A 'missing not at random' (MNAR) and 'missing at random' (MAR) growth model comparison with a buprenorphine/naloxone clinical trial

Sterling Mcpherson, Celestina Barbosa-Leiker, Mary Rose Mamey, Michael Mcdonell, Craig K. Enders, John Roll

Research output: Contribution to journalArticle

13 Citations (Scopus)

Abstract

Aims: To compare three missing data strategies: (i) the latent growth model that assumes the data are missing at random (MAR) model; (ii) the Diggle-Kenward missing not at random (MNAR) model, where dropout is a function of previous/concurrent urinalysis (UA) submissions; and (iii) the Wu-Carroll MNAR model where dropout is a function of the growth factors. Design: Secondary data analysis of a National Drug Abuse Treatment Clinical Trials Network trial that examined a 7-day versus 28-day taper (i.e. stepwise decrease in buprenorphine/naloxone) on the likelihood of submitting an opioid-positive UA during treatment. Setting: 11 out-patient treatment settings in 10 US cities. Participants: A total of 516 opioid-dependent participants. Measurements: Opioid UAs provided across the 4-week treatment period. Findings: The MAR model showed a significant effect (B=-0.45, P<0.05) of trial arm on the opioid-positive UA slope (i.e. 28-day taper participants were less likely to submit a positive UA over time) with a small effect size (d=0.20). The MNAR Diggle-Kenward model demonstrated a significant (B=-0.64, P<0.01) effect of trial arm on the slope with a large effect size (d=0.82). The MNAR Wu-Carroll model showed a significant (B=-0.41, P<0.05) effect of trial arm on the UA slope that was relatively small (d=0.31). Conclusions: This performance comparison of three missing data strategies (latent growth model, Diggle-Kenward selection model, Wu-Carrol selection model) on sample data indicates a need for increased use of sensitivity analyses in clinical trial research. Given the potential sensitivity of the trial arm effect to missing data assumptions, it is critical for researchers to consider whether the assumptions associated with each model are defensible.

Original languageEnglish (US)
Pages (from-to)51-58
Number of pages8
JournalAddiction
Volume110
Issue number1
DOIs
StatePublished - Jan 1 2015

Fingerprint

Urinalysis
Opioid Analgesics
Clinical Trials
Growth
Substance-Related Disorders
Intercellular Signaling Peptides and Proteins
Outpatients
Research Personnel
Naloxone Drug Combination Buprenorphine
Research

Keywords

  • Latent growth modeling
  • Longitudinal missing data
  • Missing not at random models
  • Randomized clinical trials
  • Sensitivity analysis
  • Substance use disorder treatment

ASJC Scopus subject areas

  • Medicine (miscellaneous)
  • Psychiatry and Mental health

Cite this

Mcpherson, S., Barbosa-Leiker, C., Mamey, M. R., Mcdonell, M., Enders, C. K., & Roll, J. (2015). A 'missing not at random' (MNAR) and 'missing at random' (MAR) growth model comparison with a buprenorphine/naloxone clinical trial. Addiction, 110(1), 51-58. https://doi.org/10.1111/add.12714

A 'missing not at random' (MNAR) and 'missing at random' (MAR) growth model comparison with a buprenorphine/naloxone clinical trial. / Mcpherson, Sterling; Barbosa-Leiker, Celestina; Mamey, Mary Rose; Mcdonell, Michael; Enders, Craig K.; Roll, John.

In: Addiction, Vol. 110, No. 1, 01.01.2015, p. 51-58.

Research output: Contribution to journalArticle

Mcpherson, S, Barbosa-Leiker, C, Mamey, MR, Mcdonell, M, Enders, CK & Roll, J 2015, 'A 'missing not at random' (MNAR) and 'missing at random' (MAR) growth model comparison with a buprenorphine/naloxone clinical trial', Addiction, vol. 110, no. 1, pp. 51-58. https://doi.org/10.1111/add.12714
Mcpherson, Sterling ; Barbosa-Leiker, Celestina ; Mamey, Mary Rose ; Mcdonell, Michael ; Enders, Craig K. ; Roll, John. / A 'missing not at random' (MNAR) and 'missing at random' (MAR) growth model comparison with a buprenorphine/naloxone clinical trial. In: Addiction. 2015 ; Vol. 110, No. 1. pp. 51-58.
@article{ec35b6e81b54446fa8229ab408b57865,
title = "A 'missing not at random' (MNAR) and 'missing at random' (MAR) growth model comparison with a buprenorphine/naloxone clinical trial",
abstract = "Aims: To compare three missing data strategies: (i) the latent growth model that assumes the data are missing at random (MAR) model; (ii) the Diggle-Kenward missing not at random (MNAR) model, where dropout is a function of previous/concurrent urinalysis (UA) submissions; and (iii) the Wu-Carroll MNAR model where dropout is a function of the growth factors. Design: Secondary data analysis of a National Drug Abuse Treatment Clinical Trials Network trial that examined a 7-day versus 28-day taper (i.e. stepwise decrease in buprenorphine/naloxone) on the likelihood of submitting an opioid-positive UA during treatment. Setting: 11 out-patient treatment settings in 10 US cities. Participants: A total of 516 opioid-dependent participants. Measurements: Opioid UAs provided across the 4-week treatment period. Findings: The MAR model showed a significant effect (B=-0.45, P<0.05) of trial arm on the opioid-positive UA slope (i.e. 28-day taper participants were less likely to submit a positive UA over time) with a small effect size (d=0.20). The MNAR Diggle-Kenward model demonstrated a significant (B=-0.64, P<0.01) effect of trial arm on the slope with a large effect size (d=0.82). The MNAR Wu-Carroll model showed a significant (B=-0.41, P<0.05) effect of trial arm on the UA slope that was relatively small (d=0.31). Conclusions: This performance comparison of three missing data strategies (latent growth model, Diggle-Kenward selection model, Wu-Carrol selection model) on sample data indicates a need for increased use of sensitivity analyses in clinical trial research. Given the potential sensitivity of the trial arm effect to missing data assumptions, it is critical for researchers to consider whether the assumptions associated with each model are defensible.",
keywords = "Latent growth modeling, Longitudinal missing data, Missing not at random models, Randomized clinical trials, Sensitivity analysis, Substance use disorder treatment",
author = "Sterling Mcpherson and Celestina Barbosa-Leiker and Mamey, {Mary Rose} and Michael Mcdonell and Enders, {Craig K.} and John Roll",
year = "2015",
month = "1",
day = "1",
doi = "10.1111/add.12714",
language = "English (US)",
volume = "110",
pages = "51--58",
journal = "Addiction",
issn = "0965-2140",
publisher = "Wiley-Blackwell",
number = "1",

}

TY - JOUR

T1 - A 'missing not at random' (MNAR) and 'missing at random' (MAR) growth model comparison with a buprenorphine/naloxone clinical trial

AU - Mcpherson, Sterling

AU - Barbosa-Leiker, Celestina

AU - Mamey, Mary Rose

AU - Mcdonell, Michael

AU - Enders, Craig K.

AU - Roll, John

PY - 2015/1/1

Y1 - 2015/1/1

N2 - Aims: To compare three missing data strategies: (i) the latent growth model that assumes the data are missing at random (MAR) model; (ii) the Diggle-Kenward missing not at random (MNAR) model, where dropout is a function of previous/concurrent urinalysis (UA) submissions; and (iii) the Wu-Carroll MNAR model where dropout is a function of the growth factors. Design: Secondary data analysis of a National Drug Abuse Treatment Clinical Trials Network trial that examined a 7-day versus 28-day taper (i.e. stepwise decrease in buprenorphine/naloxone) on the likelihood of submitting an opioid-positive UA during treatment. Setting: 11 out-patient treatment settings in 10 US cities. Participants: A total of 516 opioid-dependent participants. Measurements: Opioid UAs provided across the 4-week treatment period. Findings: The MAR model showed a significant effect (B=-0.45, P<0.05) of trial arm on the opioid-positive UA slope (i.e. 28-day taper participants were less likely to submit a positive UA over time) with a small effect size (d=0.20). The MNAR Diggle-Kenward model demonstrated a significant (B=-0.64, P<0.01) effect of trial arm on the slope with a large effect size (d=0.82). The MNAR Wu-Carroll model showed a significant (B=-0.41, P<0.05) effect of trial arm on the UA slope that was relatively small (d=0.31). Conclusions: This performance comparison of three missing data strategies (latent growth model, Diggle-Kenward selection model, Wu-Carrol selection model) on sample data indicates a need for increased use of sensitivity analyses in clinical trial research. Given the potential sensitivity of the trial arm effect to missing data assumptions, it is critical for researchers to consider whether the assumptions associated with each model are defensible.

AB - Aims: To compare three missing data strategies: (i) the latent growth model that assumes the data are missing at random (MAR) model; (ii) the Diggle-Kenward missing not at random (MNAR) model, where dropout is a function of previous/concurrent urinalysis (UA) submissions; and (iii) the Wu-Carroll MNAR model where dropout is a function of the growth factors. Design: Secondary data analysis of a National Drug Abuse Treatment Clinical Trials Network trial that examined a 7-day versus 28-day taper (i.e. stepwise decrease in buprenorphine/naloxone) on the likelihood of submitting an opioid-positive UA during treatment. Setting: 11 out-patient treatment settings in 10 US cities. Participants: A total of 516 opioid-dependent participants. Measurements: Opioid UAs provided across the 4-week treatment period. Findings: The MAR model showed a significant effect (B=-0.45, P<0.05) of trial arm on the opioid-positive UA slope (i.e. 28-day taper participants were less likely to submit a positive UA over time) with a small effect size (d=0.20). The MNAR Diggle-Kenward model demonstrated a significant (B=-0.64, P<0.01) effect of trial arm on the slope with a large effect size (d=0.82). The MNAR Wu-Carroll model showed a significant (B=-0.41, P<0.05) effect of trial arm on the UA slope that was relatively small (d=0.31). Conclusions: This performance comparison of three missing data strategies (latent growth model, Diggle-Kenward selection model, Wu-Carrol selection model) on sample data indicates a need for increased use of sensitivity analyses in clinical trial research. Given the potential sensitivity of the trial arm effect to missing data assumptions, it is critical for researchers to consider whether the assumptions associated with each model are defensible.

KW - Latent growth modeling

KW - Longitudinal missing data

KW - Missing not at random models

KW - Randomized clinical trials

KW - Sensitivity analysis

KW - Substance use disorder treatment

UR - http://www.scopus.com/inward/record.url?scp=84923169883&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=84923169883&partnerID=8YFLogxK

U2 - 10.1111/add.12714

DO - 10.1111/add.12714

M3 - Article

VL - 110

SP - 51

EP - 58

JO - Addiction

JF - Addiction

SN - 0965-2140

IS - 1

ER -