Pharmacists’ Experiences, Perceptions, and Knowledge of Direct-to-Consumer Prescription Coupons

BACKGROUND: Despite widespread use of manufacturer-sponsored prescription drug coupons and pharmacy network discount cards (i.e., direct-to-consumer prescription coupons), little is known about community pharmacists’ experiences, perceptions, and knowledge of coupon cards. OBJECTIVE: To identify community pharmacists’ experiences, perceptions, and knowledge of prescription coupons. METHODS: An 11-item telephonic survey was conducted from August 2018 to March 2019. Eligible respondents included English-speaking pharmacists employed during the survey period in a community pharmacy physically located in Connecticut. Data were analyzed via descriptive statistics and one-way analysis of variance (ANOVA). One-way ANOVAs were conducted to test the relationship between the respondents’ practice types, the average daily volume of coupons processed, and the average time needed to process each coupon. The responses were based on a 5-point Likert scale and dichotomized to enable interpretation of the results. RESULTS: There were 240 surveys completed from an eligible pool of 691 community pharmacy sites (34.7% response rate). Respondents representing 60 different businesses located across 123 of the state’s 282 major ZIP codes, representing 83.5% of the state’s population. Respondents overwhelmingly held positive perceptions of the ability of prescription coupons to increase patients’ medication access (91.7 %) and reduce out-of-pocket costs (93.3%). However, respondents also believed patients have trouble paying for prescriptions once coupons expire (70.8%). When questioned about privacy practices, 57.5% of respondents believed that it is illegal to “sell patients’ information” (i.e., with no distinction made between protected health information and any other information), while another 25.8% declined to answer, citing they did not know. Only 20.8% (n = 50) of respondents knew that community pharmacies could see lowered reimbursement from accepting network drug discount cards, and 40.4% (n = 97) knew that pharmaceutical manufacturers can cover the difference in patients’ copay costs. Approximately 10% of respondents believed (incorrectly) that discounts from pharmacy network discount cards were covered via patients’ prescription insurance and/or the third-party discount card vendor companies (7.9% and 3.3%, respectively). Respondents believed patients received prescription coupons most often from the internet or mail (77.1%), their prescribers (62.9%), or from their own community pharmacies (33.3%). Finally, on average, respondents processed 14.6 (SD 19.8) coupons per day and required 4.8 (SD 4.3) minutes for each claim. CONCLUSIONS: As far as we know, this is the first exploration of community pharmacists’ experiences, perceptions, and knowledge of direct-to-consumer prescription coupons. Results show that, while community pharmacists overwhelmingly hold positive perceptions towards prescription coupons and drug discount cards, there is an opportunity to increase general understanding of the differences in business practices between manufacturer-sponsored prescription drug coupons and pharmacy network discount cards. Community pharmacies also spend a significant amount of time processing coupon claims.

METHODS: An 11-item telephonic survey was conducted from August 2018 to March 2019. Eligible respondents included English-speaking pharmacists employed during the survey period in a community pharmacy physically located in Connecticut. Data were analyzed via descriptive statistics and one-way analysis of variance (ANOVA). One-way ANOVAs were conducted to test the relationship between the respondents' practice types, the average daily volume of coupons processed, and the average time needed to process each coupon. The responses were based on a 5-point Likert scale and dichotomized to enable interpretation of the results.
RESULTS: There were 240 surveys completed from an eligible pool of 691 community pharmacy sites (34.7% response rate). Respondents representing 60 different businesses located across 123 of the state's 282 major ZIP codes, representing 83.5% of the state's population. Respondents overwhelmingly held positive perceptions of the ability of prescription coupons to increase patients' medication access (91.7 %) and reduce out-of-pocket costs (93.3%). However, respondents also believed patients have trouble paying for prescriptions once coupons expire (70.8%). When questioned about privacy practices, 57.5% of respondents believed that it is illegal to "sell patients' information" (i.e., with no distinction made between protected health information and any other information), while another 25.8% declined to answer, citing they did not know. Only 20.8% (n = 50) of respondents knew that community pharmacies could see lowered reimbursement from accepting network drug discount cards, and 40.4% (n = 97) knew that pharmaceutical manufacturers can cover the difference in patients' copay costs. Approximately 10% of respondents believed (incorrectly) that discounts from pharmacy network discount cards were covered via patients' prescription insurance and/or the thirdparty discount card vendor companies (7.9% and 3.3%, respectively). Respondents believed patients received prescription coupons most often from the internet or mail (77.1%), their prescribers (62.9%), or from their own community pharmacies (33.3%). Finally, on average, respondents processed 14.6 (SD 19.8) coupons per day and required 4.8 (SD 4.3) minutes for each claim.
CONCLUSIONS: As far as we know, this is the first exploration of community pharmacists' experiences, perceptions, and knowledge of direct-to-consumer prescription coupons. Results show that, while community pharmacists overwhelmingly hold positive perceptions towards prescription coupons and drug discount cards, there is an opportunity to increase general understanding of the differences in business practices between manufacturer-sponsored prescription drug coupons and pharmacy R E S E A R C H P harmacies in the United States fill over 4 billion prescriptions annually from a pool of over 18,000 available drug products, and over 80% of older adults require at least 1 daily prescription. [1][2][3] Unfortunately, the high cost of U.S. prescription medications-nearly a half-trillion dollars annually-has caused the the United States to experience some of the highest cost-related medication nonadherence rates in the industrialized world. 4 Indeed, it is estimated that up to 40% of noninstitutionalized patients risk missing doses because they cannot afford their out-of-pocket prescription costs, with risks even higher among disabled and/or severely ill patients. [5][6][7][8][9] Patients frequently seek prescription price discounts by using coupons and discount cards, which can drive use of highercost brand drugs at the expense of generic or alternative drugs that are more favorably tiered on insurance plans. However, little is known about how these attempts at cost savings affect • Patients commonly use prescription coupons to lower their outof-pocket costs. • Despite the widespread use of coupons, little is known about how pharmacists perceive coupons in terms of value to the customer, effect on health care costs, and effect on pharmacy outcomes.

What is already known about this subject
• Community pharmacists overwhelmingly believe that prescription coupons increase access and reduce patients' out-of pocketexpenses but cause patients to face affordability issues once expired. • Community pharmacists are largely unaware of the difference between manufacturer-sponsored and pharmacy network discount cards. • On average, community pharmacies dedicate over an hour of staff time to process prescription coupon claims every day.

■■ Methods
An 11-item telephonic survey was conducted from August 2018 to March 2019 to identify community pharmacists' general perceptions and knowledge regarding pharmaceutical company direct-to-consumer prescription coupons and network drug discount cards (see Appendix, available in online article). Survey items were developed and beta tested by pharmacists with community pharmacy experience to ensure face validity. Eligible respondents included English-speaking community pharmacists who were employed during the survey period in a pharmacy physically located in the state of Connecticut. Any community pharmacist who practiced remotely (e.g., telepharmacy) was excluded, and only pharmacists licensed in Connecticut were eligible to participate. Pharmacy technicians, interns, and noncommunity pharmacists were excluded.
To recruit respondents, a list of community pharmacies located and actively licensed in the state of Connecticut was obtained from eLicense.ct.gov, a Connecticut database of businesses that require state licensing. A manual internet search was then completed to identify community pharmacies' publicly available phone numbers. As such, the unit of enrollment was defined as each community pharmacy's physical site, and only 1 response per site was sought. For example, if a chain operated 90 community pharmacies located in the state of Connecticut, 1 community pharmacist from each of the 90 listed would be sought for recruitment. Once either a potential respondent participated or declined participation, the site was marked as a responder or nonresponder and not contacted again. All identified pharmacies were called, and 5 attempts were made before recording the site as a nonresponder, at which no further attempts to recruit were made.
Survey calls, lasting approximately 2-3 minutes, were made anytime between 9:00 am and 9:00 pm Eastern Standard Time and were completed by investigators and 4 research assistants (RAs), including 3 PharmD students and 1 undergraduate. RAs received standardized training with investigators in research ethics, survey methods, data collection, and this study's protocol, including use of the survey and call script. The call script was used under the supervision of investigators by RAs to do the following: (a) identify potential respondents; (b) identify themselves to potential respondents; (c) confirm that respondents' met inclusion and exclusion criteria; and (4) signify the voluntary nature of the survey. The call script also oriented respondents to the survey by defining "prescription coupons" as "including any coupon made by a manufacturer for a specific product OR any nonproduct specific coupon card that may cover many different types of medications." Calls were not recorded, and respondents' answers were entered into an online Qualtrics data collection form in real time. Researchers maintained no code key to link answers with the identity of individual respondents or pharmacy location. Therefore, partially filled surveys were disregarded overall patient care, outcomes, and long-term overall health care costs. [10][11][12][13] To reduce patients' out-of-pocket costs and promote the usage of brand-name products, the pharmaceutical industry has promoted the use of direct-to-consumer prescription coupons issued by drug manufacturers and pharmaceutical companies. These coupons are applied at the point of sale against the patient's insurance copayment and thereby help reduce out-of-pocket costs. When a patient uses a pharmaceutical company direct-to-consumer prescription coupon, the pharmacist applies the coupon to the patient's existing prescription insurance via a "coordination of benefits." The prescription is first processed by the patient's primary insurance then, second, by the prescription coupon. Therefore, the prescription coupon will reimburse the community pharmacy for a specified discount after the patient's insurance has processed a copay. Pharmacists can also apply prescription coupons for patients without insurance. In this situation, the pharmaceutical company direct-to-consumer prescription coupon will only lower the patient's out-of-pocket costs up to the manufacturer's specified maximum "out-of-pocket savings." In either case, the community pharmacy is reimbursed for the discount it provided the patient.
In addition to pharmaceutical company direct-to-consumer coupons, patients often use discount cards from companies such as GoodRx and Blink. These companies (i.e., discount card vendors) create networks of pharmacies that accept their network coupons. The discount card vendors distribute network coupons to the public and use third-party contractors to negotiate with community pharmacies. In these contracts, community pharmacies agree to accept lower, out-of-pocket payments from patients who present the vendor's discount card. In exchange, the discount card vendor allows the community pharmacy to remain within its preferred network. Community pharmacies are incentivized to remain in discount card vendor networks despite revenue loss, as vendors encourage patients to move their business to contracted preferred pharmacies. When patients use drug discount cards, the community pharmacy cannot use the patient's insurance coordination-of-benefit provision. Discount card vendors and their subsidiaries may collect and store patient information when the patient signs up online to receive coupons.
Because pharmaceutical company direct-to-consumer prescription coupons and discount drug cards can only be applied at a community pharmacy, pharmacists are the health care professionals most directly affected by these programs. Therefore, community pharmacists' experiences, knowledge, and perceptions of prescription coupons and drug discount cards are important to understand. This study surveyed community pharmacists to elicit their experiences, perceptions, and knowledge of pharmaceutical company direct-to-consumer prescription coupons and drug discount network cards. and unrecoverable. If a survey was interrupted before the last 5 questions were asked, and the respondent agreed to be called back to restart the survey at a later time, the current survey was disregarded and the RAs called back to complete the entire survey, up to 5 times. RAs asked participants for better times to call and attempted calls at the suggested time. However, if a survey was interrupted before the last 5 questions and the respondent did not agree to be called back to restart the survey, the recorded responses were saved, the site was marked as a responder, and unanswered questions were treated as missing data. Items regarding general knowledge were open answer and based on industry practices that these researchers believed were commonly misunderstood or controversial among pharmacy professionals.
Open-ended answers (e.g., where participants believed discounts originated) were independently coded by 2 researchers and reconciled via discussion until consensus was reached. Codebook definitions were developed heuristically, and a log of definitions were kept. Items regarding general knowledge were open answer and based on industry practices that these researchers believed were commonly misunderstood or controversial among pharmacy professionals. Items regarding perceptions included statements followed by the typical 5-point Likert scale, where respondents rated their perceptions from "strongly disagree" to "strongly agree." Data were analyzed via descriptive statistics and one-way analyses of variance (ANOVAs) were conducted to test the relationship between site of employment and coupon volume and processing time. Respondents practicing in "other" community pharmacy settings were excluded in this analysis because of their small size. Given some small numbers reported in Likertscale data, that data were collapsed from 5 to 3 categories: (1) positive, "strongly agree" or "agree"; (2) neutral, "neither agree nor disagree"; and (3) negative, "strongly disagree" or "disagree" to run chi-square tests.
Chi-square tests were conducted to test the relationship between respondents' site of employment and their experiences and perceptions. Significance indicated an alpha < 0.05. Data regarding the respondents served were gathered (via ZIP code) from Connecticut Open Data, under the state's Office of Policy and Management and Regional Data Cooperative (DataHaven), a nonprofit organization of the National Neighborhood Indicators Partnership, in collaboration with Urban Institute in Washington, DC. 14,15 These data were similarly tested for relationships with descriptive and one-way ANOVAs. Data were analyzed in SPSS version 25 (SPSS, Inc., Chicago, IL).
This study was approved by the University of Connecticut Institutional Review Board, and as a token of appreciation for their time, respondents were invited to participate in a random drawing for 1 of 5 Amazon gift cards up to $50.

■■ Results
There were 716 community pharmacies identified as registered in Connecticut. However, when recruitment was attempted, 11 community pharmacies were no longer in business during the survey period, and another 14 attested that they were not licensed as a community pharmacy. Of the remaining 691 pharmacies, 240 surveys were completed for a response rate of 34.7%. Of nonresponders, 59.2% (n = 409) were unable to respond, citing time constraints. Respondents represented 60 different businesses located across 123 of the state's 282 major ZIP codes, representing 83.5% of the state's population ( Table 1). Demographics of practice type and population served did not differ from state averages.
Respondents overwhelmingly held positive perceptions towards prescription coupons and the ability of drug discount cards to increase patient medication access (91.7 %) and reduce out-of-pocket costs (93.3%) and believed that community pharmacies should apply prescription coupons and drug discount cards whenever possible (84.2%; Figure 1). However, respondents also generally believed that patients have trouble paying for prescriptions once prescription coupons expire (70.8%). Overall, pharmacists were more split in their belief that prescription coupons encourage the use of brand-name drugs when a less expensive generic is available (47.9%) or that their employers required them to process prescription coupons (55.5%). However, respondents working in independent community pharmacies were more likely than those in chains or grocery pharmacies to believe that prescription coupons encouraged the use of brand-name medications when a cheaper

Characteristics of Respondents' Practice Type and Patient Population Served
generic was available (76.5% vs. 56.1% and 48.2%, respectively; P < 0.05). Similarly, independent community pharmacists were significantly less likely to report that their employers required them to process prescription coupons than respondents in chain or grocery pharmacies (38.6% vs. 64.5% and 57.4%, respectively; P < 0.01).
Respondents were asked which entity is responsible for covering the copay difference (i.e., the residual copay amount) after the coupon is applied. This was an open-answer item where respondents were allowed to provide multiple responses; therefore, the sum of this item is greater than 100%. For example, 16.7% (n = 40) of respondents were unsure. Another 20.8% (n = 50) correctly responded that community pharmacies could see lowered reimbursement from accepting network drug discount cards. Similarly, 40.4% (n = 97) of pharmacists correctly responded that pharmaceutical manufacturers cover the difference in copay costs for their prescription coupons. However, 25.4% (n = 61) of responses were nonspecific, citing "coupon or card companies" without regards to the difference between prescription coupons and drug discount cards. As such, these nonspecific responses did not provide enough detail to determine the pharmacist's level of understanding about prescription coupons or drug discount cards. Another 3.3% of respondents incorrectly reported that network drug discount cards were financially covered by third-party discount card vendor companies. An additional 7.9% incorrectly reported that network drug discount cards used the patients' prescription insurance.
Respondents working in grocery community pharmacies were more likely than respondents working in chain or independent settings to correctly respond that pharmacies could see lowered reimbursement from accepting drug discount cards (32.4% vs. 17.8% and 13.6%, respectively; P < 0.05). On the other hand, respondents from independent community pharmacies were more likely than those working within chains or grocery pharmacies to correctly respond that some prescription coupons can cover the difference in copayment costs (55.9% vs. 35.6% and 36.7%, respectively; P < 0.05).
On average, respondents processed 14

Community Pharmacists' Reported Experiences, Perceptions, and Knowledge of Direct-to-Consumer Prescription Coupons
Disagree Agree Strongly Disagree Strongly Agree Neither Agree nor Disagree minutes for each claim ( Table 2). Respondents' site of employment had a significant relationship with the average number of prescription coupons processed daily [F(3,228) = 11.3; P < 0.01) with post hoc comparisons using the Tukey test, indicating that respondents in chain community pharmacies processed significantly more coupons each day than those working in independent and grocery pharmacies (22.0 (SD 25.5) vs. 12.2 (SD 11.3) and 4.2 (SD 5.2), respectively). Average time required to processes a single coupon did not differ among respondents' employment sites.
When questioned about their knowledge of prescription coupon privacy practices, 57.5% of respondents believed that it is illegal to "sell patients' information" (i.e., with no distinction made between protected health information and any other information). Another 16.7% of respondents believed the practice was legal, and the remaining 25.8% declined to answer, citing they did not know. Finally, respondents believed patients received prescription coupons most often from the internet or mail (77.1%), their prescribers (62.9%), or from their own community pharmacies (33.3%).

■■ Discussion
To our knowledge, this is the first exploration of community pharmacists' experiences, perceptions, and knowledge of prescription coupons and drug discount cards. These results show that while community pharmacists overwhelmingly hold positive perceptions of prescription coupons and drug discount cards, they are largely uninformed and unaware of the differences between prescription coupons and drug discount cards business practices. Consequently, the lack of transparency for prescription discount processes may contribute to community pharmacists' misbeliefs regarding the overall benefit to and effect on health care. Further, while community pharmacists valued the ability of prescription coupons and drug discount cards to help patients afford medications, they acknowledged that patients can face trouble paying once the discount expires. This finding coincides with previous literature suggesting that such interruptions in care could be associated with limited financial literacy among the populations using these discount cards, and use of coupons complicates chronic care management planning. 16 Similarly, managed care organization representatives overwhelmingly believe that coupons from drug manufacturers increase short-and long-term costs of plan sponsors. 12 Findings from Nemlekar et al.'s study (2013) contrast with our study findings, since community pharmacists were much less likely than managed care organizations representatives to believe that coupons promote the use of expensive brand-name medications when less costly alternatives are available. 12 Variation in respondents' perceptions by practice setting could be explained by their employers' organizational culture and/or the magnitude which coupons affect their employers' business. Whereas independent and grocery pharmacies may provide greater financial transparency to their employed community pharmacists, chain pharmacies may benefit from accepting network drug discount cards by increasing overall market share. 17 Little variation among respondents by practice setting existed, and over half of pharmacists (regardless of practice setting) believed that it was illegal for companies that deal with prescription coupons to sell patient information. However, the legality of how patient information is stored, used, and transferred by prescription coupon and drug discount card businesses and their subsidiaries is not readily transparent. Some "terms of use" and "agreements" that patients agreed to by using prescription coupons allow pharmaceutical companies and/or their affiliates to aggregate, store, and use patient-specific data, even for advertising purposes (i.e., text messages and email). [18][19][20] With prescription claims using a coordination of benefits design, the drug manufacturer and/or its subsidiary company may collect all of the prescription data used to process the claim. This includes protected health information such as the patient's name, prescription drug, days supply, number of refills, cost, prescriber, and pharmacy location data. Like pharmaceutical companies, discount card vendor companies and their subsidiaries may also collect and store patient information when the patient signs up online to receive coupons. 20,21 Some privacy policies for prescription coupons and drug discount cards do acknowledge that personal information is being collected and stored. For example, 1 vendor states that they "use personal information we receive from third parties in order to provide services to you, to run our business (including for the purposes of internal business operations and analytics, to provide, change, market or optimize our services and

Community Pharmacists' Time Spent Processing Direct-to-Consumer Prescription Coupons by Practice Setting a
Over the last decade, the use and prevalence of drug manufacturers' product coupons has become common place; whereas less than 90 medical conditions had available drug discounts in 2009, thousands of digital discount coupons are immediately accessible online and through phone applications today. 26 These discount programs are highly visible to the public, since discount card vendors' network coupons are printed in public media and physically distributed not only among health care sites such as dental and doctors' offices, but also among general, high-traffic venues such as grocery stores and gas stations. Without reforms in prescription drug pricing or e-privacy regulations, it is likely that the use of prescription discount programs will continue to grow.

Limitations
There are several limitations to this study. First, the survey was limited to 1 type of health care professional in 1 state and had a moderate response rate; results may not be readily generalizable to other settings or health care professionals. Results may not be applicable to other areas of the United States, since Connecticut differs from national averages in several social determinants of health, including median income, ethnicity, and education. We believe that these and other social determinants of health would likely affect patients' and community pharmacists' experiences, perceptions, and knowledge of prescription coupons.
Because this study was not randomized, self-selected participants may have held significantly different experiences, perceptions, and knowledge about prescription coupons than the remainder of the eligible study population. Further, because this study sought only pharmacists who directly interfaced with patients using prescription coupons and drug discount cards, responses from unlicensed pharmacists were not sought. Pharmacists in high-level director, manager, and executive leadership positions may allow their licenses to go unrenewed without their need to provide direct patient care, and inclusion of their responses may have provided alternative results, especially given the likelihood that their positions would provide them an in-depth view of prescription coupons' financial impact. Similarly, despite being the predominate force driving the use of prescription coupons, patients' experiences, perceptions, and knowledge were neither sought nor assessed.
In an effort to reduce response burden, little demographic information was collected, and only respondents' practice type could be linked to their answers (i.e., ZIP codes could not be linked to respondents' answers). As such, if other demographic data (e.g., position or years practiced) or practice site data (e.g., weekly prescripion volume) had been collected, a more complete understanding of community pharmacists' knowledge and beliefs could have been gleaned. Similarly, survey items did not require respondents to distinguish between the 2 major types of prescription discount processes (i.e., prescription coupons products) and to communicate and market to you (directly or via third parties)." 22 As such, while no claims-based personal health information is transmitted from the pharmacy directly to the discount card vendor, it appears some discount card vendor companies store similar private information that would normally be covered under the Health Insurance Portability and Accountability Act (HIPAA) during patient enrollment and/or use of vendor websites/mobile applications. 22 Consequently, while HIPAA laws apply to all entities that process medical/ prescription claims, it is unclear how these laws are applicable to or enforced on prescription coupon and drug discount card businesses and their subsidiaries when patients provide their information directly for enrollment purposes, rather than as a coordination-of-benefit design.
Further, according to one of the discount card vendor's terms of use and privacy policies, this company does collect and store data, stating that private health information is protected and not sold. However, no mention of whether information is given away (i.e., to a partner or subsidiary company). 22 Taken collectively with this study's findings, some prescription discount programs' legal disclaimers seem to indicate that information that would normally be covered under HIPAA is stored, used, and transmittable and that community pharmacists may be unknowingly fostering these practices.
Regardless of community pharmacists' personally held perceptions and knowledge regarding prescription discount programs, more than half reported that their employers required them to process coupons. This could be directly related to the business model that underpins vendor network discount cards, which in exchange for lower patient costs drive individuals to participating pharmacies. Alternatively, this finding may be explained because pharmacy employers may believe that coupons must be accepted to encourage business, despite the loss of revenue.
Along these lines, this study found that community pharmacies may endure increases in operational costs and/or lost opportunity costs from processing coupons. Specifically, this study's results can be extrapolated to estimate that an average community pharmacy spends 75 minutes of staff time processing coupon claims per day. This time could otherwise be spent not only on traditional direct revenue-generating tasks (i.e., prescription filling and immunizations), but also on safety, adherence, and wellness initiatives. This finding is particularly evident, as nearly two thirds of pharmacists contacted for this study declined participation citing time restraints.
Further, despite the stance taken by the Centers for Disease Control and Prevention that pharmacists optimize disease state management and should be used as health care providers in the wake of primary care provider shortages, "lack of time" is a leading reason why pharmacists do not provide direct patient care services such as medication therapy management. [23][24][25] and drug discount cards), so items may not have picked up on variation in knowledge and perceptions of these 2 processes within an individual respondent. Likewise, no attempts were made to distinguish between personal health information covered under U.S. health privacy laws and any other data, otherwise not covered under HIPAA. As such, respondentsbeing health care professionals-might have assumed that the item referred only to personal health information.
Survey interruptions, whereby respondents did not complete the last 5 questions, were disregarded. Therefore, the frequency of survey interruptions is not available. It is plausible that pharmacists in less busy stores had higher response rates. Finally, respondents were asked to self-report some average figures, introducing several avenues for error and biases (e.g., social desirability and recall); direct observation and/or retrospective review of prescription filling software timestamp data would have likely been more accurate.

■■ Conclusions
Community pharmacists hold positive perceptions of prescription coupons, believing that coupons improve patient access, if only temporarily. Community pharmacies also spend a significant amount of time processing prescription coupons. As such, future research should identify how prescription coupons and drug discount cards are perceived by patients and prescribers and how they affect cost-and health-related outcomes.