Trends of 5-Aminosalicylate Medication Use in Patients With Crohn Disease

Abstract

Background

5-aminosalicylate (5-ASA) medications have a long history of use for the treatment of inflammatory bowel disease and continue to be widely prescribed today. The effectiveness of 5-ASAs in ulcerative colitis is clear; however, studies have shown little benefit for induction or maintenance treatment of Crohn disease (CD). We aimed to quantify usage and examine trends in 5-ASA prescription rates in patients with CD.

Methods

Using a retrospective design, we queried a national database of commercially insured patients (Truven-Health databases) between 2009 and 2014 to identify patients with CD aged 18 to 65 years. Prescription rates for 5-ASA medications including sulfasalazine, mesalamine, olsalazine, and balsalazide were calculated for each calendar year. Regression models were used to examine year-to-year trends in prescription rates and identify patient factors associated with 5-ASA use.

Results

We identified 132,804 patients with CD, of whom 37.3% (n = 49,529) received a 5-ASA prescription during the study period. From 2009 to 2014, the overall prescription rates of 5-ASAs declined from 42.9% to 30.0% (P < 0.001). Patient factors independently associated with 5-ASA use included younger age, male sex, multimorbidity, and a health maintenance organization insurance plan, while controlling for the region of residence.

Conclusions

About 1 in 3 privately insured patients with CD received 5-ASA prescriptions despite their questionable effectiveness; however, in an encouraging trend, prescription rates significantly decreased from 2009 to 2014. This high prescription rate may reflect a gap in providers’ knowledge regarding the available evidence—an opportunity for cost savings with improved health care delivery.

INTRODUCTION

Crohn disease (CD) is a subtype of inflammatory bowel disease (IBD) that has a chronic remitting and relapsing pattern and is associated with decreased quality of life. Most patients require lifelong medical therapy, with surgery generally reserved for medically refractory or complicated disease.¹ Historically, 5-aminosalicylates (5-ASAs) have been widely used to treat patients with IBD, including CD.^2-4 However, the literature has shown that the role of 5-ASAs in the induction of remission and prevention of relapse in CD is uncertain.^{5, 6} For example, mesalamine was not consistently found to be better than placebo for induction and achieving mucosal healing in patients with active disease.^{5, 6} In general, current practice guidelines strongly recommend against the use of 5-ASA medications for treating patients with CD, including those with mild-to-moderate disease.^{1, 7} The one exception is sulfasalazine, which has shown a low level of evidence¹ in the treatment of mild Crohn colitis.^{1, 4, 8-11} Current practice guidelines recommend the use of steroids, immunomodulators, or biologics for the treatment of CD.

Despite the evidence of questionable effectiveness and more effective medical therapies we have observed that 5-ASAs are frequently prescribed for patients with CD. As such, we aimed to examine the prevalence and trends in 5-ASA medication use in patients with CD and to identify independent patient factors associated with the use of 5-ASA medications as targets for future interventions.

METHODS

Design

We performed a retrospective study to identify the prevalence of and examine trends in 5-ASA medication use in patients with CD.

Data Source

We queried a national database of commercially insured patients (Truven-Health databases) between 2009 and 2014 to identify patients with CD, because we had access to the data within the specified period. The MarketScan databases are insurance claims–based databases that contain the health information of more than 230 million privately insured patients from all over the United States. Rigorous methods are used to ensure that the data are complete, accurate, and reliable.¹² Myriad published studies have used the database to accurately evaluate health outcomes in various clinical settings.^12-15 We examined outpatient visits, inpatient admissions, services, enrollment tables, and pharmacy claims within these databases. We obtained approval from the institutional review board at the University of Michigan before the evaluation of the data (HUM00127665).

Study Sample

We used the ICD-9-CM codes to identify patients aged 18 to 65 years with a diagnosis of CD (555.x). We required the presence of at least 3 ICD-9-CM CD codes for the case definition.^{13, 15} Patients were excluded if they carried any ICD-9-CM code for ulcerative colitis (556.x). We required continuous insurance enrollment of at least 1 year to allow for enough time to capture comorbidities for patients included in the univariate and multivariable comparisons. Prescription rates of 5-ASA medications including sulfasalazine, mesalamine, olsalazine, and balsalazide were identified using National Drug Codes listed in the pharmacy claims.

Outcomes

We abstracted age, sex, Elixhauser comorbidities, region, insurance plan type, and median family income for each patient. To abstract Elixhauser comorbidities¹⁶ (congestive heart failure, valvular disease, pulmonary circulation disease, peripheral vascular disease, paralysis, other neurological disorders, chronic pulmonary disease, diabetes without chronic complications, diabetes with chronic complications, hypothyroidism, renal failure, liver disease, peptic ulcer disease, AIDS, lymphoma, metastatic cancer, solid tumor without metastasis, rheumatoid arthritis and collagen vascular disease, coagulopathy, obesity, weight loss, fluid and electrolyte disorders, chronic blood loss anemia, deficiency anemias, alcohol abuse, drug abuse, psychoses, and depression), we used an established algorithm developed by the Healthcare Cost and Utilization Project¹⁷ and subsequently calculated an Elixhauser comorbidity index for each patient. Median family income data were obtained by linking each patient’s metropolitan statistical area to income data obtained from the Federal Financial Institutions Examination Council.¹⁸ The primary endpoint was a prescription of 5-ASA in a patient with CD.

Statistical Analysis

We calculated annual prescription rates for each 5-ASA medication between 2009 and 2014. Linear regression models were used to examine trends in 5-ASA prescription rates across the calendar years. We then performed univariate unadjusted analyses to compare demographics and baseline comorbidities between those who received and did not receive a 5-ASA prescription. We used the Student t test and χ ² test for normally distributed continuous and categorical variables, respectively. The Wilcoxon rank-sum test was used for nonnormally distributed continuous variables. A multivariable logistic regression model was used to identify independent patient factors associated with 5-ASA use. We selected variables for inclusion in the multivariable model if they were significant on the univariate analyses. We used a 2-sided P value of ≤0.05 to denote statistical significance. In addition, SAS software was used to perform all analyses (version 9.4, SAS Institute Inc., Cary, NC).

In sensitivity analyses for patient factors associated with 5-ASA use, we rebuilt the logistic regression model after excluding patients with sulfasalazine because there is a low level of evidence suggesting its use in patients with mild Crohn colitis. We also examined the overall rate of 5-ASA use and annual trends after the exclusion of patients with sulfasalazine use. We also explored the rate of new 5-ASA prescriptions in patients with CD. A new prescription was defined as no prior use of 5-ASA medication during a 365-day washout period. We calculated the annual rate from the calendar year 2010 to the calendar year 2014. The rate of new prescriptions was calculated as the total number of new 5-ASA prescriptions per year divided by the annual population at risk. In addition, we examined the prevalence of patients with arthritic comorbid conditions including rheumatoid, enteropathic, and other spondyloarthritis such as psoriatic arthritis and ankylosing spondylitis among patients using sulfasalazine to examine whether these comorbid conditions were the main indication for sulfasalazine use. The prevalence of these comorbid conditions was calculated as the total number of unique patients who carried one of the diagnoses divided by the total number of unique sulfasalazine users between 2009 and 2014.

RESULTS

Between 2009 and 2014, we identified a total of 132,804 patients with CD, of whom 37.3% (n = 49,529) received a 5-ASA prescription during the study period. The overall prescription rates for 5-ASA medications declined from 42.9% (15,972/37,224) to 30.0% (4886/16,264) from 2009 to 2014 (absolute change of –12.9%, relative change of –30.0%; P < 0.01). Among patients treated with 5-ASA medications, mesalamine was the most commonly prescribed, with a rate of 90.2% (44,683/49,529). From 2009 to 2014, the annual prescription rates for mesalamine declined from 38.7% to 26.6%, followed by sulfasalazine (3.5% to 3.2%), balsalazide (1.5% to 1.2%), and olsalazine (0.2% to 0.04%). All individual medications showed statistically significant decreasing trends (P < 0.01) except for balsalazide (P value = 0.37) (Table 1).

After the exclusion of patients with a less than 1-year enrollment, a total of 101,414 patients with CD were included in the subsequent analyses. In univariate unadjusted analyses, patients prescribed a 5-ASA medication were more likely to be female and of older age (45–65 years), have fewer comorbid conditions, reside in the South, have a preferred provider organization (PPO) insurance plan type, and have a median annual income in the range of $75,000 to $85,000 (P < 0.05) (Table 2).

In the multivariable analysis, independent patient factors associated with 5-ASA use were younger age (18–29 years) (odds ratio [OR] = 1.07; 95% confidence interval [CI], 1.03-1.10), male sex (OR = 1.12; 95% CI, 1.09-1.15), multimorbidity (OR = 1.17-1.35; 95% CI, 1.14-1.40), and a health maintenance organization (HMO) insurance plan type compared with a PPO (OR = 1.07; 95% CI, 1.02-1.11), while controlling for the region of residence. Of note, median family income was not independently associated with 5-ASA use (Table 3).

In sensitivity analyses, after the exclusion of patients with sulfasalazine use (n = 4242), our findings generally remained the same. The overall rate of 5-ASA use was 35.2% (83,275/128,562). From 2009 to 2014, the overall rate of 5-ASA use declined from 40.8% to 27.3% (P < 0.01). Factors associated with 5-ASA use were younger age (18–29 years) (OR = 1.12; 95% CI, 1.09-1.16), male sex (OR = 1.12, 95% CI, 1.10-1.15), multimorbidity (OR = 1.16-1.32; 95% CI, 1.12-1.37), and an HMO compared with a PPO (OR = 1.10; 95% CI, 1.03-1.12), while controlling for the region of residence and income. The rate of new 5-ASA prescriptions declined from 15.6% (3089/19,834) in 2010 to 11.8% (1630/13,838) in 2014 (absolute change of –3.8%, relative change of –24.4%; P < 0.05) ( Supplementary Appendix 1). Of all sulfasalazine users, the prevalence of unique patients with a diagnosis of spondyloarthropathy was 8.1% (308/3803).

DISCUSSION

Among a U.S. cohort of 132,804 patients with CD, we found that approximately 1 in 3 were treated with a 5-ASA medication despite questionable evidence regarding its efficacy. Of these patients, 90.2% were treated with mesalamine, which current guidelines generally recommend against.^{1, 7} This is an important finding because it may reflect a gap in providers’ knowledge regarding the literature. This finding also represents a valuable opportunity for cost savings with improved therapy.

The current American College of Gastroenterology (ACG) guidelines of practice strongly recommend against the use of oral mesalamine for treating patients with CD, including those with mild-to-moderate disease.¹ The literature has not been consistent in showing the efficacy of mesalamine use compared to placebo use in achieving mucosal healing in patients with active CD.^{5, 6, 19} A Cochrane review also supported the same finding that there is no evidence of 5-ASA superiority to placebo for the maintenance of remission achieved with medical therapy.⁷ Although a conditional recommendation was made in the ACG guidelines for the use of mesalamine as an option for patients who undergo ileal resection for isolated ileal disease, no treatment was also considered acceptable in this setting.¹ In addition, a conditional recommendation was made, with a low level of evidence, for the use of sulfasalazine in patients with mild-to-moderate isolated colonic disease and that it should not be used in patients with perianal or fistulizing disease.^{1, 4, 8-11}

In light of this evidence, our observation that 37.3% of patients were prescribed a 5-ASA medication, of whom 90% were treated with mesalamine, is not consistent with the current standard of care. In addition to a gap in providers’ knowledge regarding the literature, this result could also be attributed to the lag between evidence and guidelines. That is an important finding especially in the presence of other more effective medical therapies including systemic and nonsystemic steroids, immunomodulators such as thiopurines and methotrexate, and biologics including anti-tumor necrosis factor agents. Sulfasalazine could be indicated for inflammatory arthritis in patients with IBD. However, of all sulfasalazine users, we found that only 8.1% reported a comorbid diagnosis of rheumatoid, enteropathic, or other spondyloarthritis. This finding could reflect that arthritides may have not been the main indication for prescribing sulfasalazine in most patients with CD in our study.

The gap between research and practice is a known issue that has been reported across multiple disciplines.^{20, 21} Reports have estimated that it takes an average of 17 years for research evidence to reach clinical practice.^{22, 23} Factors contributing to this problem include voids in communication between researchers and community practitioners and service delivery issues such as the lack of public awareness. Proposed solutions include increased engagement between researchers and community practitioners, continuing professional education, and the delivery of quality training for practitioners to enhance their skills in the adaptation, implementation, and evaluation of scientific evidence.^{20, 21}

This gap does not seem to be unique to the United States; reports from Australia and Germany have identified a similar issue of prescribing 5-ASA medications for patients with CD.^{24, 25} Gearry et al²⁴ surveyed a group of Australian gastroenterologists to examine their prescribing patterns. The survey found that 5-ASA medications were most commonly prescribed for patients with colonic (96%) and ileocolonic disease (92%). Approximately 70% of 5-ASA use was as the only medication prescribed for the induction of remission. According to the surveyed physicians, arguments for the use of 5-ASAs included anecdotal evidence of efficacy in mild disease, excellent safety profile, low cost to the patients with government subsidy, avoidance of adverse effects of alternative therapy, and a possibility for reducing colon cancer. Arguments against the use of 5-ASAs were high-quality evidence showing minimal therapeutic advantage, rare but serious adverse effects, high cost to the community in general, availability of alternative more effective medical therapies, and lack of specific evidence for reducing colon cancer in patients with CD.²⁴

Evidence against the use of 5-ASA medications was published in 2004 and 2011.^{5, 6} Guidelines with strong recommendations against the use of 5-ASAs were published in 2018.¹ Updated guidelines may take significant time to develop and reflect the updated newly available evidence. This possibility has been reported across other disciplines.²⁶ A solution for this lag could be the development of clinical pathways that allow for early adoption of evidence-based changes in practice.²⁶ Further studies are warranted to examine the trends in 5-ASA use following the publication of the 2018 ACG guidelines.

We found that a patient factor associated with 5-ASA use included having an HMO insurance plan compared with having a PPO plan. In general, HMO plans require a referral from a primary care physician to see a gastroenterologist. In addition, HMO plans may have a limited medication formulary compared with PPO plans. As such, insurance plans may play a role in determining patient access to alternative treatment options. We found that median family income was not associated with 5-ASA use. This is an important finding given that alternative treatment options such as immunomodulators and biologics are relatively more expensive.²⁴ Taken together, these findings may suggest that there is an access problem to alternative medical therapies as opposed to financial cost issues. We also found that patients with more comorbid conditions were more likely to use 5-ASAs. This finding is in line with the possibility that immunosuppressive therapy is contraindicated in patients with complex medical comorbid conditions that may increase the risk of these agents.²⁷

Despite the multifaceted problem, in an encouraging trend our findings showed that overall and new 5-ASA prescription rates declined significantly by 30% and 24.4%, respectively, from 2009 to 2014. This result may reflect a momentum toward adopting the evidence against 5-ASA use.

The limitations of this study are inherent to any large administrative database; although they provide a great opportunity for examining overall trends, they lack granularity of data. For example, we were not able to determine the duration or severity of CD, which may have otherwise helped further elucidate treatment patterns across the degrees of disease activity. In addition, we were not able to determine the anatomical location of disease or presence of perianal involvement. Likewise, we were not able to determine 5-ASA prescribers’ characteristics and specialties or specify facility-specific information such as care provided in tertiary referral IBD centers or in the community. In addition, administrative databases are generally subject to data entry errors related to erroneous billing codes. Truven-Health databases are subject to rigorous methods to ensure their reliability and accuracy,¹² and myriad published studies have used the database to accurately evaluate health outcomes in various clinical settings^12-15

Despite its limitations, the present study of >132,000 privately insured patients from all over the United States, treated over a 6-year period, provide important generalizable data on the prevalence, trends, and patient factors associated with 5-ASA use in CD. Our findings also summarize the current literature related to the use of 5-ASA medications in patients with CD, which may help further educate providers about this topic and bridge the gap between research and practice.

In conclusion, more than 1 in 3 privately insured patients with CD receive a 5-ASA prescription despite evidence of questionable efficacy. This finding underscores the significance of the gap between research and practice in treating CD. In an encouraging trend, prescription rates significantly decreased from 2009 to 2014. This represents an opportunity for cost savings with improved therapy.

Author contributions: Mohamed Noureldin: conception and design of study, acquisition of data, analysis and interpretation of data, drafting article and revising critically for important intellectual content, final approval of version to be submitted. Shirley Cohen-Mekelburg: interpretation of data, drafting article and revising critically for important intellectual content, final approval of version to be submitted. Asadullah Mahmood: interpretation of data, drafting article and revising critically for important intellectual content, final approval of version to be submitted. Ryan Stidham: conception and design of study, interpretation of data. Peter D.R. Higgins: interpretation of data, final approval of version to be submitted. Shail Govani: conception and design of study, interpretation of data, final approval of version to be submitted. Amar R. Deshpande: interpretation of data, drafting article and revising critically for important intellectual content, final approval of version to be submitted. Akbar K. Waljee: conception and design of study, acquisition of data, analysis and interpretation of data, drafting article and revising critically for important intellectual content, final approval of version to be submitted.

Supported by: No funds were received in support of this work.

Presented at: Digestive Disease Week 2019, San Diego, CA.

Conflicts of interest: No conflict of interest related to the content of the article. No benefits in any form have been or will be received from any commercial party related directly or indirectly to the subject of this article.

APPENDIX 1: ANNUAL RATES (%) OF NEW 5-ASA PRESCRIPTIONS IN PATIENTS WITH CD

REFERENCES