Utilizing a composite number needed to treat (NNT) to evaluate cost per event avoided: a more comprehensive measure of value
Author(s)
Joshua C. Toliver, BA, PharmD, PhD1, Julia Wang, PharmD2, Jigish Bhavsar, PharmD2, Sean D Sullivan, Professor3;
1Novo Nordisk Inc., Data Analyst, Austin, TX, USA, 2Novo Nordisk Incorporated, Plainsboro, NJ, USA, 3Curta Incorporated, Seattle, WA, USA
1Novo Nordisk Inc., Data Analyst, Austin, TX, USA, 2Novo Nordisk Incorporated, Plainsboro, NJ, USA, 3Curta Incorporated, Seattle, WA, USA
Presentation Documents
OBJECTIVES: Number needed to treat (NNT) and cost per event avoided are measures to represent value of treatments. NNT is the number needed to treat to prevent one negative outcome and is estimated as the inverse of Absolute Risk Reduction (ARR). Typically, the primary endpoint of a clinical trial is used for NNT calculations, however this has limitations. This research aims to overcome limitations and give a more comprehensive value of treatment by estimating NNT and cost per event avoided using multiple composite endpoints.
METHODS: Two cardiovascular outcome trials were chosen to assess NNT and cost per event avoided, SELECT (semaglutide 2.4mg) and REWIND (dulaglutide 1.5mg). Cost per major adverse cardiovascular event (MACE), cost per consistent endpoints across both trials (stroke, myocardial infarction, all-cause mortality, hospitalization for heart failure, and hospitalization for unstable angina), and cost per composite event were examined. NNT was calculated as follows; NNT=1/ARR where ARR=control event rate-experimental event rate. NNT was estimated based on the duration of each trial. Cost per event avoided was calculated as NNT*time on product or trial duration*estimated net price.
RESULTS: When assessing SELECT, cost per MACE avoided was $1,662,000 (NNT=67). However, when assessing cost per consistent endpoint avoided this was reduced to $988,200 (NNT=40) and when assessing cost per composite event avoided a reduction to $190,400 (NNT=8) was seen. In REWIND, cost per MACE avoided was $1,884,000 (NNT=72) but when assessing cost per consistent endpoint avoided this was reduced to $1,240,000 (NNT=47) and to $607,900 (NNT=23) when assessing for cost per composite event.
CONCLUSIONS: This research illustrates the limitation of using an NNT focused only on the primary endpoint, as it does not capture the comprehensive benefit of the treatment. When considering the value of a treatment through NNT or cost per event avoided, a composite endpoint should be utilized.
METHODS: Two cardiovascular outcome trials were chosen to assess NNT and cost per event avoided, SELECT (semaglutide 2.4mg) and REWIND (dulaglutide 1.5mg). Cost per major adverse cardiovascular event (MACE), cost per consistent endpoints across both trials (stroke, myocardial infarction, all-cause mortality, hospitalization for heart failure, and hospitalization for unstable angina), and cost per composite event were examined. NNT was calculated as follows; NNT=1/ARR where ARR=control event rate-experimental event rate. NNT was estimated based on the duration of each trial. Cost per event avoided was calculated as NNT*time on product or trial duration*estimated net price.
RESULTS: When assessing SELECT, cost per MACE avoided was $1,662,000 (NNT=67). However, when assessing cost per consistent endpoint avoided this was reduced to $988,200 (NNT=40) and when assessing cost per composite event avoided a reduction to $190,400 (NNT=8) was seen. In REWIND, cost per MACE avoided was $1,884,000 (NNT=72) but when assessing cost per consistent endpoint avoided this was reduced to $1,240,000 (NNT=47) and to $607,900 (NNT=23) when assessing for cost per composite event.
CONCLUSIONS: This research illustrates the limitation of using an NNT focused only on the primary endpoint, as it does not capture the comprehensive benefit of the treatment. When considering the value of a treatment through NNT or cost per event avoided, a composite endpoint should be utilized.
Conference/Value in Health Info
2025-05, ISPOR 2025, Montréal, Quebec, CA
Value in Health, Volume 28, Issue S1
Code
EE155
Topic
Economic Evaluation
Topic Subcategory
Trial-Based Economic Evaluation
Disease
SDC: Cardiovascular Disorders (including MI, Stroke, Circulatory)