Skip to main content
U.S. flag

An official website of the United States government

Dot gov

The .gov means it’s official.
Federal government websites often end in .gov or .mil. Before sharing sensitive information, make sure you’re on a federal government site.


The site is secure.
The https:// ensures that you are connecting to the official website and that any information you provide is encrypted and transmitted securely.

Cost Control for Prescription Drug Programs: Pharmacy Benefit Manager (PBM) Efforts, Effects, and Implications

Publication Date

David H. Kreling, Ph.D., R.Ph.
Sonderegger Research Center
University of Wisconsin School of Pharmacy

A background report prepared for the
Department of Health and Human Services'
Conference on Pharmaceutical Pricing Practices, Utilization and Costs

August 8-9, 2000

Leavey Conference Center, Georgetown University

Washington, DC

Final Version


With increased attention focused on prescription drug programs, expenditures, and coverage, interest also has increased in drug program cost control measures, in particular those of pharmacy benefit managers (PBMs). PBMs have had a central functional role in prescription drug coverage plans built around claims processing and program administration activities.1 A number of strategies and techniques to control cost in drug programs have been used by PBMs. The goal of this background paper is to describe and discuss different methods to control prescription drug costs that typically have been associated with PBMs. The intent of this paper is to be generally useful as a discussion stimulus to raise issues and considerations related to different cost control strategies or techniques.

Throughout this paper, the term PBM will be used generically, to refer to any prescription drug program administrator. Although PBMs technically are private firms that contract with health plans or plan sponsors and specialize in claims processing and administrative functions involved with operating a prescription drug program, HMOs, MCOs, state governments, and others may perform the same or similar functions and use similar methods or strategies for cost control. Thus, general reference to PBM infers others who may be pursuing the same or similar efforts to control costs in a drug program.

Cost control methods in prescription drug programs can be categorized as being targeted to or intending to affect pharmacies or pharmacists, drug manufacturers, prescribers, and/or consumers. Strategies directed toward pharmacies/pharmacists or manufacturers generally are price or discounting oriented, consistent with a theme of volume purchasing that serves as a basic tenet throughout managed care. Strategies directed toward consumers or prescribers can be more complex and with utilization goals in mind, sometimes achieved by manipulating the cost (monetary or non-monetary) incurred when choosing to use (prescribe) a given drug.

A how, what, when, where, why approach will be used in an attempt to describe and discuss several aspects of different cost control strategies or methods. As an initial caveat, it is important to note that although there may be a general awareness of many of the cost control methods, they are used by firms in a competitive marketplace. Competitiveness within the PBM industry limitsdivulging specifics of methods and their outcomes; they often are considered proprietary. Thus, there are limitations in attempting to provide detailed descriptions of many cost control methods, their prevalence, and evaluations of their effects. Our understanding is basedon the available research literature and anecdotal reports.

Two major sources of information about PBMs and what they do are available. Novartis Pharmaceuticals Corporation has sponsored a pharmacy benefit report for the past several years that focuses on HMO pharmacy benefits design, and has included additional data from surveys of employers and PBM pharmacy directors (Novartis 1999). Wyeth-Ayerst has sponsored a similar report based on surveys of employers (Wyeth-Ayerst 1999). Most individual PBMs produce annual reports, some with quite detailed descriptions of drug use and trends (Express Scripts 2000). A discussion of PBMs and their potential role in managing costs for a Medicare drug benefit also has been written (Kaiser 2000b). However, no industry-wide statistical compilation of cost control efforts and their effects by PBMs as an industry is available.

PBM cost control

Several cost control methods, strategies, and approaches used by PBMs are described below. The descriptions will include the technique, target, prevalence, desired effect and evidence, and potential implications. Where available, selected research articles from the literature assessing the effects of the cost control method are highlighted. A summary of different cost control methods and their effect(s) is included in Table 1.

1. Pharmacy Payments/Reimbursement -- Negotiated Prices

The standard approach for pharmacy reimbursement for dispensed prescriptions is to pay an ingredient cost amount for the drug dispensed in the prescription, plus a dispensing fee. How the ingredient cost is determined and the amount of the dispensing fee often varies depending on whether the drug dispensed is a brand name or generic drug. Ingredient cost payment for a brand name drug with an existing patent typically is determined by deducting a percentage from the drug’s average wholesale price (AWP) as an estimate of the pharmacy’s acquisition cost of the drug.2 Ingredient cost for drugs with expired patents (multisource or generic drugs) may be reimbursed the same way as brand name drugs, or based on an amount specified as a maximum allowable cost (MAC) per unit dispensed (tablet, capsule, etc.) that reflects the actual purchase cost in pharmacies.3

Since PBMs reflect aggregate purchases representing all individuals within a drug coverage program, their reimbursement formulas are established to extract volume purchase discounts from pharmacies. Levels of prices paid by PBMs generally are the lowest or some of the lowest accepted by pharmacies for any types of customers. Prices paid by cash paying customers and even Medicaid programs in many states are higher than what a PBM would pay. Thus the PBM pricing approach can be considered a negotiated price, volume discount strategy targeted at pharmacy providers.

This volume discount strategy applies to the network of pharmacies that contract with a PBM to be pharmacy providers. Restricting the network of pharmacies to a selected group of pharmacies (e.g., one chain of pharmacies) is a strategy to achieve deeper pricing discounts from the participating pharmacies. The additional savings achieved by restricting the network is balanced with trying to ensure adequate access for beneficiaries.

Evidence of market levels of PBM payments generally are not obtained directly from PBMs, but from PBM clients (e.g., employers) or similar program administrators (e.g., HMOs with in-house prescription programs), and occasionally from pharmacies. In recent years, the trend in payment rates to pharmacies has seen slight decreases.

In 1996, a study commissioned by the HCFA(now known as CMS) reported that PBM ingredient cost payments tended to fall in the range of AWP less 10 to 15%, with AWP less 13% a popular level (HCFA(now known as CMS) 1996). In addition, the PBM MAC programs tended to be more aggressive than those in state Medicaid programs, both in terms of the number of drugs with MAC amounts established and lower MAC amounts. Typical dispensing fees reported in the HCFA(now known as CMS) study ranged from $1.95 to $4.00, usually with $2.00 to $3.00 fees for brand name drug prescriptions and $3.00 to $3.50 fees for generic drug prescriptions. These AWP discounts and dispensing fees yielded lower payments than the common 10% AWP discount in Medicaid programs and average $4.12 dispensing fee across state Medicaid programs.

Three years later, a report from a survey of 375 employers (PBM clients) showed the average ingredient cost payment discount off of AWP for brand name drugs in 1998 was 13.2%, compared to 12.6% in 1997 (Wyeth-Ayerst 1999). Overall, 22% or respondents reported AWP discounts of 15% or more, and 91% of employers paid AWP less 12% in 1998 compared to 63% of employers in 1995. The average dispensing fees for brand name and generic drug prescriptions in 1998 were $2.35 and $2.44 respectively; the average brand name dispensing fee decreased from $2.50 in1995.

Similarly, the 1999 Novartis Pharmacy Benefit Report provides information on drug ingredient cost and dispensing fee payments for 108 HMOs for 1998 (Novartis 1999). Although theproportion ofHMOs using a PBM for their drug programs was not disclosed, most likely used a PBM to manage the drug benefit; only 16.6% of the respondents were staff/group model HMOs that may have managed their own drug benefit. An overall average 14.3% discount off AWP for drug ingredient cost reimbursement was reported. This average discount exceeded the 1997 and 1996 averages of 14.2% and 13.9% respectively. Dispensing fees for pharmacies in the HMO provider networks averaged $1.98 for brand name drug prescriptions and $2.06 for generic drug prescriptions.

Restricted networks yield additional discounts in prescription ingredient cost payment of one to a few percent and dispensing fee reductions of $0.50 to $1.00 or more (HCFA(now known as CMS) 1996). Given that PBM reimbursement rates are low, additional concessions for restricting the network tend to be at the margin. In addition, many states have any willing provider stipulations that preclude a PBM from excluding as a provider any pharmacy willing to accept the payment terms offered, making the choice of restricting the network up to the pharmacies as much or more so than the PBM. In general, the use of restricted networks is not widespread. Since 1995, about 20% of employers using PBMs have restricted networks for their pharmacy benefit, and the percent is not expected to change in the coming years (Wyeth-Ayerst 1999).

The potential effect of these continued compressions in pharmacy payments should be reduced prices for prescriptions and lower margins for pharmacies. The general response among pharmacies is to focus on economies of scale and production efficiencies in dispensing prescriptions. Another response by pharmacies is to shift costs to other prescriptions where payment is higher or where pricing is not established by an insurance program or payor (i.e., self-pay or indemnity-covered prescriptions). However, the proportion of prescriptions where costs can be shifted is less than one third of all prescriptions and a shrinking proportion. Ultimately pharmacies might not continue to participate as PBM providers and will either close, or redirect their business efforts on more profitable market segments or merchandise and services.

Evidence of the effect of negotiated prices from the research literature is limited and dated. Studies have shown differential revenues and/or cost shifting between payors (Kotzan & Carroll 1991; McMillan et al. 1990). Anecdotal reports of continued downward pressure on margins due to decreased PBM and insured prescription payments appear in pharmacy trade journals and even in chain drugstore annual reports (Walgreen 1999). In the pharmacy press there also are complaints by pharmacists about payment levels and remarks about cost shifting, and hints of contracts potentially being rejected and pharmacies discontinuing participation, but there is no clear evidence from systematic study of these kinds of effects. The recent anecdotal evidence would suggest there is a definite impact of reduced payments to pharmacies. However, expenditures and average prescription prices continue to increase, implying decreased pharmacy payments may not be accomplishing cost saving goals.

Continued price decreases to control cost could have quality implications. With continued pricing pressure, scale economies become paramount in pharmacy operations as a means to survive the economic crunch, and the rate of prescriptions dispensed per hour increases. This results in a corresponding decreased professional contact time, a rushed pace and abbreviated service provision, and has been associated with more dispensing errors, potentially reducing the level of quality of care. If pharmacies discontinue participating in prescription drug programs or are omitted in restricted networks, access to pharmacy services could become problematic for some consumers.

2. Generic Substitution

The goal of generic substitution is to increase the use of generic drugs and have generic drugs dispensed whenever possible. As a cost control strategy, generic substitution is multi-faceted, potentially targeted to consumers, pharmacists, and prescribers. Several different techniques/ methods are associated with generic substitution.

Consumers may face lower cost sharing requirements for prescriptions dispensed with generic drugs. The lower cost sharing may be in the form of different copayment amounts or coinsurance rates for generic and brand name drugs, or having to pay the difference in cost between the generic and brand name version of a drug if the generic is not accepted. Pharmacists can receive higher dispensing fees for generic drugs and can be paid incentives for achieving certain levels of performance in dispensing generic drugs when possible. With maximum allowable cost (MAC) programs, the pharmacy only receives reimbursement in the amount of cost of the generic drug for prescriptions, and would have to absorb the difference in cost between the brand name drug and the generic if the brand name drug was dispensed. Physician prescribing profiles and report cards can be generated to evaluate generic prescribing and identify physicians for potential educational interventions and/or sanctions.

Since the potential savings in drug cost for generic drugs compared to brand name drugs generally is large, generic substitution is common in drug plans. A general trend is toward more emphasis on stimulating or requiring generic drug use, particularly consumer initiatives. One report noted the percent of employers (as clients of PBMs) that always required generics increased from 7% in 1995 to 29% in 1998 (Wyeth-Ayerst 1999). The report also noted that efforts targeted towards pharmacists (e.g., increased dispensing fees for dispensing generic drugs or MAC programs that penalize pharmacists when they dispense a brand name drug when a generic was available) are less successful than imposing incentives on the consumer.

A similar theme appeared in a report focusing on HMOs and PBM drug benefits in HMOs. Both copayment amounts and the differential between brand name and generic prescription copayments increased between 1996 and 1998 (Novartis 1999). At the same time, the difference between brand name and generic dispensing fees for pharmacies narrowed to only $0.08, and it was suggested that differential dispensing fees may have ceased to be a motivating factor for generic dispensing.

As an overall strategy for increasing generic use and decreasing cost, generic substitution has had an impact. The percentage of prescriptions dispensed as generic drugs has grown from 33% in 1993 to 45% in 1998 (Kaiser 2000a). Clearly, this would be associated with cost savings. However, that success is moderated by the fact that the percent of sales and expenditures for generic drugs has decreased in the past few years (Kaiser 2000a; Novartis 1999). Higher prices of new drugs and new introductions likely countered gains in generic use rates to reduce the role of generic drugs overall in the cost picture.

Implications for generic substitution as a cost control strategy generally are favorable, but tempered. Potential savings from high rates of generic drug use as patents expire and generics become available are balanced by adoption of new drugs, typically at higher cost, that may offer therapeutic or quality of life improvements. To the extent that generic drug use can be fostered and maximized without compromising therapeutic or qualitative outcomes, generic substitution strategies should yield favorable cost results. An issue worth watching is whether consumer backlash develops as efforts targeted toward them to stimulate generic use intensify.

3. Manufacturer Price Concessions -- Rebates

Rebates are moneys returned by a seller to a purchaser that are delayed from the sales transaction, and can be considered a negotiated price discounting strategy targeted to drug manufacturers. Manufacturers pay a rebate based upon the amount of the firms products that are dispensed by the pharmacies providing prescription service to beneficiaries or enrollees. The rebates usually are a percent of the value (at the manufacturers transaction price) of a drug dispensed and occur separate from the claims submission/payment cycle as an after market arrangement. PBMs often negotiate and administer the rebates for drug plan sponsors and charge a fee or percent of the rebate for administering the rebates, although some sponsors (e.g., HMOs) may engage directly in the rebate process. They ultimately reduce drug program costs for sponsors.

Rebates may occur merely because the PBM (or health plan or program sponsor) represents a volume purchaser of a manufacturers product. Rebate arrangements also may have some purchase volume or market share requirements associated with them, giving credence to the notion that the discount truly reflects a volume difference. Market share stipulations associated with rebates often are connected to incentives such as formulary inclusion or pharmacist and patient incentives to influence market shares of rebated products.

Evidence of market levels and trends in rebates is sparse. Since rebate procedures, levels, and associated incentives to produce market share are considered proprietary, large scale, systematic evaluations are not common, and only general notions instead of specific amounts and reports tend to be available. In general there has been a trend for rebates more often to be dependent on or related to changes in market share, and thus, rather than rebates on all products, the number of products rebated may be on the decline (Reissman 1997, Tauber 1998). There also has been less of a tendency for rebates on generic drugs.

A report for HCFA(now known as CMS) in 1996 based on in-depth interviews with 8 PBMs reported rebates generally were lower and less universally available than within state Medicaid programs (HCFA(now known as CMS) 1996). The amount of rebate per claim was approximately $1.00 (ranging from $0.80 to $1.80 as typical amounts) representing about 5% (ranging from 3 to 9%) of drug spending (net expenditures for prescription payments after deducting patient cost sharing). A recent report noted that manufacturers have become very aggressive about requiring a measurable increase in market share before making rebate payments to a PBM (Wyeth 1999). Also, in a summary from a presentation at a PBM conference in 1998, average rebates between $0.75 and $1.50 per claim were reported, and that a plan sponsor’s size affects its abilityto negotiate favorable rebate contract terms from the PBM (PBM News 2000).

In response to the Medicaid rebate program, some new forms of rebate have been initiated. HCFA(now known as CMS) regulations require rebates on all products as a requirement for inclusion in Medicaid programs and further that Medicaid receive the best rebates available in the market. If a rebate rate better than that provided to HCFA(now known as CMS) (e.g., the HCFA(now known as CMS) minimum rebate) is provided to another purchaser, the manufacturer must submit an amount to yield an equal level of rebate to HCFA(now known as CMS) for Medicaid. To avoid paying additional rebates, yet provide additional incentives for their products to be used, manufacturers may establish different arrangements that yield economic benefits to the PBMs, such as special project funds, incentives to the PBM for the production of information (e.g., claims data sales), or educational programs (e.g., pharmacist continuing education and disease management programs) (HCFA(now known as CMS) 1996).

Rebates are intended to reduce net drug program costs and their impact can be substantial (HCFA(now known as CMS) 1995). However, some advise caution about focusing on maximizing rebates when the emphasis should be on minimizing total costs (PBM News 2000). For example, a 20% rebate may seem very attractive, yet if that is applied to a $50.00 brand name drug prescription (approximate average brand name drug prescription price), the net price would be $40.00, considerably more than what might occur if a generic drug was dispensed (with an approximate average price of $17.00). If rebates detract from the potential of the overall best cost-effective drug choices, they may lead to false economies.

Consequently, rebates as cost control strategies can produce savings for overall program costs. However, since rebates generally are associated with newer, brand name drugs, they continue to foster a new drug mind set that serves to keep a focus on newer, typically more expensive brand name drugs as our primary therapeutic agents and may lead to less emphasis on overall more cost effective therapies. It also should not be assumed that manufacturers will continue to provide rebates indefinitely and negotiate such arrangements, even with corresponding market share connections. The present rebate levels could be reduced or completely eliminated as a future business decisions by manufacturers.

4. Cost Sharing – Copayments

Cost sharing requirements in prescription drug programs require consumers to pay a portion of the cost of each prescription they obtain. As a cost control effort of PBMs, they are targeted toward consumers in an attempt to shift some responsibility to them for the cost of their utilization, and raise their sensitivity to the cost of their utilization. Effectively, copayment requirements are a component of the benefit structure for prescription drug coverage and thus can vary across health plans managed by a given PBM.4

Copayments require consumers to pay an established fixed dollar amount each time they obtain a prescription. The simplest copayment structure is a uniform copayment (e.g., $5.00) for all prescriptions. Historically, a single, uniform copayment was the original method used in prescription drug programs structured as service benefit (Prescription card) drug programs.5 Although some drug programs still use a single copayment amount for all prescriptions, most plans now have copayments that vary based upon the characteristics of the drug dispensed, i.e., whether the drug is a brand name or generic drug and whether or not the drug is a formulary (and/or preferred) drug.

Differential brand/generic copayments specify higher copayment amounts on prescriptions for brand name drugs and lower amounts on prescriptions for generic drugs. In addition to brand and generic differential copayments, some plans include an additional third tier of copayment for non-formulary (non-preferred) drugs. The third tier copayments are the highest, often sizeably more than the brand name copayment amount, since non-formulary or non-preferred drugs typically are brand name drugs without rebates.6

Research in the literature shows that copayments can reduce the number of prescriptions used and thus have a tempering effect on expenditures, and also that the effect can vary across therapeutic classes of drugs (Nelson et al. 1984; Reeder and Nelson 1985; Reeder et al. 1993). A comparison of copayment with a limit of three paid prescriptions per month for Medicaid beneficiaries showed the copayment was less effective in reducing the number of drugs used (Soumerai et al. 1987). A study investigating relationships between copayments, generic substitution, and prescription costs showed a $3.00 to $5.00 copayment was associated with a 5% reduction in the number of prescriptions, but the average ingredient cost per prescription also increased, offsetting the decrease in utilization (Smith 1993). Other studies have found that copayments can reduce the use of discretionary or non-essential drugs more than essential drugs (Poirier et al. 1998; McManus et al. 1996). One study examined the effect of a copayment change between brand and generic prescriptions. When consumers were required to pay the full cost differential between the cost of the brand name and generic drug as a copayment, the proportion of generic drugs dispensed increased (Ganther 1996). One study reported that three-tier copayments yielded estimated drug cost savings (as a result of shifts to lower cost drugs) of 2 to 5% and savings from additional cost sharing requirements of 4 to 15% (Segedin 1999). However, he reported that estimating expected savings proves difficult because of the multitude of plan designs, usage patterns, formulary structures, and other factors that are present in PBM plans and covered populations. Also, he concluded the effect of three-tiered copayments on the long term growth rate of drug costs is unclear.

Currently, copayments are the most common cost sharing requirement in prescription drug plans. One report noted that approximately 80% of drug coverage plans used by employers required beneficiaries to pay a copayment for prescriptions dispensed in retail pharmacies (Wyeth-Ayerst 1999). Recent trends have been toward increasing copayments for brand and generic drugs, and increasingly adding the third tier copayment for non-formulary drugs. A report of copayments in HMOs showed average copayments for generic (formulary) drugs was $6.17, for brand (formulary) drugs was $9.65, and for brand (non-formulary) drugs was $13.77 (Novartis 1999). In addition, although there was wide variability in copayments, the most common copayments were $10.00 for formulary brands, $5 for formulary generics, and $15 for non-formulary brands. Another survey of employers showed copayments averaging $5.25 for generic drugs, $9.75 for brand formulary drugs, and $20.50 for non-formulary brand drugs (Wyeth-Ayerst 1999). Overall approximately 6% of employers reported using this type of three-tiered copayment system. In contrast, according to HMO sources, approximately 70% of HMOs have implemented a three-tier copayment system, with $5, $15, and $25 copayments respectively for generic, brand, and non-formulary drugs (PBM News 1999). Another report of copayment systems in MCOs showed average copayments for generic, formulary branded, and non-formulary drugs of $6.63, $13.94, and $28.32, respectively (Express Scripts 2000). The percentage of MCOs offering a three-tier copayment structure increased from 36% in the spring of 1998 to 67% in the fall of 1999.

The effect of differential or tiered copayments has been to increase the cost share paid by consumers and thus reduce program costs. One report suggested moving from a $5/$10 brand/generic copayment structure to a $5/$10/$25 three-tier structure could save between 7 and 8% of a overall drug cost for a health plan (Express Scripts 2000). In addition, the lower generic and/or formulary (preferred) drug copayments can help drive drug use to lower cost drugs (generic or rebated). However, although the percent of prescriptions dispensed as generic drugs has increased to approximately 45% of all prescriptions, they represent 20% or less of drug expenditures, and the average prescription price continues to increase (Kaiser 2000a).

The implications of differential copayments can vary. The current differentials between brand and generic copayments (and possibly even tiered copayments) are relatively low and don't always incent consumers to use lower cost alternates, nor decrease an emphasis on the use of brand name drugs. As copayments increase, consumers will assume more of the cost of their prescription drug use. Increased out-of-pocket expenses can affect consumer perceptions of coverage and the quality of their prescription benefit. Conceivably, as their costs rise, consumers may be dissuaded from obtaining important prescriptions or refills, potentially increasingother health care utilization subsequent to failed therapy. If increased copayments drive use to lower cost drugs and away from more costly drugs that may be more effective or more cost effective, higher overall health plan costs may result.

5. Cost Sharing -- Coinsurance

With coinsurance cost sharing, consumers pay a percentage of the cost of each prescription dispensed. The coinsurance percent typically is fixed and does not vary by the type of drug dispensed (brand name, generic, or formulary). As the cost of the prescription dispensed increases, the amount of cost share also increases, thus there is a direct correlation between the drug resource used and the amount of patient out-of-pocket cost share (in contrast to a copayment where the amount of out-of-pocket cost is constant for a given drug type, regardless of the drug cost).

Coinsurance targets consumers as an aspect of drug coverage benefit structure. Since the amount of cost sharing per prescription varies based upon the cost of the resource (drug) used, coinsurance can sensitize consumers to differences in the costs of drugs they use, conceivably driving drug use to lower cost (generic or rebated) drugs. Coinsurance can reduce drug program costs depending on the coinsurance rate (percent) and drug utilization patterns in a drug coverage plan.

Coinsurance has been a traditional feature of indemnity-type insurance programs (often combined with a deductible), but is less common in service benefit prescription drug plans. A report of a 1998 survey of 375 employers using PBMs claimed approximately 30% of employers had a coinsurance requirement for their prescription drug coverage (Wyeth-Ayerst 1999).7 The coinsurance rate for the majority of those employers was 20%, and about one-fourth of the employers with coinsurance requirements had different coinsurance percents for brand name and generic drugs. No research on the effects of PBM coinsurance requirements on drug costs was located in a literature search, most likely because coinsurance is not as common as copayments in prescription card drug plans and only recently has begun to be considered for implementation.

Coinsurance can sensitize consumers more directly to differences in costs of drugs they use. It is the cost sharing approach that mirrors the resource allocation decision processes of uninsured consumers, but helps shield consumers from the full financial burden of their utilization. As the cost of a prescription goes up, the amount paid by the consumer increases. Consumers who use more expensive drugs pay more in absolute dollars than consumers who use less expensive (e.g. generic) drugs. However, the unpredictability of the amount of cost share for each prescription can make coinsurance cost sharing less appealing to consumers and benefit managers comfortable and familiar with copayments. It also is a challenge to apply coinsurance to mail service prescriptions because consumers likely would not be able to determine the cost share amount to remit in advance. Also, similar to increased copayments, if more costly drugs are more effective or more cost effective, coinsurance may drive use to less effective drugs or higher overall health plan costs.

6. Formularies

A simple definition of a formulary is a list of covered or reimbursable drugs. The underlying intent of a formulary is to improve prescribing and drug use quality. Formularies can be established by PBMs, health plans, sponsors, hospitals, and others. They target drug use decision-makers, including both prescribers and consumers, and can be considered a benefit structure component intended to control or influence utilization by specifying which drugs can be used. Formularies can take several forms and be labeled in various ways to reflect the breadth of drugs included and/or ease of access to drugs not listed. An open formulary includes all drugs. A close door restricted formulary only includes or covers listed drugs. Closed formularies may vary in breadth, ranging from including only one select drug within a therapeutic category or drug group to including multiple drugs within a therapeutic category. A preferred or partially restricted/closed formulary specifies the drugs covered, but allows exceptions to the list, usually with increased cost sharing (sometimes labeled incented formularies) or administrative effort (e.g., prior authorization). Drugs in a preferred formulary often are specified as the formulary drug because of a rebate arrangement with the drug’s manufacturer (although clinical parameters are, or are claimed to be, the primary determinants of potential formulary selection and status).

Evidence of the prevalence of different formulary types is mixed. About three-fourths of HMOs have preferred or closed formularies (45% preferred or partially closed and 27% closed) (Novartis 1999). There has been a slow trend toward more preferred or partially-closed formularies. Health care plans that are more closed systems, such as staff model HMOs, have higher rates of closed formularies (36.4%). In contrast, a survey of employers using PBMs revealed most employers (80%) have open formularies, with only 10% having either a closed or preferred (incented) formulary (Wyeth-Ayerst 1999). An explanation postulated in the report for the low use of closed or preferred formularies was that employers value rebates less than unrestricted access and beneficiary satisfaction. Among employers, closed or incented formularies are increasing in popularity but remain a very small proportion overall.

The research literature evaluating formularies is controversial because findings are inconsistent and often contested by proponents or antagonists of formularies. Also, studies focusing on PBM formularies are rare. The relationship among formularies, overall program costs, and therapeutic effects are complex, presenting methodologic challenges in studying formulary effects. A critical evaluation of studies of administrative restrictions in state Medicaid programs found a high prevalence of inadequate research designs, rendering the conclusions of formulary and drug exclusion studies suspect (Soumerai et al. 1993). Also, since rebates and cost sharing requirements often are connected to formularies and the formulary status of a drug, it can be difficult to isolate the formulary effect. In general, research findings from negative formulary efforts and restrictions (i.e., directing use away from a given drug by excluding it from coverage or disallowing its use) have shown unintended effects, particularly use of unexpected alternate drugs or other care (Jang 1989; Horn et al. 1996). However, there also is representation of formularies improving cost and quality of drug use in the literature. Their popularity in PBM drug programs may be a signal of their overall usefulness, real or perceived.

Formularies can be effective in limiting access to certain drugs or driving drug use to the formulary/preferred drugs and improve the quality of drug prescribing. If cost is a criterion for formulary assignment and high cost drugs are omitted, or if the formulary is associated with differential or tiered copayments and/or rebates, it can reduce program costs. At the same time, there may be potential unintended consequences associated with formularies. If changes are made in the drugs that are included or preferred on the formulary, switches in drug therapy can occur. Switches and/or interruptions in therapy have implications for therapeutic outcomes (both good and bad potentially). Formulary changes also generate the possibility of disgruntled prescribers, patients, and pharmacists who have to deal with switching drugs and therapies. A concern about formularies is that if they are driven by cost (rebate) issues, the qualitative/therapeutic outcomes may not be ideal (but, may be hard to assess). Formularies can impede therapeutic outcomes by restricting access to drugs that might be optimum for some patients; the therapeutic response to drugs can be as individual as the patients taking them. Whether and how drugs that are excluded from a formulary can be made available is an important issue in formulary design and execution.

7. Disease Management (DM) Programs

Disease management generally refers to the practice of identifying patients with specific medical conditions and providing intensive care and monitoring of drug use and effects. The goal is to maximize drug therapy effectiveness and outcomes (and minimize total treatment cost of the disease). Appropriate drug use is emphasized based on patient education and compliance with drug regimens. Pharmacists may be paid separate professional service fees for their efforts in disease management. The target of disease management programs is the consumer, through the efforts of pharmacists providing educational and therapeutic interventions for patients and prescribers.

Since PBMs typically do not have physical contact with consumers, the most common form of intervention used by PBMs in disease management programs is mailing educational materials to patients (Wyeth-Ayerst 1999). In addition, the PBMs may monitor drug utilization to ensure the targeted patients comply with desired therapeutic regimens.

Disease management programs are a relatively new service offering from PBMs. In 1996, PBMs were developing, piloting, or initiating these programs (HCFA(now known as CMS) 1996). Results from a survey of PBMs, HMOs, and employers found that 75% of PBMs offered disease management programs, and 16% and 53% of HMOs and employers respectively used the PBM programs in 1998 (Novartis 1998). Overall, 76% of HMOs reported having disease management programs in place and increasing participation in DM programs, particularly in asthma, diabetes, congestive heart failure, gastrointestinal disorders, and hypercholesterolemia.

The potential result from disease management programs is improved drug use and compliance by patients leading to enhanced health outcomes for patients and better disease control, with subsequent reduced overall health care spending for patients managed. Conceptually, this idea is appealing.

Increased drug costs can occur as a result of increased utilization from improved compliance with drug regimens. Because DM programs focus on maximizing outcomes, the therapeutic enhancements offered by new drugs are emphasized, and thus use of newer drugs increases.

Implications for disease management programs are that better outcomes can reduce overall health care costs. However, PBMs typically do not have access to other health care cost information, thus determining the impacts other than in the drug cost area is difficult. Also, the causal link of changes in overall health care costs to the disease management interventions can be weak. Critics of disease management programs contend that they merely are veiled efforts to increase use of manufacturers products since increased compliance and utilization occur. Disease management also has been criticized for focusing on only one disease at a time, and not using a more holistic approach that considers comorbidities and multiple conditions in the treatment plan.

Few evaluations of the effects of disease management programs were found in the literature. Most articles described how to establish and implement a disease management program, including advice on how they can be structured to avoid anti-kickback statute violations resulting from the involvement of pharmaceutical manufacturers in the disease management programs. One study assessed the opinions of medical directors towards disease management programs at their MCOs (Algozzine et al. 1998). In most (70%) of the 61% of MCOs with disease management programs, the medical directors were satisfied with the programs and there was agreement that improved outcomes and decreased health care costs could result. A majority of the medical directors also agreed that an independent consultant could help analyze their MCO’s prescription and medical data, and that they would be willing to accept grants or funds to initiate and support an independent program at their MCO.

8. Mail Service Prescriptions

Within PBM plans, consumers from local pharmacies or from a mail service pharmacy can obtain prescriptions. Consumers may be encouraged or required to use mail service pharmacies for obtaining prescriptions, especially prescriptions for long term, chronic therapy. Mail service pharmacies generally offer deeper discounted pricing for prescription dispensing and thus are a desirable cost control strategy (Wertheimer and Andrews 1995). Mail service pharmacies also can provide convenience and lower cost to consumers. Larger quantities of drug are dispensed and although a higher copayment typically is charged, it usually is lower than the multiple of monthly copayments that would occur for the same quantity of drug obtained via community pharmacy dispensing. In spite of lower cost sharing by consumers, the low price received from a mail service pharmacy results in lower total drug program spending by the sponsor.

Mail service prescriptions are a popular component of drug programs and practically a universal offering of PBMs. The proportion of health plans (HMOs) and employers that include mail service in the drug benefit is increasing (Novartis 1999, Wyeth-Ayerst 1999). Consequently, the percent of prescription utilization and total drug budget for mail service prescriptions are increasing. For HMOs, mail service prescriptions represented 5.6% of prescriptions and 8.8% of the total drug budget in 1998 (Novartis 1999).

Although considerable debate about the pros and cons of mail service prescriptions has occurred in the trade press and literature, research studies on mail service pharmacy are limited. Few empiric studies comparing the cost and quality of mail vs. retail pharmacy providers are available. In 1989, the HCFA(now known as CMS) commissioned a study to evaluate the use of mail service pharmacies that concluded that the estimated average charge per day supply was not substantially different from that for retail pharmacies (HCFA(now known as CMS) 1989). The results were very sensitive to assumptions about average prescription prices and average days supply and were challenged and largely discredited by the mail service pharmacy industry. Concerns about mail service pharmacies that have appeared in the literature generally are voiced by traditional community pharmacists that perceive mail service as a threat. Problems associated with mail service pharmacy include shortcomings in professional services available, lack of face-to-face communication and patient consultation, consumers being forced into receiving medications by mail, delays in receipt of medications, and physical and chemical stability and integrity of mailed drugs (Ghoshal 1996-97; Hadzija & Shrewsbury 1999). From a consumer perspective, research has found patrons of mail service pharmacies are satisfied with the services they receive, and specifically with financial aspects and technical quality of the services (Birtcher and Shepherd 1992; Johnson et al. 1997). The debate (and rhetoric) about pros and cons of mail service prescriptions will continue and it is likely that definitive studies of the impact of mail service prescriptions on costs (and quality) will remain elusive. The growth and market success of mail service firms suggest there are some real or perceived advantages as a cost control (or quality) strategy.

Many of the implications of mail service pharmacy are qualitative. Mail service prescriptions can provide convenience due to the home delivery aspect of the service. However, planning for delays between ordering and delivery is necessary to avoid gaps in therapy. Using mail service for long-term therapy prescriptions and community pharmacies for acute prescription needs can result in disjointed or incomplete patient prescription profiles. The extent to which a long-distance patient/provider interaction influences compliance and appropriate drug use and/or outcomes is unknown. In addition, the balance between decreased cost sharing, larger quantities dispensed, and lower prices can be delicate.

An additional issue is ownership of mail service pharmacies by PBMs and the potential to steer prescription business to themselves through their mail subsidiaries. Beyond this, Merck-Medco represents a fully vertical integrated channel of distribution from manufacturer to PBM to mail service pharmacy, introducing the possibility of influence and control in the market at several levels.

9. Drug Utilization Review8

Drug utilization review (DUR) is a process of systematically evaluating drug use to identify and then intervene to correct drug use problems. One goal of DUR can be to reduce costs associated with inappropriate prescribing and use of drugs. DUR can be retrospective, reviewing past claims and utilization for patterns of inappropriateness. When patterns are identified, efforts are made to intervene to change future prescribing or drug use behaviors and yield enhanced quality and/or lower cost. Alternatively, the DUR can be concurrent with the prescription dispensing process, whereby the appropriateness of the drug prescribed is assessed relative to other medications the patient is taking that are included in the PBM (or pharmacy) profile/history of the patients drug use. Concurrent DUR relies on computerized algorithms to check for drug interactions, duplications, or contraindications with the patient disease state or condition, plus screen for potentially errant drug regimen adherence (over/under consumption). Since concurrent DUR occurs during the dispensing process, via computerized linkage and alerts to the dispensing pharmacist, it has the potential of avoiding drug use problems instead of establishing actions based on previous patterns to adjust/correct future prescribing and use problems. Concurrent DUR generally is part of an on-line claims adjudication process.9

Nearly all PBS offer retrospective and concurrent DUR as a service, and they are used by a majority of clients. Concurrent DUR is more popular than retrospective DUR by both HMO and employer clients of PBS, with increasing use in the past several years. The percent of employers using PBM concurrent DUR increased from 65% in 1996 to 76% in 1998 (Wyeth-Ayerst 1999). Among HMOs, POS DUR hard edits (alerts that cannot be overridden by the dispensing pharmacist) increased between 1996 and 1997, and the most common alerts were for early refill, nonformulary notifications, drug interactions, prior authorization notices, and to adjudicate pharmacy reimbursement (Novartis 1998).

Although many reports of DUR and DUR programs appear in the literature, most are from institutional settings (hospital or nursing home). Reviews of DUR studies in outpatient settings have shown both quality and cost improvements (Kreling & Mott 1993; Kozma et al. 1993). Other studies have shown the impact of interventions targeted toward prescribers and/or pharmacists (Sleath et al. 1997; Mott et al. 1998). An assessment of concurrent DUR linked to a mail order database reported a high number of alerts that were acted on by calls to physicians to resolve the problems (Monane et al. 1998). The response of pharmacists to DUR messages also has been studied. Although interventions were not performed for all DUR messages, those related to medication overuse and drug interactions were the most useful, and the usual action by pharmacists was telephoning physicians and talking to patients (Armstrong & Markson 1997).

To the extent that DUR can avoid interactions and duplication, it can be a cost savings effort. Concurrent DUR also can be used to alert pharmacists to potential switches to formulary drugs, and thus steer drug use to preferred or less costly agents. If under-utilization is corrected subsequent to DUR, drug program costs can increase, with potential paybacks in other health care cost areas if the disease is better controlled and other costs avoided. If over-utilization is identified and averted, it

can save drug program costs (but sometimes also raise unintended access problems, if the computer system does not reflect dosing changes and increased use).

Concerns have been raised about concurrent DUR and alert messages. For example, alert messages requesting or requiring pharmacists to make changes in ordered drugs to comply with formulary, particularly if they foster a switch in a current therapy have been criticized. The number and level of importance of alerts can vary, and it is a challenge to automate the judgmental processes involved in deciphering the triviality or urgency of alerts. Responding to alerts often requires time and effort for pharmacists, patients, and prescribes, interrupting the normal flow of activities. In spite of concerns, on balance, most probably would agree that DUR has been a positive component of PBM drug programs.


Cost control strategies and PBS would not use methods if there wasn't some belief that they could be effective. Theory and logic suggest the techniques used should work, and some successes have been supported empirically. However, prescription expenditures continue to increase and have been the most rapidly growing component of health care expenditures in recent years. This indicates that cost control mechanisms are not as successful as desired, or that successes are overshadowed by other factors and forces.

For example, generic substitution policies have increased the proportion of prescriptions dispensed as generic versions of brand name drugs and thus had a favorable impact on costs. However, although the percent of prescriptions dispensed as generic drugs has increased to nearly half of all prescriptions, the percent of expenditures for generic drugs has declined in recent years. The use of higher priced new drugs has attenuated the impact of generic drugs in the overall cost picture.

Cost control efforts targeted at pharmacy providers or aimed at concentrating use with low cost providers (e.g., mail service pharmacies) have fostered the development of an efficient distribution-oriented practice of pharmacy with marginalized and minimalized emphasis on critically assessing appropriate (cost-effective) drug selection and well informed, rational drug use behaviors by consumers. At the extreme, access to pharmacists can be reduced, both literally through pharmacy closures and restricted networks, and effectively by making pharmacists too busy with mechanical prescription processing activities to be available for patient (or prescriber) questions, drug use education, or critical assessments of how and why drugs are used uniquely by patients.

On the surface, rebates on pharmaceutical products have reduced drug costs for plan sponsors. However, rebates typically are provided for new, brand name drugs and can distract attention from the potential of older, low cost generic drugs as alternate, effective therapies. At the same time, a paucity of well designed comparisons of therapeutic and quality of life outcomes for older (generic) versus newer treatments for diseases limits our vision of the potential for successful increased emphasis and use of older drugs. Knowing when and where new versus old treatments really are necessary is a challenge in prescribing decisions where science and art often are combined. Formularies might contribute to improved decisions about which drugs to use, but since rebates are a consideration in formulary decisions and preferred drug designations, the real potential of formularies may be diluted.

Since prescription drugs are consumer goods, making consumers more cost conscious purchasers is a potential strategy. There seems to be an awareness and appreciation for this in PBM cost control efforts, and perhaps an increased attention to this strategy. The development and implementation of three tiered copayments and adoption of coinsurance (instead of copayments) are movements that more fully engage consumers in the economic reality and implications of their drug use. Whether consumers are or can be equipped to make good choices regarding their health and drug use is a fundamental question and concern. There is a delicate and critical balance between avoiding unnecessary use but not avoiding necessary use.

The reasons for growth in drug use and expenditures are complex. Techniques that will empower consumers to be better informed decision makers about the need, appropriateness, and cost of their drug resource use may be better at limiting drug expenditure expansion. However, that likely will require different efforts and different targets than the traditional cost control methods PBMs have used.

Table 1: Summary of PBM Cost Control Strategies and Effects
Technique Target Effect(s) Comments
Negotiated Prices

(reduced reimbursement)

Pharmacies +/- Price for a given prescription is reduced;

May reduce access to pharmacies if network restriction is required to extract discount;

Increased efficiency (speed) in dispensing may reduce patient contact and de-emphasize evaluation of the prescription and drug use for appropriateness and cost savings

Generic Substitution

(increased dispensing fees for generic drugs, MAC programs, and/or dispensing rate incentives)

Pharmacies ++/? Decreased cost (increased dispensing of generic drugs and the difference in drug costs generally exceeds amounts spent on higher dispensing fees and/or incentives);

High rates of generic dispensing on substitutable prescriptions (from MAC and incentives), but limited by extent of prescribing for substitutable (multisource) drugs;

Differential fees or incentives too small to motivate serious pharmacist effort?

Generic Substitution

(differential copayments, lower for generic drugs)

Consumers +++/? Increased acceptance and use of generic drugs;

More program cost paid by consumers;

Relatively low difference in copayments attenuates response?

Low difference in copayments can reduce consumer awareness of real cost of brand name vs. generic drug use

Rebates Pharmaceutical Manufacturers +/- Ultimately lower program cost for rebated drug;

When combined with other incentives, increased use of rebated drugs;

May overlook total cost picture since brand name drugs are emphasized for rebates


(increased copayment amounts overall)

Consumers +/? Increased consumer sensitivity to their utilization;

Shifts additional cost to the consumer from the sponsor for each prescription;

Increased out-of-pocket expenses can affect consumer perceptions of coverage and quality of their prescription benefit


(differential brand/generic copayments)

Consumer +++/? (See differential copayments under generic substitution above)

Increased acceptance and use of generic drugs;

More program cost paid by consumers;

Low difference in copayments attenuates response and reduces awareness of real cost of brand name vs. generic drug use?


(three-tiered: generic vs. brand vs. non-formulary/non-preferred)

Consumer +++/- Increased use of formulary or preferred drugs;

More program cost paid by consumers;

Increased consumer awareness of the cost of their drug use;

Continues emphasis on brand name (rebated/formulary) drugs;

Increased consumer cost can reduce use of high cost drugs -- if they are more effective or cost-effective can yield higher overall costs

Coinsurance Consumer +++/- Increased consumer awareness of the cost of their drug use -- direct relationship between resource use and out-of-pocket cost;

More program cost paid by consumer (depending on coinsurance rate);

Unpredictable consumer cost may not be accepted;

Consumers may avoid high cost drugs when they may be needed (and? more effective or cost-effective overall);

Impractical for mail service dispensing -- cost share unable to be determined in advance

Formularies Consumer

(and prescriber/ pharmacist)

+/- Increased use of desired drugs (formulary/preferred);

Decreased program costs are possible if associated with differential cost sharing or rebates;

Can retain a focus on brand name drugs and divert from total cost picture;

If restrictiveness reduces access to cost effective drugs, can increase overall costs;

Changes in formularies can cause therapy interruptions/switches and require efforts by pharmacists and prescribers

Disease Management Programs Pharmacist and Consumer (and prescriber) +/-? Increased appropriateness of drug use (and decreased overall health care cost?);

Can increase drug use and drug program cost;

Increased pharmacist effort (and patient time) and expense

Mail Service Prescriptions Consumer +/- Decreased cost for prescriptions (deeper discounts, but may be balanced by increased quantities dispensed and decreased consumer cost sharing requirements);

Increased convenience for consumers;

Delay in receipt of prescriptions can cause therapy interruptions;

Decreased direct pharmacist interaction and patient consultation

Drug Utilization Review (DUR) Pharmacist, Prescriber, and Consumer ++/-? Increased appropriateness of drug use (decreased duplications, interactions, increased formulary/preferred drugs);

Can decrease access and interrupt therapy (if overutilization screens are too sensitive);

Increased pharmacist time to respond to alerts and prescriber time for therapy changes


Algozzine T, Pannone R, Kozma CM. Opinions of Disease Management Programs Among Medical Directors of Managed Care Organizations: American Journal of Health-System Pharmacy 55 (15 May 1998): 1029-1033.

Armstrong EP and Markson TJ. On-line DUR Messages: Pharmacists Attitudes and Actions in Response Journal of the American Pharmaceutical Association NS37 (May-June 1997): 315-320.

Birtcher KK and Shepherd MD. Users. Perceptions of Mail-service Pharmacy American Pharmacy NS32 (December 1992): 35-41.

Express Scripts. 1999 Drug Trend Report (June 2000).

Ganther JM. The Effect of a Change in Prescription Drug Insurance Benefits on Consumer Purchase of Generic Prescription Drugs unpublished MS thesis, University of Wisconsin- Madison (1996).

Ghoshal S. Mail-order Pharmacy: Good or Evil? Canadian Pharmaceutical Journal 129 Suppl (December-January 1996-1997): 11-15.

Hadzija BW and Shrewsbury RP. Stability of Four Commercial Products Delivered by Mail Order International Journal of Pharmaceutical Compounding 3 (January 1999): 59-63.

Health Care Financing Administration(now known as Centers for Medicare and Medicaid Services(CMS)). Assessment of the Impact of Pharmacy Benefit Managers Final Report by Kreling DH, Lipton HL, Collins T, and Hertz KC US Department of Health and Human Services, HCFA(now known as CMS) [NTIS pub # PB97-103683] (30 September 1996).

Health Care Financing Administration(now known as Centers for Medicare and Medicaid Services(CMS)). Impact of Medicaid Drug Rebate Program, Extramural Research Report, US Department of Health and Human Services, HCFA(now known as CMS), Office of Research and Demonstration, contract # 500-92-0022 (August 1995).

Health Care Financing Administration(now known as Centers for Medicare and Medicaid Services(CMS)). Study to Evaluate the Use of Mail Service Pharmacies Final Report by Horgan C, Goody B, Batten H, Boren J, et al.,US Department of Health and Human Services, HCFA(now known as CMS), grant #88-C98526/1-05 (21 September 1989).

Horn SD, Sharkey DM, Tracy DM, Horn CE, James B, Goodwin F. Intended and Unintended Consequences of HMO Cost-containment Strategies: Results from the Managed Care Outcomes Project American Journal of Managed Care II (March 1996): 253-264.

Jang R. Medicaid Formularies: A Critical Review of the Literature Journal of Pharmaceutical Marketing & Management 2 (Spring 1988): 39-61.

Johnson JA, Coons SJ, Hays RD, Sabers D, Langley PC, et al. Comparison of Satisfaction with Mail versus Traditional Pharmacy Services Journal of Managed Care Pharmacy 3 (May-June 1997): 327-337.

Kaiser Family Foundation. Prescription Drug Trends: A Chartbook by Kreling DH, Mott DA, Wiederholt JB, Lundy J, Levitt L. (July 2000a) [].

Kaiser Family Foundation. The Role of PBMs in Managing Drug Costs: Implications for a Medicare Drug Benefit by Cook A, Kornfield T, Gold M. (January 2000b) [].

Kotzan JA, Carroll NV. Why Medicaid Pays More: Comparison of Private Payment and Medicaid Prescription Prices, Quantities, Product Mix, and Dosage Forms Journal of Research in Pharmaceutical Economics vol. 3, no. 2 (1991): 39-56.

Kozma CM, Schulz RM, Dickson WM, Dye JT, Young TL, et al. Economic Impact of Cost-Containment Strategies in Third Party Programs in the U.S. Part 2" Pharmacoeconomics 4 (September 1993): 187-202.

Kreling DH and Mott DA. Cost Effectiveness of Drug Utilization Review in an Outpatient Setting Pharmacoeconomics 4 (December 1993): 414-436.

Lipton HL, Kreling DH, Collins T, and Hertz KC. Pharmacy Benefit Management Companies: Dimensions of Performance Annual Review of Public Health, 20 (1999): 361-401.

McManus P, Donnelly N, Henry D, Hall W, Lindner J, et al. Prescription Drug Utilization Following Patient Copayment Changes in Australia Pharmacoepidemiology & Drug Safety vol. 5, no. 6 (1996): 385-392.

McMillan JA, Carroll NV, Kotzan JA. Third-party-associated Cost-shift Pricing in Georgia Pharmacies Journal of Research in Pharmaceutical Economics vol. 2, no. 4 (1990): 33-47.

Monane M, Matthias DM, Nagle BA, Kelly MA. Improving Prescribing Patterns for the Elderly through an Online Drug Utilization Review Intervention: System Linking the Physician, Pharmacist, and Computer JAMA 280 (14 October 1998): 1249-1252.

Mott DA and Collins TM. Comparison of Targeting a DUR Letter Intervention at Pharmacists Separate From and in Addition to Physicians Journal of the American Pharmaceutical Association 38 (May-June 1998): 325-332.

Nelson AA, Reeder CE, Dickson WM. Effect of a Medicaid Drug Copayment Program on the Utilization and Cost of Prescription Services Medical Care 22 (August 1984): 724-735.

Novartis Pharmacy Benefit Report: Facts & Figures, 1999 Edition. Novartis (July 1999).

Novartis Pharmacy Benefit Report: Trends & Forecasts, 1998 Edition. Novartis (April 1998).

PBM News, Q & A vol. 4, no.1 (Spring 1999) [].

PBM News, Conference Presentations vol. 5, no.2 (Spring 2000) [].

Poirier S, LeLorier J, Page V, Lacour A. Effect of a $2 Co-payment on Prescription Refill Rates of Quebec Elderly and its Relationship to Socio-economic Status: Canadian Pharmaceutical Journal 130 (December-January 1997-1998): 30-31, 33-34.

Reeder CE and Nelson AA. Differential Impact of Copayment on Drug Use in a Medicaid Population Inquiry vol. 22, no. 4 (1985): 396-403.

Reeder CE, Lingle EW, Schulz RM, Mauch RP, Nitengale BS, et al. Economic Impact of Cost-containment Strategies in Third Party Programs in the U.S. (Part I)@ Pharmacoeconomics 4 (August 1993): 92-103.

Reissman D. Are Rebates Going Away Drug Benefit Trends 9 (August 1997): 12-13.

Segedin DA. Three-tier Co-payment plans: Design Considerations and Effectiveness Drug Benefit Trends 11 (September 1999): 43-52.

Sleath B, Collins T, Kelly HW, McCament-Mann L, Lien T. Effect of Including Both Physicians and Pharmacists in an Asthma Drug Use Review Intervention American Journal of Health-System Pharmacy 54 (1 October 1997): 2197-2200.

Smith DG. Effects of Copayments and Generic Substitution on the Use and Costs of Prescription Drugs Inquiry vol. 30, no. 2 (1993): 189-198.

Soumerai SB, Avorn J, Ross-Degnan D, Gortmaker S. Payment Restrictions for Prescription Drugs Under Medicaid: Effects on Therapy, Cost, and Equity: New England Journal of Medicine 317 (27 August 1987): 550-556.

Soumerai SB, Ross-Degnan D, Fortess EE, Abelson JA. AA Critical Analysis of Studies of State Drug Reimbursement Policies The Milbank Quarterly vol. 71, no. 2 (1993): 217-252.

Tauber B. Managed Care Rebates: Their Role in Drug Benefit Cost Control Drug Benefit Trends 10 (February 1998): 26-28.

Walgreen Company. Walgreens 1999 Annual Report (18 November 1999): p. 20.

Wertheimer AI and Andrews KB. Economics of Mail Service Pharmacy Pharmacoeconomics 7 (February 1995): 111-118.

Wyeth-Ayerst Prescription Drug Benefit Cost and Plan Design Survey Report, 1999 Edition, Pharmacy Benefit Management Institute, Inc. (1999).


1 For a description of PBMs, see Lipton et al. Pharmacy Benefit Management Companies: Dimensions of Performance Annual Review of Public Health 20 (1999): 361-301 or the proceedings of a 1996 roundtable on PBM issues sponsored by the Tufts Center for the Study of Drug Development published in the Journal of Pharmaceutical Marketing & Management 12 (1998).

2AWP is the price established by the manufacturer as a suggested list price for wholesalers selling the drug to pharmacies. Wholesalers sell the drugs to pharmacies at prices discounted from these list AWPs.

3 Note: MACs historically were established by HCFA(now known as CMS) for state Medicaid programs, and now most states establish their own MAC amounts and which drugs they set MACs on. PBMs and private insurance plans also establish MACs, often borrowing from or building on MACs for state Medicaid programs.

4Plans with no cost sharing requirements can be considered special copayment cases where the copayment amount is zero, thus no responsibility for or sensitivity to cost is shifted to the consumer.

5In service benefit drug programs, payment for the prescription is made directly to the pharmacy provider and the consumer only pays the cost sharing requirement, if any, at the time of dispensing. With indemnity prescription drug coverage, consumers pay for the prescription out-of-pocket and are reimbursed by the insurer. Typically indemnity coverage includes a deductible, and then coinsurance cost sharing after the deductible is met.

6 Some three-tiered copayment plans have a separate copayment for generic drugs, brand name drugs for which a generic drug is available, and brand name drugs without generics.

7 It is unclear what proportion of this 30% of coinsurance are indemnity or service benefit plans. It likely is a combination of both.

8Drug utilization review (DUR) also can be referred to as drug use evaluation (DUE).

9 Concurrent DUR can be referred to as prospective DUR (PRO-DUR) or point of sale (POS) DUR.