- Email: [email protected]
Drug discount cards in an era of higher prescription drug prices: A retrospective population-based study
SCIENCE AND PRACTICE Journal of the American Pharmacists Association 59 (2019) 804e808
Contents lists available at ScienceDirect
Journal of the American Pharmacists Association journal homepage: www.japha.org
RESEARCH
Drug discount cards in an era of higher prescription drug prices: A retrospective population-based study Satish Munigala, Margaret Brandon, Zackary D. Goff, Richard Sagall, Paul J. Hauptman* a r t i c l e i n f o
a b s t r a c t
Article history: Received 14 December 2018 Accepted 21 May 2019 Available online 14 August 2019
Objectives: Drug discount programs have emerged as a potential option for patients seeking greater accessibility and affordability. However, there is limited knowledge regarding program utilization and cost savings. The objective of this study was to evaluate medication prescriptions with drug discount card usage and estimate cost savings. Design: Retrospective study. Setting and participants: Using population-based prescription data, the study included patients who filled prescriptions from January 2009 to December 2016 nationwide using NeedyMeds. org drug discount cards. Outcome measures: We determined the frequency of drug discount card prescriptions (across pharmacy types, pharmacy location, and prescriber specialty), estimated cost savings using the drug discount card (average per drug discount card and total program dollars saved) and evaluated the top prescription drugs by frequency. Results: A total of 4,638,581 prescriptions with discount cards were identified (79.8% at national, 6.3% at regional, and 12.9% at local pharmacies). Most were filled at urban locations (urban clusters, 88.6%; urbanized areas, 8.4%) and in ZIP codes with lower median household incomes (62.7%). Overall, 3.62 million prescriptions (78.0% of the total) were associated with discounts, resulting in a total savings of $199,183,112 (median cost savings, $17.80 [47.8%] per prescription). Opiates were the most common class of drugs for which discount cards were used. Conclusion: The use of a drug discount program over 8 years resulted in total savings of nearly $200 million (approximately $18 per prescription) compared with the original cost. However, although patients might accrue financial benefit, there is still a lack of price transparency. Additional research is needed to better understand the impact of these programs and to evaluate ways to improve medication access at a reasonable cost to patients. © 2019 American Pharmacists Association®. Published by Elsevier Inc. All rights reserved.
Background The cost to patients for many branded and generic drugs have been increasing, contributing to higher overall out-of-pocket health care expenses.1,2 These additional costs can be a barrier to access for uninsured individuals,3-6 discourage adherence to treatment regimens,7 and contribute to prescription abandonment.8 In addition, lack of transparency in drug pricing can Disclosure: Richard Sagall currently serves as CEO for NeedyMeds.org and is the primary source for data acquisition. The remaining authors declare no conflicts of interest or financial interests in any product or service mentioned in this article. * Correspondence: Paul J. Hauptman, MD, University of Tennessee Graduate School of Medicine, 1924 Alcoa Highway, Knoxville TN 37920. E-mail address: [email protected] (P.J. Hauptman).
adversely affect adherence by patients whose options might be restricted to a retail pharmacy within a limited geographic area.9 The Medicare Prescription Drug, Improvement, and Modernization Act (MMA) of 2003 created a drug discount card program with a prescription benefit to support medication expenditures for eligible seniors.9 Pharmaceutical drug company assistance programs formed a viable alternative to the Medicare prescription benefit for patients who do not qualify for low-income subsidies. Recently, freestanding drug discount programs, such as NeedyMeds.org, GoodRx.com, and BlinkHealth.com,10-12 have been established and in some cases advertised in media; they can offer an additional option for patients. However, there is limited knowledge regarding program usage and role in medication dissemination in a general patient population.
https://doi.org/10.1016/j.japh.2019.05.021 1544-3191/© 2019 American Pharmacists Association®. Published by Elsevier Inc. All rights reserved.
SCIENCE AND PRACTICE Drug discount cards for prescription medications
Definitions Key Points Background: Drug discount programs have proliferated in recent years but a critical analysis of their impact has not been performed. Data were derived from NeedyMeds, a nonprofit organization that provides online access to a free drug discount card. Findings: In a population-based retrospective review of prescription data, a drug discount program resulted in total savings of nearly $200 million on prescription medications (approximately $18 in savings per prescription) compared with the original cost charged by pharmacies. Drug discount programs can help patients save money, but the programs exist in part because of the rising costs of medications and the absence of price transparency.
Using data from one drug discount program, we evaluated the frequency of and trends in medication prescriptions with discount card usage (by pharmacy type and location, and by prescriber specialty) and then estimated the cost savings (average per drug discount card and the total program dollar amount saved) for all prescription drugs.
Methods This retrospective study investigated prescriptions for which drug discount cards were used from January 2009 to December 2016. Data were obtained from NeedyMeds, a nationwide nonprofit organization established in 1997 with a mission to serve as a resource for information on pharmaceutical patient assistance programs. Since 2009, NeedyMeds has provided on-line access to a free drug discount card that offers a savings of up to 80% at more than 65,000 pharmacies nationwide10 and uses electronic capture of records for subsequent analysis. Data abstracted from these files include prescription details (including drug name, days’ supply, and patient cost savings), pharmacy information (e.g., name, address) and prescriber information (specialty). NeedyMeds data had prescription claims from all 50 states in the United States and for all data years during the study period. The NeedyMeds.org drug discount card can be used by anyone regardless of age, residency status, income, insurance status, and diagnosis, but it cannot be combined with insurance or any state, federal, or pharmaceutical company drug discount programs. If a drug discount card was used multiple times, we treated each record as an independent observation. Cost savings for each prescription were the total savings for a customer (patient) using the drug discount card compared to the actual price offered by the pharmacy. Prescriptions for which the drug discount card did not yield any savings (i.e., prescription price using the drug discount card was the same as the lowest price offered by the pharmacy) were excluded for the cost-savings analysis (n ¼ 1,018,781).
We classified pharmacies as national (if located in 5 states with 25 or more stores, at least 1 state not contiguous), regional (if located in 3 states with 10 or more stores), specialty (individual pharmacies affiliated with hospitals), and local businesses (located in just 1 or 2 states with fewer than 10 stores). We categorized the location of pharmacies as urban (“urbanized areas” of 50,000 or more people; “urban clusters” of at least 2500 and less than 50,000 people) or rural (encompassing all population, housing, and territory not included within an urban area) locations by mapping pharmacy ZIP code to U.S. Census Bureau ZIP code tabulation data.13,14 We also designated the ZIP codes to above- and below-median annual income level using 2015 national estimates from the 2011-2015 American Community Survey.15
Statistical analysis Categorical variables were reported as frequencies and proportions, and continuous variables were reported as median or interquartile range (IQR) where appropriate. Total number of prescriptions with a drug discount card was first identified. Prescriptions by pharmacy type, location, prescriber specialty, and median household income level were assessed. Prescriptions for which the drug discount card provided a lower price than the lowest price offered by the store were evaluated by the number of transactions, median days’ supply, and calculated cost savings (median cost saved by amount per prescription, percentage saved compared to original cost and total program dollars saved) for the entire study period and for each year (2009-2016). We evaluated the top 20 prescription drugs by frequency over the duration of the program. We also evaluated the drug discount card utilization by state, for the entire study period, and then compared the percentage of uninsured people by each state16 regarding the drugdiscount card utilization per 100,000 U.S. population by state,17 for the year 2013 (midpoint for the period under study). Data were analyzed using Statistical Analysis Software (SAS) version 9.3 (SAS Institute, Cary, NC). The Saint Louis University Institutional Review Board determined that this study did not constitute human subjects research and therefore required no additional review.
Results Prescription characteristics A total of 4,638,581 prescriptions with drug discount cards were identified during the study period. The majority were filled at national (79.8%) as opposed to regional (6.3%), local (12.9%), or hospital-affiliated (1.1%) pharmacies. Most prescriptions were filled at pharmacies located in urban clusters or urbanized areas (88.6% and 8.4%, respectively) with the remaining 3.0% from rural areas. Sixty-two percent of the prescriptions were obtained from pharmacies located in the ZIP codes with lower median household income. Prescription characteristic by prescriber specialty are illustrated in Table 1; more than 50% were in internal medicine (including specialties), family practice, or general medicine. 805
SCIENCE AND PRACTICE S. Munigala et al. / Journal of the American Pharmacists Association 59 (2019) 804e808
prescription per patient, representing a reduction of 47.8% (IQR, 21.3%-69.0%) over prescription charges at the pharmacy level.
Table 1 Characteristics of drug discount prescriptions by pharmacy type and prescriber specialty Characteristics
All drugs (N ¼ 4,638,581)
Pharmacy typea National Regional Local Affiliated with hospital Pharmacy location by ZIP code areab Rural Urban clusters Urbanized areas Median household income by pharmacy ZIP codec (per 2015 census estimate) Lower (<$56,516) Higher ($56,516) Prescriber specialty Internal medicine and subspecialties Surgery Neurology, psychiatry Gynecology Family medicine, general practice Nurse, nurse practitioner Pediatrics Emergency Medicine Other Unknown Missing percentages: (n ¼ 30,856);
a
0.2% (n ¼ 9066);
Top prescription drugs by frequency
n
%
3,694,186 289,871 595,731 49,727
79.8 6.3 12.9 1.1
137,952 4,088,566 389,952
3.0 88.6 8.4
The most prescribed medications for which drug discount cards were used are shown in Table 3. The opiate class (oxycodone, hydrocodone) was the most common; other classes of drug with 2 or more medications in the top 20 included benzodiazepines (n ¼ 3), anti-hypertensives (n ¼ 3), antidepressants (n ¼ 2), and HMG-CoA reductase inhibitors (n ¼ 2). Five of these drugs were also on the list of top 20 drugs for which no cost savings accrued. Drug discount card utilization by state
b
2,890,150 1,717,575
62.7 37.3
940,521 68,018 362,857 97,387 1,493,840 479,774 67,555 290,885 504,535 333,209
20.3 1.5 7.8 2.1 32.2 10.3 1.5 6.3 10.9 7.2
For the entire study period, drug discount card utilization was not concentrated in a specific geographical region or within few states. Texas, California, Florida, Indiana, and Ohio showed the highest drug discount card utilization rates (Supplemental Figure 1A). Although a higher percentage of uninsured people resided in Texas, Nevada, Florida, Georgia, and Alaska for the year 2013, the percentage of drug discount card utilization (per 100,000 population) was highest in New York, North Carolina, South Carolina, and Ohio (Supplemental Figures 1B and 1C). Discussion
0.5% (n ¼ 22,111); c0.7%
Prescription trends by year The number of prescriptions with drug discount cards increased from 28,878 in 2009 to 756,837 in 2016 with the highest number of prescriptions in the years 2014 and 2015. Cost savings using the drug discount card Overall, 3,619,800 prescriptions (78.0%) were associated with savings; the remainder offered no savings to patients (Table 2). Total savings for the study period was $199,183,112, with a median of $17.80 (IQR, 6.50-45.40) cost savings per
In this study, we evaluated the frequency, temporal trends and cost savings of drug discount card prescriptions using a large database from a free-standing, on-line drug discount coupon program. We found that the number of prescriptions has increased significantly, resulted in total savings of nearly $200 million on prescription medications, with an average savings of $18 at 48% savings per prescription, compared with the original cost obtained from the pharmacies (where savings were noted). Increasing prescription costs represent a significant economic burden on patients,8,18 and it can lead to decrease in adherence.7 Pricing at the consumer (patient) level has remained obtuse, but increasing attention has been paid to the complex business arrangements between drug manufacturers, pharmacy benefit managers (PBMs), insurers, wholesalers, and
Table 2 Total savings by year for transactions involving cost savings Year
2009 2010 2011 2012 2013 2014 2015 2016 Overall
All transactions
Number of transactions in which savings were noteda
n
n
%
28,878 182,512 415,145 622,298 814,866 909,575 908,470 756,837 4,638,581
18,559 133,492 295,094 453,999 620,658 733,614 727,572 636,812 3,619,800
64.3 73.1 71.1 73.0 76.2 80.7 80.1 84.1 78.0
Median patient cost savings, $ (IQR)
Median cost savings, % (IQR)
Total savings ($)
Median days’ supply (IQR)
15.4 15.5 16.3 16.3 16.7 17.7 20.6 19.0 17.8
43.7 45.0 40.6 46.5 48.6 51.6 48.5 46.9 47.7
600,712 4,326,755 10,024,670 16,729,658 25,492,066 37,878,468 58,012,341 46,118,440 199,183,112
30 30 30 30 30 30 30 30 30
(6.6-34.2) (6.7-35.2) (7.0-35.5) (7.2-37.1) (6.7-39.5) (6.9-44.4) (6.5-61.3) (4.6-57.5) (6.5-45.4)
(18.3-64.3) (19.2-64.6) (18.8-60.1) (24.2-66.2) (25.4-69.2) (26.8-70.3) (19.3-72.3) (12.2-71.7) (21.3-69.0)
Abbreviations used: IQR, interquartile range. a Prescriptions for which the drug discount card did not yield any savings were excluded for the cost saving analysis (n ¼ 1,018,781).
806
(7-30) (8-30) (10-30) (10-30) (12-30) (17-30) (20-30) (25-30) (15-30)
SCIENCE AND PRACTICE Drug discount cards for prescription medications
Table 3 Top 20 prescriptions by number of prescriptions ordered Medicationsa For which cost savings accrued Hydrocodoneeacetaminophen Alprazolam tablet Lisinopril tablet Amlodipine besylate tablet Zolpidem tartrate tablet Tramadol HCl tablet Metformin HCl tablet Lorazepam tablet Omeprazole DR capsule Gabapentin capsule Citalopram HBR tablet Simvastatin tablet Clonazepam tablet Atorvastatin tablet Oxycodoneeacetaminophen Metoprolol tartrate tablet Clonazepam 0.5 mg tablet Azithromycin 250 mg tablet VIT D2 1.25 mg (50,000 unit) Escitalopram 10 and 20 mg tablet For which cost savings did not accrue Levothyroxine tablet Lisinopril tablet Metformin HCl tablet Metoprolol tartrate tablet Citalopram HBR tablet Pravastatin sodium tablet Hydrochlorothiazide CP Prednisone tablet Carvedilol tablet Lisinopril-HCTZ tablet Fluoxetine HCl capsule Cyclobenzaprine tablet Trazodone tablet Amoxicillin capsule Tramadol HCl tablet Ibuprofen tablet Meloxicam tablet Furosemide tablet Warfarin sodium tablet Ciprofloxacin HCl tablet
No. of prescriptions 128,022 101,990 89,316 65,876 65,154 52,882 48,732 48,244 46,491 45,413 44,824 43,531 42,868 42,589 39,040 37,094 36,823 35,703 33,163 31,739 48,810 37,507 32,709 23,652 22,322 19,203 17,921 17,401 17,315 17,009 15,459 13,867 12,780 12,060 11,077 10,458 10,284 10,146 9,608 9,513
a Drugs with different doses were combined to obtain single-frequency estimate.
the pharmacies themselves. Moreover, with PBMs and drug wholesalers playing a prominent but largely hidden role in prescription drug plans, the lack of pricing transparency19 limits the ability of patients to navigate as consumers.7,20 Therefore, prescribers and patients need to consider alternate strategies to make medications affordable.21 Our data show that although the drug discount program is widely available, use was focused in urban areas and in the ZIP codes with median household incomes lower than the national average. One inference is that patients theoretically in greatest need of cost savings are accessing the site. In addition, 85% of prescriptions were filled at large, national retail pharmacies. Whether this reflects a business decision to drive traffic to particular pharmacies cannot be determined. Although we noticed an increasing trend in the number of drug discount card prescriptions from 2009 to 2016, there was a slight decrease in the number of prescriptions in 2016
compared with 2014 and 2015, which might reflect a change in Walmart’s policy on accepting drug discount cards22 or increasing competition from other nationally advertised freestanding drug discount programs. More recently, several states and the U.S. Congress passed “gag clauses bills” that seek to block commercial PBM or health insurer contracts that could prohibit pharmacies from disclosing cheaper options to customers.23 Patients who have Medicare Part D, Medicaid, or any state or federal prescription insurance can use the discount only if they choose not to use their government-sponsored drug plan. Prior studies have shown that patients with insurance status also use drug discount programs, but might chose not to present their prescription insurance cards or inform the pharmacy to bill through their insurance.24 Our study is limited by the lack of patient demographic data, which is how NeedyMeds.org data capture works by design. Although our study shows that the drug discount program is widely focused in urban areas, we do not know the proportion of utilization of discount cards compared with other forms of payment. Because of the nature of the database, we were unable to evaluate the use of drug discount cards for brand versus generic drugs. Of more than 4.6 million prescriptions, only 214,385 prescriptions were labeled as branded, but this variable is not recorded consistently. Each drug card can be used more than once, and more than one person can use the same drug card. In addition, a single individual can use more than one card (e.g., if card is lost). Unfortunately, there are scant findings in this area. In one study, racial and ethnic disparities in perception toward generic prescription drugs were observed but had no effect on utilization of generic drug discount programs.25 We cannot assess the business decisions underlying the discounting (neither the magnitude of the cost savings nor the reasons why 22% of prescriptions were associated with no net financial benefit to the patient). We also cannot determine whether there are any significant differences in the drug discount card program pattern related to a specific disease condition or drug class; however, in a subanalysis of medications that can be prescribed for heart failure, we found no differences in terms of temporal trends or cost savings (data not shown). Because there are no existing criteria for defining the type of pharmacies, we used the pharmacy location and number of stores in defining national chain and other types of pharmacy businesses; ours may be a conservative estimate. NeedyMeds data had prescription claims from all 50 states in the United States, for all data years during the study period. However, we did not see a correlation between the percentage of uninsured and the use of the NeedyMed cards. Whether this lack of correlation persists over time will require additional research.
Conclusion Our data suggest the drug discount prescriptions in an online program have increased since 2009, resulting in a total savings of several hundred millions of dollars on prescription medications compared with the original cost obtained from the pharmacies. Although the drug discount programs might help patients, the programs exist in part because of the rising costs of medications and the absence of price transparency.
807
SCIENCE AND PRACTICE S. Munigala et al. / Journal of the American Pharmacists Association 59 (2019) 804e808
The availability of the coupons and the magnitude of the discount may be predicated on other financial agreements that are made between the different parties (e.g., pharmaceutical companies, PBMs, wholesalers and pharmacies) involved in the journey that a medication takes before it reaches the patient; this may be particularly germane with for-profit drug discount programs. The potential implications for regulatory oversight are significant if improved patient access to medications at a reasonable cost is a policy goal. The effects of repealing contracted “gag clauses” on drug discount card utilization and the relationships between drug discount card utilization and population dynamics and insurance status should also be evaluated in future studies. Acknowledgment A standard data use agreement was developed between Saint Louis University and NeedyMeds to allow the database to be used for research purposes and subsequent publication of the findings. Source of funding: no funding was requested or received for the development of this study and manuscript. References 1. Sachs R, Bagley N, Lakdawalla DN. Innovative contracting for pharmaceuticals and Medicaid’s best-price rule. J Health Polit Policy Law. 2018;43(1):5e18. 2. The Henry J Kaiser Family Foundation and Health Research & Education Trust. Employer health benefits: 2012 annual survey. Section 9: prescription drug benefits. September 27, 2011. Available at: http://ehbs.kff. org/pdf/2012/8345.pdf. Accessed December 10, 2018. 3. Heisler M, Wagner TH, Piette JD. Patient strategies to cope with high prescription medication costs: Who is cutting back on necessities, increasing debt, or underusing medications? J Behav Med. 2005;28(1):43e51. 4. Wagner TH, Heisler M, Piette JD. Prescription drug co-payments and costrelated medication underuse. Health Econ Policy Law. 2008;3(Pt 1):51e67. 5. Felland LE, Reschovsky JD. More nonelderly Americans face problems affording prescription drugs. Track Rep. 2009;(22):1e4. 6. Pelaez S, Lamontagne AJ, Collin J, Gauthier A, Grad RM, Blais L, et al. Patients’ perspective of barriers and facilitators to taking long-term controller medication for asthma: A novel taxonomy. BMC Pulm Med. 2015;(15):42. 7. Hauptman PJ, Goff ZD, Vidic A, Chibnall JT, Bleske BE. Variability in retail pricing of generic drugs for heart failure. JAMA Intern Med. 2017;177(1): 126e128. 8. Shrank WH, Choudhry NK, Fischer MA, et al. The epidemiology of prescriptions abandoned at the pharmacy. Ann Intern Med. 2010;153(10): 633e640. 9. Havrda DE, Omundsen BA, Bender W, Kirkpatrick MA. Impact of the Medicare modernization act on low-income persons. Ann Intern Med. 2005;143:600e608.
808
10. NeedyMeds. Available at: http://www.needymeds.org. Accessed December 10, 2018. 11. GoodRx. Available at: https://www.goodrx.com/discount-card. Accessed December 10, 2018. 12. BlinkHealth. Available at: https://www.blinkhealth.com/about-us. Accessed December 10, 2018. 13. U.S. Census Bureau. Geography, Urban and Rural. Available at: https:// www.census.gov/geo/reference/urban-rural.html. Accessed December 10, 2018. 14. U.S. Census Bureau. Census summary file 1, table P2; using American FactFinder. Available at: https://factfinder.census.gov/faces/tableservices/jsf/ pages/productview.xhtml?src¼bkmk; 2010. Accessed December 10, 2018. 15. U.S. Census Bureau. 2011-2015 American Community Survey 5-Year Estimates, table B19013; using American FactFinder. Available at: https://factfinder.census.gov/faces/tableservices/jsf/pages/productview. xhtml?src¼bkmk; 2010. Accessed December 10, 2018. 16. U.S. Census Bureau. Health Insurance Coverage in the United States: 2017. Available at: https://www.census.gov/content/dam/Census/library/ publications/2018/demo/p60-264.pdf. Accessed April 28, 2019. 17. U.S. Census Bureau. Annual Estimates of the Resident Population by Sex, Race, and Hispanic Origin for the United States, States, and Counties: April 1, 2010 to July 1, 2013. Available at: https://factfinder.census.gov/ faces/tableservices/jsf/pages/productview.xhtml?src¼bkmk. Accessed April 28, 2019. 18. Rizzo JA, Zeckhauser R. Generic script share and the price of brand-name drugs: the role of consumer choice. Int J Health Care Fi. 2009;9(3): 291e316. 19. Fein AJ. How drug channels benefits our readers. Available at: http:// www.drugchannels.net/p/about-blog.html. Accessed December 10, 2018. 20. Pociask S. You can blame pharmacy benefit managers for higher drug prices. Available at: http://www.realclearhealth.com/articles/2017/03/ 28/you_can_blame_pharmacy_benefit_managers_for_higher_drug_prices_ 110516.html. Accessed December 10, 2018. 21. Shaw CR. Reducing the burden of medication costs to improve medication adherence. Nurse Pract. 2014;39(7):43e47. 22. PS Card. Available at: http://www.pscard.com/index.cfm/discountpharmacy/walmart-discount-pharmacy/. Accessed December 10, 2018. 23. States Regulating Pharmaceutical Benefit Managers. National Conference of State Legislatures. Available at: http://www.ncsl.org/research/health/ pbm-state-legislation.aspx. Accessed February 27, 2019. 24. Patel HK, Dwibedi N, Omojasola A, Sansgiry SS. Impact of generic drug discount programs on managed care organizations. Am J Pharm Benefits. 2011;3(1):45e53. 25. Omojasola A, Hernandez M, Sansgiry S, Jones L. Perception of generic prescription drugs and utilization of generic drug discount programs. Ethn Dis. 2012;22(4):479e485. Satish Munigala, MBBS, MPH, Saint Louis University Center for Outcomes Research, St. Louis, MO Margaret Brandon, BS, MS, Saint Louis University School of Medicine, St. Louis, MO Zackary D. Goff, MD, Saint Louis University School of Medicine, St. Louis, MO; currently at Department of Medicine, Johns Hopkins University, Baltimore, MD Richard Sagall, MD, NeedyMeds, Gloucester, MA Paul J. Hauptman, MD, University of Tennessee Graduate School of Medicine, Knoxville TN
SCIENCE AND PRACTICE Drug discount cards for prescription medications
Supplementary data
Supplemental Figure 1. Drug discount card utilization and the proportion of uninsured population by state. A) Entire study period (2009 to 2016); B) Uninsured by state; C) 2013.
808.e1
Contents lists available at ScienceDirect
Journal of the American Pharmacists Association journal homepage: www.japha.org
RESEARCH
Drug discount cards in an era of higher prescription drug prices: A retrospective population-based study Satish Munigala, Margaret Brandon, Zackary D. Goff, Richard Sagall, Paul J. Hauptman* a r t i c l e i n f o
a b s t r a c t
Article history: Received 14 December 2018 Accepted 21 May 2019 Available online 14 August 2019
Objectives: Drug discount programs have emerged as a potential option for patients seeking greater accessibility and affordability. However, there is limited knowledge regarding program utilization and cost savings. The objective of this study was to evaluate medication prescriptions with drug discount card usage and estimate cost savings. Design: Retrospective study. Setting and participants: Using population-based prescription data, the study included patients who filled prescriptions from January 2009 to December 2016 nationwide using NeedyMeds. org drug discount cards. Outcome measures: We determined the frequency of drug discount card prescriptions (across pharmacy types, pharmacy location, and prescriber specialty), estimated cost savings using the drug discount card (average per drug discount card and total program dollars saved) and evaluated the top prescription drugs by frequency. Results: A total of 4,638,581 prescriptions with discount cards were identified (79.8% at national, 6.3% at regional, and 12.9% at local pharmacies). Most were filled at urban locations (urban clusters, 88.6%; urbanized areas, 8.4%) and in ZIP codes with lower median household incomes (62.7%). Overall, 3.62 million prescriptions (78.0% of the total) were associated with discounts, resulting in a total savings of $199,183,112 (median cost savings, $17.80 [47.8%] per prescription). Opiates were the most common class of drugs for which discount cards were used. Conclusion: The use of a drug discount program over 8 years resulted in total savings of nearly $200 million (approximately $18 per prescription) compared with the original cost. However, although patients might accrue financial benefit, there is still a lack of price transparency. Additional research is needed to better understand the impact of these programs and to evaluate ways to improve medication access at a reasonable cost to patients. © 2019 American Pharmacists Association®. Published by Elsevier Inc. All rights reserved.
Background The cost to patients for many branded and generic drugs have been increasing, contributing to higher overall out-of-pocket health care expenses.1,2 These additional costs can be a barrier to access for uninsured individuals,3-6 discourage adherence to treatment regimens,7 and contribute to prescription abandonment.8 In addition, lack of transparency in drug pricing can Disclosure: Richard Sagall currently serves as CEO for NeedyMeds.org and is the primary source for data acquisition. The remaining authors declare no conflicts of interest or financial interests in any product or service mentioned in this article. * Correspondence: Paul J. Hauptman, MD, University of Tennessee Graduate School of Medicine, 1924 Alcoa Highway, Knoxville TN 37920. E-mail address: [email protected] (P.J. Hauptman).
adversely affect adherence by patients whose options might be restricted to a retail pharmacy within a limited geographic area.9 The Medicare Prescription Drug, Improvement, and Modernization Act (MMA) of 2003 created a drug discount card program with a prescription benefit to support medication expenditures for eligible seniors.9 Pharmaceutical drug company assistance programs formed a viable alternative to the Medicare prescription benefit for patients who do not qualify for low-income subsidies. Recently, freestanding drug discount programs, such as NeedyMeds.org, GoodRx.com, and BlinkHealth.com,10-12 have been established and in some cases advertised in media; they can offer an additional option for patients. However, there is limited knowledge regarding program usage and role in medication dissemination in a general patient population.
https://doi.org/10.1016/j.japh.2019.05.021 1544-3191/© 2019 American Pharmacists Association®. Published by Elsevier Inc. All rights reserved.
SCIENCE AND PRACTICE Drug discount cards for prescription medications
Definitions Key Points Background: Drug discount programs have proliferated in recent years but a critical analysis of their impact has not been performed. Data were derived from NeedyMeds, a nonprofit organization that provides online access to a free drug discount card. Findings: In a population-based retrospective review of prescription data, a drug discount program resulted in total savings of nearly $200 million on prescription medications (approximately $18 in savings per prescription) compared with the original cost charged by pharmacies. Drug discount programs can help patients save money, but the programs exist in part because of the rising costs of medications and the absence of price transparency.
Using data from one drug discount program, we evaluated the frequency of and trends in medication prescriptions with discount card usage (by pharmacy type and location, and by prescriber specialty) and then estimated the cost savings (average per drug discount card and the total program dollar amount saved) for all prescription drugs.
Methods This retrospective study investigated prescriptions for which drug discount cards were used from January 2009 to December 2016. Data were obtained from NeedyMeds, a nationwide nonprofit organization established in 1997 with a mission to serve as a resource for information on pharmaceutical patient assistance programs. Since 2009, NeedyMeds has provided on-line access to a free drug discount card that offers a savings of up to 80% at more than 65,000 pharmacies nationwide10 and uses electronic capture of records for subsequent analysis. Data abstracted from these files include prescription details (including drug name, days’ supply, and patient cost savings), pharmacy information (e.g., name, address) and prescriber information (specialty). NeedyMeds data had prescription claims from all 50 states in the United States and for all data years during the study period. The NeedyMeds.org drug discount card can be used by anyone regardless of age, residency status, income, insurance status, and diagnosis, but it cannot be combined with insurance or any state, federal, or pharmaceutical company drug discount programs. If a drug discount card was used multiple times, we treated each record as an independent observation. Cost savings for each prescription were the total savings for a customer (patient) using the drug discount card compared to the actual price offered by the pharmacy. Prescriptions for which the drug discount card did not yield any savings (i.e., prescription price using the drug discount card was the same as the lowest price offered by the pharmacy) were excluded for the cost-savings analysis (n ¼ 1,018,781).
We classified pharmacies as national (if located in 5 states with 25 or more stores, at least 1 state not contiguous), regional (if located in 3 states with 10 or more stores), specialty (individual pharmacies affiliated with hospitals), and local businesses (located in just 1 or 2 states with fewer than 10 stores). We categorized the location of pharmacies as urban (“urbanized areas” of 50,000 or more people; “urban clusters” of at least 2500 and less than 50,000 people) or rural (encompassing all population, housing, and territory not included within an urban area) locations by mapping pharmacy ZIP code to U.S. Census Bureau ZIP code tabulation data.13,14 We also designated the ZIP codes to above- and below-median annual income level using 2015 national estimates from the 2011-2015 American Community Survey.15
Statistical analysis Categorical variables were reported as frequencies and proportions, and continuous variables were reported as median or interquartile range (IQR) where appropriate. Total number of prescriptions with a drug discount card was first identified. Prescriptions by pharmacy type, location, prescriber specialty, and median household income level were assessed. Prescriptions for which the drug discount card provided a lower price than the lowest price offered by the store were evaluated by the number of transactions, median days’ supply, and calculated cost savings (median cost saved by amount per prescription, percentage saved compared to original cost and total program dollars saved) for the entire study period and for each year (2009-2016). We evaluated the top 20 prescription drugs by frequency over the duration of the program. We also evaluated the drug discount card utilization by state, for the entire study period, and then compared the percentage of uninsured people by each state16 regarding the drugdiscount card utilization per 100,000 U.S. population by state,17 for the year 2013 (midpoint for the period under study). Data were analyzed using Statistical Analysis Software (SAS) version 9.3 (SAS Institute, Cary, NC). The Saint Louis University Institutional Review Board determined that this study did not constitute human subjects research and therefore required no additional review.
Results Prescription characteristics A total of 4,638,581 prescriptions with drug discount cards were identified during the study period. The majority were filled at national (79.8%) as opposed to regional (6.3%), local (12.9%), or hospital-affiliated (1.1%) pharmacies. Most prescriptions were filled at pharmacies located in urban clusters or urbanized areas (88.6% and 8.4%, respectively) with the remaining 3.0% from rural areas. Sixty-two percent of the prescriptions were obtained from pharmacies located in the ZIP codes with lower median household income. Prescription characteristic by prescriber specialty are illustrated in Table 1; more than 50% were in internal medicine (including specialties), family practice, or general medicine. 805
SCIENCE AND PRACTICE S. Munigala et al. / Journal of the American Pharmacists Association 59 (2019) 804e808
prescription per patient, representing a reduction of 47.8% (IQR, 21.3%-69.0%) over prescription charges at the pharmacy level.
Table 1 Characteristics of drug discount prescriptions by pharmacy type and prescriber specialty Characteristics
All drugs (N ¼ 4,638,581)
Pharmacy typea National Regional Local Affiliated with hospital Pharmacy location by ZIP code areab Rural Urban clusters Urbanized areas Median household income by pharmacy ZIP codec (per 2015 census estimate) Lower (<$56,516) Higher ($56,516) Prescriber specialty Internal medicine and subspecialties Surgery Neurology, psychiatry Gynecology Family medicine, general practice Nurse, nurse practitioner Pediatrics Emergency Medicine Other Unknown Missing percentages: (n ¼ 30,856);
a
0.2% (n ¼ 9066);
Top prescription drugs by frequency
n
%
3,694,186 289,871 595,731 49,727
79.8 6.3 12.9 1.1
137,952 4,088,566 389,952
3.0 88.6 8.4
The most prescribed medications for which drug discount cards were used are shown in Table 3. The opiate class (oxycodone, hydrocodone) was the most common; other classes of drug with 2 or more medications in the top 20 included benzodiazepines (n ¼ 3), anti-hypertensives (n ¼ 3), antidepressants (n ¼ 2), and HMG-CoA reductase inhibitors (n ¼ 2). Five of these drugs were also on the list of top 20 drugs for which no cost savings accrued. Drug discount card utilization by state
b
2,890,150 1,717,575
62.7 37.3
940,521 68,018 362,857 97,387 1,493,840 479,774 67,555 290,885 504,535 333,209
20.3 1.5 7.8 2.1 32.2 10.3 1.5 6.3 10.9 7.2
For the entire study period, drug discount card utilization was not concentrated in a specific geographical region or within few states. Texas, California, Florida, Indiana, and Ohio showed the highest drug discount card utilization rates (Supplemental Figure 1A). Although a higher percentage of uninsured people resided in Texas, Nevada, Florida, Georgia, and Alaska for the year 2013, the percentage of drug discount card utilization (per 100,000 population) was highest in New York, North Carolina, South Carolina, and Ohio (Supplemental Figures 1B and 1C). Discussion
0.5% (n ¼ 22,111); c0.7%
Prescription trends by year The number of prescriptions with drug discount cards increased from 28,878 in 2009 to 756,837 in 2016 with the highest number of prescriptions in the years 2014 and 2015. Cost savings using the drug discount card Overall, 3,619,800 prescriptions (78.0%) were associated with savings; the remainder offered no savings to patients (Table 2). Total savings for the study period was $199,183,112, with a median of $17.80 (IQR, 6.50-45.40) cost savings per
In this study, we evaluated the frequency, temporal trends and cost savings of drug discount card prescriptions using a large database from a free-standing, on-line drug discount coupon program. We found that the number of prescriptions has increased significantly, resulted in total savings of nearly $200 million on prescription medications, with an average savings of $18 at 48% savings per prescription, compared with the original cost obtained from the pharmacies (where savings were noted). Increasing prescription costs represent a significant economic burden on patients,8,18 and it can lead to decrease in adherence.7 Pricing at the consumer (patient) level has remained obtuse, but increasing attention has been paid to the complex business arrangements between drug manufacturers, pharmacy benefit managers (PBMs), insurers, wholesalers, and
Table 2 Total savings by year for transactions involving cost savings Year
2009 2010 2011 2012 2013 2014 2015 2016 Overall
All transactions
Number of transactions in which savings were noteda
n
n
%
28,878 182,512 415,145 622,298 814,866 909,575 908,470 756,837 4,638,581
18,559 133,492 295,094 453,999 620,658 733,614 727,572 636,812 3,619,800
64.3 73.1 71.1 73.0 76.2 80.7 80.1 84.1 78.0
Median patient cost savings, $ (IQR)
Median cost savings, % (IQR)
Total savings ($)
Median days’ supply (IQR)
15.4 15.5 16.3 16.3 16.7 17.7 20.6 19.0 17.8
43.7 45.0 40.6 46.5 48.6 51.6 48.5 46.9 47.7
600,712 4,326,755 10,024,670 16,729,658 25,492,066 37,878,468 58,012,341 46,118,440 199,183,112
30 30 30 30 30 30 30 30 30
(6.6-34.2) (6.7-35.2) (7.0-35.5) (7.2-37.1) (6.7-39.5) (6.9-44.4) (6.5-61.3) (4.6-57.5) (6.5-45.4)
(18.3-64.3) (19.2-64.6) (18.8-60.1) (24.2-66.2) (25.4-69.2) (26.8-70.3) (19.3-72.3) (12.2-71.7) (21.3-69.0)
Abbreviations used: IQR, interquartile range. a Prescriptions for which the drug discount card did not yield any savings were excluded for the cost saving analysis (n ¼ 1,018,781).
806
(7-30) (8-30) (10-30) (10-30) (12-30) (17-30) (20-30) (25-30) (15-30)
SCIENCE AND PRACTICE Drug discount cards for prescription medications
Table 3 Top 20 prescriptions by number of prescriptions ordered Medicationsa For which cost savings accrued Hydrocodoneeacetaminophen Alprazolam tablet Lisinopril tablet Amlodipine besylate tablet Zolpidem tartrate tablet Tramadol HCl tablet Metformin HCl tablet Lorazepam tablet Omeprazole DR capsule Gabapentin capsule Citalopram HBR tablet Simvastatin tablet Clonazepam tablet Atorvastatin tablet Oxycodoneeacetaminophen Metoprolol tartrate tablet Clonazepam 0.5 mg tablet Azithromycin 250 mg tablet VIT D2 1.25 mg (50,000 unit) Escitalopram 10 and 20 mg tablet For which cost savings did not accrue Levothyroxine tablet Lisinopril tablet Metformin HCl tablet Metoprolol tartrate tablet Citalopram HBR tablet Pravastatin sodium tablet Hydrochlorothiazide CP Prednisone tablet Carvedilol tablet Lisinopril-HCTZ tablet Fluoxetine HCl capsule Cyclobenzaprine tablet Trazodone tablet Amoxicillin capsule Tramadol HCl tablet Ibuprofen tablet Meloxicam tablet Furosemide tablet Warfarin sodium tablet Ciprofloxacin HCl tablet
No. of prescriptions 128,022 101,990 89,316 65,876 65,154 52,882 48,732 48,244 46,491 45,413 44,824 43,531 42,868 42,589 39,040 37,094 36,823 35,703 33,163 31,739 48,810 37,507 32,709 23,652 22,322 19,203 17,921 17,401 17,315 17,009 15,459 13,867 12,780 12,060 11,077 10,458 10,284 10,146 9,608 9,513
a Drugs with different doses were combined to obtain single-frequency estimate.
the pharmacies themselves. Moreover, with PBMs and drug wholesalers playing a prominent but largely hidden role in prescription drug plans, the lack of pricing transparency19 limits the ability of patients to navigate as consumers.7,20 Therefore, prescribers and patients need to consider alternate strategies to make medications affordable.21 Our data show that although the drug discount program is widely available, use was focused in urban areas and in the ZIP codes with median household incomes lower than the national average. One inference is that patients theoretically in greatest need of cost savings are accessing the site. In addition, 85% of prescriptions were filled at large, national retail pharmacies. Whether this reflects a business decision to drive traffic to particular pharmacies cannot be determined. Although we noticed an increasing trend in the number of drug discount card prescriptions from 2009 to 2016, there was a slight decrease in the number of prescriptions in 2016
compared with 2014 and 2015, which might reflect a change in Walmart’s policy on accepting drug discount cards22 or increasing competition from other nationally advertised freestanding drug discount programs. More recently, several states and the U.S. Congress passed “gag clauses bills” that seek to block commercial PBM or health insurer contracts that could prohibit pharmacies from disclosing cheaper options to customers.23 Patients who have Medicare Part D, Medicaid, or any state or federal prescription insurance can use the discount only if they choose not to use their government-sponsored drug plan. Prior studies have shown that patients with insurance status also use drug discount programs, but might chose not to present their prescription insurance cards or inform the pharmacy to bill through their insurance.24 Our study is limited by the lack of patient demographic data, which is how NeedyMeds.org data capture works by design. Although our study shows that the drug discount program is widely focused in urban areas, we do not know the proportion of utilization of discount cards compared with other forms of payment. Because of the nature of the database, we were unable to evaluate the use of drug discount cards for brand versus generic drugs. Of more than 4.6 million prescriptions, only 214,385 prescriptions were labeled as branded, but this variable is not recorded consistently. Each drug card can be used more than once, and more than one person can use the same drug card. In addition, a single individual can use more than one card (e.g., if card is lost). Unfortunately, there are scant findings in this area. In one study, racial and ethnic disparities in perception toward generic prescription drugs were observed but had no effect on utilization of generic drug discount programs.25 We cannot assess the business decisions underlying the discounting (neither the magnitude of the cost savings nor the reasons why 22% of prescriptions were associated with no net financial benefit to the patient). We also cannot determine whether there are any significant differences in the drug discount card program pattern related to a specific disease condition or drug class; however, in a subanalysis of medications that can be prescribed for heart failure, we found no differences in terms of temporal trends or cost savings (data not shown). Because there are no existing criteria for defining the type of pharmacies, we used the pharmacy location and number of stores in defining national chain and other types of pharmacy businesses; ours may be a conservative estimate. NeedyMeds data had prescription claims from all 50 states in the United States, for all data years during the study period. However, we did not see a correlation between the percentage of uninsured and the use of the NeedyMed cards. Whether this lack of correlation persists over time will require additional research.
Conclusion Our data suggest the drug discount prescriptions in an online program have increased since 2009, resulting in a total savings of several hundred millions of dollars on prescription medications compared with the original cost obtained from the pharmacies. Although the drug discount programs might help patients, the programs exist in part because of the rising costs of medications and the absence of price transparency.
807
SCIENCE AND PRACTICE S. Munigala et al. / Journal of the American Pharmacists Association 59 (2019) 804e808
The availability of the coupons and the magnitude of the discount may be predicated on other financial agreements that are made between the different parties (e.g., pharmaceutical companies, PBMs, wholesalers and pharmacies) involved in the journey that a medication takes before it reaches the patient; this may be particularly germane with for-profit drug discount programs. The potential implications for regulatory oversight are significant if improved patient access to medications at a reasonable cost is a policy goal. The effects of repealing contracted “gag clauses” on drug discount card utilization and the relationships between drug discount card utilization and population dynamics and insurance status should also be evaluated in future studies. Acknowledgment A standard data use agreement was developed between Saint Louis University and NeedyMeds to allow the database to be used for research purposes and subsequent publication of the findings. Source of funding: no funding was requested or received for the development of this study and manuscript. References 1. Sachs R, Bagley N, Lakdawalla DN. Innovative contracting for pharmaceuticals and Medicaid’s best-price rule. J Health Polit Policy Law. 2018;43(1):5e18. 2. The Henry J Kaiser Family Foundation and Health Research & Education Trust. Employer health benefits: 2012 annual survey. Section 9: prescription drug benefits. September 27, 2011. Available at: http://ehbs.kff. org/pdf/2012/8345.pdf. Accessed December 10, 2018. 3. Heisler M, Wagner TH, Piette JD. Patient strategies to cope with high prescription medication costs: Who is cutting back on necessities, increasing debt, or underusing medications? J Behav Med. 2005;28(1):43e51. 4. Wagner TH, Heisler M, Piette JD. Prescription drug co-payments and costrelated medication underuse. Health Econ Policy Law. 2008;3(Pt 1):51e67. 5. Felland LE, Reschovsky JD. More nonelderly Americans face problems affording prescription drugs. Track Rep. 2009;(22):1e4. 6. Pelaez S, Lamontagne AJ, Collin J, Gauthier A, Grad RM, Blais L, et al. Patients’ perspective of barriers and facilitators to taking long-term controller medication for asthma: A novel taxonomy. BMC Pulm Med. 2015;(15):42. 7. Hauptman PJ, Goff ZD, Vidic A, Chibnall JT, Bleske BE. Variability in retail pricing of generic drugs for heart failure. JAMA Intern Med. 2017;177(1): 126e128. 8. Shrank WH, Choudhry NK, Fischer MA, et al. The epidemiology of prescriptions abandoned at the pharmacy. Ann Intern Med. 2010;153(10): 633e640. 9. Havrda DE, Omundsen BA, Bender W, Kirkpatrick MA. Impact of the Medicare modernization act on low-income persons. Ann Intern Med. 2005;143:600e608.
808
10. NeedyMeds. Available at: http://www.needymeds.org. Accessed December 10, 2018. 11. GoodRx. Available at: https://www.goodrx.com/discount-card. Accessed December 10, 2018. 12. BlinkHealth. Available at: https://www.blinkhealth.com/about-us. Accessed December 10, 2018. 13. U.S. Census Bureau. Geography, Urban and Rural. Available at: https:// www.census.gov/geo/reference/urban-rural.html. Accessed December 10, 2018. 14. U.S. Census Bureau. Census summary file 1, table P2; using American FactFinder. Available at: https://factfinder.census.gov/faces/tableservices/jsf/ pages/productview.xhtml?src¼bkmk; 2010. Accessed December 10, 2018. 15. U.S. Census Bureau. 2011-2015 American Community Survey 5-Year Estimates, table B19013; using American FactFinder. Available at: https://factfinder.census.gov/faces/tableservices/jsf/pages/productview. xhtml?src¼bkmk; 2010. Accessed December 10, 2018. 16. U.S. Census Bureau. Health Insurance Coverage in the United States: 2017. Available at: https://www.census.gov/content/dam/Census/library/ publications/2018/demo/p60-264.pdf. Accessed April 28, 2019. 17. U.S. Census Bureau. Annual Estimates of the Resident Population by Sex, Race, and Hispanic Origin for the United States, States, and Counties: April 1, 2010 to July 1, 2013. Available at: https://factfinder.census.gov/ faces/tableservices/jsf/pages/productview.xhtml?src¼bkmk. Accessed April 28, 2019. 18. Rizzo JA, Zeckhauser R. Generic script share and the price of brand-name drugs: the role of consumer choice. Int J Health Care Fi. 2009;9(3): 291e316. 19. Fein AJ. How drug channels benefits our readers. Available at: http:// www.drugchannels.net/p/about-blog.html. Accessed December 10, 2018. 20. Pociask S. You can blame pharmacy benefit managers for higher drug prices. Available at: http://www.realclearhealth.com/articles/2017/03/ 28/you_can_blame_pharmacy_benefit_managers_for_higher_drug_prices_ 110516.html. Accessed December 10, 2018. 21. Shaw CR. Reducing the burden of medication costs to improve medication adherence. Nurse Pract. 2014;39(7):43e47. 22. PS Card. Available at: http://www.pscard.com/index.cfm/discountpharmacy/walmart-discount-pharmacy/. Accessed December 10, 2018. 23. States Regulating Pharmaceutical Benefit Managers. National Conference of State Legislatures. Available at: http://www.ncsl.org/research/health/ pbm-state-legislation.aspx. Accessed February 27, 2019. 24. Patel HK, Dwibedi N, Omojasola A, Sansgiry SS. Impact of generic drug discount programs on managed care organizations. Am J Pharm Benefits. 2011;3(1):45e53. 25. Omojasola A, Hernandez M, Sansgiry S, Jones L. Perception of generic prescription drugs and utilization of generic drug discount programs. Ethn Dis. 2012;22(4):479e485. Satish Munigala, MBBS, MPH, Saint Louis University Center for Outcomes Research, St. Louis, MO Margaret Brandon, BS, MS, Saint Louis University School of Medicine, St. Louis, MO Zackary D. Goff, MD, Saint Louis University School of Medicine, St. Louis, MO; currently at Department of Medicine, Johns Hopkins University, Baltimore, MD Richard Sagall, MD, NeedyMeds, Gloucester, MA Paul J. Hauptman, MD, University of Tennessee Graduate School of Medicine, Knoxville TN
SCIENCE AND PRACTICE Drug discount cards for prescription medications
Supplementary data
Supplemental Figure 1. Drug discount card utilization and the proportion of uninsured population by state. A) Entire study period (2009 to 2016); B) Uninsured by state; C) 2013.
808.e1