• Center on Health Equity & Access
  • Clinical
  • Health Care Cost
  • Health Care Delivery
  • Insurance
  • Policy
  • Technology
  • Value-Based Care

Impact of Co-pay Assistance on Patient, Clinical, and Economic Outcomes

Publication
Article
The American Journal of Managed CareMay 2022
Volume 28
Issue 5

Limited evidence from a literature review suggests that co-pay assistance was associated with improved treatment persistence/adherence across various diseases, with indirect evidence suggesting improvements in clinical outcomes.

ABSTRACT

Objectives: Patient assistance programs (eg, co-pay assistance) may reduce patients’ out-of-pocket costs for prescription medicines, providing financial assistance to access medicines for reduced or no cost. A literature review to identify peer-reviewed articles on studies evaluating the impact of co-pay assistance on clinical, patient, and economic outcomes was conducted.

Study Design: A literature review was conducted by searching Embase and MEDLINE.

Methods: The population of interest was patients who had received co-pay assistance; the intervention was co-pay assistance; comparator was no co-pay assistance; and outcomes were treatment adherence, compliance, discontinuation, interruption, barriers to adherence, and specific therapeutic outcomes. Articles from the United States published between January 2015 and June 2021 were included.

Results: A total of 1249 initial articles were identified, of which 19 published articles representing 12 studies were included. Most studies were retrospective claims analyses (n = 10); there was also 1 randomized controlled trial and 1 prospective and observational study. One article assessed the association between co-pay assistance and patient-reported outcomes, 7 explored the relationship between co-pay assistance and clinical outcomes, and 6 assessed the impact of policy/program changes on co-pay assistance. Co-pay assistance was associated with improved treatment persistence/adherence across various diseases, with limited indirect evidence of this translating into clinical outcomes improvements. Lack of long-term outcomes and uncertainty around program sustainment from co-pay assistance programs are limitations.

Conclusions: Limited evidence suggests a potential link between co-pay assistance and clinical outcomes; future research addressing study design challenges in measuring the effects of co-pay assistance is needed.

Am J Manag Care. 2022;28(5):e189-e197. https://doi.org/10.37765/ajmc.2022.89151

_____

Takeaway Points

This literature review focuses on the impact of patient assistance programs and outcomes. The findings include the following:

  • Some evidence of a positive association between co-pay assistance and adherence, which may suggest an indirect association with outcomes.
  • Limited direct evidence on clinical outcomes, and thus further research is needed to understand the impact of co-pay assistance on clinical outcomes.

_____

Almost half the individuals in the United States are prescribed medicine and nearly a quarter take 3 or more medications.1 However, many people struggle to take their medications as prescribed. Across clinical conditions and populations, approximately half of patients do not adhere to their prescribed treatments.1,2 Such adherence is highly important to achieve the clinical benefits of prescription medicines. There are, however, many barriers to adherence,1,2 with cost being a key driver of adherence problems. A number of studies have demonstrated that patients are sensitive to even small adjustments in their out-of-pocket (OOP) costs, which can result in nonadherence.3,4 Patient OOP costs have increased over time and may affect an individual’s ability to access prescription medications.5,6

Patient assistance programs (PAPs) have been proposed as a potential method to reduce OOP costs. They provide financial assistance, enabling patients to access medications for reduced or no cost. PAPs are sponsored by different entities, including pharmaceutical manufacturers, governments, and other third parties,7,8 and may apply to both generic and branded drugs.8 Commonly, PAPs are operationalized by the use of prescription discount co-payment (co-pay) cards and coupons distributed by providers or directly to patients, which provide a discount on cost sharing. Alternatively, assistance may include access to free drugs and cash subsidies. Drug manufacturer co-pay cards/coupons have been criticized for potentially increasing total costs.9,10 In response to potential cost impacts of the use of co-pay cards, payers have recently offered programs discouraging their use via co-pay accumulator or maximizer initiatives. These discourage the use of coupons by disallowing the value of the coupon to count against a patient’s deductible allowance.

Beyond the immediate impact on patient OOP costs, co-pay assistance may affect treatment utilization, such as adherence, with downstream impacts on clinical, economic, or humanistic outcomes (Figure 111). Given the increased attention paid to co-pay assistance and patient OOP costs,12 a literature review was conducted to identify peer-reviewed, published literature evaluating the impact of co-pay assistance on the clinical outcomes of adherence to and persistence with treatment, in addition to economic and patient outcomes.

METHODS

This literature review was conducted to identify peer-reviewed articles on studies evaluating the impact of co-pay assistance on clinical, patient, and economic outcomes.

Eligibility Criteria

The literature search was guided by a Population, Intervention, Comparator, and Outcomes, or PICO, approach; the population was defined as patients who received co-pay assistance; the intervention and comparator were co-pay assistance and no co-pay assistance, respectively; and the outcomes were treatment adherence, compliance, discontinuation, interruption, barriers to adherence, and specific therapeutic outcomes.

Search results were limited to articles from the United States published between January 2015 and June 2021. Studies not primarily related to co-pay assistance for medications (eg, vouchers incentivizing smoking cessation, co-pay assistance for nondrug services) were excluded. Articles reporting only impacts of co-pay assistance on OOP costs were also excluded, as were those not providing sufficient detail on at least 1 specific outcome of interest. Given that the focus was on original research, review articles, editorials, letters, notes, commentaries, case studies, and non-English publications were excluded. As the investigation sought to assess the situation in the United States only, studies conducted outside the United States were excluded.

Data Sources and Search Strategy

The electronic literature databases Embase and MEDLINE were searched for papers and conference abstracts using ProQuest Dialog software. The search string used is shown in Figure 2.

Search terms were identified based on studies known prior to the review.13-17 Terms were reviewed for completeness, and the search string was modified until all previously known studies were captured. A review identifying conceptual approaches to adherence research was used to inform the search terms employed describing medication-taking behavior.11

Records identified via other sources included those found through reviewing references and email notifications from journals of new publications.

Study Selection

Article titles and abstracts were reviewed by 2 independent reviewers (K.D.P. and W.B.W.), with discussions as necessary during the review process to ensure comparable understanding of selection criteria. Disagreements were resolved by consensus between the 2 reviewers, with a third reviewer being the decision maker if no resolution could be achieved. Articles were classified as relevant, not relevant but related, or not relevant. Those considered not relevant but related (eg, commentary papers) were considered for contextual discussion points. Full-text review was conducted for abstracts that met the initial screening criteria.

Data Collection, Data Items, Summary Measures, and Synthesis of Results

Data for key study design, patient population, demographics, end points, and outcomes of interest were extracted into a spreadsheet, allowing the authors to summarize the key themes.

RESULTS

Study Selection

The search yielded a total of 1249 results, of which 19 published manuscripts/conference abstracts, representing 12 studies, were included in the final sample (Figure 2); 12 articles were relevant and are covered here, and others informed the discussion. A large proportion of studies were excluded because they were not based in the United States (n = 351) or because the outcome (ie, no clinical/humanistic outcome, or economic outcome consists of OOP costs only) or the population (ie, non–medication-related cost sharing) was not of interest (n = 358 and n = 443, respectively). There was 1 conflicting review, which was resolved after discussion and additional scrutiny of the article by both reviewers concurrently.

Study Characteristics

A summary of studies in the final sample is shown in Table 1 [part A, part B, and part C].13-28 The majority of studies were retrospective claims analyses (n = 10/12; 83%), 1 study was a randomized controlled trial, and 1 was prospective and observational. The association between co-pay assistance and the following outcomes was reported: patient-reported outcomes (n = 1), clinical outcomes (n = 7, including 5 measuring treatment persistence/discontinuation or adherence), co-pay assistance policy (n = 6), and co-pay accumulators and maximizers (n = 1). Table 214,20,23,29,30 provides a summary of the studies identified that addressed direct or indirect measurements of co-pay assistance on outcomes.

Impact of Co-pay Assistance on Patient-Reported Outcomes

De Souza et al18 examined the effects of patient-level interventions, including co-pay assistance, patient navigators, social workers, financial counselors, support groups, and transportation vouchers, on financial toxicity, defined as an adverse event of cancer treatments. This prospective observational study found that individuals with cancer who received co-pay assistance for their treatments had a significant improvement in financial toxicity over time, with 89% of patients experiencing an improvement over 3 months. However, improvements in financial toxicity did not clearly translate into significant improvements in health-related quality of life (HRQOL), probably owing to the multifaceted nature of HRQOL (ie, effects of disease outweighing improvement in financial toxicity).

Impact of Co-pay Assistance on Clinical Outcomes

The impact of co-pay assistance on clinical outcomes was assessed in 7 articles. The impact on objective clinical outcomes is likely to be mediated by intermediate outcomes, such as adherence to or persistence with treatment (Figure 111). Five studies related to the ARTEMIS trial (NCT02406677) were identified that measured associations between co-pay assistance and intermediate outcomes, such as treatment adherence and persistence, as well as hard clinical outcomes, such as major adverse cardiovascular events (MACE). In ARTEMIS, patients using P2Y12 inhibitors were randomized to receive vouchers (with a median value of $137 for a 30-day medication supply) or usual care with no vouchers. The use of vouchers to offset the cost of medication co-payments for P2Y12 inhibitors significantly increased patient-reported medication persistence at 1 year compared with the control group (87.0% vs 83.8%; P < .001; observed increase, 3.3%; 95% CI, 1.0%-5.5%); however, voucher use did not significantly decrease MACE. A post hoc analysis of ARTEMIS found that the effect of vouchers on medication persistence and MACE was dependent on the likelihood of voucher use, with the greatest improvements in medication persistence and reduction in MACE seen in patients most likely to use them.20 A secondary analysis from this study demonstrated that nonpersistence was associated with a higher risk of MACE at 12 months than persistence.21

One study examined the impact of coupons on incident statin utilization and costs through a retrospective claims analysis.13 The study found that coupon users had $5 lower average monthly OOP costs at 1 year than coupon nonusers; however, this difference was not statistically significant. Statin utilization and switching were similar for coupon users and coupon nonusers, but coupon users were 11.1% less likely to terminate statin therapy than coupon nonusers at 4 years. Additionally, this study identified an association between the level of coupon use and statin utilization, with high levels of coupon use resulting in greater statin utilization and a lower likelihood of treatment switching and discontinuation.

A retrospective cohort study of oral anaplastic lymphoma kinase inhibitors (ALKis) in patients with ALK-positive non–small cell lung cancer (NSCLC) found associations between co-pay assistance for incident ALKi users and outcomes, such as the proportion of patients with a prescription abandonment, time to prescription pickup, and treatment persistence.14 Co-pay assistance decreased OOP costs for a patient’s first ALKi prescription by an average of $1930. In an adjusted analysis, patients with co-pay assistance were more likely to pick up their ALKi prescriptions than those without it (88.2% lower risk of prescription abandonment of their first approved claim), picked up their prescriptions approximately 3 weeks sooner than those without co-pay assistance (2.6 days vs 25.7 days), and were more likely to remain on treatment, with a 24.3% lower risk of discontinuing therapy and an increase in median time to discontinuation of 43 days compared with those without co-pay assistance.

Finally, a retrospective claims analysis for immunosuppressant medications suggested that co-pay assistance was positively associated with adherence in transplant recipients (odds ratio, 1.33; P = .09).24

Impact of Co-pay Assistance Policies

The impact of co-pay assistance policies on clinical outcomes was assessed in 6 articles,15,17,25-28 with a focus on Massachusetts, which was one of the first states to prohibit manufacturer co-pay assistance under its antikickback statute. In July 2012, the antikickback statute was amended to allow temporary manufacturer co-pay assistance for medicines without a generic equivalent. A study examining the coupon ban found that coupons were associated with increased use of branded drugs facing entry of generic therapeutic alternatives,25 whereas 2 studies assessing the impact of the policy amendment on drug utilization found that it was associated with increased use of autoimmune biologics17 and branded drugs with close generic therapeutic substitutes.26 A third study assessed medical costs and health care resource utilization associated with the policy across 6 disease areas and found that in the 5-year period following the amendment, the rise in availability of co-pay assistance programs was associated with a significant reduction in the annual number of hospitalizations and annual medical costs in 4 of the 6 disease areas compared with before the antikickback statute amendment.15 In contrast to some of the findings in Massachusetts, Rome et al27 found no significant changes in generic substitution in the first year after the California ban on co-payment coupons for brand-name drugs with generic competition, and Brouwer et al28 found that PAP use within a regional integrated health system had no effect on drug demand.

Impact of Co-pay Accumulators and Maximizers

Co-pay accumulators and maximizers have been relatively new tools for pharmacy benefit managers. These discourage the use of co-pay cards by restricting the amount on the card that counts against a patient’s deductible and OOP cost maximum. Only 1 study on co-pay accumulators was identified, which found associated reductions in adherence and persistence.16 Sherman et al16 found that the implementation of a co-pay accumulator adjustment program (CAAP) was associated with significant reductions in autoimmune specialty drug adherence and that the risk of treatment discontinuation was significantly higher following CAAP implementation than before it.

DISCUSSION

The findings of this literature review highlight the limited evidence and the heterogeneous nature of studies in this area. Results of the studies identified suggest that co-pay assistance may have a positive impact on persistence with and adherence to treatment across a range of diseases, with some indirect evidence indicating improvements in clinical outcomes. Although previous reviews on this topic have examined adherence or PAPs as part of a suite of services,31,32 our review is unique in that we focused specifically on the effect of PAPs alone on clinical, economic, or humanistic outcomes, and the dearth of evidence indicates that future research is warranted, especially in diverse patient populations and settings.

Impact of Co-pay Assistance on Clinical Outcomes

Adherence to long-term treatment is a complex behavior influenced by a variety of factors and associated with improved outcomes.33 Although the study on ALKis in NSCLC lacked a direct measurement of clinical outcomes,15 subsequent studies revealed a potential indirect association (Table 214,20,23,29,30) between the impact of co-pay assistance and progression-free survival and overall survival. A delay of greater than 3 weeks in filling prescriptions for patients without co-pay assistance (compared with those with co-pay assistance) may be associated with shorter overall survival and a 2-fold increase in the risk of death.30 Furthermore, poor treatment persistence in those without co-pay assistance has been shown to have a clinical bearing. The authors of one retrospective database study reported that each additional month on ALKi therapy was associated with a 13% lower risk of disease progression or death and a 32% lower risk of death.29 Applying these reduced risks per month to the difference of 43 days in treatment duration observed by Seetasith et al,14 co-pay assistance for ALKi therapy may be associated with an 18% lower risk of disease progression or death and a 43% lower risk of death.

The lack of impact on clinical outcomes in the ARTEMIS study highlights the challenges of studying the effects of co-pay assistance. A number of potential reasons were noted for the lack of alignment between persistence with treatment and clinical outcomes in this study.34 The introduction of vouchers led to only limited differences in guideline-preferred therapies between groups, and relatively low adherence levels were detected based on prescription fill data; both may have contributed to challenges in detecting differences in clinical outcomes, especially in the context of the sample sizes and adherence levels. Additionally, in this study, 27% of patients randomized to the intervention group failed to use their vouchers. Taken together, co-pay assistance was shown to be a factor in improved persistence with P2Y12 therapy. However, a multifactorial approach including co-pay assistance may be required to improve clinical outcomes in this disease area.

Limitations and Implications for Future Research

This review was limited to the past 6 years, given the rapid changes in the policy environment, and focused specifically on co-pay assistance programs and policies, which form part of a larger body of literature examining other related aspects. Several studies have demonstrated that higher OOP costs are associated with prescription abandonment, lower adherence, and/or higher rates of discontinuation.5,35,36 Furthermore, studies have examined different mechanisms to reduce or eliminate co-payments, but the effectiveness (and practicality of implementing these) have varied. For example, Volpp et al37 found no effect of eliminating antihypertensive co-pays via post hoc rebates on blood pressure control. In contrast, studies examining the impact of patient support programs38 or financial navigators,39 both of which may include co-pay assistance, have found associations with improved quality of life, adherence/persistence, and costs. Given the multifaceted nature of these interventions, however, the extent to which these effects are attributable to co-pay assistance is unclear. Nonetheless, co-pay assistance programs represent a relatively practical mechanism to achieve lower OOP costs, and thus are the focus of this review.

The small number of studies identified is a limitation, indicating the need for future research, particularly on longer-term outcomes. However, a number of obstacles to collection and analysis of these data remain, including access and study design challenges, such as selection of the appropriate control group. For instance, despite ARTEMIS being a prospective randomized controlled trial, there were challenges in evaluating the impact of co-pay assistance because of difficulties selecting the appropriate study population (ie, individuals likely to use co-pay assistance in the real world). Use of secondary data sources may be a more practical approach to evaluate effects of co-pay assistance, but this is not without challenges. These include (1) few data sources, mostly limited to open claims, to identify co-pay assistance while also making it challenging to measure adherence/persistence due to the lack of enrollment data; and (2) claims data linked to electronic health records may be required, although sufficient sample sizes may be challenging. Lastly, identification of the appropriate counterfactual cohort may be challenging (ie, individuals who may be eligible for co-pay assistance but did not receive it) because data for variables that might indicate eligibility, such as income, are often unavailable. Future work on enhancing and linking data sets is required to capture the relevant data points needed to advance research in this area.

Lastly, additional research may support ongoing health policy discussions around access to co-pay assistance programs. Studies have found conflicting results on the impact of generic utilization,26,27 and further research is needed to inform future policies. Additionally, several policy changes have been made in response to the co-pay accumulators, with some states banning the use of such programs. Evolving policy includes the HHS Notice of Benefit and Payment Parameters for 2021,40 which allows health plans flexibility in whether to count manufacturer co-pay assistance toward deductibles or OOP maximums—a reversal of the previous year’s policy of requiring plans to count co-pay assistance toward deductibles and OOP maximums. Well-designed research would inform these evolving policies.

CONCLUSIONS

Although existing literature on the relationship between co-pay assistance and outcomes is limited, there is a signal that co-pay assistance can have a positive impact on treatment persistence and adherence across clinical conditions, with indirect evidence suggesting clinical outcome benefits. These findings should be considered in context, as studies have yet to evaluate long-term outcomes and program sustainment. Additional data measuring the direct link between co-pay assistance and clinical outcomes are needed. Future research should address the numerous study design challenges involved in measuring the effects of co-pay assistance.

Acknowledgments

The authors thank Rebecca Hornby, PhD, of Oxford PharmaGenesis, Oxford, UK, for providing medical writing support, which was sponsored by Genentech, Inc, in accordance with Good Publication Practice guidelines.

Author Affiliations: Genentech, Inc (KDP, WBW), South San Francisco, CA; Department of Population Health Sciences, Duke University School of Medicine (LLZ), Durham, NC; Center of Innovation to Accelerate Discovery and Practice Transformation, Durham Veterans Affairs Health Care System (LLZ), Durham, NC.

Source of Funding: This study was funded by Genentech, Inc.

Author Disclosures: Drs Parekh and Wong are employed by Genentech (a member of the Roche Group) and own Roche stock; Genentech offers co-pay assistance programs. Dr Zullig is employed by the Department of Veterans Affairs and Duke University; has received grants from PhRMA Foundation and Proteus Digital Health; and has received honoraria from Novartis and Pfizer.

Authorship Information: Concept and design (KDP, WBW, LLZ); acquisition of data (WBW); analysis and interpretation of data (KDP, WBW); drafting of the manuscript (KDP, WBW, LLZ); critical revision of the manuscript for important intellectual content (KDP, WBW, LLZ); and administrative, technical, or logistic support (WBW).

Address Correspondence to: Krupa D. Parekh, PharmD, MPH, 105 S Hill Rd, Colonia, NJ 07067. Email: krupa.parekh@roche.com.

REFERENCES

1. Brown MT, Bussell JK. Medication adherence: WHO cares? Mayo Clin Proc. 2011;86(4):304-314. doi:10.4065/mcp.2010.0575

2. Gast A, Mathes T. Medication adherence influencing factors—an (updated) overview of systematic reviews. Syst Rev. 2019;8(1):112. doi:10.1186/s13643-019-1014-8

3. Dusetzina SB, Winn AN, Abel GA, Huskamp HA, Keating NL. Cost sharing and adherence to tyrosine kinase inhibitors for patients with chronic myeloid leukemia. J Clin Oncol. 2014;32(4):306-311. doi:10.1200/JCO.2013.52.9123

4. Pham HH, Alexander GC, O’Malley AS. Physician consideration of patients’ out-of-pocket costs in making common clinical decisions. Arch Intern Med. 2007;167(7):663-668. doi:10.1001/archinte.167.7.663

5. Doshi JA, Li P, Huo H, Pettit AR, Armstrong KA. Association of patient out-of-pocket costs with prescription abandonment and delay in fills of novel oral anticancer agents. J Clin Oncol. 2018;36(5):476-482. doi:10.1200/JCO.2017.74.5091

6. Dusetzina SB, Huskamp HA, Keating NL. Specialty drug pricing and out-of-pocket spending on orally administered anticancer drugs in Medicare part D, 2010 to 2019. JAMA. 2019;321(20):2025-2027. doi:10.1001/jama.2019.4492

7. Choudhry NK, Lee JL, Agnew-Blais J, Corcoran C, Shrank WH. Drug company-sponsored patient assistance programs: a viable safety net? Health Aff (Millwood). 2009;28(3):827-834. doi:10.1377/hlthaff.28.3.827

8. Sen AP, Kang SY, Rashidi E, Ganguli D, Anderson G, Alexander GC. Characteristics of copayment offsets for prescription drugs in the United States. JAMA Intern Med. 2021;181(6):758-764. doi:10.1001/jamainternmed.2021.0733

9. Dafny LS, Ody CJ, Schmitt MA. Undermining value-based purchasing—lessons from the pharmaceutical industry. N Engl J Med. 2016;375(21):2013-2015. doi:10.1056/NEJMp1607378

10. Howard DH. Drug companies’ patient-assistance programs – helping patients or profits? N Engl J Med. 2014;371(2):97-99. doi:10.1056/NEJMp1401658

11. Vrijens B, De Geest S, Hughes DA, et al; ABC Project Team. A new taxonomy for describing and defining adherence to medications. Br J Clin Pharmacol. 2012;73(5):691-705. doi:10.1111/j.1365-2125.2012.04167.x

12. Bosworth HB, Granger BB, Mendys P, et al. Medication adherence: a call for action. Am Heart J. 2011;162(3):412-424. doi:10.1016/j.ahj.2011.06.007

13. Daubresse M, Andersen M, Riggs KR, Alexander GC. Effect of prescription drug coupons on statin utilization and expenditures: a retrospective cohort study. Pharmacotherapy. 2017;37(1):12-24. doi:10.1002/phar.1802

14. Seetasith A, Wong W, Tse J, Burudpakdee C. The impact of copay assistance on patient out-of-pocket costs and treatment rates with ALK inhibitors. J Med Econ. 2019;22(5):414-420. doi:10.1080/13696998.2019.1580200

15. Seetasith A, Wu N, Wong W. The impact of manufacturers’ copay card ban lift on disease-related hospitalizations and medical costs in Massachusetts in six disease areas. J Manag Care Spec Pharm. 2019;25(suppl 10-a):S96. doi:10.18553/jmcp.2019.25.10-a.s1

16. Sherman BW, Epstein AJ, Meissner B, Mittal M. Impact of a co-pay accumulator adjustment program on specialty drug adherence. Am J Manag Care. 2019;25(7):335-340.

17. Sullivan P, Skup M, Mittal M, Terasawa E, Macaulay D, Chao J. The impact of pharmaceutical manufacturer copay cards on patient access to biologics. J Manag Care Spec Pharm. 2016;22(suppl 4-a):S124. doi:10.18553/jmcp.2016.22.4.S1

18. De Souza JA, Proussaloglou E, Nicholson L, Wang Y. Evaluating financial toxicity (FT) interventions. J Clin Oncol. 2017;35(suppl 15):e21673. doi:10.1200/JCO.2017.35.15_suppl.e21673

19. Doll JA, Kaltenbach LA, Anstrom KJ, et al. Impact of a copayment reduction intervention on medication persistence and cardiovascular events in hospitals with and without prior medication financial assistance programs. J Am Heart Assoc. 2020;9(8):e014975. doi:10.1161/JAHA.119.014975

20. Fanaroff AC, Peterson ED, Kaltenbach LA, et al. Copayment reduction voucher utilization and associations with medication persistence and clinical outcomes: findings from the ARTEMIS trial. Circ Cardiovasc Qual Outcomes. 2020;13(5):e006182. doi:10.1161/CIRCOUTCOMES.119.006182

21. Fanaroff AC, Peterson ED, Kaltenbach LA, et al. Agreement and accuracy of medication persistence identified by patient self-report vs pharmacy fill: a secondary analysis of the cluster randomized ARTEMIS trial. JAMA Cardiol. 2020;5(5):532-539. doi:10.1001/jamacardio.2020.0125

22. Fanaroff AC, Peterson ED, Kaltenbach LA, et al. Association of a P2Y12 inhibitor copayment reduction intervention with persistence and adherence with other secondary prevention medications: a post hoc analysis of the ARTEMIS cluster-randomized clinical trial. JAMA Cardiol. 2020;5(1):38-46. doi:10.1001/jamacardio.2019.4408

23. Wang TY, Kaltenbach LA, Cannon CP, et al. Effect of medication co-payment vouchers on P2Y12 inhibitor use and major adverse cardiovascular events among patients with myocardial infarction: the ARTEMIS randomized clinical trial. JAMA. 2019;321(1):44-55. doi:10.1001/jama.2018.19791

24. Cherikh WS, Wilk AR, Maghirang J, et al. Immunosuppressant adherence in adult heart transplant recipients: an analysis of pharmacy claims and organ procurement and transplantation network data. Am J Transplant. 2019;19(suppl 3):1146-1147.

25. Dafny L, Ody C, Schmitt M. When discounts raise costs: the effect of copay coupons on generic utilization. Am Econ J Econ Policy. 2017;9(2):91-123. doi:10.1257/pol.20150588

26. Huang Y, Sadownik S, Nasuti L, Mattingly TJ II, Auerbach D, Seltz D. Utilization and spending impact of prescription drug coupons in Massachusetts. Health Serv Res. 2020;55(S1):10. doi:10.1111/1475-6773.13334

27. Rome BN, Gagne JJ, Kesselheim AS. Association of California’s prescription drug coupon ban with generic drug use. JAMA. 2021;325(23):2399-2402. doi:10.1001/jama.2021.6568

28. Brouwer E, Yeung K, Barthold D, Hansen R. Characterizing patient assistance program use and patient responsiveness to specialty drug price for multiple sclerosis in a mid-size integrated health system. J Manag Care Spec Pharm. 2021;27(6):732-742. doi:10.18553/jmcp.2021.27.6.732

29. Sheinson D, Wu N, Seetasith A, Wong W. Does longer treatment persistence in the real-world translate to better clinical outcomes? a case study of ALK+ non-small cell lung cancer (NSCLC). Value Health. 2020;23(suppl 1):S26. doi:10.1016/j.jval.2020.04.1530

30. Sheinson D, Wong WB, Wu N, Mansfield AS. Impact of delaying initiation of anaplastic lymphoma kinase inhibitor treatment on survival in patients with advanced non-small-cell lung cancer. Lung Cancer. 2020;143:86-92. doi:10.1016/j.lungcan.2020.03.005

31. Felder TM, Palmer NR, Lal LS, Mullen PD. What is the evidence for pharmaceutical patient assistance programs? a systematic review. J Health Care Poor Underserved. 2011;22(1):24-49. doi:10.1353/hpu.2011.0003

32. Hung A, Blalock DV, Miller J, et al. Impact of financial medication assistance on medication adherence: a systematic review. J Manag Care Spec Pharm. 2021;27(7):924-935. doi:10.18553/jmcp.2021.27.7.924

33. Kronish IM, Ye S. Adherence to cardiovascular medications: lessons learned and future directions. Prog Cardiovasc Dis. 2013;55(6):590-600. doi:10.1016/j.pcad.2013.02.001

34. Jackevicius CA, Ko DT. Medication co-payment vouchers, adherence with antiplatelet therapy, and adverse cardiovascular events after myocardial infarction. JAMA. 2019;321(1):37-39. doi:10.1001/jama.2018.20396

35. Rome BN, Gagne JJ, Avorn J, Kesselheim AS. Non-warfarin oral anticoagulant copayments and adherence in atrial fibrillation: a population-based cohort study. Am Heart J. 2021;233:109-121. doi:10.1016/j.ahj.2020.12.010

36. Chernew ME, Shah MR, Wegh A, et al. Impact of decreasing copayments on medication adherence within a disease management environment. Health Aff (Millwood). 2008;27(1):103-112. doi:10.1377/hlthaff.27.1.103

37. Volpp KG, Troxel AB, Long JA, et al. A randomized controlled trial of co-payment elimination: the CHORD trial. Am J Manag Care. 2015;21(8):e455-464.

38. Rubin DT, Mittal M, Davis M, Johnson S, Chao J, Skup M. Impact of a patient support program on patient adherence to adalimumab and direct medical costs in Crohn’s disease, ulcerative colitis, rheumatoid arthritis, psoriasis, psoriatic arthritis, and ankylosing spondylitis. J Manag Care Spec Pharm. 2017;23(8):859-867. doi:10.18553/jmcp.2017.16272

39. Knight TG, Aguiar M, Robinson M, et al. Financial toxicity intervention improves quality of life in hematologic malignancy patients. Blood. 2020;136(suppl 1):21. doi:10.1182/blood-2020-136578

40. CMS, HHS. Patient Protection and Affordable Care Act; HHS notice of benefit and payment parameters for 2021; notice requirement for non-federal governmental plans. Fed Regist. 2020;85(25):7088-7159.

Related Videos
James Chambers, PhD
1 KOL is featured in this series.
1 KOL is featured in this series.
Justin Oldham, MD, MS, an expert on IPF
Mei Wei, MD, an oncologist specializing in breast cancer at Huntsman Cancer Institute at the University of Utah.
Dr Bonnie Qin
Screenshot of an interview with Ruben Mesa, MD
Screenshot of Adam Colborn, JD during an interview
Justin Oldham, MD, MS, an expert on IPF
Related Content
© 2024 MJH Life Sciences
AJMC®
All rights reserved.