2533https://doi.org/10.1038/ajg.2009.322
        • Karve S.
        • Candrilli S.
        • Kappelman M.D.
        • Tolleson-Rinehart S.
        • Tennis P.
        • Andrews E.
        Healthcare utilization and comorbidity burden among children and young adults in the United States with systemic lupus erythematosus or inflammatory bowel disease.
        J Pediatr. 2012; 161 (e662): 662-670https://doi.org/10.1016/j.jpeds.2012.03.045
        • Prenzler A.
        • Bokemeyer B.
        • von der Schulenburg J.M.
        • Mittendorf T.
        Health care costs and their predictors of inflammatory bowel diseases in Germany.
        Eur J Health Econ. 2011; 12: 273-283https://doi.org/10.1007/s10198-010-0281-z
        • Wilke T.
        • Groth A.
        • Fuchs A.
        • et al.
        Real life treatment of diabetes mellitus type 2 patients: an analysis based on a large sample of 394,828 German patients.
        Diabetes Res Clin Pract. 2014; 106: 275-285https://doi.org/10.1016/j.diabres.2014.08.002
        • Groth A.
        • Wilke T.
        • Borghs S.
        • Gille P.
        • Joeres L.
        Real life pharmaceutical treatment patterns for adult patients with focal epilepsy in Germany: a longitudinal and cross-sectional analysis of recently approved anti-epileptic drugs.
        Ger Med Sci. 2017; 15: Doc09https://doi.org/10.3205/000250
        • McLean L.P.
        • Cross R.K.
        Adverse events in IBD: to stop or continue immune suppressant and biologic treatment.
        Expert Rev Gastroenterol Hepatol. 2014; 8: 223-240https://doi.org/10.1586/17474124.2014.881715
        • McAuliffe M.E.
        • Lanes S.
        • Leach T.
        • et al.
        Occurrence of adverse events among patients with inflammatory bowel disease in the HealthCore Integrated Research Database.
        Curr Med Res Opin. 2015; 31: 1655-1664https://doi.org/10.1185/03007995.2015.1065242
      6. Vergütung steigt um 620 Millionen Euro [press release]. Published August 23, 2018. Accessed April 18, 2019. https://www.kbv.de/html/1150_36611.php.

      7. Volkswirtschaftliche Kosten durch Arbeitsunfähigkeit 2016. Bundesanstalt fur Arbeitsschutz und Arbeitsmedizin, 2016: 1-7
        • Scribano M.L.
        Vedolizumab for inflammatory bowel disease: from randomized controlled trials to real-life evidence.
        World J Gastroenterol. 2018; 24: 2457-2467https://doi.org/10.3748/wjg.v24.i23.2457
        • Mocko P.
        • Kawalec P.
        • Pilc A.
        Safety profile of biologic drugs in the treatment of inflammatory bowel diseases: a systematic review and network meta-analysis of randomized controlled trials.
        Clin Drug Investig. 2017; 37: 25-37https://doi.org/10.1007/s40261-016-0459-y
        • Wilke T.
        • Groth A.
        • Long G.H.
        • Tatro A.R.
        • Sun D.
        Adverse events associated with the treatment of inflammatory bowel disease in Germany.
        Value Health. 2018; 21 (Abstract PGI1.): S142https://doi.org/10.1016/j.jval.2018.09.844
        • Stark R.
        • Konig H.H.
        • Leidl R.
        Costs of inflammatory bowel disease in Germany.
        Pharmacoeconomics. 2006; 24: 797-814https://doi.org/10.2165/00019053-200624080-00006
        • Prenzler A.
        • Yen L.
        • Mittendorf T.
        • von der Schulenburg J.M.
        Cost effectiveness of ulcerative colitis treatment in Germany: a comparison of two oral formulations of mesalazine.
        BMC Health Serv Res. 2011; 11: 157https://doi.org/10.1186/1472-6963-11-157