Analysis of binary outcomes with missing data: missing = smoking, last observation carried forward, and a little multiple imputation

Hedeker, Donald; Mermelstein, Robin J.; Demirtaş, Hakan

doi:10.1111/j.1360-0443.2007.01946.x

Cited by 146 publications

(175 citation statements)

References 25 publications

Supporting

Mentioning

169

Contrasting

Unclassified

Order By: Relevance

“…Third, although a key advantage of the mixed-effects modeling approach is that it can be applied when participants are not measured at the same number of time points, some fraction of participants might be missing because they were not remitted. We repeated the analyses presented in Tables 3 and 5 with missing outcomes recoded as nonremission (43) and found similar results, which suggests the robustness of the findings.…”

Section: Discussionsupporting

confidence: 56%

Continuing Care and Long-Term Substance Use Outcomes in Managed Care: Early Evidence for a Primary Care-Based Model

Felicia¹,

Parthasarathy²,

Mertens³

et al. 2011

Psychiatric Services

View full text Add to dashboard Cite

Objectives How best to provide ongoing services to patients with substance use disorders to sustain long-term recovery is a significant clinical and policy question that has not been adequately addressed. Analyzing nine years of prospective data for 991 adults who entered substance abuse treatment in a private, nonprofit managed care health plan, this study aimed to examine the components of a continuing care model (primary care, specialty substance abuse treatment, and psychiatric services) and their combined effect on outcomes over nine years after treatment entry. Methods In a longitudinal observational study, follow-up measures included self-reported alcohol and drug use, Addiction Severity Index scores, and service utilization data extracted from the health plan databases. Remission, defined as abstinence or non-problematic use, was the outcome measure. Results A mixed-effects logistic random intercept model controlling for time and other covariates found that yearly primary care, and specialty care based on need as measured at the prior time point, were positively associated with remission over time. Persons receiving continuing care (defined as having yearly primary care and specialty substance abuse treatment and psychiatric services when needed) had twice the odds of achieving remission at follow-ups (p<.001) as those without. Conclusions Continuing care that included both primary care and specialty care management to support ongoing monitoring, self-care, and treatment as needed was important for long-term recovery of patients with substance use disorders.

show abstract

Section: Discussionsupporting

confidence: 56%

Continuing Care and Long-Term Substance Use Outcomes in Managed Care: Early Evidence for a Primary Care-Based Model

Felicia¹,

Parthasarathy²,

Mertens³

et al. 2011

Psychiatric Services

View full text Add to dashboard Cite

show abstract

“…Despite the conservative nature of this assumption, the assumption that loss to follow up meant participants were still smoking only contradicted one finding from the primary analyses (i.e., those achieving a reduction of greater than 50% at week 16). It thus appears that this assumption has the potential to under-estimate the beneficial effects of the intervention [30][31][32] , and although this approach has been widely advocated, more research into dealing with missing data in smoking trials is justified.…”

Section: Discussionmentioning

confidence: 99%

“…multiple imputation of missing data), which may provide more reliable estimates of treatment effects [30][31][32] .…”

Section: Downloaded Frommentioning

confidence: 99%

An Exploratory Analysis of the Smoking and Physical Activity Outcomes From a Pilot Randomized Controlled Trial of an Exercise Assisted Reduction to Stop Smoking Intervention in Disadvantaged Groups

Thompson¹,

Greaves²,

Ayres³

et al. 2015

NICTOB

View full text Add to dashboard Cite

“…For the primary outcome, to assess the influence of the assumption that missing data equals 'smoking' on the effect size, we used the Hedeker method to test various scenarios of the association between smoking and having missing data. 84 Other outcomes for smoking cessation were analysed in a similar way.…”

Section: Sample Sizementioning

confidence: 99%

The London Exercise And Pregnant smokers (LEAP) trial: a randomised controlled trial of physical activity for smoking cessation in pregnancy with an economic evaluation

Ussher

Lewis

Aveyard

et al. 2015

Health Technol Assess

View full text Add to dashboard Cite

BackgroundSmoking during pregnancy is the main preventable cause of poor birth outcomes. Improved methods are needed to help women to stop smoking during pregnancy. Pregnancy provides a compelling rationale for physical activity (PA) interventions as cessation medication is contraindicated or ineffective, and an effective PA intervention could be highly cost-effective.ObjectiveTo examine the effectiveness and cost-effectiveness of a PA intervention plus standard behavioural support for smoking cessation relative to behavioural support alone for achieving smoking cessation at the end of pregnancy.DesignMulticentre, two-group, pragmatic randomised controlled trial and economic evaluation with follow-up at the end of pregnancy and 6 months postnatally. Randomisation was stratified by centre and a computer-generated sequence was used to allocate participants using a 1 : 1 ratio.Setting13 hospitals offering antenatal care in the UK.ParticipantsWomen between 10 and 24 weeks’ gestation smoking five or more cigarettes a day before pregnancy and one or more during pregnancy.InterventionsParticipants were randomised to behavioural support for smoking cessation (control) or behavioural support plus a PA intervention consisting of supervised treadmill exercise plus PA consultations. Neither participants nor researchers were blinded to treatment allocation.Main outcome measuresThe primary outcome was self-reported, continuous smoking abstinence between a quit date and end of pregnancy, validated by expired carbon monoxide and/or salivary cotinine. Secondary outcomes were maternal weight, depression, birth outcomes, withdrawal symptoms and urges to smoke. The economic evaluation investigated the costs of the PA intervention compared with the control intervention.ResultsIn total, 789 women were randomised (n = 394 PA,n = 395 control). Four were excluded post randomisation (two had been enrolled twice in sequential pregnancies and two were ineligible and randomised erroneously). The intention-to-treat analysis comprised 785 participants (n = 392 PA,n = 393 control). There was no significant difference in the rate of abstinence at the end of pregnancy between the PA group (7.7%) and the control group (6.4%) [odds ratio for PA group abstinence 1.21, 95% confidence interval (CI) 0.70 to 2.10]. For the PA group compared with the control group, there was a 33% (95% CI 14% to 56%), 28% (95% CI 7% to 52%) and 36% (95% CI 12% to 65%) significantly greater increase in self-reported minutes of moderate- and vigorous-intensity PA from baseline to 1 week, 4 weeks and 6 weeks respectively. Accelerometer data showed that there was no significant difference in PA levels between the groups. There were no significant differences between the groups for change in maternal weight, depression, withdrawal symptoms or urges to smoke. Adverse events and birth outcomes were similar between the groups except for there being significantly more caesarean births in the control group than in the PA group (28.7% vs. 21.3%;p < 0.023). The PA intervention was less costly than the control intervention by £35 per participant. This was mainly attributable to increased health-care usage in the control group. However, there was considerable statistical uncertainty around this estimate.ConclusionsDuring pregnancy, offering an intervention combining supervised exercise and PA counselling does not add to the effectiveness of behavioural support for smoking cessation. Only 10% of participants had PA levels accessed by accelerometer and it is, therefore, unclear whether or not the lack of an effect on the primary outcome is the result of insufficient increases in PA. Research is needed to identify the smoking populations most suitable for PA interventions and methods for increasing PA adherence.Trial registrationCurrent Controlled Trials ISRCTN48600346.FundingThis project was funded by the NIHR Health Technology Assessment programme and will be published in full inHealth Technology Assessment; Vol. 19, No. 84. See the NIHR Journals Library website for further project information.

show abstract

Analysis of binary outcomes with missing data: missing = smoking, last observation carried forward, and a little multiple imputation

Cited by 146 publications

References 25 publications

Continuing Care and Long-Term Substance Use Outcomes in Managed Care: Early Evidence for a Primary Care-Based Model

Continuing Care and Long-Term Substance Use Outcomes in Managed Care: Early Evidence for a Primary Care-Based Model

An Exploratory Analysis of the Smoking and Physical Activity Outcomes From a Pilot Randomized Controlled Trial of an Exercise Assisted Reduction to Stop Smoking Intervention in Disadvantaged Groups

The London Exercise And Pregnant smokers (LEAP) trial: a randomised controlled trial of physical activity for smoking cessation in pregnancy with an economic evaluation

Contact Info

Product

Resources

About