TY - JOUR
AU1 - MD, Michelle V. Prosberg,
AU2 - PhD, Marianne K. Vester-Andersen, MD,
AU3 - MSc, Mikael Andersson,
AU4 - DMSci, Tine Jess, MD,
AU5 - MD, Jon T. Andersen,
AU6 - PhD, Ida Vind, MD,
AU7 - DMSci, Flemming Bendtsen, MD,
AB - Background Noncompliance to long-term medical therapy is a well-known problem among patients treated for ulcerative colitis, but studies of long-term consequences in unselected patients are lacking. The authors aimed to determine the risk of recurrence according to long-term compliance with oral 5-aminosalicylic acid among unselected patients with ulcerative colitis. Methods The authors conducted a 7-year follow-up study of a population-based inception cohort of 243 Danish patients with ulcerative colitis diagnosed from 2003 to 2004. Compliance was defined as consumption of ≥80% of prescribed oral 5-aminosalicylic acid. Data were collected from medical records and the Danish National Prescription Database. They performed Cox regression analysis with adjustments for demographic and clinical characteristics to examine risk of recurrence (defined by increased use of oral 5-Aminosalicylic Acid, other additional treatment, or colectomy) in compliant versus noncompliant patients. Results In total, 182 patients (75%) experienced at least 1 recurrence during follow-up. For the first year after diagnosis, risk of recurrence did not differ significantly between compliant and noncompliant patients. For 1 to 3 years (hazard ratio: 0.46, 95% CI, 0.33–0.63) and 3 to 8 years (hazard ratio: 0.42, 95% CI, 0.32–0.55) after diagnosis, risk of recurrence was significantly decreased among noncompliant patients compared with that of compliant patients. Conclusions This unselected cohort study revealed a reverse association between compliance and recurrence of ulcerative colitis. This is unlikely to be explained by severe confounding because the authors were able to adjust for several demographic and clinical factors. Results may instead reflect that patients during recurrence-free periods through self-management choose not to take their medication. compliance, ulcerative colitis, recurrence Noncompliance to therapy is a widespread medical problem and highly prevalent particularly among patients treated for a chronic condition. According to the World Health Organization, compliance with long-term therapies in the general population of developed countries is around 50%.1 Noncompliance has been shown to be a problem in quiescent diseases like ulcerative colitis (UC).2,3 The reason might, in remission periods, be the lack of continuing symptoms that normally would encourage patients to take their medication.4 UC is characterized by alternating periods of relapses and remission, and only a minor number of patients suffer from chronic continuous disease activity. The extent of the inflammation progresses in an unpredictable manner from minor rectal inflammation and proctitis with only few symptoms to left-sided colitis, and it may progress or even start as a severe extensive colitis, which in worst cases is not responding to conventional medical treatment and hence leads to rescue therapy or ultimately colectomy.5 5-Aminosalicylic acid (5-ASA) is the first line treatment for the induction and maintenance of remission in mild-to-moderate UC, most often administered as tablets. In case of recurrence, the daily dose of 5-ASA is increased or treatment with corticosteroids and/or topical therapy is added. In patients presenting with recurrence or persistent symptoms of active UC without a period of remission or steroid dependency, additional treatment with immunosuppressants and subsequent biological therapy is often initiated.6,7 In a systematic review, nonadherence to oral medication among patients with inflammatory bowel disease (IBD) ranged from 7% to 72% of which most of the 17 included studies reported a nonadherence rate of 30% to 45%.8 Regarding UC, compliance studies have revealed noncompliance rates (defined as taking <80% of the oral medication) ranging from 40% to 60% with 5-ASA therapy.2,9 In the last decades, compliance to therapy among UC patients, and IBD patients in general, has been investigated prospectively and retrospectively in several studies by using either urine samples,10 questionnaires,11 pharmacy prescriptions12,–14 or a combination thereof.3,9,15,16 The study designs are dominated by either clinical trials17 or observational studies based on registry data.2,3,10,12,–14,18,19 Results from clinical trials may not be representative of daily clinical practice, because patients are monitored more closely than patients outside a study setting. Further, studies using pharmacy prescription records from registries often lack medical history and clinical information of the included patients12,–14,18,20 which may bias the result. Nor do these studies have information regarding the reasons for discontinuation with medication or further changes in therapy in agreement with the practitioners because it may not be registered in pharmacy claims databases. Most compliance studies conducted in the United States are based on insurance databases which may represent selected patient groups. Studies evaluating the impact of compliance on outcome of UC in unselected patient populations with detailed information on prescriptions both from medical records and pharmacy data are lacking. Based on the hypothesis that noncompliance is associated with a higher risk of recurrence throughout the disease course, the authors aimed to determine the risk of recurrence of UC according to long-term compliance with any oral 5-ASA in an unselected population-based inception cohort of UC patients. Materials and Methods Study Design and Participants The authors performed the study on a population-based, historical, inception cohort previously described in detail.21 The inception cohort comprised all patients diagnosed with IBD during the years 2003 and 2004 in Copenhagen City and County, including Frederiksberg. Patients initially diagnosed with IBD comprised a total of 562 patients (326 patients with UC, 209 patients with Crohn's disease and 27 with unclassified IBD) during the period January 1, 2003 to December 31, 2004 from a total population of 1,211,634 residents (23% of the entire Danish population) in a well-defined geographical area.21 During the period between November 1, 2011 and November 30, 2012, 2 of the investigators (MVA and MVP) performed a systematic review of the eligible patient medical records from 2003 to 2004 (time of diagnosis) to December 31, 2011 thereby ensuring an observation period of minimum 7 years. The original diagnosis was reassessed after retrieval of medical records, leading to the total of 300 UC and 213 Crohn's disease patients.22 In cases of doubt, the patient's medical history and records were further discussed and assessed by the 2 IBD experts (FB and IV). The inclusion criteria were: Patients confirmed to be diagnosed with UC between January 1, 2003 and December 31, 2004, and Prescribed treatment with oral 5-ASA during the follow-up period Variables The primary outcome of interest was recurrence. Medical recurrence was defined as UC-related treatment with increased dose of oral 5-ASA (defined as a new daily dose ≥2400 mg), additional treatment with rectal 5-ASA, steroids or oral steroids (defined as any dose, and identified by the Anatomical Therapeutic Chemical [ATC] classification system code [A07EA]), a change in treatment regime from 5-ASA to immunosuppressive drug (defined as azathioprine/6-mercaptopurine at any dose, and identified by the ATC code [L04AX01/L01BB02]) or biological treatment with anti-tumor necrosis factor alpha agents, defined as infliximab/adalimumab at any dose identified by the ATC code (L04AB04/L04AB02) and ultimately colectomy (defined by the national surgery code [SKS code KJFH]). The definition of recurrence is a modification of the term applied in the studies by Romberg-Camps et al,23 Wolters et al24 and Hoie et al.25 Compliance with oral 5-ASA was defined as the extent to which patients act in accordance with the prescribed medical regime proposed by the International Society for Pharmacoeconomics and Outcomes Research.26 The authors chose compliance to be the explanatory variable, because they aimed to explain variance in the risk of recurrence with respect to the proportion of medication on hand compared with the prescribed dosage. They chose to measure compliance as a dichotomous value by categorizing a patient as being compliant (possession of ≥80% of the prescribed doses of 5-ASA) or noncompliant (possession of <80%). Treatment with oral 5-ASA included sulfasalazine and 5-ASA defined by the ATC code A07EC01-04. We did not distinguish between 5-ASA preparations. Independent variables included demographic and clinical data. Demographic data consisted of age (<30; 30–49; 50–69; ≥70 years), sex (male/female), living conditions (alone/cohabiting), more than 13 years of schooling (yes/no), income (high, medium or low tertiles) and smoking status (never, current, or former smoker). We chose the above allocations because we intended to obtain an even distribution in each of the subgroups for the included patients. Clinical data included disease extent, disease course, disease activity, colectomy (no/yes) and information on hospitalization. The disease extent was based on the patients' endoscopic examinations and defined as E1: proctitis (proximal extent to the sigmoid colon), E2: left-sided (to the splenic flexure), or E3: extensive disease (beyond the splenic flexure). The disease course was a dynamic variable based on the clinical activity over time and defined as indolent (no recurrence in any year after diagnosis), intermittent (have had ≥1 recurrence, but no recurrence for the past year), or continuous (≥1 recurrence during each year of follow-up). The definition of disease course is a modification of the definitions used in the studies by Munkholm et al and Langholz et al.27,28 Disease activity was based on the requirement of additional treatment options and classified as high activity (treatment with steroids ≥50 mg, immunosuppressive, or biological treatment) or low activity (treatment with 5-ASA or steroids <50 mg). Information on hospitalizations included all admissions regardless of diagnosis and specialty, the date of admission, and the date of discharge. Data Sources Medical Records—The Danish Crohn and Colitis Database From the medical records, we were able to gather information on UC therapy (oral and rectal 5-ASA, azathioprine/6-mercaptopurine, biological treatments, or systemic steroids), start and end date of prescriptions, the prescribed dosage and any further changes in the prescribed therapy. Information on the patients' clinical history including, disease extent, disease course, and disease activity was also collected. All the information was registered on predesigned data sheets and scanned into the Danish Crohn and Colitis Database. Pharmacy Records—The Danish National Prescription Registry The Danish National Prescription Registry29 contains information on all prescriptions dispensed from pharmacies in Denmark since the January of 1995. All prescriptions are registered at an individual level using the unique civil registration number (CPR-number) and have been linked to the medical data included in this study. All pharmacies are obligated to register all prescriptions dispensed in this nationwide registry. Furthermore, all residents in Denmark are covered by the national healthcare security system and, therefore, the cost of certain drugs is partly reimbursed. The registry is unique on a global scale, since it is the only registry which covers the consumption of medicinal products for the entire population of a country over many years.30 For each oral 5-ASA medication prescribed for the treatment of UC, confirmed by linkage with medical records, we extracted the information needed to calculate compliance with oral 5-ASA in terms of ATC code, date of sale, strength per unit, number of units per packet, number of packets, and reiteration number. The Danish National Patient Registry, Central Person Registry and Statistics Denmark Denmark has a tax-supported welfare system which guarantees the entire population of 5.6 million free healthcare services. Every contact to the hospital system has since 1977 (1994 for outpatient visits) been registered at an individual level in the Danish National Patient Registry using the CPR number.31 We retrieved information on hospitalization for the entire period of follow-up. Furthermore, demographic data were obtained from the Central Person Registry and Statistics Denmark at baseline. Measurement Compliance was calculated as the redeemed dose relative to the prescribed dose of oral 5-ASA. The dosage was measured in milligrams. The prescribed dose was calculated as the sum of daily doses from inclusion until a change in dose, recurrence, or censoring based on the prescription data from the medical records. The redeemed dose was calculated by multiplying the total amount of tablets with the potency of each tablet handed out to the patient at any pharmacy in the entire country from inclusion until recurrence or censuring. Thus, with respect to oral 5-ASA, the authors were able to account for a change in dosage (due to increase or decrease), agreed patient-physician discontinuation periods of oral 5-ASA, overlapping pharmacy claims (e.g., an early refill) and refilling prescriptions at different pharmacies. Furthermore, the authors used the information on hospitalizations to calculate the length of admission so that they could exclude this period in the measurement of compliance status. They chose to do so since admitted patients do not administer their own medication while being hospitalized. In terms of measuring compliance using pharmacy records it was assumed that a prescription picked up and paid indicated that the medicine was taken. Statistical Methods To estimate the effect of compliance on recurrence, unadjusted and adjusted Cox regression models were used with time since diagnosis as underlying timescale. Individual compliance status was assessed using longitudinal information on prescription dates, intervals, and doses. In other words, compliance was handled as a dynamic variable taking each prescription into account All adjustment factors, listed in Table 1, were selected a priori and complete case analysis was used in the models. Of note, 40 patients (16%) had missing data with respect to smoking status, income, or education. To make sure the Cox proportional hazards assumption was fulfilled, the effect of all factors was divided into 4 time periods (91 to <365.25 d, 1 to <3 yr, 3 yr to <8 yr and 8+ yr), but the last period was disregarded because of sparse data. Those factors having a significant change in effect on recurrence over time were kept with a time interaction and the others were kept unchanged in the final adjusted model. To overcome the influence of disease activity at diagnosis, it was defined that patients could not have a recurrence the first 3 months from the date of diagnosis regardless of initial treatment. Therefore, follow-up started at 91 days after UC diagnosis and ended on the last day of the study (December 31, 2011), or because of emigration, death, or colectomy. All cohort members were considered at risk on the day after a previous recurrence. The time at which the medical records stated that the person was off 5-ASA treatment was not included. Because the patients could get the outcome several times, the sandwich estimator was used. All tests were Wald tests and associations were considered significant for P-values < 0.05. Table 1. Demographic and Clinical Data at Baselinea     View Large Table 1. Demographic and Clinical Data at Baselinea     View Large All data management and statistical analyses were performed using SAS software, version 9.3 (SAS Institute, Cary, NC). Ethical Considerations The Regional Ethics Committee (No. H-1-2011-088) and the Danish Data Protection Agency (No. 01769-HVH-2012-027) approved the study, and permission was obtained from the Danish Medicines Agency (No. 704006). Results From the original cohort a total of 243 patients were included in the analysis. Figure 1 describes the selection process. Five patients were censured due to emigration less than 3 months after diagnosis. Figure 1. View largeDownload slide Flowchart of sample selection. Figure 1. View largeDownload slide Flowchart of sample selection. The demographic and clinical characteristics at baseline are listed in Table 1. The distribution of patients and number of cases of recurrence with respect to compliance status are listed in Table 2. A total of 182 patients of 243 patients (75%) experienced at least one recurrence of UC during follow-up. A total of 693 cases of recurrences were registered on an average of 3.8 recurrences per patient (693/182). Table 2. Number of Patients and Cases of Recurrence with Respect to Compliance Status     View Large Table 2. Number of Patients and Cases of Recurrence with Respect to Compliance Status     View Large As we measured compliance over time each patient could dynamically switch between being compliant or noncompliant to oral 5-ASA. Because of the dynamic nature of compliance, 216 of the 243 included patients (88%) were at some point during follow-up registered as compliant. Likewise, 218 (89%) patients were at some point registered as noncompliant during follow-up. Of the 216 compliant patients, 150 patients (69%) experienced at least 1 recurrence during follow-up, on average 3 recurrences per patient. Only 106 of the 218 noncompliant patients (49%) suffered from at least one recurrence, on average 2 recurrences during follow-up. In Table 3, we have listed the distribution of recurrences with respect to compliance status divided into 3 periods during follow-up. The fourth period (above 8 yr) was not judged to be of statistical relevance because only 7 patients were included in this group. Of interest, almost twice as many recurrences are registered among compliant versus noncompliant patients in the first year after diagnosis (106/59). With respect to time periods, the recurrence rate per year was highest during the first year from diagnosis of UC in both groups, with 106 out of a total of 461 cases of recurrence (23%) among compliant patients versus 59 cases of a total of 232 cases of recurrences (25%) among noncompliant patients decreasing to 38.6 (8%) and 20.0 (9%) cases of recurrences per year on average in 3 to 8 years after diagnosis for compliant and noncompliant patients respectively. Table 3. Number of Recurrence Cases with Respect to Compliance Status Over Time     View Large Table 3. Number of Recurrence Cases with Respect to Compliance Status Over Time     View Large The authors' primary outcome was the risk of recurrence associated with compliance status over time, as listed in Table 4. With respect to time periods, the risk of recurrence did not differ significantly between compliant and noncompliant patients in the first year after diagnosis (P = 0.71). However, in 1 to 3 and 3 to 8 years after diagnosis, the risk of recurrence unexpectedly seemed to decrease significantly among noncompliant patients compared with compliant patients (adjusted hazard ratio [HR]: 0.46, 95% CI, 0.33–0.63; adjusted HR: 0.42, 95% CI, 0.32–0.55, respectively). The results were adjusted for all demographic and clinical factors listed in Table 1. Of note, the adjusted results of compliance and the risk of recurrence were stable when comparing to the unadjusted analysis (unadjusted HR [1–3 yr]: 0.57, 95% CI, 0.42–0.77; unadjusted HR [3–8 yr]: 0.51, 95% CI, 0.39–0.66). Table 4. Risk of Recurrence and the Relationship with Noncompliance     View Large Table 4. Risk of Recurrence and the Relationship with Noncompliance     View Large The risk of recurrence with respect to demographic and clinical factors is listed in Tables 5 and 6. Age showed to be time-dependent and therefore we accounted for this by comparing age groups within the 3 periods. As shown in Table 5, there was a statistically significant relationship between age and recurrence 1 to 3 and 3 to 8 years after diagnosis, where patients younger than 30 years of age had a higher risk of recurrence than patients with higher age (adjusted HR: 1.27, 95% CI, 0.90–1.79 and adjusted HR: 1.43, 95% CI, 1.07–1.93, respectively). Further, education was associated with risk of recurrence because more than 13 years of schooling decreased the risk of recurrence significantly (adjusted HR: 0.78, 95% CI, 0.63–0.96). Smoking status was also related to the risk of recurrence, as former smokers had an increased risk of recurrence compared with that of nonsmokers (adjusted HR: 1.36, 95% CI, 1.10–1.69), whereas current smokers seemed to have decreased risk of recurrence (adjusted HR: 0.75, 95% CI, 0.56–0.997). Sex, income and living conditions had no significant impact on recurrence. Table 5. Risk of Recurrence and the Relationship to Demographic Factors     View Large Table 5. Risk of Recurrence and the Relationship to Demographic Factors     View Large The relationship between clinical variables and risk of recurrence are listed in Table 6. Disease extent and disease activity did not influence recurrence risk significantly in the adjusted analysis. However, disease course was as expected significantly associated with risk of recurrence. Table 6. Risk of Recurrence and the Relationship to Clinical Factors     View Large Table 6. Risk of Recurrence and the Relationship to Clinical Factors     View Large Discussion This 7-year follow-up of a population-based inception cohort with detailed data from medical records and national prescription information, revealed somewhat surprisingly a statistically significant association between noncompliance and risk of recurrence 1 to 3 and 3 to 8 years after diagnosis of UC. The authors had hypothesized that noncompliance was associated with a higher risk of recurrence during follow-up. Contradictorily, these results implied a greater risk of recurrence among compliant patients compared with noncompliant patients. Using the prescription claims database from the Veterans Affairs Healthcare system, a recent nationwide retrospective cohort study examined the long-term adherence to mesalazine and the risk of recurrence of UC among 13,062 patients. Conflicting to the findings of this study, low adherers (consumption <50%) had a significantly increased risk of recurrence compared with high adherers (consumption ≥80%) (HR: 1.7 and P < 0.001).13 However, these measurements are not directly comparable as the authors divided compliance into noncompliant (consumption <80%) or compliant (consumption ≥80%) patient groups. To define recurrence, Khan et al have used a prescription of 40 mg per day or more of oral prednisolone or any dose of parenteral methylprednisolone as a proxy for recurrence, whereas any additional dose of steroids was defined as recurrent in the present study, which may lead to a higher number of recurrences. Khan et al lack information on clinical data, such as disease extent and activity, which could bias their results. The demographic data presented in the study by Khan et al showed that the study population was predominated by men (94%)13 compared to 50.2% men in this study. A study by Kane et al3 which followed 99 consecutive patients' outcome from an outpatient clinic at the University of Chicago Medical Center. Patients were included if they had UC in remission for more than 6 months and were taking mesalamine on maintenance. Nonadherence was defined based on pharmacy records as filling less than 80% of prescribed medication. Kane et al found an 89% chance of maintaining remission among adherent patients compared with only 39% among nonadherent patients (P = 0.001). Thus, nonadherence accounted for a 5-fold risk of recurrence compared with adherent patients (HR: 5.5 and P < 0.001). More recently, Kawakami et al19 have followed 104 outpatients with UC in remission prospectively over 1 year to examine nonadherence and its relationship with recurrence using self-administered questionnaire and medical records. The nonadherent patients (defined as taking <80% of the prescribed dosage of 5-ASA) showed a higher rate of 1-year relapse compared with adherent patients with a 2.3-fold increased risk of relapse (95% CI, 1.004–5.24, P = 0.04). This discrepancy between these studies and the authors' study may be explained by different patient populations because the former studies only included patients with a low risk of recurrence (remission >6 mo, no usage of steroid or immunomodulators <6 mo, and no hospitalization ≤12 mo), whereas in this unselected study most patients (82.7%) had either left-sided or extensive disease with a potential higher risk of recurrence. Higgins et al32 have drawn attention to the need for studies based on real-world settings and which follow patients over an extended time period to gauge true adherence or compliance to detect long-term outcomes including recurrence. Strengths and Limitations The major strength of the study is the design, which is based on an unselected, inception cohort using a national prescription database and linkage to prescription details from the medical records. Hence, we were able to account for misclassification in relation to compliance, as we could differentiate patients who did not take the medication as directed (discontinued the medication) versus patients who simply followed their prescribers' directions correctly (agreed on decrease in dosage or a pause in therapy). The fact that compliance was measured retrospectively reduces the impact of the Hawthorne effect which is defined in prospective observational studies as participants changing their performance in response to being observed. The use of registries prevents, to some extent, differentiated misclassification, because the data are registered unrelated to our hypothesis. With respect to coding errors, it is reasonable to assume that these are randomly distributed among the included patients and therefore do not bias the results of this study in any way.29 The use of prescription claim database does not account for whether or not the patients actually took the medication and how they took it. Nevertheless, compliance is unlikely in patients who do not redeem their prescriptions on a regular basis. The cost of certain drugs is only partly reimbursed, therefore patients who do obtain regular refills are highly likely to be compliant because most patients do not discard or stockpile expensive medication. The authors believe that the present approach is a reliable way to measure compliance and the closest one can get to measure long-term compliance in a real-world clinical practice setting. The authors were further able to evaluate the impact of long-term compliance on the risk of recurrence over time and to adjust for several demographic and clinical factors (age, sex, living conditions, education, income, smoking status, disease extent, disease course, disease activity, colectomy, and information on hospitalization). Finally, the observed association between disease course and recurrence demonstrates the internal validity of data. The authors are aware that this study had some limitations. The retrospective nature of the study makes it vulnerable to missing or unidentified confounders and modifiers. Still, they were able to account for several of these factors. Of note, they did not find a statistically significant association between disease activity and recurrence, which could be because the authors used a surrogate marker for disease activity based on additional treatment regimes. However, confounding by disease activity was judged to be minor. Due to the retrospective character of our study, the authors were not able to strictly define the date of remission for every patient, thus the 3-month period was arbitrarily chosen. In their opinion, 3 months were a reasonable time interval for patients to obtain remission and begin maintenance treatment of oral 5-ASA. With respect to switchers, the authors did not distinguish between 5-ASA preparations within the same class. If a patient initiated a new 5-ASA preparation, they assumed that the remaining tablets from the previous prescription were left unused because the patient would immediately begin to take the new 5-ASA preparations. Noncompliant patients may also have included patients with a degree of self-management, e.g., self-treating their mild disease successfully based on an on-demand approach. The authors were, however, not able to account for self-management, because they did not have information on how the patients followed the prescription. Of note, Hawthorne et al33 have emphasized that educational and behavioral interventions tailored to individual patients with UC can optimize medication adherence. In conclusion, this is the first study to evaluate the effect of compliance on the risk of recurrence of UC based on an inception cohort with linkage to medical records and a national prescription database. The major findings revealed a reverse and conflicting association between compliance and recurrence, compared with other studies although they adjusted for several demographic and clinical factors including previous disease activity. The results of this study may therefore reflect that some noncompliant patients self-treat their mild disease successfully based on an on-demand approach and they choose not to take their medication in long recurrence-free periods. Acknowledgments Guarantor of the article: F. Bendtsen, Professor, MD, DMSci. Author contributions: M. V. Prosberg contributed to the concept and study design, acquisition and interpretation of data, drafting and critical revision of the manuscript. M. K. Vester-Andersen contributed to the concept and study design, acquisition and interpretation of data, and critical revision of the manuscript. M. Andersson contributed to statistical analysis, interpretation of data, and critical revision of the manuscript. T. Jess contributed to interpretation of data and critical revision of the manuscript. J. T. Andersen contributed to the analysis of data and critical revision of the manuscript. I. Vind contributed to the concept and study design, and critical revision of the manuscript. F. Bendtsen contributed to the concept and study design, interpretation of data, and critical revision of the manuscript. All authors have made final approval of the version to be submitted. References 1. WHO. Adherence to Long-term Therapies—Evidens for Action. 2003. Available at: http://www.who.int/chp/knowledge/publications/adherence_full_report.pdf. Accessed September 16, 2015. 2. Kane SV, Cohen RD, Aikens JE, et al.  . Prevalence of nonadherence with maintenance mesalamine in quiescent ulcerative colitis. Am J Gastroenterol . 2001; 96: 2929– 2933. Google Scholar CrossRef Search ADS PubMed  3. Kane S, Huo D, Aikens J, et al.  . Medication nonadherence and the outcomes of patients with quiescent ulcerative colitis. Am J Med . 2003; 114: 39– 43. Google Scholar CrossRef Search ADS PubMed  4. Feagan BG, Macdonald JK. Oral 5-aminosalicylic acid for maintenance of remission in ulcerative colitis. Cochrane Database Syst Rev.  2012; 10:CD000544. 5. Cosnes J, Gower-Rousseau C, Seksik P, et al.  . Epidemiology and natural history of inflammatory bowel diseases. Gastroenterology . 2011; 140: 1785– 1794. Google Scholar CrossRef Search ADS PubMed  6. Dignass A, Lindsay JO, Sturm A, et al.  . Second European evidence-based consensus on the diagnosis and management of ulcerative colitis part 2: current management. J Crohns Colitis . 2012; 6: 991– 1030. Google Scholar CrossRef Search ADS PubMed  7. Järnerot G, Hertervig E, Friis-Liby I, et al.  . Infliximab as rescue therapy in severe to moderately severe ulcerative colitis: a randomized, placebo-controlled study. Gastroenterology . 2005; 128: 1805– 1811. Google Scholar CrossRef Search ADS PubMed  8. Jackson CA, Clatworthy J, Robinson A, et al.  . Factors associated with non-adherence to oral medication for inflammatory bowel disease: a systematic review. Am J Gastroenterol . 2010; 105: 525– 539. Google Scholar CrossRef Search ADS PubMed  9. Shale MJ, Riley SA. Studies of compliance with delayed-release mesalazine therapy in patients with inflammatory bowel disease. Aliment Pharmacol Ther . 2003; 18: 191– 198. Google Scholar CrossRef Search ADS PubMed  10. Gifford AE, Berg AH, Lahiff C, et al.  . A random urine test can identify patients at risk of mesalamine non-adherence: a prospective study. Am J Gastroenterol . 2013; 108: 249– 255. Google Scholar CrossRef Search ADS PubMed  11. Bernal I, Domènech E, Garcia-Planella E, et al.  . Medication-taking behavior in a cohort of patients with inflammatory bowel disease. Dig Dis Sci . 2006; 51: 2165– 2169. Google Scholar CrossRef Search ADS PubMed  12. Kane SV, Sumner M, Solomon D, et al.  . Twelve-month persistency with oral 5-aminosalicylic acid therapy for ulcerative colitis: results from a large pharmacy prescriptions database. Dig Dis Sci . 2011; 56: 3463– 3470. Google Scholar CrossRef Search ADS PubMed  13. Khan N, Abbas AM, Bazzano LA, et al.  . Long-term oral mesalazine adherence and the risk of disease flare in ulcerative colitis: nationwide 10-year retrospective cohort from the veterans affairs healthcare system. Aliment Pharmacol Ther . 2012; 36: 755– 764. Google Scholar CrossRef Search ADS PubMed  14. Kane SV, Accortt NA, Magowan S, et al.  . Predictors of persistence with 5-aminosalicylic acid therapy for ulcerative colitis. Aliment Pharmacol Ther . 2009; 29: 855– 862. Google Scholar CrossRef Search ADS PubMed  15. Elkjaer M, Shuhaibar M, Burisch J, et al.  . E-health empowers patients with ulcerative colitis: a randomised controlled trial of the web-guided “Constant-care” approach. Gut . 2010; 59: 1652– 1661. Google Scholar CrossRef Search ADS PubMed  16. Trindade AJ, Ehrlich A, Kornbluth A, et al.  . Are your patients taking their medicine? Validation of a new adherence scale in patients with inflammatory bowel disease and comparison with physician perception of adherence. Inflamm Bowel Dis . 2011; 17: 599– 604. Google Scholar CrossRef Search ADS PubMed  17. Ford AC, Khan KJ, Sandborn WJ, et al.  . Once-daily dosing vs. conventional dosing schedule of mesalamine and relapse of quiescent ulcerative colitis: systematic review and meta-analysis. Am J Gastroenterol.  2011; 106: 2070– 2077; quiz 2078. Google Scholar CrossRef Search ADS PubMed  18. Yen L, Wu J, Hodgkins PL, et al.  . Medication use patterns and predictors of nonpersistence and nonadherence with oral 5-aminosalicylic acid therapy in patients with ulcerative colitis. J Manag Care Pharm . 2012; 18: 701– 712. Google Scholar PubMed  19. Kawakami A, Tanaka M, Nishigaki M, et al.  . Relationship between non-adherence to aminosalicylate medication and the risk of clinical relapse among Japanese patients with ulcerative colitis in clinical remission: a prospective cohort study. J Gastroenterol . 2013; 48: 1006– 1015. Google Scholar CrossRef Search ADS PubMed  20. Lachaine J, Yen L, Beauchemin C, et al.  . Medication adherence and persistence in the treatment of Canadian ulcerative colitis patients: analyses with the RAMQ database. BMC Gastroenterol . 2013; 13: 23. Google Scholar CrossRef Search ADS PubMed  21. Vind I, Riis L, Jess T, et al.  . Increasing incidences of inflammatory bowel disease and decreasing surgery rates in Copenhagen City and County, 2003-2005: a population-based study from the Danish Crohn colitis database. Am J Gastroenterol . 2006; 101: 1274– 1282. Google Scholar CrossRef Search ADS PubMed  22. Vester-Andersen MK, Prosberg MV, Jess T, et al.  . Disease course and surgery rates in inflammatory bowel disease: a population-based, 7-year follow-up study in the era of immunomodulating therapy. Am J Gastroenterol . 2014; 109: 705– 714. Google Scholar CrossRef Search ADS PubMed  23. Romberg-Camps MJL, Dagnelie PC, Kester ADM, et al.  . Influence of phenotype at diagnosis and of other potential prognostic factors on the course of inflammatory bowel disease. Am J Gastroenterol . 2009; 104: 371– 383. Google Scholar CrossRef Search ADS PubMed  24. Wolters FL, Russel MG, Sijbrandij J, et al.  . Phenotype at diagnosis predicts recurrence rates in Crohn's disease. Gut . 2006; 55: 1124– 1130. Google Scholar CrossRef Search ADS PubMed  25. Höie O, Wolters F, Riis L, et al.  . Ulcerative colitis: patient characteristics may predict 10-yr disease recurrence in a European-wide population-based cohort. Am J Gastroenterol . 2007; 102: 1692– 1701. Google Scholar CrossRef Search ADS PubMed  26. ISPOR Medication compliance and persistence special interest group. Standardizing Definition of Terms. 2012. Available at: http://www.ispor.org/sigs/mcp_accomplishments.asp. Accessed September 16, 2015. 27. Munkholm P. Crohn's disease–occurrence, course and prognosis. An epidemiologic cohort-study. Dan Med Bull . 1997; 44: 287– 302. Google Scholar PubMed  28. Langholz E. Ulcerative colitis. An epidemiological study based on a regional inception cohort, with special reference to disease course and prognosis. Dan Med Bull . 1999; 46: 400– 415. Google Scholar PubMed  29. Wallach Kildemoes H, Toft Sorensen H, Hallas J. The danish national prescription registry. Scand J Public Health . 2011; 39: 38– 41. Google Scholar CrossRef Search ADS PubMed  30. Johannesdottir SA, Horváth-Puhó E, Ehrenstein V, et al.  . Existing data sources for clinical epidemiology: the danish national database of reimbursed prescriptions. Clin Epidemiol . 2012; 4: 303– 313. Google Scholar CrossRef Search ADS PubMed  31. Pedersen CB. The danish civil registration system. Scand J Public Health . 2011; 39: 22– 25. Google Scholar CrossRef Search ADS PubMed  32. Higgins PDR, Rubin DT, Kaulback K, et al.  . Systematic review: impact of non-adherence to 5-aminosalicylic acid products on the frequency and cost of ulcerative colitis flares. Aliment Pharmacol Ther . 2009; 29: 247– 257. Google Scholar CrossRef Search ADS PubMed  33. Hawthorne AB, Rubin G, Ghosh S. Review article: medication non-adherence in ulcerative colitis–strategies to improve adherence with mesalazine and other maintenance therapies. Aliment Pharmacol Ther . 2008; 27: 1157– 1166. Google Scholar CrossRef Search ADS PubMed  Copyright © 2016 Crohn's & Colitis Foundation of America, Inc.
TI - Long-term Compliance with Oral 5-aminosalicylic Acid Therapy and Risk of Disease Recurrence in Patients with Ulcerative Colitis: A Population-based Cohort Study
JF - Inflammatory Bowel Diseases
DO - 10.1097/MIB.0000000000000700
DA - 2016-04-01
UR - https://www.deepdyve.com/lp/oxford-university-press/long-term-compliance-with-oral-5-aminosalicylic-acid-therapy-and-risk-Fi8pv88iSN
SP - 925
EP - 932
VL - 22
IS - 4
DP - DeepDyve
ER -