ABSTRACT

Objectives: To examine the association between treatment compliance with biologics approved for rheumatoid arthritis (RA) and 1-year total and RA-specific nonbiologic healthcare costs.

Study Design: Retrospective analysis of MarketScan Commercial Database claims data from July 1, 2009, to December 31, 2013.

Methods: Non-elderly adults (aged 18-63 years) with RA initiating treatment (index date) with a first-line biologic, with continuous plan enrollment for 6 months preindex and 12 months post index, were eligible. Outcomes included compliance (using proportion of days covered [PDC]) and persistence. Associations between compliance and total and RA-specific nonbiologic costs were assessed.

Results: Data from 14,696 patients were analyzed. Mean PDC (SD) was 0.65 (0.31); 46.8% of patients were persistent on index biologic. Mean total healthcare costs were $44,387 for intravenous abatacept, $40,434 for subcutaneous abatacept, $33,422 for adalimumab, $36,599 for certolizumab pegol, $33,214 for etanercept, $34,381 for golimumab, and $40,188 for infliximab. Compared with poorly compliant patients (PDC <0.2), total nonbiologic incremental costs for more compliant patients with PDC 0.2 to <0.4, 0.4 to <0.6, 0.6 to <0.8, and ≥0.8, respectively, were lower at –$1678 (cost ratio [CR] = 0.91; 95% CI, 0.86-0.97; P = .0025), –$4158 (CR = 0.78; 95% CI, 0.74-0.83; P <.0001), –$5127 (CR = 0.73; 95% CI, 0.69-0.78; P <.0001), and –$7961 (CR = 0.58; 95% CI, 0.56-0.62; P <.0001). The respective RA-related incremental nonbiologic costs, compared to poorly compliant patients, were $186 (CR = 1.08; 95% CI, 1.02-1.15; P = .0067), –$168 (CR = 0.92; 95% CI, 0.87-0.98; P = .0126), –$225 (CR = 0.90; 95% CI, 0.85-0.95; P = .0002), and –$560 (CR = 0.75; 95% CI, 0.71-0.79; P <.0001).

Conclusions: Better compliance was associated with lower overall and RA-related nonbiologic costs.

                                                                                         Am J Pharm Benefits. 2017;9(5):84-90

Rheumatoid arthritis (RA) is a systemic, inflammatory, autoimmune disease that primarily affects the linings (synovial membranes) of the joints. Approximately 1.3 million adults in the United States are estimated to have RA.1 A recent study of the total societal burden of RA estimates an annual healthcare cost of $8.4 billion and total societal cost of $19.3 billion per year.2 Left untreated, 20% to 30% of people diagnosed with RA become work-disabled within 3 years.3 Patients with RA suffer significant impairments in health-related quality of life, which can be alleviated by effective treatment.4

The American College of Rheumatology (ACR) recommends treatment with nonbiologic disease-modifying antirheumatic drugs (DMARDs) for early RA patients with low disease activity and an absence of poor prognostic features, and for early RA patients with moderate disease activity with poor prognostic features.5 ACR recommendations also suggest a more aggressive approach to inhibit the progression of joint damage and other complications from RA that may develop soon after diagnosis. These recommendations include the use of a biologic tumor necrosis factor (TNF) blocker with or without methotrexate in patients who have high disease activity with poor prognostic features. For patients with established RA (disease duration ≥6 months), biologic DMARDs are recommended if they have had an inadequate response to combination therapy with nonbiologic DMARDs. Biologic DMARDs approved by the FDA for first-line treatment of RA include the TNF blockers adalimumab, certolizumab pegol, etanercept, golimumab, and infliximab, and a T-cell co-stimulation blocker, abatacept.

Medication compliance/adherence is an important aspect of any treatment. An analysis of pharmacy claims data across 6 chronic conditions reported adherence rates ranging from 28% to 66%.6 Noncompliance can lead to increased healthcare use, including hospitalizations, fractures, cardiovascular events, and emergency department visits.7-10 Notably, increased healthcare use due to medication noncompliance is associated with increased costs.11-13 In patients using antihypertensive medications, for instance, the cost of using more drugs by compliant patients was less than spending on acute cardiovascular events in noncompliant patients.13

The objective of the current study was to examine and quantify the association between treatment compliance with first-line RA biologics and total and RA-specific nonbiologic healthcare costs over 1 year. Medication cost is a dominant component of healthcare cost among RA patients. Although increasing compliance to RA therapies would increase RA medication costs, some other components of healthcare costs may be reduced, because patients’ improved outcomes may result in lower likelihood of future hospitalizations, physician visits, or disease-related complications.

PATIENTS AND METHODS

Data Source

Data for this analysis were obtained from the Truven Health Analytics MarketScan® Claims and Encounters Database (Commercial). The Commercial Database contains inpatient medical, outpatient medical, and outpatient prescription drug information of employees and their dependents covered under a variety of fee-for-service and managed care health plans, including exclusive provider organizations, preferred provider organizations, point-of-service plans, indemnity plans, and health maintenance organizations. It provides detailed cost, use, and outcomes data for healthcare services performed in both inpatient and outpatient settings, including outpatient prescription drug claims and person-level enrollment data through the use of unique enrollee identifiers.

Study Design

The study used data from July 1, 2009, to December 31, 2013, spanning the pre-index and postindex period for all eligible patients. The index date was the date of the first claim for abatacept, adalimumab, certolizumab pegol, etanercept, golimumab, or infliximab between January 1, 2010, and January 1, 2013. The baseline period included the 6 months prior to the index date (pre-index period). The follow-up period encompassed the 12 months following and including the index date (postindex period).

Patient Selection

To be eligible for inclusion in the analysis, patients had ≥1 claim for first-line biologic therapy for RA between January 1, 2010, and January 1, 2013; were aged 18 to 63 years; had continuous enrollment in the claims database with medical and pharmacy benefits during the pre-index and postindex periods; and had ≥1 nondiagnostic medical claim with a primary or secondary diagnosis of RA (International Classification of Diseases, Ninth Revision, Clinical Modification [ICD-9-CM] diagnosis code 714.0x) during the pre-index period or on the index date. Patients were excluded from the analysis if they had ≥1 nondiagnostic medical claim with a diagnosis for other biologic indications at any time during the pre-index period (including the index date), including plaque psoriasis (ICD-9-CM code 696.1x), psoriatic arthritis (696.0x), ankylosing spondylitis (720.0x), juvenile idiopathic arthritis (714.3x), Crohn’s disease (555.xx), ulcerative colitis (556.xx), non-Hodgkin lymphoma (200.xx, 202.xx), or chronic lymphocytic leukemia (204.1x); ≥1 claim for any of the index drugs during the pre-index period (not including the index date); ≥1 claim for any biologic agent indicated for second-line treatment of RA (or not indicated for RA) or tofacitinib during the pre-index period (including the index date); ≥1 claim for ustekinumab during the study; claims for ≥2 different first-line RA biologics on the index date; any claim for tocilizumab prior to the FDA approval date (January 11, 2010); any claim for a first-line RA treatment administered only through subcutaneous (SC) injection with a Healthcare Common Procedure Coding System (HCPCS) code of that treatment (non-standard administration) during the study; or any claim for a first-line RA treatment administered only through intravenous (IV) infusion with a National Drug Code (NDC) of that treatment (non-standard administration) during the study.

Study Outcomes

Outcomes included rates of compliance, rates of persistence, and the total nonbiologic healthcare and RA-related nonbiologic healthcare cost over 1 year.

Compliance was determined by the proportion of days covered (PDC) for the index biologic during the 12-month follow-up period. Low compliance of self-administered biologics (abatacept, adalimumab, certolizumab pegol, etanercept, golimumab) was defined as proportion of days covered <80% during the post index period. Low compliance of infused biologics (abatacept, infliximab) was defined as insufficient infusions corresponding to the lower end of the permissible dosing schedules (14-18 infusions per year for abatacept or 7-10 infusions per year for infliximab) over the postindex period. Persistence was measured based on days on index therapy. Patients were considered persistent on index therapy until a 45-day gap was observed in claims for the index biologic or a switch to another biologic.

Total net healthcare cost was defined as all-cause healthcare cost (including both medical and pharmacy cost), excluding RA-related biologic cost, over 12 months of follow-up. RA-related net healthcare cost was defined as RA-related healthcare cost (medical and pharmacy cost), excluding RA-related biologic cost. Inpatient medical claims with a primary diagnosis of RA (ICD-9-CM: 714.0x) and outpatient medical claims with a primary or secondary diagnosis of RA constituted RA-related medical cost. Medical costs included all services rendered during a hospital stay, emergency department visits, physician visits, laboratory tests, medical procedures, and radiological examinations. RA-related pharmacy cost included both drug and administration costs using NDC or HCPCS codes that were indicative of oral- or joint-administered glucocorticoids, conventional DMARDs, biologic DMARDs, and tofacitinib. Total healthcare costs were the sum of patient and health plan costs.

Statistical Considerations

Descriptive analyses were conducted for categorical variables (number, percentage) and continuous variables (mean, SD, range, and 95% CIs). Treatment exposure (dose and number of administrations) was accounted for using a validated effectiveness algorithm14 that has been applied to this database in previous analyses.15 Generalized linear models with a gamma distribution and a log link were specified to assess the association between compliance and the various cost variables. Additionally, treatment-specific comparisons in cost were explored within these models. Models were adjusted for patient demographics (age, gender, health plan type, index year), baseline clinical characteristics and indices of health status (pre-index medications including antibiotics, oral corticosteroids and joint injections, selected comorbid conditions, number of ICD-9-CM codes, number of prescriptions, number of unique medications), baseline treatments (index biologic), and pre-index costs. Adjusted incremental costs for each level of compliance were estimated using recycled predictions.16 The incremental analysis was conducted by comparing each subsequent PDC cohort with patients with PDC <0.2. Costs were adjusted to the most recent year of data available for the study (2013).

RESULTS

Patients

A total of 14,696 patients were eligible to be included in the study (eAppendix Table 1; [eAppendices available at ajpb.com]). The mean age (SD) was 49.3 (10.0) years, most patients were female (78.3%), and most patients were in a managed care plan (87.1%) (Table 1). During the pre-index period, the mean number (SD) of unique ICD-9-CM codes was 10.0 (6.7), the mean number (SD) of prescription claims with a unique NDC code was 21.2 (15.9), and the mean number (SD) of unique drug claims was 8.4 (5.3).

Compliance and Persistence

Across all RA biologic treatments, the mean (SD) PDC was 0.65 (0.31) for the index medication. Compliance based on PDC >0.8 was achieved by 44.4% of patients during the 12-month postindex period (eAppendix Table 2). Overall persistence on index medication during the 12-month post index period was 46.8%.

Costs

The mean net healthcare cost (SD) for all patients was $34,863 ($29,221), and mean RA-related healthcare cost (SD) was $22,665 ($11,571) during the postindex period (eAppendix Table 3).

Association Between Compliance and Costs

Across all index medication cohorts, compliant patients had consistently lower total net healthcare costs compared with noncompliant patients (Figure). In general, increasing PDC led to lower all-cause and RA-related nonbiologic costs (Table 2). Decreases in incremental total net healthcare costs and RA-related nonbiologic costs were observed with increasing compliance (Table 3).

Based on generalized linear models, compliance was a significant predictor of all-cause nonbiologic costs and RA-related nonbiologic costs (Table 4). Compared with patients with poor compliance (PDC <0.20), the all-cause nonbiologic cost ratio for compliant patients (PDC ≥0.80) was 0.58 (95% confidence limits [CL], 0.56, 0.62; P <.0001) and RA-related nonbiologic cost ratio for compliant patients was 0.75 (95% CL, 0.71, 0.79; P <.0001). Compared with etanercept, cost ratios were significantly higher for abatacept (IV or SC), certolizumab pegol, golimumab, and infliximab (Table 4). Regression models for total healthcare costs, nonbiologic costs, RA-related costs, RA-related nonbiologic costs, and biologic costs are provided in eAppendix Tables 4 to 8.

DISCUSSION

The results of this study showed that across all index medication cohorts, compliant patients had consistently lower all-cause nonbiologic healthcare costs compared with those of noncompliant patients. Medication cost represents a large component of healthcare cost among RA patients. Intuitively, increasing compliance would increase RA medication costs, but understanding whether there may be a reduction in other healthcare costs due to improved outcomes is important in the overall care of the patient. Showing that increasing compliance with biologics for RA reduces other healthcare costs can help a health plan justify new programs to improve compliance and patient outcomes.

Studies have shown that nonadherence to medication can lead to accelerated progression of some diseases, increasing the need for more aggressive treatment and associated complications, and therefore increasing healthcare utilization and cost.17,18 Compliance with nonbiologic and biologic DMARDs for the treatment of RA is notoriously suboptimal.19 A prior study comparing medication adherence and cost of adalimumab and etanercept using a claims database showed that greater adherence to medication led to significantly fewer inpatient visits for etanercept (P = .0005) and a trend toward fewer inpatient visits for adalimumab (P = .0536).20 Similar to the results of our study, adherence to both medications led to significantly lower total costs (P ≤.0001). Compliance with RA medications over many years has the potential to lead to substantial savings in long-term nonbiologic costs for RA by further reducing disease progression and disability. Comparisons of compliance with oral DMARDs (eg, tofacitinib) versus compliance with injectable DMARDS (eg, biologics) are warranted.

Limitations

This analysis was limited in that it evaluated only the impact of treatment compliance on direct medical costs; future research should include productivity costs, as these may also be impacted by biologic treatment compliance. Limitations that are commonly identified with the use of administrative claims databases for clinical research also apply. The database used in this study consists of adjudicated claims that are used to facilitate payment; while it is likely that coding errors are rare, claims that facilitate payment are more likely to be present and complete in the data. Patients were required to meet pre-index enrollment criteria; results from this study may not be generalizable to patients with shorter or intermittent insurance coverage. Some bias arising from dropping patients without complete data may have occurred. Reasons for discontinuing or changing therapy are not available in the databases. Clinical information about disease severity is not available in claims databases. Furthermore, other unmeasured confounders (demographic and clinical characteristics, physician preferences/education, etc) could have affected the association between compliance and total cost. These findings may not be generalizable to uninsured individuals, non-US patients, and pediatric patients with RA.

CONCLUSIONS

Patient compliance was associated with lower overall and RA-related nonbiologic costs. In addition to the clinical benefits of biologic therapies for RA (including improvement in symptoms to achieve low disease activity or remission and associated improvements in quality of life4), payers and patients who are highly compliant with their biologic therapies can realize economic benefits compared with patients who are poorly compliant. Efforts to encourage medication compliance in this patient population should always be included in a treatment plan.

Author Affiliations: Truven Health Analytics, Cambridge, MA (MB, BHJ); Amgen Inc, Thousand Oaks, CA (DHT, DJH, BSS).
Source of Funding: This study was sponsored by Immunex, a wholly owned subsidiary of Amgen Inc, and by Wyeth, which was acquired by Pfizer in October 2009.
Author Disclosures: MB and BHJ are employees of Truven Health Analytics, which received funding from Amgen Inc for this study. DHH and BSS are employees and shareholders of Amgen Inc. DT is a former employee and current shareholder of Amgen Inc.
Authorship Information: Concept and design (MB, BHJ, DHT, DJH); acquisition of data (MB, BHJ); analysis and interpretation of data (MB, BHJ, DHT, DJH, BSS); drafting of the manuscript (MB); critical revision of the manuscript for important intellectual content (MB, BHJ, DHT, DJH, BSS); statistical analysis (MB, BHJ); provision of patients or study materials (MB, BHJ); and supervision (DHT, BSS).
Address Correspondence to: Machaon Bonafede, Truven Health Analytics, 150 Cambridge Park Dr, Cambridge, MA 02140. E-mail: machaon.bonafede@truvenhealth.com.

REFERENCES
1. Helmick CG, Felson DT, Lawrence RC, et al. Estimates of the prevalence of arthritis and other rheumatic conditions in the United States. Part I. Arthritis Rheum. 2008;58(1):15-25. doi: 10.1002/art.23177.
2. Birnbaum H, Pike C, Kaufman R, Marynchenko M, Kidolezi Y, Cifaldi M. Societal cost of rheumatoid arthritis patients in the US. Curr Med Res Opin. 2010;26(1):77-90. doi: 10.1185/03007990903422307.
3. Sokka T. Work disability in early rheumatoid arthritis. Clin Exp Rheumatol. 2003;21(5 Suppl 31):S71-S74.
4. Strand V, Sharp V, Koenig AS, et al. Comparison of health-related quality of life in rheumatoid arthritis, psoriatic arthritis and psoriasis and effects of etanercept treatment. Ann Rheum Dis. 2012;71(7):1143-1150. doi: 10.1136/annrheumdis-2011-200387. 
5. Singh JA, Furst DE, Bharat A, et al. 2012 update of the 2008 American College of Rheumatology recommendations for the use of disease-modifying antirheumatic drugs and biologic agents in the treatment of rheumatoid arthritis. Arthritis Care Res (Hoboken). 2012;64(5):625-639. doi: 10.1002/acr.21641.
6. Yeaw J, Benner JS, Walt JG, Sian S, Smith DB. Comparing adherence and persistence across 6 chronic medication classes. J Manag Care Pharm. 2009;15(9):728-740.
7. Sun P, Lian J. Treatment adherence in newly diagnosed type 2 diabetes: patient characteristics and long-term impact of adherence on inpatient care utilization. Postgrad Med. 2016;128(4):338-45. doi: 10.1080/00325481.2016.1151326.
8. Ho SC, Chong HY, Chaiyakunapruk N, Tangiisuran B, Jacob SA. Clinical and economic impact of non-adherence to antidepressants in major depressive disorder: a systematic review. J Affect Disord. 2016;193:1-10. doi: 10.1016/j.jad.2015.12.029.
9. Chodick G, Moser SS, Goldshtein I. Non-adherence with bisphosphonates among patients with osteoporosis: impact on fracture risk and healthcare cost. Expert Rev Pharmacoecon Outcomes Res. 2016;16(3):359-70. doi: 10.1586/14737167.2016.1171145.
10. Thomas NP, Curkendall S, Farr AM, Yu E, Hurley D. The impact of persistence with therapy on inpatient admissions and emergency room visits in the US among patients with multiple sclerosis. J Med Econ. 2016;19(5):497-505. doi: 10.3111/13696998.2015.1134546. 
11. Vekeman F, Sasane M, Cheng WY, et al. Adherence to iron chelation therapy and associated healthcare resource utilization and costs in Medicaid patients with sickle cell disease and thalassemia. J Med Econ. 2016;19(3):292-303.  doi: 10.3111/13696998.2015.1117979.
12. Dilokthornsakul P, Thoopputra T, Patanaprateep O, Kongsakon R, Chaiyakunapruk N. Effects of medication adherence on hospitalizations and healthcare costs in patients with schizophrenia in Thailand. SAGE Open Med. 2016;4:2050312116637026. doi: 10.1177/2050312116637026.
13. Yang Z, Howard DH, Will J, Loustalot F, Ritchey M, Roy K. Association of antihypertensive medication adherence with healthcare use and Medicaid expenditures for acute cardiovascular events. Med Care. 2016;54(5):504-511. doi: 10.1097/MLR.0000000000000515.
14. Curtis JR, Baddley JW, Yang S, et al. Derivation and preliminary validation of an administrative claims-based algorithm for the effectiveness of medications for rheumatoid arthritis. Arthritis Res Ther. 2011;13(5):R155. doi: 10.1186/ar3471. 
15. Bonafede M, Johnson BH, Princic N, Shah N, Harrison DJ. Cost per patient-year in response using a claims-based algorithm for the 2 years following biologic initiation in patients with rheumatoid arthritis. J Med Econ. 2015;18(5):376-389. doi: 10.3111/13696998.2014.1001849.
16. Basu A, Rathouz PJ. Estimating marginal and incremental effects on health outcomes using flexible link and variance function models. Biostatistics. 2005;6(1):93-109.
17. Urquhart J. Pharmacoeconomic consequences of variable patient
compliance with prescribed drug regimens. Pharmacoeconomics. 1999;15(3):217-228.
18. Hughes DA, Bagust A, Haycox A, Walley T. The impact of non-compliance on the cost-effectiveness of pharmaceuticals: a review of the literature. Health Econ. 2001;10(7):601-615.
19. De Vera MA, Mailman J, Galo JS. Economics of non-adherence to biologic therapies in rheumatoid arthritis. Curr Rheumatol Rep. 2014;16(11):460. doi: 10.1007/s11926-014-0460-5.
20. Borah BJ, Huang X, Zarotsky V, Globe D. Trends in RA patients’ adherence to subcutaneous anti-TNF therapies and costs. Curr Med Res Opin. 2009;25(6):1365-1377. doi: 10.1185/03007990902896386.