Tailoring CONSORT-SPI to improve the reporting of smoking cessation intervention trials: An expert consensus study.

BACKGROUND AND AIMS: Inadequate reporting of smoking cessation intervention trials is common and leads to significant challenges for researchers. The aim of this study was to tailor CONSORT (Consolidated Standards of Reporting Trials)-SPI (Social and Psychological Interventions) guidelines to improve reporting of trials of behavioural interventions to promote smoking cessation. METHOD: Informed by missing data from the IC-SMOKE (Intervention and Comparison group support provided in SMOKing cEssation) systematic review project, this study used a multi-stage Delphi process to examine which items could be added or modified to improve the reporting of smoking cessation trials. The first stage involved an on-line survey of 17 international experts in smoking cessation and trial methodology voting on the importance of items for inclusion in the updated guidelines. This was followed by a face-to-face expert consensus meeting attended by 15 of these experts, where the final inclusion and exclusion of new items and modifications were agreed upon. A nine-point Likert scale was used to establish consensus, with suggested modifications requiring agreement of 75% or more. Disagreements in the first stage were presented again at the second stage for discussion and a second round of voting. Only items which reached the threshold for agreement were included. RESULTS: The experts agreed on the inclusion of 10 new items and the specification of 12 existing items. This included modifications that could apply to trials more widely (e.g. the rationale for the comparator), but also modifications that were very specific to smoking cessation trials (e.g. the reporting of smoking cessation outcomes). CONCLUSIONS: A Delphi study has developed a modified CONSORT-SPI guideline (CONSORT-SPI-SMOKE) to improve the reporting of trials of behavioural interventions to promote smoking cessation.

Estimating the true effectiveness of smoking cessation interventions under variable comparator conditions: A systematic review and meta-regression.

BACKGROUND AND AIMS: Behavioural smoking cessation trials have used comparators that vary considerably between trials. Although some previous meta-analyses made attempts to account for variability in comparators, these relied on subsets of trials and incomplete data on comparators. This study aimed to estimate the relative effectiveness of (individual) smoking cessation interventions while accounting for variability in comparators using comprehensive data on experimental and comparator interventions. METHODS: A systematic review and meta-regression was conducted including 172 randomised controlled trials with at least 6 months follow-up and biochemically verified smoking cessation. Authors were contacted to obtain unpublished information. This information was coded in terms of active content and attributes of the study population and methods. Meta-regression was used to create a model predicting smoking cessation outcomes. This model was used to re-estimate intervention effects, as if all interventions have been evaluated against the same comparators. Outcome measures included log odds of smoking cessation for the meta-regression models and smoking cessation differences and ratios to compare relative effectiveness. RESULTS: The meta-regression model predicted smoking cessation rates well (pseudo R2 = 0.44). Standardising the comparator had substantial impact on conclusions regarding the (relative) effectiveness of trials and types of intervention. Compared with a 'no support comparator', self-help was 1.33 times (95% CI = 1.16-1.49), brief physician advice 1.61 times (95% CI = 1.31-1.90), nurse individual counselling 1.76 times (95% CI = 1.62-1.90), psychologist individual counselling 2.04 times (95% CI = 1.95-2.15) and group psychologist interventions 2.06 times (95% CI = 1.92-2.20) more effective. Notably, more elaborate experimental interventions (e.g. psychologist counselling) were typically compared with more elaborate comparators, masking their effectiveness. CONCLUSIONS: Comparator variability and underreporting of comparators obscures the interpretation, comparison and generalisability of behavioural smoking cessation trials. Comparator variability should, therefore, be taken into account when interpreting and synthesising evidence from trials. Otherwise, policymakers, practitioners and researchers may draw incorrect conclusions about the (cost) effectiveness of smoking cessation interventions and their constituent components.

Practitioner and digitally delivered interventions for reducing hazardous and harmful alcohol consumption in people not seeking alcohol treatment: a systematic review and network meta-analysis.

AIM: To compare the effectiveness of practitioner versus digitally delivered interventions for reducing hazardous and harmful alcohol consumption. DESIGN: Systematic review and network meta-analysis comprising comprehensive search for randomised controlled trials, robust screening and selection methods and appraisal with the Cochrane Risk of Bias tool. Network meta-analyses were conducted in Stata using random effects, frequentist models. The confidence in network meta-analysis (CINeMA) tool was used to assess confidence in effect sizes. SETTING: Online or community or health settings where the intervention was immediately accessible without referral. PARTICIPANTS: Non treatment-seeking hazardous or harmful drinkers. MEASUREMENTS: Primary outcome was mean difference in alcohol consumption (g/wk); secondary outcome was number of single high intensity drinking episodes. Baseline consumption was analysed as a covariate. FINDINGS: Of 201 included trials (94 753 participants), 152 reported a consumption outcome that could be converted to grams/week; 104 reported number of single high intensity drinking episodes. At 1 and 6 months, practitioner delivered interventions reduced consumption more than digitally delivered interventions (1 month: -23 g/wk (95% CI, -43 to -2); 6 months: -14 g/wk [95% CI, -25 to -3]). At 12 months there was no evidence of difference between practitioner and digitally delivered interventions (-6 g/wk [95% CI, -24 to 12]). There was no evidence of a difference in single high intensity drinking episodes between practitioner and digitally delivered interventions at any time point. Effect sizes were small, but could impact across a population with relatively high prevalence of hazardous and harmful drinking. Heterogeneity was a concern. Some inconsistency was indicated at 1 and 6 months, but little evidence was apparent at 12 months. CONCLUSION: Practitioner delivered interventions for reducing hazardous and harmful alcohol consumption are more effective than digitally delivered interventions up to 6 months; at 12 months there is no evidence of a difference.

Using machine learning to extract information and predict outcomes from reports of randomised trials of smoking cessation interventions in the Human Behaviour-Change Project.

Background  Using reports of randomised trials of smoking cessation interventions as a test case, this study aimed to develop and evaluate machine learning (ML) algorithms for extracting information from study reports and predicting outcomes as part of the Human Behaviour-Change Project. It is the first of two linked papers, with the second paper reporting on further development of a prediction system. Methods  Researchers manually annotated 70 items of information ('entities') in 512 reports of randomised trials of smoking cessation interventions covering intervention content and delivery, population, setting, outcome and study methodology using the Behaviour Change Intervention Ontology. These entities were used to train ML algorithms to extract the information automatically. The information extraction ML algorithm involved a named-entity recognition system using the 'FLAIR' framework. The manually annotated intervention, population, setting and study entities were used to develop a deep-learning algorithm using multiple layers of long-short-term-memory (LSTM) components to predict smoking cessation outcomes. Results  The F1 evaluation score, derived from the false positive and false negative rates (range 0-1), for the information extraction algorithm averaged 0.42 across different types of entity (SD=0.22, range 0.05-0.88) compared with an average human annotator's score of 0.75 (SD=0.15, range 0.38-1.00). The algorithm for assigning entities to study arms ( e.g., intervention or control) was not successful. This initial ML outcome prediction algorithm did not outperform prediction based just on the mean outcome value or a linear regression model. Conclusions  While some success was achieved in using ML to extract information from reports of randomised trials of smoking cessation interventions, we identified major challenges that could be addressed by greater standardisation in the way that studies are reported. Outcome prediction from smoking cessation studies may benefit from development of novel algorithms, e.g., using ontological information to inform ML (as reported in the linked paper 3).

Pragmatic evaluation of methods for retrieving unpublished information on comparator interventions in a systematic review of smoking cessation trials.

OBJECTIVE: Reporting of the content and delivery characteristics of comparator interventions in published articles is often incomplete. This study examines the feasibility and validity of two methods for collecting additional information on comparator interventions from trial authors. METHODS & MEASURES: In a systematic review of smoking cessation trials (IC-Smoke), all trial authors were asked to send unpublished comparator intervention materials and complete a specially-developed comparator intervention checklist. All published and additionally obtained information from authors were coded for behaviour change techniques (BCTs) and other characteristics (type of comparator, provider, provider training, delivery mode and treatment duration). To assess representativeness, we assessed the amount of additional information obtained from trial authors compared with the amount that was published. We examined known-group and convergent validity of comparator intervention data when using only published or also unpublished information. RESULTS: Additional information were obtained from 91/136 (67%) of trial authors. Representativeness, known-group and convergent validity improved substantially based on the data collected by means of the comparator intervention checklist, but not by requesting authors to send any existing comparator materials. CONCLUSIONS: Requesting authors for unpublished comparator intervention data, using specially-developed checklists and unpublished materials, substantially improves the quality of data available for systematic reviews.

Predictors of recruitment and retention in randomized controlled trials of behavioural smoking cessation interventions: a systematic review and meta-regression analysis.

AIM: To investigate predictors of participant eligibility, recruitment and retention in behavioural randomized controlled trials (RCTs) for smoking cessation. METHOD: Systematic review and pre-specified meta-regression analysis of behavioural RCTs for smoking cessation including adult (≥ 18-year-old) smokers. The pre-specified predictors were identified through a literature review and experts' consultation and included participant, trial and intervention characteristics and recruitment and retention strategies. Outcome measures included eligibility rates (proportion of people eligible for the trials), recruitment rates, retention rates and differential retention rates. RESULTS: A total of 172 RCTs with 89 639 participants. Eligibility [median 57.6%; interquartile range (IQR) = 34.7-83.7], recruitment (median 66.4%; IQR = 42.7-85.2) and retention rates (median 80.5%; IQR = 68.5-89.5) varied considerably across studies. For eligibility rates, the recruitment strategy appeared not to be associated with eligibility rates. For recruitment rates, use of indirect recruitment strategies (e.g. public announcements) [odds ratio (OR) = 0.30, 95% confidence interval (CI) = 0.11-0.82] and self-help interventions (OR = 0.14, 95% CI = 0.03-0.67) were associated with lower recruitment rates. For retention rates, higher retention was seen if the sample had ongoing physical health condition/s (OR = 1.66, 95% CI = 1.04-2.63), whereas lower retention was seen amongst primarily female samples (OR = 0.83, 95% CI = 0.71-0.98) and those motivated to quit smoking (OR = 0.74, 95% CI = 0.55-0.99) when indirect recruitment methods were used (OR = 0.60, 95% CI = 0.38-0.97) and at longer follow-up assessments (OR = 0.83, 95% CI = 0.79-0.87). For differential retention, higher retention in the intervention group occurred when the intervention but not comparator group received financial incentives for smoking cessation (OR = 1.35, 95% CI = 1.02-1.77). CONCLUSIONS: In randomized controlled trials of behavioural smoking cessation interventions, recruitment and retention rates appear to be higher for smoking cessation interventions that include a person-to-person rather than at-a-distance contact; male participants, smokers with chronic conditions, smokers not initially motivated to quit and shorter follow-up assessments seems to be associated with improved retention; financial incentive interventions improve retention in groups receiving them relative to comparison groups.

Assessing Markers of Reproducibility and Transparency in Smoking Behaviour Change Intervention Evaluations.

INTRODUCTION: Activities promoting research reproducibility and transparency are crucial for generating trustworthy evidence. Evaluation of smoking interventions is one area where vested interests may motivate reduced reproducibility and transparency. AIMS: Assess markers of transparency and reproducibility in smoking behaviour change intervention evaluation reports. METHODS: One hundred evaluation reports of smoking behaviour change intervention randomised controlled trials published in 2018-2019 were identified. Reproducibility markers of pre-registration; protocol sharing; data, material, and analysis script sharing; replication of a previous study; and open access publication were coded in identified reports. Transparency markers of funding and conflict of interest declarations were also coded. Coding was performed by two researchers, with inter-rater reliability calculated using Krippendorff's alpha. RESULTS: Seventy-one percent of reports were open access, and 73% were pre-registered. However, there are only 13% provided accessible materials, 7% accessible data, and 1% accessible analysis scripts. No reports were replication studies. Ninety-four percent of reports provided a funding source statement, and eighty-eight percent of reports provided a conflict of interest statement. CONCLUSIONS: Open data, materials, analysis, and replications are rare in smoking behaviour change interventions, whereas funding source and conflict of interest declarations are common. Future smoking research should be more reproducible to enable knowledge accumulation. This study was pre-registered: https://osf.io/yqj5p.

Effectiveness of training stop-smoking advisers to deliver cessation support to the UK national proposed standard versus usual care in Malaysia: a two-arm cluster-randomized controlled trial.

AIMS: To assess the effectiveness of training stop smoking services providers in Malaysia to deliver support for smoking cessation based on the UK National Centre for Smoking Cessation and Training (NCSCT) standard treatment programme compared with usual care. DESIGN: Two-arm cluster-randomized controlled effectiveness trial across 19 sites with follow-up at 4-week, 3-month, and 6-month. SETTING: Stop smoking services operating in public hospitals in Malaysia. PARTICIPANTS: Five hundred and two smokers [mean ± standard deviation (SD), age 45.6 (13.4) years; 97.4% male] attending stop smoking services in hospital settings in Malaysia: 330 in 10 hospitals in the intervention condition and 172 in nine hospitals in the control condition. INTERVENTION AND COMPARATOR: The intervention consisted of training stop-smoking practitioners to deliver support and follow-up according to the NCSCT Standard Treatment Programme. The comparator was usual care (brief support and follow-up). MEASUREMENTS: The primary outcome was continuous tobacco smoking abstinence up to 6 months in smokers who received smoking cessation treatment, verified by expired-air carbon monoxide (CO) concentration. Secondary outcomes were continuous CO-verified tobacco smoking abstinence up to 4 weeks and 3 months. RESULTS: Follow-up rates at 4 weeks, 3 months and 6 months were 80.0, 70.6 and 53.3%, respectively, in the intervention group and 48.8, 30.8 and 23.3%, respectively, in the control group. At 6-month follow-up, 93 participants in the intervention group and 19 participants in the control group were abstinent from smoking, representing 28.2 versus 11.0% in an intention-to-treat (ITT) analysis assuming that participants with missing data had resumed smoking, and 52.8 versus 47.5% in a follow-up-only (FUO) analysis. Unadjusted odds ratios (accounting for clustering) were 5.04, (95% confidence interval (CI) = 1.22-20.77, P = 0.025) and 1.70, (95% CI = 0.25-11.53, P = 0.589) in the ITT and FUO analyses, respectively. Abstinence rates at 4 week and 3 month follow-ups were significantly higher in the intervention versus control group in the ITT but not the FUO analysis. CONCLUSIONS: On an intention-to-treat analysis with missing-equals-smoking imputation, training Malaysian stop smoking service providers in the UK National Centre for Smoking Cessation and Training standard treatment programme appeared to increase 6 month continuous abstinence rates in smokers seeking help with stopping compared with usual care. However, the effect may have been due to increasing follow-up rates.

Underreporting of the active content of behavioural interventions: a systematic review and meta-analysis of randomised trials of smoking cessation interventions.

Despite its importance, underreporting of the active content of experimental and comparator interventions in published literature has not been previously examined for behavioural trials. We assessed completeness and variability in reporting in 142 randomised controlled trials of behavioural interventions for smoking cessation published between 1/1996 and 11/2015. Two coders reliably identified the potential active components of experimental and comparator interventions (activities targeting behaviours key to smoking cessation and qualifying as behaviour change techniques, BCTs) in published, and in unpublished materials obtained from study authors directly. Unpublished materials were obtained for 129/204 (63%) experimental and 93/142 (65%) comparator groups. For those, only 35% (1200/3403) of experimental and 26% (491/1891) of comparator BCTs could be identified in published materials. Reporting quality (#published BCTs/#total BCTs) varied considerably between trials and between groups within trials. Experimental (vs. comparator) interventions were better reported (B(SE) = 0.34 (0.11), p < .001). Unpublished materials were more often obtained for recent studies (B(SE) = 0.093 (0.03), p = .003) published in behavioural (vs. medical) journals (B(SE) = 1.03 (0.41), p = .012). This high variability in underreporting of active content compromises reader's ability to interpret the effects of individual trials, compare and explain intervention effects in evidence syntheses, and estimate the additional benefit of an experimental intervention in other settings.

Obesity prevention in the early years: A mapping study of national policies in England from a behavioural science perspective.

BACKGROUND: Evidence indicates that early life is critical for determining future obesity risk. A sharper policy focus on pregnancy and early childhood could help improve obesity prevention efforts. This study aimed to systematically identify and categorise policy levers used in England with potential to influence early life course (pregnancy, 0-5 years) and identify how these interface with energy balance behaviours. The objective is to identify gaps and where further policy actions could most effectively focus. METHODS: A behavioural science approach was taken using the Capability-Opportunity-Motivation-Behaviour (COM-B) model and Behaviour Change Wheel (BCW) framework. The key determinants of energy balance in the early years were identified from the Foresight Systems Map. Policy actions were scoped systematically from available literature, including any health or non-health policies which could impact on energy balance behaviours. Foresight variables and policy actions were considered in terms of COM-B and the BCW to determine approaches likely to be effective for obesity prevention and treatment. Existing policies were overlaid across the map of key risk factors to identify gaps in obesity prevention and treatment provision. RESULTS: A wide range of policy actions were identified (n = 115) to address obesity-relevant risk factors. These were most commonly educational or guidelines relating to environmental restructuring (i.e. changing the physical or social context). Scope for strengthening policies relating to the food system (e.g. the market price of food) and psychological factors contributing to obesity were identified. Policies acted via all aspects of the COM-B model, but there was scope for improving policies to increase capability through skills acquisition and both reflective and automatic motivation. CONCLUSIONS: There is substantial policy activity to address early years obesity but much is focused on education. Scope exists to strengthen actions relating to upstream policies which act on food systems and those targeting psychological factors contributing to obesity risk.

Are population trends in high-risk alcohol consumption in smokers associated with trends in quit attempts and quit success? A time-series analysis.

OBJECTIVES: Monthly changes in the prevalence of high-risk drinking and smoking in England appear to be positively correlated. This study aimed to assess how far monthly changes in high-risk drinking were specifically associated with attempts to stop smoking and the success of quit attempts. DESIGN: Data were used from the Alcohol and Smoking Toolkit Studies between April 2014 and June 2018. These involve monthly household face-to-face surveys of representative samples of ~1800 adults. SETTING: England. PARTICIPANTS: Data were aggregated on 17 560 past-year smokers over the study period. PRIMARY AND SECONDARY OUTCOME MEASURES: Autoregressive integrated moving average with exogenous input (ARIMAX) modelling was used to assess the association over time between monthly prevalence of high-risk drinking among smokers and (a) prevalence of attempts to quit smoking and (b) prevalence of successful quit attempts in those attempting to quit. Bayes factors (BF) were calculated to compare the null hypothesis with the hypothesis of an effect sufficiently large (ß=0.6) to explain the established association between overall prevalence in smoking and high-risk drinking. RESULTS: No statistically significant associations were found between monthly changes in prevalence of high-risk drinking among smokers and attempts to quit smoking (ß=0.156, 95% CI -0.079 to 0.391, p=0.194) or quit success (ß=0.066, 95% CI -0.524 to 0.655, p=0.827). BF indicated that the data were insensitive but suggested there is weak evidence for the null hypothesis in the case of both quit attempts (BF=0.80) and quit success (BF=0.53). CONCLUSIONS: Monthly changes in prevalence of high-risk alcohol consumption in England are not clearly associated with changes in quit attempt or quit success rates.

Exploratory Analyses of the Popularity and Efficacy of Four Behavioral Methods of Gradual Smoking Cessation.

INTRODUCTION: Around half of smokers attempt to stop by cutting-down first. Evidence suggests that this results in similar quit rates to abrupt quitting. Evidence for the effectiveness and popularity of different gradual cessation methods is sparse. METHODS: Secondary, exploratory, analyses of a randomized trial of gradual versus abrupt smoking cessation. Gradual participants (N = 342) chose between four methods of cutting-down over 2 weeks: cutting-out the easiest cigarettes first (HR-E); cutting-out the most difficult cigarettes first (HR-D); smoking on an increasing time schedule (SR); and not smoking during particular periods (SFP). Nicotine replacement therapy and behavioral support were provided before and after quit day. We used logistic and linear regression modeling to test whether the method chosen was associated with smoking reduction, quit attempts, and abstinence, while adjusting for potential confounders. RESULTS: Participants were on average 49 years old, smoked 20 cigarettes per day, and had a Fagerstrom Test for Cigarette Dependence score of 6. 14.9% (51/342) chose HR-E, 2.1% (7/342) HR-D, 46.2% (158/342) SFP, and 36.8% (126/342) SR. We found no evidence of adjusted or unadjusted associations between method and successful 75% reduction in cigarette consumption, reduction in percentage cigarettes per day or exhaled carbon monoxide, quit attempts, or abstinence at 4-week or 6-month follow-up. CONCLUSIONS: Future research and practice could focus more heavily on the SR and SFP methods as these appeared notably more popular than HR. There was substantial imprecision in the efficacy data, which should be treated with caution; however, none of the gradual cessation methods showed clear evidence of being more efficacious than others. IMPLICATIONS: There is evidence that people who would like to quit smoking gradually should be supported to do so. However, as this is relatively new thinking and there is large potential for variation in methods, guidance on the best way to offer support is sparse. This article is an exploratory analysis of the popularity and efficacy of various methods in an attempt to move the topic forward and inform the implementation of gradual smoking cessation methods in practice. The identified popularity of some methods over others signposts directions for future research.

Behaviour change techniques associated with smoking cessation in intervention and comparator groups of randomized controlled trials: a systematic review and meta-regression.

AIMS: To estimate the strengths of associations between use of behaviour change techniques (BCTs) and clusters of BCTs in behavioural smoking cessation interventions and comparators with smoking cessation rates. METHOD: Systematic review and meta-regression of biochemically verified smoking cessation rates on BCTs in interventions and comparators in randomized controlled trials, adjusting for a priori-defined potential confounding variables, together with moderation analyses. Studies were drawn from the Cochrane Tobacco Addiction Group Specialised Register. Data were extracted from published and unpublished (i.e. obtained from study authors) study materials by two independent coders. Adequately described intervention (k = 143) and comparator (k = 92) groups were included in the analyses (n = 43 992 participants). Using bivariate mixed-effects meta-regressions, while controlling for key a priori confounders, we regressed smoking cessation on (a) three BCT groupings consistent with dual-process theory (i.e. associative, reflective motivational and self-regulatory), (b) 17 expert-derived BCT groupings (i.e. BCT taxonomy version 1 clusters) and (c) individual BCTs from the BCT taxonomy version 1. RESULTS: Among person-delivered interventions, higher smoking cessation rates were predicted by BCTs targeting associative and self-regulatory processes (B = 0.034, 0.041, P < 0.05), and by three individual BCTs (prompting commitment, social reward, identity associated with changed behaviour). Among written interventions, BCTs targeting taxonomy cluster 10a (rewards) predicted higher smoking cessation (B = 0.394, P < 0.05). Moderation effects were observed for nicotine dependence, mental health status and mode of delivery. CONCLUSIONS: Among person-delivered behavioural smoking cessation interventions, specific behaviour change techniques and clusters of techniques are associated with higher success rates.

Variability and effectiveness of comparator group interventions in smoking cessation trials: a systematic review and meta-analysis.

AIMS: To examine variability and effectiveness of interventions provided to comparator (control) groups in smoking cessation trials. METHODS: Systematic review with meta-analysis of randomized controlled trials (RCTs) of behavioral interventions for smoking cessation, with or without stop-smoking medication. We searched the Cochrane Tobacco Addiction Group Specialized Register for RCTs with objective outcomes measured at ≥ 6 months. Study authors were contacted to obtain comprehensive descriptions of their comparator interventions. Meta-regression analyses examined the relationships of smoking cessation rates with stop-smoking medication and behavior change techniques. RESULTS: One hundred and four of 142 eligible comparator groups (n = 23 706) had complete data and were included in analyses. There was considerable variability in the number of behavior change techniques delivered [mean = 15.97, standard deviation (SD) = 13.54, range = 0-45] and the provision of smoking cessation medication (43% of groups received medication) throughout and within categories of comparator groups (e.g. usual care, brief advice). Higher smoking cessation rates were predicted by provision of medication [B = 0.334, 95% confidence interval (CI) = 0.030-0.638, P = 0.031] and number of behavior change techniques included (B = 0.020, 95% CI = 0.008-0.032, P < 0.001). Modelled cessation rates in comparator groups that received the most intensive support were 15 percentage points higher than those that received the least (23 versus 8%). CONCLUSIONS: Interventions delivered to comparator groups in smoking cessation randomized controlled trials vary considerably in content, and cessation rates are strongly predicted by stop-smoking medication and number of behavior change techniques delivered.

Exploring Mechanisms of Action: Using a Testing Typology to Understand Intervention Performance in an HIV Self-Testing RCT in England and Wales.

SELPHI involves two interventions: (A) It provides one HIV self-testing (HIVST) kit; (B) It offers 3-monthly repeat HIVST kits if participants report ongoing risk. A logic model underpinned by the Behaviour Change Wheel informed the design of the intervention. SELPHI recruited 10,135 cis-men and trans people in England and Wales, all reporting anal sex with a man. This paper explores how the interventions were experienced and the pathways to impact for different groups of trial participants. In-depth interviews with 37 cis-men who have sex with men (MSM) were used to inductively categorise participants based on sexual and HIV testing histories. Themes relating to intervention experiences and impacts were mapped onto SELPHI-hypothesised intermediate outcomes to consider intervention impacts. Three groups were identified: 'Inexperienced testers' engaged with SELPHI to overcome motivational and social and physical opportunity testing barriers. For 'pro self-testers', testing frequency was constrained by psychological and social barriers and lack of opportunity. 'Opportunistic adopters' engaged in HIVST for novelty and convenience. Perceived impacts for inexperienced testers were most closely aligned with the logic model, but for opportunistic adopters there was little evidence of impact. Distinctive groups were discernible with divergent intervention experiences. Using COM-B as a model for understanding behaviour change in relation to HIVST, our results indicate how HIVST interventions could be adapted to respond to different needs based on the target population's demographic and behavioural features.

Knowledge Extraction and Prediction from Behavior Science Randomized Controlled Trials: A Case Study in Smoking Cessation.

Due to the fast pace at which randomized controlled trials are published in the health domain, researchers, consultants and policymakers would benefit from more automatic ways to process them by both extracting relevant information and automating the meta-analysis processes. In this paper, we present a novel methodology based on natural language processing and reasoning models to 1) extract relevant information from RCTs and 2) predict potential outcome values on novel scenarios, given the extracted knowledge, in the domain of behavior change for smoking cessation.

Association of prevalence of electronic cigarette use with smoking cessation and cigarette consumption in England: a time-series analysis between 2006 and 2017.

AIMS: To provide up-to-date estimates of how changes in the prevalence of electronic cigarette (e-cigarette) use in England have been associated with changes in smoking cessation activities and daily cigarette consumption among smokers in England. DESIGN: Time-series analysis of population trends. SETTING: England. PARTICIPANTS: Participants came from the Smoking Toolkit Study, which involves repeated, cross-sectional household surveys of individuals aged 16 years and older in England. Data were aggregated on approximately 1200 past-year smokers each quarter (total n = 50 498) between 2007 and 2017. MEASUREMENTS: Prevalence of e-cigarette use in current smokers was used to predict (a) prevalence of quit attempts among last-year smokers, (b) overall quit rates among last-year smokers and (c) mean cigarette consumption per day among current smokers. Prevalence of e-cigarette use during a quit attempt among last-year smokers was used to predict (a) quit success rate among last-year smokers and (b) overall quit rates among last-year smokers. FINDINGS: Overall quit rates increased by 0.054% [95% confidence interval (CI) = 0.032-0.076, P < 0.001] and 0.050% (95% CI = 0.031-0.069, P < 0.001) respectively for every 1% increase in the prevalence of e-cigarette use by smokers and e-cigarette use during a quit attempt. Quit success rates increased by 0.060% (95% CI = 0.043-0.078, P < 0.001) for every 1% increase in the prevalence of e-cigarette use during a quit attempt. No clear evidence was found for an association between e-cigarette use and either prevalence of quit attempt (BAdj = 0.011, 95% CI = -0.046 to 0.069, P = 0.698) or cigarette consumption (BAdj = 0.019, 95% CI = -0.043 to 0.082, P = 0.542). CONCLUSION: Changes in prevalence of e-cigarette use in England have been positively associated with the overall quit rates and quit success rates but not clearly associated with the prevalence of quit attempts and mean cigarette consumption.

Is the use of e-cigarettes for smoking cessation associated with alcohol consumption? A population-level survey of successful quitters in England.

OBJECTIVE: To examine associations between the use of e-cigarettes for smoking cessation and levels of alcohol consumption, high-risk drinking, and attempts to cut down alcohol consumption compared with use of nicotine replacement therapy (NRT) or no aid. METHODS: Cross-sectional survey of adults (≥16 years) in England. The sample included a total of 961 people who had quit smoking with the use of either e-cigarettes (n = 425), NRT (n = 116), or no aid (n = 421) within the past year and were still abstinent at the survey. Drinking behaviour was assessed with the AUDIT. RESULTS: Mean (SD) alcohol consumption among those who quit smoking with e-cigarettes, NRT, and no aid was 7.78 (13.41), 7.12 (13.85), and 5.55 (8.70) units/week, respectively. The prevalence of high-risk drinking was 43.3% (n = 184), 32.2% (n = 37), and 36.8% (n = 155), respectively. Among high-risk drinkers, the prevalence of attempts to cut down alcohol consumption was 22.3% (n = 41), 18.9% (n = 7), and 27.7% (n = 43), respectively. After adjustment for covariates, those who quit with e-cigarettes had significantly higher alcohol consumption than those who quit unaided (B = 1.69, 95%CI 0.21-3.17), but there was no significant difference relative to those who quit with NRT. Differences in high-risk drinking and attempts to cut down were not significant, but Bayes factors indicated the data were insensitive (range: 0.47-0.95). CONCLUSIONS: Recent ex-smokers who used e-cigarettes to help them quit consumed around two more units of alcohol each week than those who quit unaided, but their alcohol consumption was similar to those who quit with NRT. Data on differences in high-risk drinking and attempts to cut down alcohol consumption among high-risk drinkers were inconclusive.

Prevalence and correlates of long-term e-cigarette and nicotine replacement therapy use: a prospective study in England.

OBJECTIVES: To examine the prevalence of, and sociodemographic and smoking-related characteristics associated with, long-term e-cigarette use compared with long-term nicotine replacement therapy (NRT) use. DESIGN: Cross-sectional and prospective survey, the Smoking Toolkit Study, with baseline data collected between September 2014 and September 2016 and follow-ups at 6 and 12 months. SETTING: England. PARTICIPANTS: Population representative sample of 40 933 adults aged 16+ years. MAIN OUTCOME MEASURES: Prevalence of long-term (≥12 months) use of e-cigarettes and NRT by retrospective self-report among baseline respondents (all adults, n=40 933; smokers, n=8406) and current use at baseline, 6 months and 12 months in a subsample of smokers who responded to follow-up (n=733). RESULTS: Of baseline respondents, 1.5% (95% CI 1.4% to 1.6%, n=604) of adults and 3.9% (95% CI 3.5% to 4.3%, n=327) of smokers were long-term e-cigarette users and 0.5% (95% CI 0.4% to 0.6%, n=205) of adults and 1.3% (95% CI 1.1% to 1.5%, n=112) of smokers were long-term NRT users. Assessed prospectively, 13.4% (95% CI 10.9% to 15.9%, n=100) of smokers were long-term e-cigarette users and 1.9% (95% CI 0.9% to 2.9%, n=14) were long-term NRT users. Among all adults, long-term use by never smokers of either e-cigarettes (0.1%, n=27) or NRT (0.0%, n=7) was rare. Among past-year smokers, long-term e-cigarette and NRT use was higher among older smokers compared with those who were 16-34 years old (OR range=1.55-5.21). Long-term e-cigarette use only was lower in smokers who were less educated (OR=0.63, 95% CI 0.49 to 0.81), from social grades C2DE (OR=0.66, 95% CI 0.52 to 0.84) and with children in the household (OR=0.66, 95% CI 0.51 to 0.85). Long-term e-cigarette use and long-term NRT use were higher among smokers more motivated to quit (OR=2.05, 95% CI 1.63 to 2.60 and OR=2.33, 95% CI 1.57 to 3.46). CONCLUSIONS: In the adult population in England, long-term use of e-cigarettes and long-term use of NRT are almost exclusively by current or ex-smokers. Only a minority of past-year smokers retrospectively report long-term e-cigarette or NRT use, but this figure may be an underestimate, especially for e-cigarette use, which is more than threefold higher when assessed prospectively.

Comparison of Trends in Self-reported Cigarette Consumption and Sales in England, 2011 to 2018.

Importance: Population cigarette consumption is declining in many countries. Accurate estimates of long- and short-term changes are vital for policy evaluation and planning. Survey data and sales data that are used to make these estimates each have important potential biases, so triangulation using different methods is required for robust estimation. Objectives: To compare monthly estimates of cigarette consumption in England from a nationally representative survey and recorded cigarette sales and to triangulate an accurate estimate of changes in cigarette consumption since 2011. Design, Setting, and Participants: This study used time series analyses based on survey data and recorded cigarette sales to estimate and compare trends in population cigarette consumption in England from 2011 to 2018. Survey participants were representative samples of 1700 people aged 16 years or older each month in England. Main Outcomes and Measures: Monthly cigarette retail sales data from August 2011 through February 2018 were obtained from a data agency. Monthly self-reports of cigarette consumption were collected over the same period using the Smoking Toolkit Study. Results: A total of 136 677 individuals (51.1% female; mean [SD] age, 46.7 [18.8] years) were surveyed. Over the study period, mean monthly cigarette consumption in England was 2.85 billion (95% CI, 2.78 billion to 2.93 billion) cigarettes based on survey data compared with 3.08 billion (95% CI, 3.03 billion to 3.13 billion) estimated from sales data. Over the whole period, cigarette consumption declined by 24.4% based on survey data and 24.1% based on sales data. This equated to 118.4 million and 117.4 million fewer cigarettes consumed per month (or approximately 1.4 billion per year) based on survey data and sales data, respectively. After adjusting for underlying trends, month-by-month changes in cigarette consumption were closely aligned: a 1% change in survey-estimated cigarette consumption was associated with a 0.98% (95% CI, 0.53%-1.44%) change in sales estimates. Conclusions and Relevance: Survey data and sales data were closely aligned in showing that overall cigarette sales in England have declined by almost a quarter since 2011, amounting to more than 1 billion fewer cigarettes smoked each year. The alignment between the 2 methods provides increased confidence in the accuracy of parameters provided by the Smoking Toolkit Study and sales data. It indicates that estimated changes in cigarette consumption are robust and provide a meaningful basis for policy evaluation and planning.