When does attrition lead to biased estimates of alcohol consumption? Bias analysis for loss to follow-up in 30 longitudinal cohorts

Katherine M Keyes; Justin Jager; Jonathan Platt; Caroline Rutherford; Megan E Patrick; Deborah D Kloska; John Schulenberg

doi:10.1002/mpr.1842

Back

When does attrition lead to biased estimates of alcohol consumption? Bias analysis for loss to follow-up in 30 longitudinal cohorts

Journal article

Open access

Peer reviewed

When does attrition lead to biased estimates of alcohol consumption? Bias analysis for loss to follow-up in 30 longitudinal cohorts

Katherine M Keyes, Justin Jager, Jonathan Platt, Caroline Rutherford, Megan E Patrick, Deborah D Kloska and John Schulenberg

International journal of methods in psychiatric research, Vol.29(4), pp.1-9

12/2020

DOI: 10.1002/mpr.1842

PMCID: PMC7723204

PMID: 32656917

Files and links (1)

url

https://doi.org/10.1002/mpr.1842View

Published (Version of record) Open Access

Abstract

Survey nonresponse has increased across decades, making the amount of attrition a focus in generating inferences from longitudinal data. Use of inverse probability weights [IPWs] and other statistical approaches are common, but residual bias remains a threat. Quantitative bias analysis for nonrandom attrition as an adjunct to IPW may yield more robust inference. Data were drawn from the Monitoring the Future panel studies [twelfth grade, base-year: 1976-2005; age 29/30 follow-up: 1987-2017, N = 73,298]. We then applied IPW imputation in increasing percentages, assuming varying risk differences [RDs] among nonresponders. Measurements included past-two-week binge drinking at base-year and every follow-up. Demographic and other correlates of binge drinking contributed to IPW estimation. Attrition increased: 31.14%, base-year 1976; 61.33%, base-year 2005. The magnitude of bias depended not on attrition rate but on prevalence of binge drinking and RD among nonrespondents. The probable range of binge drinking among nonresponders was 12-45%. In every scenario, base-year and follow-up binge drinking were associated. The likely range of true RDs was 0.14 [95% CI: 0.11-0.17] to 0.28 [95% CI: 0.25-0.31]. When attrition is present, the amount of attrition alone is insufficient to understand contribution to effect estimates. We recommend including bias analysis in longitudinal analyses.

alcohol

adolescents

attrition

Details

Title: Subtitle: When does attrition lead to biased estimates of alcohol consumption? Bias analysis for loss to follow-up in 30 longitudinal cohorts
Creators: Katherine M Keyes - Department of Epidemiology, Mailman School of Public Health, Columbia University, New York, New York, USA
Justin Jager - School of Social and Family Dynamics, Arizona State University, Tempe, Arizona, USA
Jonathan Platt - Department of Epidemiology, Mailman School of Public Health, Columbia University, New York, New York, USA
Caroline Rutherford - Department of Epidemiology, Mailman School of Public Health, Columbia University, New York, New York, USA
Megan E Patrick - Institute of Child Development, University of Minnesota, Minneapolis, Minnesota, USA
Deborah D Kloska - Institute for Social Research, University of Michigan, Ann Arbor, Michigan, USA
John Schulenberg - Department of Psychology, University of Michigan, Ann Arbor, Michigan, USA
Resource Type: Journal article
Publication Details: International journal of methods in psychiatric research, Vol.29(4), pp.1-9
DOI: 10.1002/mpr.1842
PMID: 32656917
PMCID: PMC7723204
NLM abbreviation: Int J Methods Psychiatr Res
ISSN: 1049-8931
eISSN: 1557-0657
Grant note: R01DA037902 / NIDA NIH HHS R01 DA001411 / NIDA NIH HHS R01AA026861 / NIDA NIH HHS R01 DA016575 / NIDA NIH HHS
Language: English
Date published: 12/2020
Academic Unit: Epidemiology; Injury Prevention Research Center
Record Identifier: 9984214702902771

Metrics

6 Record Views

16 Times Cited - Web of Science