Constructing a confidence interval for the fraction who benefit from treatment, using randomized trial data

Emily J. Huang, Xingyuan Fang, Daniel F. Hanley, Michael Rosenblum

Research output: Contribution to journalArticle

Abstract

The fraction who benefit from treatment is the proportion of patients whose potential outcome under treatment is better than that under control. Inference on this parameter is challenging since it is only partially identifiable, even in our context of a randomized trial. We propose a new method for constructing a confidence interval for the fraction, when the outcome is ordinal or binary. Our confidence interval procedure is pointwise consistent. It does not require any assumptions about the joint distribution of the potential outcomes, although it has the flexibility to incorporate various user-defined assumptions. Our method is based on a stochastic optimization technique involving a second-order, asymptotic approximation that, to the best of our knowledge, has not been applied to biomedical studies. This approximation leads to statistics that are solutions to quadratic programs, which can be computed efficiently using optimization tools. In simulation, our method attains the nominal coverage probability or higher, and can have narrower average width than competitor methods. We apply it to a trial of a new intervention for stroke.

Original languageEnglish (US)
JournalBiometrics
DOIs
StateAccepted/In press - Jan 1 2019

Fingerprint

Randomized Trial
Potential Outcomes
Confidence interval
confidence interval
Confidence Intervals
Second-order Asymptotics
Quadratic Program
Coverage Probability
Stochastic Optimization
Asymptotic Approximation
Statistics
Stroke
Joint Distribution
Simulation Methods
Optimization Techniques
Categorical or nominal
Proportion
Therapeutics
Flexibility
methodology

All Science Journal Classification (ASJC) codes

  • Statistics and Probability
  • Biochemistry, Genetics and Molecular Biology(all)
  • Immunology and Microbiology(all)
  • Agricultural and Biological Sciences(all)
  • Applied Mathematics

Cite this

@article{7f97483e4c12445a89e1f35f365d895e,
title = "Constructing a confidence interval for the fraction who benefit from treatment, using randomized trial data",
abstract = "The fraction who benefit from treatment is the proportion of patients whose potential outcome under treatment is better than that under control. Inference on this parameter is challenging since it is only partially identifiable, even in our context of a randomized trial. We propose a new method for constructing a confidence interval for the fraction, when the outcome is ordinal or binary. Our confidence interval procedure is pointwise consistent. It does not require any assumptions about the joint distribution of the potential outcomes, although it has the flexibility to incorporate various user-defined assumptions. Our method is based on a stochastic optimization technique involving a second-order, asymptotic approximation that, to the best of our knowledge, has not been applied to biomedical studies. This approximation leads to statistics that are solutions to quadratic programs, which can be computed efficiently using optimization tools. In simulation, our method attains the nominal coverage probability or higher, and can have narrower average width than competitor methods. We apply it to a trial of a new intervention for stroke.",
author = "Huang, {Emily J.} and Xingyuan Fang and Hanley, {Daniel F.} and Michael Rosenblum",
year = "2019",
month = "1",
day = "1",
doi = "10.1111/biom.13101",
language = "English (US)",
journal = "Biometrics",
issn = "0006-341X",
publisher = "Wiley-Blackwell",

}

Constructing a confidence interval for the fraction who benefit from treatment, using randomized trial data. / Huang, Emily J.; Fang, Xingyuan; Hanley, Daniel F.; Rosenblum, Michael.

In: Biometrics, 01.01.2019.

Research output: Contribution to journalArticle

TY - JOUR

T1 - Constructing a confidence interval for the fraction who benefit from treatment, using randomized trial data

AU - Huang, Emily J.

AU - Fang, Xingyuan

AU - Hanley, Daniel F.

AU - Rosenblum, Michael

PY - 2019/1/1

Y1 - 2019/1/1

N2 - The fraction who benefit from treatment is the proportion of patients whose potential outcome under treatment is better than that under control. Inference on this parameter is challenging since it is only partially identifiable, even in our context of a randomized trial. We propose a new method for constructing a confidence interval for the fraction, when the outcome is ordinal or binary. Our confidence interval procedure is pointwise consistent. It does not require any assumptions about the joint distribution of the potential outcomes, although it has the flexibility to incorporate various user-defined assumptions. Our method is based on a stochastic optimization technique involving a second-order, asymptotic approximation that, to the best of our knowledge, has not been applied to biomedical studies. This approximation leads to statistics that are solutions to quadratic programs, which can be computed efficiently using optimization tools. In simulation, our method attains the nominal coverage probability or higher, and can have narrower average width than competitor methods. We apply it to a trial of a new intervention for stroke.

AB - The fraction who benefit from treatment is the proportion of patients whose potential outcome under treatment is better than that under control. Inference on this parameter is challenging since it is only partially identifiable, even in our context of a randomized trial. We propose a new method for constructing a confidence interval for the fraction, when the outcome is ordinal or binary. Our confidence interval procedure is pointwise consistent. It does not require any assumptions about the joint distribution of the potential outcomes, although it has the flexibility to incorporate various user-defined assumptions. Our method is based on a stochastic optimization technique involving a second-order, asymptotic approximation that, to the best of our knowledge, has not been applied to biomedical studies. This approximation leads to statistics that are solutions to quadratic programs, which can be computed efficiently using optimization tools. In simulation, our method attains the nominal coverage probability or higher, and can have narrower average width than competitor methods. We apply it to a trial of a new intervention for stroke.

UR - http://www.scopus.com/inward/record.url?scp=85071608321&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=85071608321&partnerID=8YFLogxK

U2 - 10.1111/biom.13101

DO - 10.1111/biom.13101

M3 - Article

C2 - 31206586

AN - SCOPUS:85071608321

JO - Biometrics

JF - Biometrics

SN - 0006-341X

ER -