In budget-impact analyses, a budget scenario in which the new drug is not in the treatment mix requires estimating the treatment shares over the analysis time hori- zon for the current treatment mix. However, it is equally important to create budget scenarios in which the new drug is in the treatment mix, which requires estimating the treatment shares in the reimbursement-eligible population that are expected for the new drug and for other competing treatments when the new drug is added to the formulary. There are no standard data sources for these estimates. The treatment
65
shares for the new drug are typically based on manufacturer or researcher projec- tions. Alternatively, they can be estimated using observed uptake rates for previously introduced new drugs in the condition or even using diffusion modeling techniques.
Generally, uptake of the new drug will be assumed to increase over time. The mag- nitude of the uptake of the new drug over time will likely depend on the unmet need for treatment and the efficacy and safety of the new drug compared with current treatments. It is important that the assumptions made are both credible to and change- able by the users of the analysis to reflect their own expectations for the new drug.
Care should be taken when estimating uptake for the new drug to ensure it is estimated for the population included in the analysis. If that population is limited to the reimbursement-eligible population and if current treatment share data are avail- able for this population, then the uptake estimates should also refer to the reimbursement- eligible population. However, if current treatment share data are only available for the full condition population, then the budget-impact analysis should focus on the full condition population, and estimates of treatment share for the new drug should also reflect its treatment share in the full condition population even though its use may be restricted to those with specific treatment history or condition characteristics.
There are limited data sources for estimating uptake of a new drug. Uptake will depend on the degree of unmet need and the efficacy and price of the new drug.
Frequently, forecasts developed by the manufacturer of the new drug or estimates by clinical experts are used. Because of the uncertainty in these forecasts, alterna- tive values should be tested in sensitivity analyses. The forecasted uptake will also depend on whether the estimate is for the total condition population or just for the reimbursement-eligible population.
In Box 4.3, we show the relationship between uptake of a new drug in the reimbursement- eligible population and in the total condition population.
Once the estimates of the treatment shares for the new drug have been deter- mined for the chosen population, then the impact of the uptake of the new drug on the shares of the current treatments must be estimated. The clinical data and drug
Box 4.3 Estimating Uptake of New Drug in the Total Condition Population or in the Reimbursement-Eligible Population
A new biologic drug is indicated and reimbursed only for those with severe chronic plaque psoriasis for whom both conventional treatment and tumor necrosis factor-alpha (TNF-α) inhibitor biologic treatment have failed. There are limited treatment alternatives in the reimbursement- eligible population.
The expected uptake in the reimbursement-eligible population is 30%. If the reimbursement-eligible population accounts for only 10% of those with mod- erate or severe disease, then what is the expected uptake in all those with moderate or severe chronic plaque psoriasis?
Expected uptake can be calculated as 30% × 10% = 3%.
4 Estimating the Treatment Mix
indication will generally determine whether the new drug will be added to current treatments, substituted for the current treatments, or a mix of both. If the new drug is indicated for add-on therapy, then the researcher has to determine which drugs it will be added to. Entries in the treatment mix will need to be added to represent all the new drug combinations so that their price and clinical efficacy can be appropri- ately used in the analysis. If the new drug replaces currently used treatments, then estimates from which current drugs the treatment shares for the new drug will be taken will be needed. If there are no clinical or other data to guide these estimates, one possibility is to assume that the treatment share will be taken equi- proportionately from all current treatments. Another possibility, perhaps based on expert opinion, is to assume that it will be taken equi-proportionately from only a subset of current treatments (e.g., only from branded treatments or only from a specific class of drugs in the current treatment mix). Once again, the magnitude of these estimates will vary depending on whether the treatment shares are for the total condition popula- tion or for only the reimbursement-eligible population.
In Box 4.4, we present a hypothetical example of the treatment shares needed for a new drug that is indicated for both monotherapy and add-on therapy as well as a hypothetical example of a new drug taking treatment share from only the subset of current drugs in the same class.
Box 4.4 Examples of Impact of New Drug on Current Treatment Shares New Drug Indicated for Monotherapy or Add-On Therapy
In this hypothetical example, the new drug C is indicated for both mono- therapy and for combination therapy with drug A or drug B. The table below indicates the treatment share data required for the budget-impact analysis.
In this example, both drug A and B composed 50% of the market before the introduction of drug C. Once drug C was added to formulary, it was estimated that the projected uptake of drug C monotherapy would be 10%. The uptake of drug C in combination therapy was estimated at 50% where it was added to
Hypothetical treatment shares for a new drug indicated for both monotherapy and combination therapy
Drug regimen
Current treatment shares without drug C (%)
Treatment shares with drug C (%)
Drug A 50 20
Drug B 50 20
Drug C 0 10
Drug A + Drug C 0 25
Drug B + Drug C 0 25
67
drugs A and B equally. Shares for this new monotherapy and combination therapy were projected to be taken equi-proportionately from drugs A and B.
New Drug in One Drug Class Used to Treat Relapsing-Remitting Multiple Sclerosis
Peginterferon beta-1a is a pegylated interferon indicated for the treatment of relapsing-remitting multiple sclerosis. In this hypothetical budget-impact analysis, we assumed that the treatment share for peginterferon beta-1a would be taken equi-proportionately only from the other interferons currently used to treat relapsing-remitting multiple sclerosis. We assumed that the treatment shares of the drugs in different drug classes indicated for relapsing-remitting
multiple sclerosis would not be affected by the addition of peginterferon beta- 1a to the treatment mix.
However, the budget impact of a new drug is often quite sensitive to the assumptions about from which drugs the treatment share for the new drug is taken. Alternative assumptions about the source of the new drug’s treatment shares should be tested in sensitivity analyses, and the treatment mix should always allow a user to change the default assumptions. Thus, the treatment shares of other drugs for relapsing-remitting multiple sclerosis should be included in the treatment mix to allow for sensitivity analyses where treat- ment share is taken from multiple drug classes and not only the interferons.
Hypothetical treatment shares over time with and without new interferon indicated for relapsing- remitting multiple sclerosis
Drug
2016 with PEG-IFN (%)
2016 without PEG-IFN (%)
Peginterferon beta-1a 5.00 0.00
Delayed-release dimethyl fumarate 25.00 25.00
IFNβ-1a (intramuscular) 3.33 5.00
IFNβ-1a (subcutaneous) 44 μg 3.33 5.00
IFNβ-1b (generic) 0.67 1.00
IFNβ-1b (branded) 2.67 4.00
Glatiramer acetate 20 mg 5.00 5.00
Glatiramer acetate 40 mg 25.00 25.00
Natalizumab 5.00 5.00
Fingolimod 15.00 15.00
Teriflunomide 14 mg 10.00 10.00
Total 100.00 100.00
IFNβ interferon beta, PEG-IFN peginterferon beta-1a 4 Estimating the Treatment Mix