Applications of asymptotic confidence intervals with continuity corrections for asymmetric comparisons in noninferiority trials

Julia N. Soulakova, Brianna C. Bright

Research output: Contribution to journalArticle

3 Citations (Scopus)

Abstract

A large-sample problem of illustrating noninferiority of an experimental treatment over a referent treatment for binary outcomes is considered. The methods of illustrating noninferiority involve constructing the lower two-sided confidence bound for the difference between binomial proportions corresponding to the experimental and referent treatments and comparing it with the negative value of the noninferiority margin. The three considered methods, Anbar, Falk-Koch, and Reduced Falk-Koch, handle the comparison in an asymmetric way, that is, only the referent proportion out of the two, experimental and referent, is directly involved in the expression for the variance of the difference between two sample proportions. Five continuity corrections (including zero) are considered with respect to each approach. The key properties of the corresponding methods are evaluated via simulations. First, the uncorrected two-sided confidence intervals can, potentially, have smaller coverage probability than the nominal level even for moderately large sample sizes, for example, 150 per group. Next, the 15 testing methods are discussed in terms of their Type I error rate and power. In the settings with a relatively small referent proportion (about 0.4 or smaller), the Anbar approach with Yates' continuity correction is recommended for balanced designs and the Falk-Koch method with Yates' correction is recommended for unbalanced designs. For relatively moderate (about 0.6) and large (about 0.8 or greater) referent proportion, the uncorrected Reduced Falk-Koch method is recommended, although in this case, all methods tend to be over-conservative. These results are expected to be used in the design stage of a noninferiority study when asymmetric comparisons are envisioned.

Original languageEnglish (US)
Pages (from-to)147-155
Number of pages9
JournalPharmaceutical Statistics
Volume12
Issue number3
DOIs
StatePublished - May 1 2013

Fingerprint

Continuity Correction
Non-inferiority
Confidence interval
Confidence Intervals
Proportion
Unbalanced Designs
Balanced Design
Confidence Bounds
Binary Outcomes
Type I Error Rate
Coverage Probability
Sample Size
Margin
Categorical or nominal
Tend
Testing
Zero

Keywords

  • binomial distribution
  • coverage probability
  • difference between proportions
  • variance estimation

ASJC Scopus subject areas

  • Statistics and Probability
  • Pharmacology
  • Pharmacology (medical)

Cite this

Applications of asymptotic confidence intervals with continuity corrections for asymmetric comparisons in noninferiority trials. / Soulakova, Julia N.; Bright, Brianna C.

In: Pharmaceutical Statistics, Vol. 12, No. 3, 01.05.2013, p. 147-155.

Research output: Contribution to journalArticle

@article{6191c3b33efb4255ba1c8c614e053424,
title = "Applications of asymptotic confidence intervals with continuity corrections for asymmetric comparisons in noninferiority trials",
abstract = "A large-sample problem of illustrating noninferiority of an experimental treatment over a referent treatment for binary outcomes is considered. The methods of illustrating noninferiority involve constructing the lower two-sided confidence bound for the difference between binomial proportions corresponding to the experimental and referent treatments and comparing it with the negative value of the noninferiority margin. The three considered methods, Anbar, Falk-Koch, and Reduced Falk-Koch, handle the comparison in an asymmetric way, that is, only the referent proportion out of the two, experimental and referent, is directly involved in the expression for the variance of the difference between two sample proportions. Five continuity corrections (including zero) are considered with respect to each approach. The key properties of the corresponding methods are evaluated via simulations. First, the uncorrected two-sided confidence intervals can, potentially, have smaller coverage probability than the nominal level even for moderately large sample sizes, for example, 150 per group. Next, the 15 testing methods are discussed in terms of their Type I error rate and power. In the settings with a relatively small referent proportion (about 0.4 or smaller), the Anbar approach with Yates' continuity correction is recommended for balanced designs and the Falk-Koch method with Yates' correction is recommended for unbalanced designs. For relatively moderate (about 0.6) and large (about 0.8 or greater) referent proportion, the uncorrected Reduced Falk-Koch method is recommended, although in this case, all methods tend to be over-conservative. These results are expected to be used in the design stage of a noninferiority study when asymmetric comparisons are envisioned.",
keywords = "binomial distribution, coverage probability, difference between proportions, variance estimation",
author = "Soulakova, {Julia N.} and Bright, {Brianna C.}",
year = "2013",
month = "5",
day = "1",
doi = "10.1002/pst.1566",
language = "English (US)",
volume = "12",
pages = "147--155",
journal = "Pharmaceutical Statistics",
issn = "1539-1604",
publisher = "John Wiley and Sons Ltd",
number = "3",

}

TY - JOUR

T1 - Applications of asymptotic confidence intervals with continuity corrections for asymmetric comparisons in noninferiority trials

AU - Soulakova, Julia N.

AU - Bright, Brianna C.

PY - 2013/5/1

Y1 - 2013/5/1

N2 - A large-sample problem of illustrating noninferiority of an experimental treatment over a referent treatment for binary outcomes is considered. The methods of illustrating noninferiority involve constructing the lower two-sided confidence bound for the difference between binomial proportions corresponding to the experimental and referent treatments and comparing it with the negative value of the noninferiority margin. The three considered methods, Anbar, Falk-Koch, and Reduced Falk-Koch, handle the comparison in an asymmetric way, that is, only the referent proportion out of the two, experimental and referent, is directly involved in the expression for the variance of the difference between two sample proportions. Five continuity corrections (including zero) are considered with respect to each approach. The key properties of the corresponding methods are evaluated via simulations. First, the uncorrected two-sided confidence intervals can, potentially, have smaller coverage probability than the nominal level even for moderately large sample sizes, for example, 150 per group. Next, the 15 testing methods are discussed in terms of their Type I error rate and power. In the settings with a relatively small referent proportion (about 0.4 or smaller), the Anbar approach with Yates' continuity correction is recommended for balanced designs and the Falk-Koch method with Yates' correction is recommended for unbalanced designs. For relatively moderate (about 0.6) and large (about 0.8 or greater) referent proportion, the uncorrected Reduced Falk-Koch method is recommended, although in this case, all methods tend to be over-conservative. These results are expected to be used in the design stage of a noninferiority study when asymmetric comparisons are envisioned.

AB - A large-sample problem of illustrating noninferiority of an experimental treatment over a referent treatment for binary outcomes is considered. The methods of illustrating noninferiority involve constructing the lower two-sided confidence bound for the difference between binomial proportions corresponding to the experimental and referent treatments and comparing it with the negative value of the noninferiority margin. The three considered methods, Anbar, Falk-Koch, and Reduced Falk-Koch, handle the comparison in an asymmetric way, that is, only the referent proportion out of the two, experimental and referent, is directly involved in the expression for the variance of the difference between two sample proportions. Five continuity corrections (including zero) are considered with respect to each approach. The key properties of the corresponding methods are evaluated via simulations. First, the uncorrected two-sided confidence intervals can, potentially, have smaller coverage probability than the nominal level even for moderately large sample sizes, for example, 150 per group. Next, the 15 testing methods are discussed in terms of their Type I error rate and power. In the settings with a relatively small referent proportion (about 0.4 or smaller), the Anbar approach with Yates' continuity correction is recommended for balanced designs and the Falk-Koch method with Yates' correction is recommended for unbalanced designs. For relatively moderate (about 0.6) and large (about 0.8 or greater) referent proportion, the uncorrected Reduced Falk-Koch method is recommended, although in this case, all methods tend to be over-conservative. These results are expected to be used in the design stage of a noninferiority study when asymmetric comparisons are envisioned.

KW - binomial distribution

KW - coverage probability

KW - difference between proportions

KW - variance estimation

UR - http://www.scopus.com/inward/record.url?scp=84878148007&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=84878148007&partnerID=8YFLogxK

U2 - 10.1002/pst.1566

DO - 10.1002/pst.1566

M3 - Article

C2 - 23554217

AN - SCOPUS:84878148007

VL - 12

SP - 147

EP - 155

JO - Pharmaceutical Statistics

JF - Pharmaceutical Statistics

SN - 1539-1604

IS - 3

ER -