Years-needed-to-treat to add 1 year of life: a new metric to estimate treatment effects in randomized trials

Wayne C Levy; Dariush Mozaffarian; David T Linker; Kenneth W Kenyon; John G F Cleland; Michel Komajda; Willem J Remme; Christian Torp-Pedersen; Marco Metra; Philip A Poole-Wilson; COMET Investigators; Wayne C Levy; Dariush Mozaffarian; David T Linker; Kenneth W Kenyon; John G F Cleland; Michel Komajda; Willem J Remme; Christian Torp-Pedersen; Marco Metra; Philip A Poole-Wilson; COMET Investigators

doi:10.1093/eurjhf/hfn048

Years-needed-to-treat to add 1 year of life: a new metric to estimate treatment effects in randomized trials

Wayne C Levy, Dariush Mozaffarian, David T Linker, Kenneth W Kenyon, John G F Cleland, Michel Komajda, Willem J Remme, Christian Torp-Pedersen, Marco Metra, Philip A Poole-Wilson, COMET Investigators, Wayne C Levy, Dariush Mozaffarian, David T Linker, Kenneth W Kenyon, John G F Cleland, Michel Komajda, Willem J Remme, Christian Torp-Pedersen, Marco MetraPhilip A Poole-Wilson, COMET Investigators

8 Citations (Scopus)

Abstract

AIMS: A standard metric to estimate absolute treatment effects is numbers-needed-to-treat (NNT), which implicitly assumes that all benefits reverse at trial-end. However, in-trial survival benefits typically do not reverse until long after trial-end, so that NNT will substantially underestimate lifetime benefits. METHODS AND RESULTS: We developed a new concept, years-needed-to-treat (YNT) to add 1 year of life, that quantifies the expected average life expectancy for two treatments including the estimated years of life remaining post-trial. Numbers-needed-to-treat and YNT were calculated in the COMET trial, in which carvedilol vs. metoprolol tartrate resulted in 17% lower mortality over 4.8 years. A multivariate Cox model was used to predict survival. Remaining years of life were estimated using the mortality-life-table method. At trial-end, survival was 9% higher in the carvedilol arm. Assuming that patients remained on the same therapy post-trial, the average total years of life for carvedilol vs. metoprolol were 10.63 +/- 0.19 vs. 9.48 +/- 0.18 (P < 0.0001) or 1.15 (95% confidence interval 0.64-1.66) additional years of life. The YNT was 9.2, indicating that 9.2 person-years of treatment added 1 person-year of life, compared with NNT of 59. CONCLUSION: Compared with NNT, the YNT method more accurately accounts for potential long-term benefits of interventions in randomized trials.

Original language	English
Journal	European Journal of Heart Failure
Volume	11
Issue number	3
Pages (from-to)	256-63
Number of pages	8
ISSN	1388-9842
DOIs	https://doi.org/10.1093/eurjhf/hfn048 https://doi.org/10.1093/eurjhf/hfn048
Publication status	Published - 1 Mar 2009

UN SDGs

This output contributes to the following UN Sustainable Development Goals (SDGs)

Access to Document

Cite this

Levy, W. C., Mozaffarian, D., Linker, D. T., Kenyon, K. W., Cleland, J. G. F., Komajda, M., Remme, W. J., Torp-Pedersen, C., Metra, M., Poole-Wilson, P. A., COMET Investigators, Levy, W. C., Mozaffarian, D., Linker, D. T., Kenyon, K. W., Cleland, J. G. F., Komajda, M., Remme, W. J., Torp-Pedersen, C., ... COMET Investigators (2009). Years-needed-to-treat to add 1 year of life: a new metric to estimate treatment effects in randomized trials. European Journal of Heart Failure, 11(3), 256-63. https://doi.org/10.1093/eurjhf/hfn048, https://doi.org/10.1093/eurjhf/hfn048

Levy, WC, Mozaffarian, D, Linker, DT, Kenyon, KW, Cleland, JGF, Komajda, M, Remme, WJ, Torp-Pedersen, C, Metra, M, Poole-Wilson, PA, COMET Investigators, Levy, WC, Mozaffarian, D, Linker, DT, Kenyon, KW, Cleland, JGF, Komajda, M, Remme, WJ, Torp-Pedersen, C, Metra, M, Poole-Wilson, PA & COMET Investigators 2009, 'Years-needed-to-treat to add 1 year of life: a new metric to estimate treatment effects in randomized trials', European Journal of Heart Failure, vol. 11, no. 3, pp. 256-63. https://doi.org/10.1093/eurjhf/hfn048, https://doi.org/10.1093/eurjhf/hfn048

@article{e183bb90785e11df928f000ea68e967b,

title = "Years-needed-to-treat to add 1 year of life: a new metric to estimate treatment effects in randomized trials",

abstract = "AIMS: A standard metric to estimate absolute treatment effects is numbers-needed-to-treat (NNT), which implicitly assumes that all benefits reverse at trial-end. However, in-trial survival benefits typically do not reverse until long after trial-end, so that NNT will substantially underestimate lifetime benefits. METHODS AND RESULTS: We developed a new concept, years-needed-to-treat (YNT) to add 1 year of life, that quantifies the expected average life expectancy for two treatments including the estimated years of life remaining post-trial. Numbers-needed-to-treat and YNT were calculated in the COMET trial, in which carvedilol vs. metoprolol tartrate resulted in 17% lower mortality over 4.8 years. A multivariate Cox model was used to predict survival. Remaining years of life were estimated using the mortality-life-table method. At trial-end, survival was 9% higher in the carvedilol arm. Assuming that patients remained on the same therapy post-trial, the average total years of life for carvedilol vs. metoprolol were 10.63 +/- 0.19 vs. 9.48 +/- 0.18 (P < 0.0001) or 1.15 (95% confidence interval 0.64-1.66) additional years of life. The YNT was 9.2, indicating that 9.2 person-years of treatment added 1 person-year of life, compared with NNT of 59. CONCLUSION: Compared with NNT, the YNT method more accurately accounts for potential long-term benefits of interventions in randomized trials.",

author = "Levy, {Wayne C} and Dariush Mozaffarian and Linker, {David T} and Kenyon, {Kenneth W} and Cleland, {John G F} and Michel Komajda and Remme, {Willem J} and Christian Torp-Pedersen and Marco Metra and Poole-Wilson, {Philip A} and {COMET Investigators} and Levy, {Wayne C} and Dariush Mozaffarian and Linker, {David T} and Kenyon, {Kenneth W} and Cleland, {John G F} and Michel Komajda and Remme, {Willem J} and Christian Torp-Pedersen and Marco Metra and Poole-Wilson, {Philip A} and Torp-Pedersen, {Christian Tobias}",

note = "Keywords: Adrenergic beta-Antagonists; Carbazoles; Confidence Intervals; Follow-Up Studies; Heart Failure; Humans; Kaplan-Meiers Estimate; Life Expectancy; Metoprolol; Odds Ratio; Propanolamines; Survival Rate; Time Factors; Treatment Outcome",

year = "2009",

month = mar,

day = "1",

doi = "10.1093/eurjhf/hfn048",

language = "English",

volume = "11",

pages = "256--63",

journal = "European Journal of Heart Failure, Supplement",

issn = "1567-4215",

publisher = "JohnWiley & Sons Ltd",

number = "3",

}

TY - JOUR

T1 - Years-needed-to-treat to add 1 year of life: a new metric to estimate treatment effects in randomized trials

AU - Levy, Wayne C

AU - Mozaffarian, Dariush

AU - Linker, David T

AU - Kenyon, Kenneth W

AU - Cleland, John G F

AU - Komajda, Michel

AU - Remme, Willem J

AU - Torp-Pedersen, Christian

AU - Metra, Marco

AU - Poole-Wilson, Philip A

AU - COMET Investigators

AU - Levy, Wayne C

AU - Mozaffarian, Dariush

AU - Linker, David T

AU - Kenyon, Kenneth W

AU - Cleland, John G F

AU - Komajda, Michel

AU - Remme, Willem J

AU - Torp-Pedersen, Christian

AU - Metra, Marco

AU - Poole-Wilson, Philip A

AU - COMET Investigators

N1 - Keywords: Adrenergic beta-Antagonists; Carbazoles; Confidence Intervals; Follow-Up Studies; Heart Failure; Humans; Kaplan-Meiers Estimate; Life Expectancy; Metoprolol; Odds Ratio; Propanolamines; Survival Rate; Time Factors; Treatment Outcome

PY - 2009/3/1

Y1 - 2009/3/1

N2 - AIMS: A standard metric to estimate absolute treatment effects is numbers-needed-to-treat (NNT), which implicitly assumes that all benefits reverse at trial-end. However, in-trial survival benefits typically do not reverse until long after trial-end, so that NNT will substantially underestimate lifetime benefits. METHODS AND RESULTS: We developed a new concept, years-needed-to-treat (YNT) to add 1 year of life, that quantifies the expected average life expectancy for two treatments including the estimated years of life remaining post-trial. Numbers-needed-to-treat and YNT were calculated in the COMET trial, in which carvedilol vs. metoprolol tartrate resulted in 17% lower mortality over 4.8 years. A multivariate Cox model was used to predict survival. Remaining years of life were estimated using the mortality-life-table method. At trial-end, survival was 9% higher in the carvedilol arm. Assuming that patients remained on the same therapy post-trial, the average total years of life for carvedilol vs. metoprolol were 10.63 +/- 0.19 vs. 9.48 +/- 0.18 (P < 0.0001) or 1.15 (95% confidence interval 0.64-1.66) additional years of life. The YNT was 9.2, indicating that 9.2 person-years of treatment added 1 person-year of life, compared with NNT of 59. CONCLUSION: Compared with NNT, the YNT method more accurately accounts for potential long-term benefits of interventions in randomized trials.

AB - AIMS: A standard metric to estimate absolute treatment effects is numbers-needed-to-treat (NNT), which implicitly assumes that all benefits reverse at trial-end. However, in-trial survival benefits typically do not reverse until long after trial-end, so that NNT will substantially underestimate lifetime benefits. METHODS AND RESULTS: We developed a new concept, years-needed-to-treat (YNT) to add 1 year of life, that quantifies the expected average life expectancy for two treatments including the estimated years of life remaining post-trial. Numbers-needed-to-treat and YNT were calculated in the COMET trial, in which carvedilol vs. metoprolol tartrate resulted in 17% lower mortality over 4.8 years. A multivariate Cox model was used to predict survival. Remaining years of life were estimated using the mortality-life-table method. At trial-end, survival was 9% higher in the carvedilol arm. Assuming that patients remained on the same therapy post-trial, the average total years of life for carvedilol vs. metoprolol were 10.63 +/- 0.19 vs. 9.48 +/- 0.18 (P < 0.0001) or 1.15 (95% confidence interval 0.64-1.66) additional years of life. The YNT was 9.2, indicating that 9.2 person-years of treatment added 1 person-year of life, compared with NNT of 59. CONCLUSION: Compared with NNT, the YNT method more accurately accounts for potential long-term benefits of interventions in randomized trials.

U2 - 10.1093/eurjhf/hfn048

DO - 10.1093/eurjhf/hfn048

M3 - Journal article

C2 - 19164422

SN - 1567-4215

VL - 11

SP - 256

EP - 263

JO - European Journal of Heart Failure, Supplement

JF - European Journal of Heart Failure, Supplement

IS - 3

ER -

Years-needed-to-treat to add 1 year of life: a new metric to estimate treatment effects in randomized trials

Abstract

UN SDGs

Access to Document

Fingerprint

Cite this