Does accounting for seizure frequency variability increase clinical trial power?

Daniel M. Goldenholz, Shira R. Goldenholz, Robert Moss, Jacqueline French, Daniel Lowenstein, Ruben Kuzniecky, Sheryl R. Haut, Sabrina Cristofaro, Kamil Detyniecki, John Hixson, Philippa Karoly, Mark Cook, Alex Strashny, William H. Theodore, Carl Pieper

Research output: Contribution to journalArticle

7 Citations (Scopus)

Abstract

Objective: Seizure frequency variability is associated with placebo responses in randomized controlled trials (RCT). Increased variability can result in drug misclassification and, hence, decreased statistical power. We investigated a new method that directly incorporated variability into RCT analysis, ZV. Methods: Two models were assessed: the traditional 50%-responder rate (RR50), and the variability-corrected score, ZV. Each predicted seizure frequency upper and lower limits using prior seizures. Accuracy was defined as percentage of time-intervals when the observed seizure frequencies were within the predicted limits. First, we tested the ZV method on three datasets (SeizureTracker: n=3016, Human Epilepsy Project: n=107, and NeuroVista: n=15). An additional independent SeizureTracker validation dataset was used to generate a set of 200 simulated trials each for 5 different sample sizes (total N=100 to 500 by 100), assuming 20% dropout and 30% drug efficacy. "Power" was determined as the percentage of trials successfully distinguishing placebo from drug (p<0.05). Results: Prediction accuracy across datasets was, ZV: 91-100%, RR50: 42-80%. Simulated RCT ZV analysis achieved >90% power at N=100 per arm while RR50 required N=200 per arm. Significance: ZV may increase the statistical power of an RCT relative to the traditional RR50.

Original languageEnglish (US)
JournalEpilepsy Research
DOIs
StateAccepted/In press - 2017

Fingerprint

Seizures
Clinical Trials
Randomized Controlled Trials
Placebos
Pharmaceutical Preparations
Sample Size
Epilepsy
Datasets

Keywords

  • Clinical trials
  • Epilepsy
  • Natural variability
  • Placebo effect
  • Prediction
  • Seizure frequency

ASJC Scopus subject areas

  • Neurology
  • Clinical Neurology

Cite this

Goldenholz, D. M., Goldenholz, S. R., Moss, R., French, J., Lowenstein, D., Kuzniecky, R., ... Pieper, C. (Accepted/In press). Does accounting for seizure frequency variability increase clinical trial power? Epilepsy Research. https://doi.org/10.1016/j.eplepsyres.2017.07.013

Does accounting for seizure frequency variability increase clinical trial power? / Goldenholz, Daniel M.; Goldenholz, Shira R.; Moss, Robert; French, Jacqueline; Lowenstein, Daniel; Kuzniecky, Ruben; Haut, Sheryl R.; Cristofaro, Sabrina; Detyniecki, Kamil; Hixson, John; Karoly, Philippa; Cook, Mark; Strashny, Alex; Theodore, William H.; Pieper, Carl.

In: Epilepsy Research, 2017.

Research output: Contribution to journalArticle

Goldenholz, DM, Goldenholz, SR, Moss, R, French, J, Lowenstein, D, Kuzniecky, R, Haut, SR, Cristofaro, S, Detyniecki, K, Hixson, J, Karoly, P, Cook, M, Strashny, A, Theodore, WH & Pieper, C 2017, 'Does accounting for seizure frequency variability increase clinical trial power?', Epilepsy Research. https://doi.org/10.1016/j.eplepsyres.2017.07.013
Goldenholz, Daniel M. ; Goldenholz, Shira R. ; Moss, Robert ; French, Jacqueline ; Lowenstein, Daniel ; Kuzniecky, Ruben ; Haut, Sheryl R. ; Cristofaro, Sabrina ; Detyniecki, Kamil ; Hixson, John ; Karoly, Philippa ; Cook, Mark ; Strashny, Alex ; Theodore, William H. ; Pieper, Carl. / Does accounting for seizure frequency variability increase clinical trial power?. In: Epilepsy Research. 2017.
@article{8b457825e1bd4b99bc550bdda3232c18,
title = "Does accounting for seizure frequency variability increase clinical trial power?",
abstract = "Objective: Seizure frequency variability is associated with placebo responses in randomized controlled trials (RCT). Increased variability can result in drug misclassification and, hence, decreased statistical power. We investigated a new method that directly incorporated variability into RCT analysis, ZV. Methods: Two models were assessed: the traditional 50{\%}-responder rate (RR50), and the variability-corrected score, ZV. Each predicted seizure frequency upper and lower limits using prior seizures. Accuracy was defined as percentage of time-intervals when the observed seizure frequencies were within the predicted limits. First, we tested the ZV method on three datasets (SeizureTracker: n=3016, Human Epilepsy Project: n=107, and NeuroVista: n=15). An additional independent SeizureTracker validation dataset was used to generate a set of 200 simulated trials each for 5 different sample sizes (total N=100 to 500 by 100), assuming 20{\%} dropout and 30{\%} drug efficacy. {"}Power{"} was determined as the percentage of trials successfully distinguishing placebo from drug (p<0.05). Results: Prediction accuracy across datasets was, ZV: 91-100{\%}, RR50: 42-80{\%}. Simulated RCT ZV analysis achieved >90{\%} power at N=100 per arm while RR50 required N=200 per arm. Significance: ZV may increase the statistical power of an RCT relative to the traditional RR50.",
keywords = "Clinical trials, Epilepsy, Natural variability, Placebo effect, Prediction, Seizure frequency",
author = "Goldenholz, {Daniel M.} and Goldenholz, {Shira R.} and Robert Moss and Jacqueline French and Daniel Lowenstein and Ruben Kuzniecky and Haut, {Sheryl R.} and Sabrina Cristofaro and Kamil Detyniecki and John Hixson and Philippa Karoly and Mark Cook and Alex Strashny and Theodore, {William H.} and Carl Pieper",
year = "2017",
doi = "10.1016/j.eplepsyres.2017.07.013",
language = "English (US)",
journal = "Epilepsy Research",
issn = "0920-1211",
publisher = "Elsevier",

}

TY - JOUR

T1 - Does accounting for seizure frequency variability increase clinical trial power?

AU - Goldenholz, Daniel M.

AU - Goldenholz, Shira R.

AU - Moss, Robert

AU - French, Jacqueline

AU - Lowenstein, Daniel

AU - Kuzniecky, Ruben

AU - Haut, Sheryl R.

AU - Cristofaro, Sabrina

AU - Detyniecki, Kamil

AU - Hixson, John

AU - Karoly, Philippa

AU - Cook, Mark

AU - Strashny, Alex

AU - Theodore, William H.

AU - Pieper, Carl

PY - 2017

Y1 - 2017

N2 - Objective: Seizure frequency variability is associated with placebo responses in randomized controlled trials (RCT). Increased variability can result in drug misclassification and, hence, decreased statistical power. We investigated a new method that directly incorporated variability into RCT analysis, ZV. Methods: Two models were assessed: the traditional 50%-responder rate (RR50), and the variability-corrected score, ZV. Each predicted seizure frequency upper and lower limits using prior seizures. Accuracy was defined as percentage of time-intervals when the observed seizure frequencies were within the predicted limits. First, we tested the ZV method on three datasets (SeizureTracker: n=3016, Human Epilepsy Project: n=107, and NeuroVista: n=15). An additional independent SeizureTracker validation dataset was used to generate a set of 200 simulated trials each for 5 different sample sizes (total N=100 to 500 by 100), assuming 20% dropout and 30% drug efficacy. "Power" was determined as the percentage of trials successfully distinguishing placebo from drug (p<0.05). Results: Prediction accuracy across datasets was, ZV: 91-100%, RR50: 42-80%. Simulated RCT ZV analysis achieved >90% power at N=100 per arm while RR50 required N=200 per arm. Significance: ZV may increase the statistical power of an RCT relative to the traditional RR50.

AB - Objective: Seizure frequency variability is associated with placebo responses in randomized controlled trials (RCT). Increased variability can result in drug misclassification and, hence, decreased statistical power. We investigated a new method that directly incorporated variability into RCT analysis, ZV. Methods: Two models were assessed: the traditional 50%-responder rate (RR50), and the variability-corrected score, ZV. Each predicted seizure frequency upper and lower limits using prior seizures. Accuracy was defined as percentage of time-intervals when the observed seizure frequencies were within the predicted limits. First, we tested the ZV method on three datasets (SeizureTracker: n=3016, Human Epilepsy Project: n=107, and NeuroVista: n=15). An additional independent SeizureTracker validation dataset was used to generate a set of 200 simulated trials each for 5 different sample sizes (total N=100 to 500 by 100), assuming 20% dropout and 30% drug efficacy. "Power" was determined as the percentage of trials successfully distinguishing placebo from drug (p<0.05). Results: Prediction accuracy across datasets was, ZV: 91-100%, RR50: 42-80%. Simulated RCT ZV analysis achieved >90% power at N=100 per arm while RR50 required N=200 per arm. Significance: ZV may increase the statistical power of an RCT relative to the traditional RR50.

KW - Clinical trials

KW - Epilepsy

KW - Natural variability

KW - Placebo effect

KW - Prediction

KW - Seizure frequency

UR - http://www.scopus.com/inward/record.url?scp=85028085955&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=85028085955&partnerID=8YFLogxK

U2 - 10.1016/j.eplepsyres.2017.07.013

DO - 10.1016/j.eplepsyres.2017.07.013

M3 - Article

C2 - 28781216

AN - SCOPUS:85028085955

JO - Epilepsy Research

JF - Epilepsy Research

SN - 0920-1211

ER -