A systematic approach to subgroup analyses in a smoking cessation trial

Arthur N. Westover, T. Michael Kashner, Theresa M. Winhusen, Richard M. Golden, Paul A. Nakonezny, Bryon Adinoff, Steven S. Henley

Research output: Contribution to journalArticle

11 Citations (Scopus)

Abstract

Background: Traditional approaches to subgroup analyses that test each moderating factor as a separate hypothesis can lead to erroneous conclusions due to the problems of multiple comparisons, model misspecification, and multicollinearity. Objective: To demonstrate a novel, systematic approach to subgroup analyses that avoids these pitfalls. Methods: A Best Approximating Model (BAM) approach that identifies multiple moderators and estimates their simultaneous impact on treatment effect sizes was applied to a randomized, controlled, 11-week, double-blind efficacy trial on smoking cessation of adult smokers with attention-deficit/hyperactivity disorder (ADHD), randomized to either OROS-methylphenidate (n = 127) or placebo (n = 128), and treated with nicotine patch. Binary outcomes measures were prolonged smoking abstinence and point prevalence smoking abstinence. Results: Although the original clinical trial data analysis showed no treatment effect on smoking cessation, the BAM analysis showed significant subgroup effects for the primary outcome of prolonged smoking abstinence: (1) lifetime history of substance use disorders (adjusted odds ratio [AOR] 0.27; 95% confidence interval [CI] 0.10-0.74), and (2) more severe ADHD symptoms (baseline score >36; AOR 2.64; 95% CI 1.17-5.96). A significant subgroup effect was also shown for the secondary outcome of point prevalence smoking abstinence-age 18 to 29 years (AOR 0.23; 95% CI 0.07-0.76). Conclusions: The BAM analysis resulted in different conclusions about subgroup effects compared to a hypothesis-driven approach. By examining moderator independence and avoiding multiple testing, BAMs have the potential to better identify and explain how treatment effects vary across subgroups in heterogeneous patient populations, thus providing better guidance to more effectively match individual patients with specific treatments.

Original languageEnglish (US)
Pages (from-to)498-507
Number of pages10
JournalAmerican Journal of Drug and Alcohol Abuse
Volume41
Issue number6
DOIs
StatePublished - Nov 2 2015

Fingerprint

Smoking Cessation
Smoking
Odds Ratio
Confidence Intervals
Attention Deficit Disorder with Hyperactivity
Tobacco Use Cessation Products
Methylphenidate
Therapeutics
Substance-Related Disorders
Placebos
Outcome Assessment (Health Care)
Clinical Trials
Population

Keywords

  • Attention deficit hyperactivity disorder
  • methylphenidate
  • modeling
  • statistics
  • subgroup analysis
  • tobacco

ASJC Scopus subject areas

  • Medicine (miscellaneous)
  • Psychiatry and Mental health
  • Clinical Psychology

Cite this

A systematic approach to subgroup analyses in a smoking cessation trial. / Westover, Arthur N.; Kashner, T. Michael; Winhusen, Theresa M.; Golden, Richard M.; Nakonezny, Paul A.; Adinoff, Bryon; Henley, Steven S.

In: American Journal of Drug and Alcohol Abuse, Vol. 41, No. 6, 02.11.2015, p. 498-507.

Research output: Contribution to journalArticle

Westover, Arthur N. ; Kashner, T. Michael ; Winhusen, Theresa M. ; Golden, Richard M. ; Nakonezny, Paul A. ; Adinoff, Bryon ; Henley, Steven S. / A systematic approach to subgroup analyses in a smoking cessation trial. In: American Journal of Drug and Alcohol Abuse. 2015 ; Vol. 41, No. 6. pp. 498-507.
@article{e41d7719fd5c4430b1c7c289a2d9690e,
title = "A systematic approach to subgroup analyses in a smoking cessation trial",
abstract = "Background: Traditional approaches to subgroup analyses that test each moderating factor as a separate hypothesis can lead to erroneous conclusions due to the problems of multiple comparisons, model misspecification, and multicollinearity. Objective: To demonstrate a novel, systematic approach to subgroup analyses that avoids these pitfalls. Methods: A Best Approximating Model (BAM) approach that identifies multiple moderators and estimates their simultaneous impact on treatment effect sizes was applied to a randomized, controlled, 11-week, double-blind efficacy trial on smoking cessation of adult smokers with attention-deficit/hyperactivity disorder (ADHD), randomized to either OROS-methylphenidate (n = 127) or placebo (n = 128), and treated with nicotine patch. Binary outcomes measures were prolonged smoking abstinence and point prevalence smoking abstinence. Results: Although the original clinical trial data analysis showed no treatment effect on smoking cessation, the BAM analysis showed significant subgroup effects for the primary outcome of prolonged smoking abstinence: (1) lifetime history of substance use disorders (adjusted odds ratio [AOR] 0.27; 95{\%} confidence interval [CI] 0.10-0.74), and (2) more severe ADHD symptoms (baseline score >36; AOR 2.64; 95{\%} CI 1.17-5.96). A significant subgroup effect was also shown for the secondary outcome of point prevalence smoking abstinence-age 18 to 29 years (AOR 0.23; 95{\%} CI 0.07-0.76). Conclusions: The BAM analysis resulted in different conclusions about subgroup effects compared to a hypothesis-driven approach. By examining moderator independence and avoiding multiple testing, BAMs have the potential to better identify and explain how treatment effects vary across subgroups in heterogeneous patient populations, thus providing better guidance to more effectively match individual patients with specific treatments.",
keywords = "Attention deficit hyperactivity disorder, methylphenidate, modeling, statistics, subgroup analysis, tobacco",
author = "Westover, {Arthur N.} and Kashner, {T. Michael} and Winhusen, {Theresa M.} and Golden, {Richard M.} and Nakonezny, {Paul A.} and Bryon Adinoff and Henley, {Steven S.}",
year = "2015",
month = "11",
day = "2",
doi = "10.3109/00952990.2015.1044605",
language = "English (US)",
volume = "41",
pages = "498--507",
journal = "American Journal of Drug and Alcohol Abuse",
issn = "0095-2990",
publisher = "Informa Healthcare",
number = "6",

}

TY - JOUR

T1 - A systematic approach to subgroup analyses in a smoking cessation trial

AU - Westover, Arthur N.

AU - Kashner, T. Michael

AU - Winhusen, Theresa M.

AU - Golden, Richard M.

AU - Nakonezny, Paul A.

AU - Adinoff, Bryon

AU - Henley, Steven S.

PY - 2015/11/2

Y1 - 2015/11/2

N2 - Background: Traditional approaches to subgroup analyses that test each moderating factor as a separate hypothesis can lead to erroneous conclusions due to the problems of multiple comparisons, model misspecification, and multicollinearity. Objective: To demonstrate a novel, systematic approach to subgroup analyses that avoids these pitfalls. Methods: A Best Approximating Model (BAM) approach that identifies multiple moderators and estimates their simultaneous impact on treatment effect sizes was applied to a randomized, controlled, 11-week, double-blind efficacy trial on smoking cessation of adult smokers with attention-deficit/hyperactivity disorder (ADHD), randomized to either OROS-methylphenidate (n = 127) or placebo (n = 128), and treated with nicotine patch. Binary outcomes measures were prolonged smoking abstinence and point prevalence smoking abstinence. Results: Although the original clinical trial data analysis showed no treatment effect on smoking cessation, the BAM analysis showed significant subgroup effects for the primary outcome of prolonged smoking abstinence: (1) lifetime history of substance use disorders (adjusted odds ratio [AOR] 0.27; 95% confidence interval [CI] 0.10-0.74), and (2) more severe ADHD symptoms (baseline score >36; AOR 2.64; 95% CI 1.17-5.96). A significant subgroup effect was also shown for the secondary outcome of point prevalence smoking abstinence-age 18 to 29 years (AOR 0.23; 95% CI 0.07-0.76). Conclusions: The BAM analysis resulted in different conclusions about subgroup effects compared to a hypothesis-driven approach. By examining moderator independence and avoiding multiple testing, BAMs have the potential to better identify and explain how treatment effects vary across subgroups in heterogeneous patient populations, thus providing better guidance to more effectively match individual patients with specific treatments.

AB - Background: Traditional approaches to subgroup analyses that test each moderating factor as a separate hypothesis can lead to erroneous conclusions due to the problems of multiple comparisons, model misspecification, and multicollinearity. Objective: To demonstrate a novel, systematic approach to subgroup analyses that avoids these pitfalls. Methods: A Best Approximating Model (BAM) approach that identifies multiple moderators and estimates their simultaneous impact on treatment effect sizes was applied to a randomized, controlled, 11-week, double-blind efficacy trial on smoking cessation of adult smokers with attention-deficit/hyperactivity disorder (ADHD), randomized to either OROS-methylphenidate (n = 127) or placebo (n = 128), and treated with nicotine patch. Binary outcomes measures were prolonged smoking abstinence and point prevalence smoking abstinence. Results: Although the original clinical trial data analysis showed no treatment effect on smoking cessation, the BAM analysis showed significant subgroup effects for the primary outcome of prolonged smoking abstinence: (1) lifetime history of substance use disorders (adjusted odds ratio [AOR] 0.27; 95% confidence interval [CI] 0.10-0.74), and (2) more severe ADHD symptoms (baseline score >36; AOR 2.64; 95% CI 1.17-5.96). A significant subgroup effect was also shown for the secondary outcome of point prevalence smoking abstinence-age 18 to 29 years (AOR 0.23; 95% CI 0.07-0.76). Conclusions: The BAM analysis resulted in different conclusions about subgroup effects compared to a hypothesis-driven approach. By examining moderator independence and avoiding multiple testing, BAMs have the potential to better identify and explain how treatment effects vary across subgroups in heterogeneous patient populations, thus providing better guidance to more effectively match individual patients with specific treatments.

KW - Attention deficit hyperactivity disorder

KW - methylphenidate

KW - modeling

KW - statistics

KW - subgroup analysis

KW - tobacco

UR - http://www.scopus.com/inward/record.url?scp=84944277089&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=84944277089&partnerID=8YFLogxK

U2 - 10.3109/00952990.2015.1044605

DO - 10.3109/00952990.2015.1044605

M3 - Article

C2 - 26065433

AN - SCOPUS:84944277089

VL - 41

SP - 498

EP - 507

JO - American Journal of Drug and Alcohol Abuse

JF - American Journal of Drug and Alcohol Abuse

SN - 0095-2990

IS - 6

ER -