Health Affairs, 22, no. 3 (2003): 149-158 doi: 10.1377/hlthaff.22.3.149 © 2003 by Project HOPE	New Online   Senate Health Reform Bill   Rewarding Providers   Public Option Policy Brief   Health Reform & Abortion   Delivery System Reform

This Article Abstract Figures Only Reprint (PDF) Submit a response to this article Alert me when this article is cited Alert me when Comments are posted Alert me if a correction is posted Services E-mail this article to a friend Similar articles in this journal Similar articles in PubMed Alert me to new issues of the journal Add to My Personal Archive Download to Citation Manager Reprints & Permissions Citing Articles Citing Articles via HighWire Citing Articles via Web of Science (21) Citing Articles via Google Scholar Google Scholar Articles by Huskamp, H. A. Articles by Blumenthal, D. Search for Related Content PubMed PubMed Citation Articles by Huskamp, H. A. Articles by Blumenthal, D. Related Collections Medicare Military Medicine Pharmaceuticals

Health Tracking

MARKETWATCH

The Impact Of A National Prescription Drug Formulary On Prices, Market Share, And Spending: Lessons For Medicare?

Abstract

 
Several recent bills in Congress to add a Medicare prescription drug benefit would allow the use of formularies to control costs. However, there is little empirical evidence of the impact of formularies among elderly and disabled populations. We assess the effect of a closed formulary implemented by the Veterans Health Administration (VHA) in 1997 on drug prices, market share, and drug spending. We find that the VHA National Formulary was effective at shifting prescribing behavior toward the selected drugs, achieving sizable price reductions from manufacturers, and greatly decreasing drug spending.

Formularies are lists of drugs that are preferred by a health plan or employer. The level of financial incentives for prescribing from the list varies depending on the type of formulary. "Open" formularies list drugs favored by the organization and serve an educational role. Typically there are no incentives to prescribe from the list. "Closed" formularies are lists of drugs that are available for coverage by the organization. Coverage for medications not on the list may be available only if the physician believes that the drug is clearly preferable and obtains a waiver. An intermediate arrangement is an incentive formulary (such as a three-tier formulary) that provides financial incentives for patients and their physicians to select preferred drugs while still providing some level of coverage for nonpreferred drugs.

In 1995 the Veterans Health Administration (VHA), the largest health care system in the country, established its own pharmacy benefit manager, the VHA Pharmacy Benefits Management Strategic Healthcare Group (hereafter, the VHA PBM).³ The VHA PBM implemented a national formulary in June 1997.

The objective of this paper is to assess the impact of the VHA formulary on market share, drug prices, and outpatient drug spending and to estimate total savings in that spending resulting from the National Formulary. The formulary represents a "real-world" national experiment with the use of a closed formulary for a large population of primarily elderly and disabled people. As a result, this experiment may shed some light on issues that would arise if formularies were used in the context of a Medicare drug benefit.

Tools for cost savings. Formularies typically achieve cost savings in two primary ways. First, most formularies attempt to steer patients toward lower-price products by using financial incentives such as lower copayments for preferred drugs. An extensive literature on the impact of standard drug copayments shows that raising enrollees’ out-of-pocket price results in lower drug use and spending.⁴

Second, formularies can increase bargaining power with drug manufacturers and stimulate price competition among brand-name drugs.⁵ Formularies that provide stronger financial incentives for patients and their physicians to choose drugs preferred by the plan or organization (for example, closed formularies) generally have a greater ability to direct the volume of prescriptions among competing products (or in industry jargon, "move market share") according to price and thus extract larger price discounts from manufacturers.

Only a few empirical studies have assessed the effect of closed or other incentive formularies.⁶ All of these studies focused on privately insured populations of employees and their dependents, and thus generalizability to elderly or disabled populations is limited.

The VHA National Formulary

Top The VHA National Formulary Data And Methods Study Results Discussion NOTES

 
More than 3.8 million veterans received care from VHA facilities in 2000.⁷ The VHA population is older on average (45 percent were age 65 or older, and 37 percent were ages 45–64, when the formulary was implemented in 1997) and has a higher level of disability than most employed populations have.⁸

Standardization. Before the National Formulary was implemented, VHA facilities made local formulary decisions under a decentralized system. As a result, there was much variation across the country in the availability of drugs. The formulary was adopted in part to standardize the pharmacy benefit and increase its portability across the country.⁹ The VHA also felt that a standardized benefit would help to consolidate its bargaining power with drug makers, resulting in lower prices and lower spending. Because either minimal ($2 per prescription) or no copayments are required for outpatient prescription drugs in the VHA, it cannot rely on financial incentives to steer patients toward preferred drugs. However, the VHA’s fixed budget creates general incentives for physicians to be cost-conscious in their prescribing.

Approaches. Like most formularies, the VHA formulary involves a combination of approaches, including closed, open, and preferred contracts. Most drug classes are open. From 1997 through early 2000, only six classes (angiotensin-converting enzyme ([ACE] inhibitors, alpha blockers, HMG-CoA reductase inhibitors, H2 blockers, luteinizing hormone-releasing hormone agonists [LHRHs], and proton pump inhibitors [PPIs]) had been closed, although two of these were subsequently reclassified (Exhibit 1). However, the four classes that remained closed as of early 2000 (ACE inhibitors, HMGs, LHRHs, and PPIs) accounted for approximately 13 percent of all VHA drug spending in 1999.¹⁰

Recommendations about formulary status for particular drug classes are made jointly by the VHA PBM, a panel of physician advisers, and the local formulary leaders for each of the twenty-two regional Veterans Integrated Service Networks, or VISNs, after extensive review of the medical literature for the class considered. Recommendations are made on the basis of safety, efficacy, and cost.¹¹

For closed classes, only a limited number of drugs (usually one or two) in a class are included in the National Formulary (hereafter referred to as "closed drugs"). Although the VHA uses a national formulary, each of the 22 VISNs and 163 medical centers continues to maintain its own formulary. For the closed classes, however, VISNs and hospitals are prohibited from adding any non–National Formulary drugs to their formularies.

Waiver processes. The VHA requires all facilities to use a formulary waiver process to permit the use of nonformulary agents in closed classes if there is a contraindication, adverse reaction, or therapeutic failure of a formulary agent or if a patient has previously responded well to a nonformulary drug and changing drugs is considered risky.¹² Although local waiver processes vary, these requirements allow for much flexibility in physicians’ prescribing decisions. An informal survey of all institutions by the VHA PBM in December 1998 found that 88 percent of waiver requests were approved.¹³ The VHA PBM is just beginning to collect standardized data on formulary waiver requests across the system.¹⁴

Data And Methods

Top The VHA National Formulary Data And Methods Study Results Discussion NOTES

 
Data. We obtained from the VHA aggregate monthly market share data for six drug classes: three classes that remained closed during the study period (ACE inhibitors, HMGs, and PPIs), one class that was initially closed and subsequently reopened (H2 blockers), one class that was initially closed and subsequently reclassified as preferred (alpha blockers), and one class that was designated as preferred throughout the period (CCBs). We also obtained monthly VISN-level price data and aggregate spending data for each drug product in these classes. The study period is January 1995 through July 1999, twenty-nine months before adoption and twenty-six months after. Thus, we have 1,210 (55 months x 22 VISNs) observations for each drug.

Analytic approach. We hypothesized that there would be shifts in prescribing toward the closed or preferred products, resulting in a greater market share for these products; price concessions compared with the preformulary period; and lower outpatient pharmacy spending than would have occurred absent the policies of the National Formulary. Market share was calculated as the number of prescriptions for a particular drug divided by the total number of prescriptions for the class. Using monthly price data, we calculated the average change in price per pill across VISNs for the most popular product form of the closed or preferred agent in each class after the class was added to the National Formulary.

We conducted multiple regression analyses to estimate the impact of the National Formulary on outpatient pharmacy spending per VHA outpatient user for each closed and preferred class, controlling for secular trends in the VHA system and changes in the age and sex composition of patients using the system over time. A separate regression model is estimated for each class, so the unit of analysis is a drug class–VISN–month. The dependent variable in the regression models is the logarithm of outpatient spending per VHA outpatient user for a class.¹⁵ Independent variables include the mean age and proportion female of the VISN’s user population during a given month; dummy variables indicating the closed, preferred, or open status of the specific drugs during a given month; and a set of dummy variables for each of the VISNs to control for time-invariant characteristics of the VISNs and their user populations. Also, a quadratic time trend is included to account for the overall trend in drug-class spending over time.

To estimate savings attributable to the six classes studied, we use the regression model described above to estimate outpatient pharmacy spending per user with and without the National Formulary in place. The estimated savings are the differences between the two estimates multiplied by the number of users in each month, summed over the number of months a drug class was closed or preferred.

Study Results

Top The VHA National Formulary Data And Methods Study Results Discussion NOTES

 
Effect on prescribing patterns. Given the relatively flexible formulary waiver process in place across the VHA system, high compliance with the National Formulary was not assured. Exhibit 2 illustrates the change in market share for one closed class, PPIs. Just before implementation, market share for Prilosec was approximately 84 percent, and market share for Prevacid was approximately 16 percent. In less than six months market share for the selected drug in the class, Prevacid, had reached almost 95 percent. Exhibit 3 illustrates the change in market share for HMGs.

View larger version (11K): [in this window] [in a new window]   EXHIBIT 2 Market Shares In The Veterans Health Administration (VHA) National Formulary For Two Proton Pump Inhibitors (Closed Class), October 1994–July 1999

View larger version (20K): [in this window] [in a new window]   EXHIBIT 3 Market Shares In The Veterans Health Administration (VHA) National Formulary For HMG-CoA Reductase Inhibitors (Closed Class), October 1994–July 1999

Discussion

Top The VHA National Formulary Data And Methods Study Results Discussion NOTES

 
Our results show that the VHA National Formulary has been effective at changing prescribing patterns and moving market share to closed drugs within closed classes despite a flexible waiver process. As a result, the VHA has realized sizable price reductions from manufacturers of drugs in closed classes. It is important to note that these are savings below the Federal Supply Schedule prices, which are themselves among the lowest in the market.¹⁷ According to our findings, the National Formulary was associated with reductions in average outpatient pharmacy spending per out-patient VHA user and estimated savings of greater than $80 million for the approximately two-year period following formulary adoption for the five closed classes studied.

Although the VHA has reviewed dozens of drug classes for possible closed or preferred status, it has focused class closure on a small number of classes, which together account for a sizable share of VHA outpatient pharmacy spending and which include multiple brand-name drugs with relatively high therapeutic similarity. The VHA also periodically reevaluates the formulary status of drug classes on the National Formulary and has reclassified some closed classes to preferred or open status in cases where the PBM and its medical advisers no longer felt that closure was necessary or appropriate to achieve the VHA’s goals.¹⁸

Impact on quality of care. Although the VHA National Formulary has achieved savings in outpatient drug spending, it has nevertheless been politically controversial. Because of the large size of the VHA user population and the levels of compliance with the formulary described above, drug manufacturers may face large losses of market share if their product is not selected by the VHA as the closed agent in a closed class. Manufacturers have opposed the National Formulary, arguing, along with some veterans’ advocates, that it may reduce quality of care by restricting veterans’ access to a full range of drug treatments. The Institute of Medicine concluded that there was no evidence to suggest that the National Formulary had negatively affected quality of care but acknowledged that data available at the time were inadequate to conclusively make this assessment.¹⁹ Peter Glassman and colleagues’ survey of attending physicians in the VHA system approximately one year after National Formulary implementation found that most thought they could prescribe needed drugs (63 percent) and that their patients could obtain nonformulary drugs when needed (65 percent).²⁰ Twenty-nine percent disagreed with the statement that the formulary enhanced their ability to provide high-quality care to their patients, which may reflect generally negative perceptions of cost control techniques that restrict clinical autonomy, as found in earlier work.²¹

Potential for Medicare. Formularies are one potential management strategy for a Medicare drug benefit, and most believe that their use will be allowed under any such benefit passed by Congress. In fact, five recent bills in Congress (House-passed plan H.R. 4954, Rangel plan H.R. 5019, Graham plan S. 2625 and Amendment, tripartisan plan S. 2729, and Hagel plan S. 2736) either allow or require the use of formularies. There is likely to be much more debate in Congress over how such formularies should be designed. For example, two bills require the use of incentive formularies (H.R. 5019 and S. 2625 and Amendment), while the other three do not specify allowable formulary structures.²² Some bills specify a minimum or maximum number of drugs that can receive preferred status, while others are less prescriptive.

If Congress ultimately allows the use of formularies in a Medicare benefit, Congress and the Centers for Medicare and Medicaid Services (CMS) will face decisions similar to those faced by the VHA in designing its formulary. In particular, they must decide how restrictive the formulary should be; how many and which drug classes will be included; and how class boundaries will be defined. These decisions involve clear trade-offs between recognizing heterogeneity in the properties of different drugs and patients’ preferences on the one hand, and the greater cost savings possible from a more restrictive formulary on the other.

The rules set by Congress and the CMS regarding formulary structure in combination with the risk arrangement between the government and the contracting entity that delivers the benefit will no doubt affect the flexibility of any formularies that are developed and the interest of potential entities in participating. Under some legislative proposals, the contracting entity would bear financial risk only for meeting performance standards, while other proposals would require the entity to assume substantial financial risk. If an entity faces much financial risk for benefit costs, it has an incentive to implement a more restrictive formulary and to adopt an inflexible waiver process. If the participation rules also limit formulary restrictiveness, fewer plans may be willing to participate. The formulary approach undertaken in response to the risk arrangement will affect the ultimate costs of the benefit as well as the generosity of coverage for enrollees.

Reasons for caution. Comparisons between the VHA National Formulary and the possible use of formularies in Medicare should be made with caution because of the unique nature of the VHA system. VHA physicians are typically salaried, so they do not have a direct financial incentive to control drug costs, but they may have a greater commitment to working in a system that achieves maximum benefit for fixed resources. A Medicare benefit is likely to require much greater cost sharing on the part of enrollees than required of VHA users. A Medicare incentive formulary with large copayment differences between preferred and nonpreferred drugs might produce effects on utilization patterns similar to those of the VHA’s closed formulary. It is also important to note that the VHA National Formulary has been accompanied by a series of quality improvement activities, including an expansion of pharmacists’ role in patient education, increased use of pharmacist advisers in clinical practice, and programs in which pharmacists monitor adherence to drug treatment guidelines using integrated electronic records systems.

The VHA’s experience suggests that a narrowly focused formulary can lead to substantial changes in practice patterns, sizable price reductions, and decreased drug spending in a population with a large proportion of elderly and disabled people. Further study of the effects of various formulary arrangements on quality of care and medication compliance among the elderly is critical to ensuring that cost savings are not achieved at the expense of patients’ health and quality of care.

Editor's Notes

 
Haiden Huskamp is an assistant professor of health economics, Department of Health Care Policy, at the Harvard Medical School. Arnold Epstein is the John H. Foster Professor, Department of Health Policy and Management, at the Harvard School of Public Health. David Blumenthal is a professor of medicine and health care policy at the Harvard Medical School.

The authors are grateful for the financial support of a subcontract from the Institute of Medicine (IOM), the Alfred P. Sloan Foundation, the Agency for Healthcare Research and Quality (5 P01 HS10803-2), and the National Institute of Mental Health (1 K01 MH 66109). Richard Frank, Roger Herdman, Christine Coussens, and members of the IOM Committee on VA Pharmacy Formulary Analysis offered a multitude of constructive suggestions and advice. Kristina Hanson provided helpful comments on an earlier draft.

NOTES

Top The VHA National Formulary Data And Methods Study Results Discussion NOTES

 

1. K. Levit, et al., "Trends in U.S. Health Care Spending, 2001," Health Affairs (Jan/Feb 2003): 154–164.
2. Scott-Levin, 2000 Managed Care Formulary Audit (Newtown, Pa.: Scott-Levin, 2001).
3. K.W. Kizer, J.E. Ogden, and J.E. Ray, "Establishing Pharmacy Benefits Management in the Veterans Health Care System," Drug Benefit Trends 9, no. 8 (1997): 24–27.
4. For example, see A. Liebowitz, W.G. Manning, and J.P. Newhouse, "The Demand for Prescription Drugs as a Function of Cost-Sharing, Social Science and Medicine 21, no. 10 (1985): 251–277; C.E. Reeder and A.A. Nelson, "The Differential Impact of Copayment on Drug Use in a Medicaid Population," Inquiry 22, no. 4 (1985): 396–403[Medline]; and S.B. Soumerai et al., "Payment Restrictions for Prescription Drugs under Medicaid: Effects on Therapy, Cost, and Equity," New England Journal of Medicine 317, no. 9 (1987): 550–556.[Abstract]
5. R.G. Frank, "Prescription Drug Prices: Why Do Some Pay More than Others Do?" Health Affairs (Mar/Apr 2001): 115–128.
6. See, for example, B.R. Motheral and R. Henderson, "The Effect of a Closed Formulary on Prescription Drug Use and Costs," Inquiry 36, no. 4 (1999/2000): 481–491; S.D. Horn, P.D. Sharkey, and C. Phillips-Harris, "Formulary Limitations and the Elderly: Results from the Managed Care Outcomes Project," American Journal of Managed Care 4, no. 8 (1998): 1105–1113[Medline]; B. Motheral and K.A. Fairman, "Effect of a Three-Tier Prescription Copay on Pharmaceutical and Other Medical Utilization," Medical Care 39, no. 12 (2001): 1293–1304[Medline]; and G.F. Joyce et al., "Employer Drug Benefit Plans and Spending on Prescription Drugs," Journal of the American Medical Association 288, no. 14 (2002): 1733–1774.[Abstract/Free Full Text]
7. "Facts about the Department of Veterans Affairs," May 2001, www.va.gov/pressrel/vafact01.htm (20 February 2003).
8. Ibid.
9. VHA Directive 97-047, 1997.
10. U.S. General Accounting Office, PrescriptionDrugs: Expanding Access to Federal Prices Could Cause Other Price Changes, Pub. no. GAO/HEHS-00-118 (Washington: GAO, August 2000).
11. Institute of Medicine, Description and Analysis of the VA National Formulary (Washington: National Academies Press, June 2000).
12. VHA Directive 97-047.
13. IOM, Description and Analysis.
14. Joseph Canzolino, VHA PBM, personal communication, 9 October 2002.
15. A logarithmic transformation is used to address skewness in the distribution of outpatient spending per user and to facilitate interpretation of the coefficients (that is, as a percentage change).
16. Michael Valentino, associate chief consultant, VHA PBM Strategic Healthcare Group, personal communication, 11 January 2000.
17. A study conducted by the Congressional Budget Office found that federal facilities paid 58 percent of the average invoice price paid by retail pharmacies for 100 brand-name drugs in 1994, compared with 91 percent for hospitals and 82 percent for HMOs. Congressional Budget Office, How Increased Competition from Generic Drugs Has Affected Prices and Returns in the Pharmaceutical Industry (Washington: CBO, July 1998).
18. For example, after a generic version of the popular H2 blocker Zantac became available, the VHA PBM no longer felt that class closure was necessary to stimulate price competition in the class.
19. IOM, Description and Analysis.
20. P.A. Glassman et al., "Physician Perceptions of a National Formulary," American Journal of Managed Care 7, no. 3 (2001): 241–250.[Medline]
21. Ibid; and J.M. Schectman, E.G. Elinsky, and N.K. Kanwal, "HMO Physician Attitudes toward Drug Cost Containment Strategies," HMO Practice 9, no. 3 (1995): 116–119.
22. Health Policy Alternatives Inc. for the Henry J. Kaiser Family Foundation, "Prescription Drug Coverage for Medicare Beneficiaries: A Side-by-Side Comparison of Selected Proposals," July 2002, www.kff.org/content/2002/6053/6053v7.pdf (7 March 2003).

                      What's this?

This article has been cited by other articles:

				M. A. Fischer, C. Vogeli, M. Stedman, T. Ferris, M. A. Brookhart, and J. S. Weissman Effect of Electronic Prescribing With Formulary Decision Support on Medication Use and Cost Arch Intern Med, December 8, 2008; 168(22): 2433 - 2439. [Abstract] [Full Text] [PDF]

				A. B. Frakt, S. D. Pizer, and A. M. Hendricks Controlling Prescription Drug Costs: Regulation and the Role of Interest Groups in Medicare and the Veterans Health Administration Journal of Health Politics Policy and Law, December 1, 2008; 33(6): 1079 - 1106. [Abstract] [PDF]

				J. J. Mongan, T. G. Ferris, and T. H. Lee Options for Slowing the Growth of Health Care Costs N. Engl. J. Med., April 3, 2008; 358(14): 1509 - 1514. [Full Text] [PDF]

				D. P. Goldman, G. F. Joyce, and Y. Zheng Prescription Drug Cost Sharing: Associations With Medication and Medical Utilization and Spending and Health JAMA, July 4, 2007; 298(1): 61 - 69. [Abstract] [Full Text] [PDF]

				N. K. Choudhry and A. S. Detsky A Perspective on US Drug Reimportation JAMA, January 19, 2005; 293(3): 358 - 362. [Full Text] [PDF]

				G. N. Nugent, A. Hendricks, L. Nugent, and M. L. Render Value for Taxpayers' Dollars: What VA Care Would Cost at Medicare Prices Med Care Res Rev, December 1, 2004; 61(4): 495 - 508. [Abstract] [PDF]

				M. M. Mello, D. M. Studdert, and T. A. Brennan The Pharmaceutical Industry versus Medicaid -- Limits on State Initiatives to Control Prescription-Drug Costs N. Engl. J. Med., February 5, 2004; 350(6): 608 - 613. [Full Text] [PDF]

				H. A. Huskamp, P. A. Deverka, A. M. Epstein, R. S. Epstein, K. A. McGuigan, and R. G. Frank The Effect of Incentive-Based Formularies on Prescription-Drug Utilization and Spending N. Engl. J. Med., December 4, 2003; 349(23): 2224 - 2232. [Abstract] [Full Text] [PDF]