Nicotine & Tobacco Research, 2015, 1–9 doi:10.1093/ntr/ntv055 Original investigation © The Author 2015. Published by Oxford University Press on behalf of the Society for Research on Nicotine and Tobacco. All rights reserved. For permissions, please e-mail: journals.permissions@oup.com. 1 Introduction In this study, the validity of common approaches to handle missing smoking status data, such as penalized imputation (PI) 1 —sometimes referred to as “missing=smoking”—and multiple imputation (MI) 2 is evaluated for internet-based smoking cessation trials. Missing smoking status data are often caused by participant drop-out before follow-up measurements took place. Randomized controlled trials of internet-based interventions including those for smoking cessation typically face higher drop-out rates than face-to-face psychological intervention trials or drug trials. 3 High drop-out rates are sometimes even considered a typical characteristic of internet interventions, Original investigation The Missing=Smoking Assumption: A Fallacy in Internet-Based Smoking Cessation Trials? Matthijs Blankers PhD 1,2,3 , Eline Suzanne Smit PhD 4,5 , Peggy van der Pol PhD 6 , Hein de Vries PhD 5 , Ciska Hoving PhD 5 , Margriet van Laar PhD 1,6 1 Netherlands Expertise Centre on Tobacco Control (NET), Trimbos Institute, Utrecht, the Netherlands; 2 Department of Psychiatry, Academic Medical Centre, University of Amsterdam, Amsterdam, the Netherlands; 3 Department of Research, Arkin, Amsterdam, the Netherlands; 4 Amsterdam School of Communication Research/ASCoR, Department of Communication Science, University of Amsterdam, Amsterdam, the Netherlands; 5 CAPHRI School for Public Health and Primary Care, Department of Health Promotion, Maastricht University, Maastricht, the Netherlands; 6 Department of Drug Monitoring, Trimbos Institute, Utrecht, the Netherlands Corresponding Author: Matthijs Blankers, PhD, Trimbos-instituut, PO Box 725, 3500 AS, Utrecht, the Netherlands. Telephone: +31-30-297-11-00; Fax: +31-30-297-11-11; E-mail: mblankers@trimbos.nl Abstract Introduction: In this study, penalized imputation (PI), a common approach to handle missing smok- ing status data and sometimes referred to as “missing=smoking,” is compared with other missing data approaches using data from internet-based smoking cessation trials. Two hypotheses were tested: (1) PI leads to more conservative effect estimates than complete observations analysis; and (2) PI and multiple imputation (MI) lead to similar effect estimates under balanced (equal missing- ness proportions among the trial arms) and unbalanced missingness. Methods: First, the outcomes of 22 trials included in a recent Cochrane review on internet-based smoking cessation interventions were reanalyzed using only the complete observations, and after applying PI. Second, in a simulation study outcomes under PI, complete observations analysis, and two types of MI were compared. For this purpose, individual patient data from one of the Cochrane review trials were used. Results of the missing data approaches were compared with reference data without missing observations, upon which balanced and unbalanced missingness scenarios were imposed. Results: In the reanalysis of 22 trials, relative risks (RR = 1.15 [1.00; 1.33]) after PI were nearly identical to those under complete observations analysis (RR = 1.14 [0.98; 1.32]). In the simulation study, PI was the only approach that led to deviations from the reference data beyond its 95% confdence interval. Conclusions: Analyses after PI led to pooled results equivalent to complete observations analy- ses. PI also led to signifcant deviations from the reference in the simulation studies. PI biases the reported effects of interventions, favoring the condition with the lowest proportion of missingness. Therefore, more sophisticated missing data approaches than PI should be applied. Nicotine & Tobacco Research Advance Access published April 3, 2015 at Universiteit van Amsterdam on April 7, 2015 http://ntr.oxfordjournals.org/ Downloaded from