Formulary

Formulary in Medicine Coverage and Prescribing Decisions

A formulary is the approved list of medicines that a payer or provider will fund or stock, and it turns a funding decision into a working rule for prescribers, pharmacists and patients. Each entry records whether a medicine is available, for which patients, at what charge to the patient and after which checks. Health economists study formularies because they decide how a fixed drug budget is spent and change whether people start and stay on treatment. The same word covers an NHS local formulary that follows NICE technology appraisals and a tiered drug list in a United States insurance plan, and the two work in different ways.

What a formulary records beyond the product list

A listing on its own says little about real access. The record attached to each medicine sets the conditions of use, so two formularies that list the same products can produce very different prescribing and spending.

A formulary is one part of a payer's wider coverage policy and, in insurance, of its benefit design. Each inclusion, removal or change of status is a coverage decision.

Local formularies in the NHS

In England, NICE medicines practice guideline MPG1 (Developing and updating local formularies, published 2014 and last updated October 2015) defines a local formulary as the output of processes that support the managed introduction, use or withdrawal of healthcare treatments within a health economy, service or organisation. Decisions are made by formally constituted groups such as trust formulary groups, drug and therapeutics committees and area prescribing committees, which MPG1 asks to document the rationale for each decision, publish formulary information online and run a transparent appeals process.

National appraisal and local formularies are linked by law. Regulation 7 of the NICE (Constitution and Functions) and NHS England (Information Functions) Regulations 2013 says that a technology appraisal recommendation sets a period for relevant health bodies to provide funding, and that this period ends 3 months after publication unless NICE specifies a longer one, for example where training or infrastructure must first be in place. MPG1 accordingly asks local groups to adopt a medicine recommended by a NICE technology appraisal into the formulary automatically within 3 months, where clinically appropriate, and not to duplicate NICE's evidence assessment or challenge its recommendations. MPG1 adds that the NHS Constitution gives patients a right to medicines NICE has appraised positively, and to an evidence-based local decision on medicines it has not.

Local formulary work therefore matters most for medicines and indications outside the NICE technology appraisal programme. For these, MPG1 asks groups to apply standard criteria through a multi-criteria decision tool covering patient safety, clinical effectiveness, cost effectiveness or resource impact, strength of evidence, place in therapy, national guidance and priorities, local health priorities, equity of access and stakeholder views. Biosimilar medicines are named in MPG1 as an example of resource impact, and biosimilar substitution is a frequent formulary topic.

Insurer formularies and Medicare Part D

In United States insurance, HealthCare.gov defines a formulary as a list of prescription drugs covered by a prescription drug plan or another insurance plan offering drug benefits, also called a drug list. Federal rules for Medicare Part D at 42 CFR 423.120 show what a regulated insurer formulary has to include. A Part D formulary must be developed and reviewed by a pharmacy and therapeutic committee with a majority of practising physicians or pharmacists and at least one independent physician and one independent pharmacist free of conflict with the plan and with manufacturers.

The same section requires the committee to base clinical decisions on scientific evidence, including peer-reviewed literature and pharmacoeconomic studies, and to review and approve every prior authorisation criterion, step therapy protocol and quantity limit. Each therapeutic category and class must include at least two drugs that are not therapeutically equivalent and bioequivalent, unless the class has only one drug, or CMS approves the plan's case that only two exist and one is clinically superior. Plans must also run a transition process for enrolees whose current drug is off the formulary or newly restricted.

How economic evidence enters formulary decisions

A formulary group asks two economic questions about a new medicine. The first is whether its health gain justifies its cost compared with what it would displace, which is the question addressed by health technology assessment and cost-effectiveness analysis. The second is whether the payer can absorb the spending in the budget period, which is the job of budget impact analysis. Where a NICE appraisal exists, the first question has been answered nationally and the local task in MPG1 is implementation, and the guideline names budget identification and service design among the barriers that can delay adoption.

Prices complicate both questions. When a managed entry agreement or a confidential discount applies, the list price is not the price the payer pays, so a comparison of listed prices can rank products wrongly. A listing also does not fix uptake, which depends on restrictions, prescriber behaviour and patient charges.

Tiers, cost sharing and the evidence on patient response

A tiered formulary assigns medicines to tiers with different patient charges, usually a low copayment for generics and higher charges or coinsurance for non-preferred brands and specialty drugs. The aim is to move use towards preferred products by changing the price the patient pays, which makes tier design a form of cost sharing.

Huskamp and colleagues compared two employer plans that changed formulary design with comparison groups in the same insurers. In the plan that replaced a single tier with three tiers and raised all copayments, 49% of enrolees initially taking tier 3 statins switched to tier 1 or tier 2 medications, against 17% of the comparison group, and 21% stopped statins entirely, against 11%. In the plan with more moderate changes, enrolees were more likely to switch but not to stop the class.

A 2007 review of 132 studies by Goldman, Joyce and Zheng found that higher cost sharing was associated with lower drug use and worse medication adherence, and for some chronic conditions with more use of medical services, so a pharmacy saving can be partly returned as hospital and outpatient costs.

Worked example of moving a brand statin up one tier

All numbers in this example are illustrative. A plan has 1,000 members taking brand statin B, each filling 12 prescriptions a year. The allowed cost per fill is USD 250 for B and USD 15 for generic G. Before the change B sits on tier 2 with a copayment of USD 30, and G sits on tier 1 with a copayment of USD 10. The plan moves B to tier 3 with a copayment of USD 60. It is assumed that 500 members stay on B, 400 switch to G and 100 stop treatment, and that each member who stops generates USD 600 a year of extra medical care paid by the plan.

Annual plan spending on the class is

$$S = \sum_{j} N_j \, F \, (C_j - K_j) + M$$

where $S$ is the plan's annual spending, $N_j$ is the number of members using product $j$, $F$ is the number of fills per member per year, $C_j$ is the allowed cost per fill of product $j$, $K_j$ is the member copayment per fill for product $j$, and $M$ is the extra medical spending paid by the plan for members who stop treatment.

1. Spending before the change. The plan pays 250 minus 30, or USD 220, per fill of B. Plan spending is 1,000 × 12 × 220 = USD 2,640,000 and member payments are 1,000 × 12 × 30 = USD 360,000, so total spending is USD 3,000,000.

2. Spending after the change. Members staying on B cost the plan 500 × 12 × 190 = USD 1,140,000 and pay 500 × 12 × 60 = USD 360,000. Members on G cost the plan 400 × 12 × 5 = USD 24,000 and pay 400 × 12 × 10 = USD 48,000. Extra medical care costs 100 × 600 = USD 60,000. Plan spending is 1,140,000 + 24,000 + 60,000 = USD 1,224,000, a fall of USD 1,416,000, and member payments rise to USD 408,000, an increase of USD 48,000.

3. Where the plan's saving comes from. The higher copayment paid by the 500 who stay on B gives 500 × 12 × 30 = USD 180,000. Switching to G gives 400 × 12 × (220 minus 5) = USD 1,032,000. Stopping gives 100 × 12 × 220 = USD 264,000. These sum to USD 1,476,000 in drug spending, and subtracting the USD 60,000 of medical care leaves the USD 1,416,000 found in step 2.

4. What happens to total spending. Plan and member spending together fall from USD 3,000,000 to 1,140,000 + 360,000 + 24,000 + 48,000 + 60,000 = USD 1,632,000, a reduction of USD 1,368,000. Of this, USD 1,128,000 (400 × 12 × 235) is due to switching to a cheaper product and USD 240,000 (100 × 12 × 250 minus 60,000) is due to people stopping treatment.

5. Interpretation. The plan's saving exceeds the fall in total spending by USD 48,000, the net extra amount paid by members. The USD 180,000 paid by members who stay on B is a transfer, paid by patients to the plan, and saves no resources. The switching saving is a real efficiency gain if G works as well as B for these patients. The USD 240,000 from stopping comes with a health loss that no budget shows, which a full evaluation would value in QALYs.

Restrictions, substitution and switching

Utilisation management adds conditions to a listing. Prior authorisation requires approval before the payer pays, step therapy requires a trial of a preferred medicine first, and quantity limits cap the amount supplied in a period. Each can steer use towards cheaper or better-evidenced options, and each adds administrative cost and can delay treatment for patients who meet the criteria.

Substitution policies change the product dispensed. Generic substitution and automatic substitution replace a product with an equivalent version of the same medicine, whereas therapeutic interchange replaces it with a different medicine judged clinically comparable. The Part D rules list generic substitution and therapeutic interchange among the utilisation management policies the committee must review. A formulary change that moves a stable patient to another product for reasons other than clinical response is non-medical switching, and its monitoring costs and any loss of disease control belong in the evaluation.

Reading formulary effects in economic evaluation

The perspective of an analysis decides how a formulary change looks. A payer perspective counts the plan's spending only, so moving costs to patients appears as a saving. A health system perspective that includes out-of-pocket costs treats that shift as a transfer, as the worked example shows. Studies that report pharmacy spending alone miss medical cost offsets and health effects.

In decision models, formulary status enters through the available comparators, the eligible population, uptake and treatment sequences. A restriction such as second-line use changes the population treated and so the expected cost-effectiveness, which is why a model built for a formulary decision should reflect the listing as written, because a listing can be narrower than the licensed indication.

Common misreadings

Several readings of formularies recur in policy debate. Each fails against the rules and evidence above.

1. Listing means access. A listed medicine may still require prior authorisation, a failed first-line trial or a high patient charge.
2. A preferred tier signals clinical superiority. Tier placement can reflect net price as well as evidence, and the Part D rules require the committee to consider whether a tier position has any therapeutic advantages in safety and efficacy.
3. Lower pharmacy spending means better value. Part of the saving may be a transfer to patients, and part may return as medical costs or lost health.
4. A local formulary can override a NICE appraisal. In England, a positive technology appraisal must be funded within the set period and MPG1 asks local groups not to challenge it.

Sources

• Goldman DP, Joyce GF, Zheng Y. Prescription drug cost sharing: associations with medication and medical utilization and spending and health. JAMA. 2007;298(1):61-69. doi:10.1001/jama.298.1.61
• HealthCare.gov. Glossary: formulary. Accessed 2 November 2026. https://www.healthcare.gov/glossary/formulary/
• Huskamp HA, Deverka PA, Epstein AM, Epstein RS, McGuigan KA, Frank RG. The effect of incentive-based formularies on prescription-drug utilization and spending. New England Journal of Medicine. 2003;349(23):2224-2232. doi:10.1056/NEJMsa030954
• National Institute for Health and Care Excellence. Developing and updating local formularies. Medicines practice guideline MPG1. Published 14 March 2014, last updated 28 October 2015. https://www.nice.org.uk/guidance/mpg1
• The National Institute for Health and Care Excellence (Constitution and Functions) and NHS England (Information Functions) Regulations 2013, SI 2013/259, regulation 7. https://www.legislation.gov.uk/uksi/2013/259/regulation/7
• US Code of Federal Regulations. Title 42, section 423.120, Access to covered Part D drugs. eCFR, accessed 2 November 2026. https://www.ecfr.gov/current/title-42/section-423.120