What to do after smoking relapse? A sequential multiple assignment randomized trial of chronic care smoking treatments.

AIM: To compare effects of three post-relapse interventions on smoking abstinence. DESIGN: Sequential three-phase multiple assignment randomized trial (SMART). SETTING: Eighteen Wisconsin, USA, primary care clinics. PARTICIPANTS: A total of 1154 primary care patients (53.6% women, 81.2% White) interested in quitting smoking enrolled from 2015 to 2019; 582 relapsed and were randomized to relapse recovery treatment. INTERVENTIONS: In phase 1, patients received cessation counseling and 8 weeks nicotine patch. Those who relapsed and agreed were randomized to a phase 2 relapse recovery group: (1) reduction counseling + nicotine mini-lozenges + encouragement to quit starting 1 month post-randomization (preparation); (2) repeated encouragement to quit starting immediately post-randomization (recycling); or (3) advice to call the tobacco quitline (control). The first two groups could opt into phase 3 new quit treatment [8 weeks nicotine patch + mini-lozenges plus randomization to two treatment factors (skill training and supportive counseling) in a 2 × 2 design]. Phase 2 and 3 interventions lasted ≤ 15 months. MEASUREMENTS: The study was powered to compare each active phase 2 treatment with the control on the primary outcome: biochemically confirmed 7-day point-prevalence abstinence 14 months post initiating phase 2 relapse recovery treatment. Exploratory analyses tested for phase 3 counseling factor effects. FINDINGS: Neither skill training nor supportive counseling (each on versus off) increased 14-month abstinence rates; skills on versus off 9.3% (14/151) versus 5.2% (8/153), P = 0.19; support on versus off 6.6% (10/152) versus 7.9% (12/152), P = 0.73. Phase 2 preparation did not produce higher 14-month abstinence rates than quitline referral; 3.6% (8/220) versus 2.1% [3/145; risk difference = 1.5%, 95% confidence interval (CI) = -1.8-5.0%, odds ratio (OR) = 1.8, 95% CI = 0.5-6.9]. Recycling, however, produced higher abstinence rates than quitline referral; 6.9% (15/217) versus 2.1% (three of 145; risk difference, 4.8%, 95% CI = 0.7-8.9%, OR = 3.5, 95% CI = 1.0-12.4). Recycling produced greater entry into new quit treatment than preparation: 83.4% (181/217) versus 55.9% (123/220), P < 0.0001. CONCLUSIONS: Among people interested in quitting smoking, immediate encouragement post-relapse to enter a new round of smoking cessation treatment ('recycling') produced higher probability of abstinence than tobacco quitline referral. Recycling produced higher rates of cessation treatment re-engagement than did preparation/cutting down using more intensive counseling and pharmacotherapy.

Predicting response time on self-report rating scale assessments of noncognitive constructs.

Methodological studies of response time on noncognitive assessments have separately demonstrated the relevance of content trait level and response styles as predictive factors. In this paper we examine the simultaneous relevance of both factors as well as the potential for omitted predictor bias when ignoring either factor. Using response time data from several different noncognitive assessments, we demonstrate how a multilevel regression model that attends simultaneously to content and response style factors leads to consistent findings that support the simultaneous relevance of both factors. The average effects of response style consistently emerge as stronger, although also show greater respondent-level variability, possibly due to the multiple different underlying causes of response style behavior. Some implications for the use of response times in noncognitive measurement are considered.

Unipolar IRT and the Author Recognition Test (ART).

Item response theory (IRT) analyses are often used to evaluate measurement error in educational and psychological test instruments. In such contexts, the latent traits/proficiencies are typically assumed normally distributed and a cumulative normal/logistic measurement link function is applied. Such choices are consistent with constructs that are viewed as bipolar in nature and play a critical role in defining the latent proficiency metric against which the measurement error in the test is evaluated. Recently, alternative models that portray the construct as unipolar have been highlighted as being more appropriate for certain psychopathology and personality constructs. In this paper we extend consideration of unipolar IRT models for a recognition task measure, using several example datasets from various versions of the Author Recognition Test (ART), a measure of print exposure. We show how the decision between unipolar versus bipolar IRT modeling has substantial implications for the quantification and interpretation of measurement error in the ART. In sharp contrast to prior bipolar IRT analyses of the ART, under unipolar IRT measurement error in the ART is minimized at low levels of latent print exposure, and increases as latent print exposure increases. Implications for consideration of unipolar IRT with other constructs and measures (e.g., vocabulary, specialized forms of knowledge) that reflect a similar type of response process are considered in the discussion.

Using Item Scores and Distractors to Detect Test Speededness.

Test speededness refers to a situation in which examinee performance is inadvertently affected by the time limit of the test. Because speededness has the potential to severely bias both person and item parameter estimates, it is crucial that speeded examinees are detected. In this article, we develop a change-point analysis (CPA) procedure for detecting test speededness. Our procedure distinguishes itself from existing CPA procedures by using information from both item scores and distractors. Using detailed simulations, we show that under most conditions, the new CPA procedure improves the detection of speeded examinees and produces more accurate change-point estimates. It therefore seems there is a considerable amount of information to be gained from the item distractors, which, quite notably are available in all multiple-choice data. A real data example is also provided.

Relative Robustness of CDMs and (M)IRT in Measuring Growth in Latent Skills.

Previous studies have demonstrated evidence of latent skill continuity even in tests intentionally designed for measurement of binary skills. In addition, the assumption of binary skills when continuity is present has been shown to potentially create a lack of invariance in item and latent ability parameters that may undermine applications. In this article, we examine measurement of growth as one such application, and consider multidimensional item response theory (MIRT) as a competing alternative. Motivated by prior findings concerning the effects of skill continuity, we study the relative robustness of cognitive diagnostic models (CDMs) and (M)IRT models in the measurement of growth under both binary and continuous latent skill distributions. We find CDMs to be a less robust way of quantifying growth under misspecification, and subsequently provide a real-data example suggesting underestimation of growth as a likely consequence. It is suggested that researchers should regularly attend to the assumptions associated with the use of latent binary skills and consider (M)IRT as a potentially more robust alternative if unsure of their discrete nature.

Rejoinder to Commentaries on Lyu, Bolt and Westby's "Exploring the Effects of Item Specific Factors in Sequential and IRTree Models".

We respond to the commentaries on Lyu, Bolt and Westby's "Exploring the effects of item specific factors in sequential and IRTree models." The commentaries raise important points that allow us to clarify our theoretical expectation for item specific factors in many educational and psychological test items. At the same time, we agree with the commentaries in acknowledging challenges associated with providing empirical evidence for their presence and reflect on strategies that might support their estimation. We maintain that the principal concern is the ambiguity item specific factors create in attempting to interpret or use the parameters beyond the first node.

Exploring the Effects of Item-Specific Factors in Sequential and IRTree Models.

Test items for which the item score reflects a sequential or IRTree modeling outcome are considered. For such items, we argue that item-specific factors, although not empirically measurable, will often be present across stages of the same item. In this paper, we present a conceptual model that incorporates such factors. We use the model to demonstrate how the varying conditional distributions of item-specific factors across stages become absorbed into the stage-specific item discrimination and difficulty parameters, creating ambiguity in the interpretations of item and person parameters beyond the first stage. We discuss implications in relation to various applications considered in the literature, including methodological studies of (1) repeated attempt items; (2) answer change/review, (3) on-demand item hints; (4) item skipping behavior; and (5) Likert scale items. Our own empirical applications, as well as several examples published in the literature, show patterns of violations of item parameter invariance across stages that are highly suggestive of item-specific factors. For applications using sequential or IRTree models as analytical models, or for which the resulting item score might be viewed as outcomes of such a process, we recommend (1) regular inspection of data or analytic results for empirical evidence (or theoretical expectations) of item-specific factors; and (2) sensitivity analyses to evaluate the implications of item-specific factors for the intended inferences or applications.

A machine learning analysis of correlates of mortality among patients hospitalized with COVID-19.

It is vital to determine how patient characteristics that precede COVID-19 illness relate to COVID-19 mortality. This is a retrospective cohort study of patients hospitalized with COVID-19 across 21 healthcare systems in the US. All patients (N = 145,944) had COVID-19 diagnoses and/or positive PCR tests and completed their hospital stays from February 1, 2020 through January 31, 2022. Machine learning analyses revealed that age, hypertension, insurance status, and healthcare system (hospital site) were especially predictive of mortality across the full sample. However, multiple variables were especially predictive in subgroups of patients. The nested effects of risk factors such as age, hypertension, vaccination, site, and race accounted for large differences in mortality likelihood with rates ranging from about 2-30%. Subgroups of patients are at heightened risk of COVID-19 mortality due to combinations of preadmission risk factors; a finding of potential relevance to outreach and preventive actions.

The Relationship of COVID-19 Vaccination with Mortality Among 86,732 Hospitalized Patients: Subpopulations, Patient Factors, and Changes over Time.

BACKGROUND: Information on COVID-19 vaccination effects on mortality among patients hospitalized with COVID-19 could inform vaccination outreach efforts and increase understanding of patient risk. OBJECTIVE: Determine the associations of vaccination status with mortality in adult patients hospitalized with COVID-19. DESIGN: This retrospective cohort study assessed the characteristics and mortality rates of adult patients hospitalized with COVID-19 across 21 healthcare systems in the USA from January 1, 2021, to January 31, 2022. PARTICIPANTS: Adult patients admitted to participating hospitals who had COVID-19 diagnoses and/or positive PCR tests and completed their hospital stay via discharge or death. MAIN MEASURE: In-hospital mortality vs. discharge (outcome) and patient age, sex, race, ethnicity, BMI, insurance status, comorbidities, and vaccination status extracted from the electronic health record (EHR). KEY RESULTS: Of 86,732 adult patients hospitalized with COVID-19, 45,082 (52%) were female, mean age was 60 years, 20,800 (24%) were Black, and 22,792 (26.3%) had one or more COVID-19 vaccinations. Statistically adjusted mortality rates for unvaccinated and vaccinated patients were 8.3% (95% CI, 8.1-8.5) and 5.1% (95% CI, 4.8-5.4) respectively (7.9% vs. 4.5% with no immune compromise). Vaccination was associated with especially large reductions in mortality for obese (OR = 0.67; 95% CI 0.56-0.80) and severely obese (OR = 0.52; 95% CI, 0.41-0.67) patients and for older patients (OR = 0.99; 95% CI, 0.98-0.99). Mortality likelihood was higher later in the study period (August 2021-January 31, 2022) than earlier (January 1, 2021-July 30, 2021) (OR = 1.10; 95% CI = 1.04-1.17) and increased significantly for vaccinated patients from 4.6% (95% CI, 3.9-5.2%) to 6.5% (95% CI, 6.2-6.9%). CONCLUSIONS: Patients vaccinated for COVID-19 had reduced mortality, especially for obese/severely obese and older individuals. Vaccination's protective effect against mortality declined over time and hospitalized obese and older individuals may derive especially great benefit from prior vaccination against SARS-CoV-2.

Smoking Status, Nicotine Medication, Vaccination, and COVID-19 Hospital Outcomes: Findings from the COVID EHR Cohort at the University of Wisconsin (CEC-UW) Study.

INTRODUCTION: Available evidence is mixed concerning associations between smoking status and COVID-19 clinical outcomes. Effects of nicotine replacement therapy (NRT) and vaccination status on COVID-19 outcomes in smokers are unknown. METHODS: Electronic health record data from 104 590 COVID-19 patients hospitalized February 1, 2020 to September 30, 2021 in 21 U.S. health systems were analyzed to assess associations of smoking status, in-hospital NRT prescription, and vaccination status with in-hospital death and ICU admission. RESULTS: Current (n = 7764) and never smokers (n = 57 454) did not differ on outcomes after adjustment for age, sex, race, ethnicity, insurance, body mass index, and comorbidities. Former (vs never) smokers (n = 33 101) had higher adjusted odds of death (aOR, 1.11; 95% CI, 1.06-1.17) and ICU admission (aOR, 1.07; 95% CI, 1.04-1.11). Among current smokers, NRT prescription was associated with reduced mortality (aOR, 0.64; 95% CI, 0.50-0.82). Vaccination effects were significantly moderated by smoking status; vaccination was more strongly associated with reduced mortality among current (aOR, 0.29; 95% CI, 0.16-0.66) and former smokers (aOR, 0.47; 95% CI, 0.39-0.57) than for never smokers (aOR, 0.67; 95% CI, 0.57, 0.79). Vaccination was associated with reduced ICU admission more strongly among former (aOR, 0.74; 95% CI, 0.66-0.83) than never smokers (aOR, 0.87; 95% CI, 0.79-0.97). CONCLUSIONS: Former but not current smokers hospitalized with COVID-19 are at higher risk for severe outcomes. SARS-CoV-2 vaccination is associated with better hospital outcomes in COVID-19 patients, especially current and former smokers. NRT during COVID-19 hospitalization may reduce mortality for current smokers. IMPLICATIONS: Prior findings regarding associations between smoking and severe COVID-19 disease outcomes have been inconsistent. This large cohort study suggests potential beneficial effects of nicotine replacement therapy on COVID-19 outcomes in current smokers and outsized benefits of SARS-CoV-2 vaccination in current and former smokers. Such findings may influence clinical practice and prevention efforts and motivate additional research that explores mechanisms for these effects.

Relations of Current and Past Cancer with Severe Outcomes among 104,590 Hospitalized COVID-19 Patients: The COVID EHR Cohort at the University of Wisconsin.

BACKGROUND: There is mixed evidence about the relations of current versus past cancer with severe COVID-19 outcomes and how they vary by patient and cancer characteristics. METHODS: Electronic health record data of 104,590 adult hospitalized patients with COVID-19 were obtained from 21 United States health systems from February 2020 through September 2021. In-hospital mortality and ICU admission were predicted from current and past cancer diagnoses. Moderation by patient characteristics, vaccination status, cancer type, and year of the pandemic was examined. RESULTS: 6.8% of the patients had current (n = 7,141) and 6.5% had past (n = 6,749) cancer diagnoses. Current cancer predicted both severe outcomes but past cancer did not; adjusted odds ratios (aOR) for mortality were 1.58 [95% confidence interval (CI), 1.46-1.70] and 1.04 (95% CI, 0.96-1.13), respectively. Mortality rates decreased over the pandemic but the incremental risk of current cancer persisted, with the increment being larger among younger vs. older patients. Prior COVID-19 vaccination reduced mortality generally and among those with current cancer (aOR, 0.69; 95% CI, 0.53-0.90). CONCLUSIONS: Current cancer, especially among younger patients, posed a substantially increased risk for death and ICU admission among patients with COVID-19; prior COVID-19 vaccination mitigated the risk associated with current cancer. Past history of cancer was not associated with higher risks for severe COVID-19 outcomes for most cancer types. IMPACT: This study clarifies the characteristics that modify the risk associated with cancer on severe COVID-19 outcomes across the first 20 months of the COVID-19 pandemic. See related commentary by Egan et al., p. 3.

The first 20 months of the COVID-19 pandemic: Mortality, intubation and ICU rates among 104,590 patients hospitalized at 21 United States health systems.

MAIN OBJECTIVE: There is limited information on how patient outcomes have changed during the COVID-19 pandemic. This study characterizes changes in mortality, intubation, and ICU admission rates during the first 20 months of the pandemic. STUDY DESIGN AND METHODS: University of Wisconsin researchers collected and harmonized electronic health record data from 1.1 million COVID-19 patients across 21 United States health systems from February 2020 through September 2021. The analysis comprised data from 104,590 adult hospitalized COVID-19 patients. Inclusion criteria for the analysis were: (1) age 18 years or older; (2) COVID-19 ICD-10 diagnosis during hospitalization and/or a positive COVID-19 PCR test in a 14-day window (+/- 7 days of hospital admission); and (3) health system contact prior to COVID-19 hospitalization. Outcomes assessed were: (1) mortality (primary), (2) endotracheal intubation, and (3) ICU admission. RESULTS AND SIGNIFICANCE: The 104,590 hospitalized participants had a mean age of 61.7 years and were 50.4% female, 24% Black, and 56.8% White. Overall risk-standardized mortality (adjusted for age, sex, race, ethnicity, body mass index, insurance status and medical comorbidities) declined from 16% of hospitalized COVID-19 patients (95% CI: 16% to 17%) early in the pandemic (February-April 2020) to 9% (CI: 9% to 10%) later (July-September 2021). Among subpopulations, males (vs. females), those on Medicare (vs. those on commercial insurance), the severely obese (vs. normal weight), and those aged 60 and older (vs. younger individuals) had especially high mortality rates both early and late in the pandemic. ICU admission and intubation rates also declined across these 20 months. CONCLUSIONS: Mortality, intubation, and ICU admission rates improved markedly over the first 20 months of the pandemic among adult hospitalized COVID-19 patients although gains varied by subpopulation. These data provide important information on the course of COVID-19 and identify hospitalized patient groups at heightened risk for negative outcomes. TRIAL REGISTRATION: ClinicalTrials.gov Identifier: NCT04506528 (https://clinicaltrials.gov/ct2/show/NCT04506528).

The Potential for Interpretational Confounding in Cognitive Diagnosis Models.

Binary examinee mastery/nonmastery classifications in cognitive diagnosis models may often be an approximation to proficiencies that are better regarded as continuous. Such misspecification can lead to inconsistencies in the operational definition of "mastery" when binary skills models are assumed. In this paper we demonstrate the potential for an interpretational confounding of the latent skills when truly continuous skills are treated as binary. Using the DINA model as an example, we show how such forms of confounding can be observed through item and/or examinee parameter change when (1) different collections of items (such as representing different test forms) previously calibrated separately are subsequently calibrated together; and (2) when structural restrictions are placed on the relationships among skill attributes (such as the assumption of strictly nonnegative growth over time), among other possibilities. We examine these occurrences in both simulation and real data studies. It is suggested that researchers should regularly attend to the potential for interpretational confounding by studying differences in attribute mastery proportions and/or changes in item parameter (e.g., slip and guess) estimates attributable to skill continuity when the same samples of examinees are administered different test forms, or the same test forms are involved in different calibrations.

Item Complexity: A Neglected Psychometric Feature of Test Items?

Despite its frequent consideration in test development, item complexity receives little attention in the psychometric modeling of item response data. In this address, I consider how variability in item complexity can be expected to emerge in the form of item characteristic curve (ICC) asymmetry, and how such effects may significantly influence applications of item response theory, especially those that assume interval-level properties of the latent proficiency metric and groups that vary substantially in mean proficiency. One application is the score gain deceleration phenomenon often observed in vertical scaling contexts, especially in subject areas like math or second language acquisition. It is demonstrated how the application of symmetric IRT models in the presence of complexity-induced positive ICC asymmetry can be a likely cause. A second application concerns the positive correlation between DIF and difficulty commonly seen in verbal proficiency (and other subject area) tests where problem-solving complexity is minimal and proficiency-related guessing effects are likely pronounced. Here we suggest negative ICC asymmetry as a probable cause and apply sensitivity analyses to demonstrate the ease with which such correlations disappear when allowing for negative ICC asymmetry. Unfortunately, the presence of systematic forms of ICC asymmetry is easily missed due to the considerable flexibility afforded by latent trait metrics in IRT. Speculation is provided regarding other applications for which attending to ICC asymmetry may prove useful.

Noncompensatory MIRT For Passage-Based Tests.

We consider a multidimensional noncompensatory approach for binary items in passage-based tests. The passage-based noncompensatory model (PB-NM) emphasizes two underlying components in solving passage-based test items: a passage-related component and a passage-independent component. An advantage of the PB-NM model over commonly applied compensatory models (e.g., bifactor model) is that the two components are parameterized in relation to difficulty as opposed to discrimination parameters. As a result, while simultaneously accounting for passage-related local item dependence, the model permits the assessment of how items based on the same passage may require varying levels of passage comprehension (as well as varying levels of passage-independent proficiency) to obtain a correct response. Through a simulation study, we evaluate the comparative fit of the PB-NM against the bifactor model and also illustrate the relationship between the difficulty parameters of the PB-NM and the discrimination parameters of the bifactor model. We further apply the PB-NM to an actual reading comprehension test to demonstrate the relevance of the model in understanding variation in the relative difficulty of the two components across different item types.

A psychometric model for respondent-level anchoring on self-report rating scale instruments.

Among the various forms of response bias that can emerge with self-report rating scale assessments are those related to anchoring, the tendency for respondents to select categories in close proximity to the rating category used for the immediately preceding item. In this study we propose a psychometric model based on a multidimensional nominal model for response style that also simultaneously accommodates a respondent-level anchoring tendency. The model is estimated using a fully Bayesian estimation procedure. By applying this model to a real test data set measuring extraversion, we explore a theory that both response styles and anchoring might be viewed as evidence of a lack of effortful responding. Empirical results show that there is a positive correlation between the strength of midpoint response style and the anchoring effect; further, responses indicative of either anchoring or response style both negatively correlate with response time, consistent with a theory that both phenomena reflect reduced respondent effort. The results support attending to both anchoring and midpoint response style as ways of assessing respondent engagement.

Understanding contexts of smoking and vaping among dual users: analysis of ecological momentary assessment data.

AIMS: To understand dual users' cigarette and e-cigarette use patterns, including the contexts in which they vape versus smoke and to understand how environmental and internal contexts and smoking patterns differ between dual users and exclusive smokers. DESIGN: Longitudinal observational trial. SETTING: Research center in Wisconsin, USA. PARTICIPANTS: Adult dual users (n = 162) and adults who exclusively smoked (n = 143), with no plans to quit smoking or vaping in the next 30 days. MEASUREMENTS: Participants carried smartphones for 2 weeks at baseline to record each use event for the two products and report on the context of their product use. The percentage of mornings where participants vaped first versus smoked were used to compute e-cigarette dependence. FINDINGS: Hierarchical linear regression models with random slopes and intercepts examined the within- and between-subject effects of context on the likelihood of vaping (versus smoking); significant fixed effects were tested for moderation by e-cigarette dependence. Dual users reported significantly more puffs/cigarette [mean = 13.1, standard deviation (SD) = 10.2] than puffs/vape event (mean = 11.7, SD = 11.5; P = 0.01). E-cigarette dependence moderated the influence of social cues (t-ratio = 2.4, P = 0.02) and smoking restrictions (t-ratio = 3.1, P = 0.003) on the likelihood of vaping versus smoking [odds ratio (OR) = 2.30, P = 0.02]. Context was more related to which product was used in those of low versus higher e-cigarette dependence. Reports of strong cravings to smoke and positive expectancies for cigarettes were associated with a reduced likelihood of vaping, whereas strong cravings to vape and positive vaping expectancies were related to increased likelihood of vaping. CONCLUSIONS: Among dual users of cigarettes and e-cigarettes with no motivation to quit, vaping appears to be related to internal cues and more highly linked with social contexts and smoking restrictions (i.e. under stronger external stimulus control) among those with low to moderate e-cigarette dependence compared with high e-cigarette dependence. These findings illustrate the importance of contextual factors in tobacco product use among dual users with the influence of context being reduced at high levels of e-cigarette dependence.

Effects of Combined Varenicline With Nicotine Patch and of Extended Treatment Duration on Smoking Cessation: A Randomized Clinical Trial.

Importance: Smoking cessation medications are routinely used in health care. Research suggests that combining varenicline with the nicotine patch, extending the duration of varenicline treatment, or both, may increase cessation effectiveness. Objective: To compare combinations of varenicline plus the nicotine or placebo patch vs combinations used for either 12 weeks (standard duration) or 24 weeks (extended duration). Design, Settings, and Participants: Double-blind, 2 × 2 factorial randomized clinical trial conducted from November 11, 2017, to July 9, 2020, at 1 research clinic in Madison, Wisconsin, and at 1 clinic in Milwaukee, Wisconsin. Of the 5836 adults asked to participate in the study, 1251 who smoked 5 cigarettes/d or more were randomized. Interventions: All participants received cessation counseling and were randomized to 1 of 4 medication groups: varenicline monotherapy for 12 weeks (n = 315), varenicline plus nicotine patch for 12 weeks (n = 314), varenicline monotherapy for 24 weeks (n = 311), or varenicline plus nicotine patch for 24 weeks (n = 311). Main Outcomes and Measures: The primary outcome was carbon monoxide-confirmed self-reported 7-day point prevalence abstinence at 52 weeks. Results: Among 1251 patients who were randomized (mean [SD] age, 49.1 [11.9] years; 675 [54.0%] women), 751 (60.0%) completed treatment and 881 (70.4%) provided final follow-up. For the primary outcome, there was no significant interaction between the 2 treatment factors of medication type and medication duration (odds ratio [OR], 1.03 [95% CI, 0.91 to 1.17]; P = .66). For patients randomized to 24-week vs 12-week treatment duration, the primary outcome occurred in 24.8% (154/622) vs 24.3% (153/629), respectively (risk difference, -0.4% [95% CI, -5.2% to 4.3%]; OR, 1.01 [95% CI, 0.89 to 1.15]). For patients randomized to varenicline combination therapy vs varenicline monotherapy, the primary outcome occurred in 24.3% (152/625) vs 24.8% (155/626), respectively (risk difference, 0.4% [95% CI, -4.3% to 5.2%]; OR, 0.99 [95% CI, 0.87 to 1.12]). Nausea occurrence ranged from 24.0% to 30.9% and insomnia occurrence ranged from 24.4% to 30.5% across the 4 groups. Conclusions and Relevance: Among adults smoking 5 cigarettes/d or more, there were no significant differences in 7-day point prevalence abstinence at 52 weeks among those treated with combined varenicline plus nicotine patch therapy vs varenicline monotherapy, or among those treated for 24 weeks vs 12 weeks. These findings do not support the use of combined therapy or of extended treatment duration. Trial Registration: ClinicalTrials.gov Identifier: NCT03176784.

Data Missing Not at Random in Mobile Health Research: Assessment of the Problem and a Case for Sensitivity Analyses.

BACKGROUND: Missing data are common in mobile health (mHealth) research. There has been little systematic investigation of how missingness is handled statistically in mHealth randomized controlled trials (RCTs). Although some missing data patterns (ie, missing at random [MAR]) may be adequately addressed using modern missing data methods such as multiple imputation and maximum likelihood techniques, these methods do not address bias when data are missing not at random (MNAR). It is typically not possible to determine whether the missing data are MAR. However, higher attrition in active (ie, intervention) versus passive (ie, waitlist or no treatment) conditions in mHealth RCTs raise a strong likelihood of MNAR, such as if active participants who benefit less from the intervention are more likely to drop out. OBJECTIVE: This study aims to systematically evaluate differential attrition and methods used for handling missingness in a sample of mHealth RCTs comparing active and passive control conditions. We also aim to illustrate a modern model-based sensitivity analysis and a simpler fixed-value replacement approach that can be used to evaluate the influence of MNAR. METHODS: We reanalyzed attrition rates and predictors of differential attrition in a sample of 36 mHealth RCTs drawn from a recent meta-analysis of smartphone-based mental health interventions. We systematically evaluated the design features related to missingness and its handling. Data from a recent mHealth RCT were used to illustrate 2 sensitivity analysis approaches (pattern-mixture model and fixed-value replacement approach). RESULTS: Attrition in active conditions was, on average, roughly twice that of passive controls. Differential attrition was higher in larger studies and was associated with the use of MAR-based multiple imputation or maximum likelihood methods. Half of the studies (18/36, 50%) used these modern missing data techniques. None of the 36 mHealth RCTs reviewed conducted a sensitivity analysis to evaluate the possible consequences of data MNAR. A pattern-mixture model and fixed-value replacement sensitivity analysis approaches were introduced. Results from a recent mHealth RCT were shown to be robust to missing data, reflecting worse outcomes in missing versus nonmissing scores in some but not all scenarios. A review of such scenarios helps to qualify the observations of significant treatment effects. CONCLUSIONS: MNAR data because of differential attrition are likely in mHealth RCTs using passive controls. Sensitivity analyses are recommended to allow researchers to assess the potential impact of MNAR on trial results.