HARMONY (HARM reduction for Opiates, Nicotine and You): Statistical Analysis Plan for a Randomised Controlled Trial of the Effectiveness of Vaporised Nicotine Products for Tobacco Smoking Cessation amongst NSW Opiate Agonist Treatment Clients

Background and rationale

• Smoking continues to be the main cause of preventable disease in Australia linked to cancer, respiratory disease and cardiovascular disease. Although smoking rates have decreased in the general population in Australia (13%), they continue to be disproportionately high amongst lower socioeconomic status groups.

• Tobacco smoking is highly prevalent amongst opiate agonist treatment clients. The highest rates of smoking amongst people with substance use disorders are seen in those on opiate agonist treatment (OAT): as high as 73% to 94% smoking prevalence.

• Clients in OAT are greatly interested in quitting smoking, especially if provided in their treatment clinic.

• While smoking cessation appears possible in OAT smokers, few interventions have demonstrated efficacy in this population of tobacco users.

• Most NRT interventions decreased cigarette use, but sustained abstinence rates remained low. High relapse rates in this population may be due to factors related to addiction (e.g. nicotine enhancing the opioid withdrawal reductions due to OAT), lack of cessation support, and the high levels of smoking in their social network.

• Vaporised nicotine products (VNPs) are a broad range of battery-powered devices that deliver an aerosol of propylene glycol and/or glycerine, nicotine and flavours.

• Large scale population surveys in the US (n=161,054) and UK (n= 170,490) suggest that VNP use is significantly associated with smoking cessation.

• A small (n=12), single group trial in the USA among people on OAT found reductions in cigarettes smoked per day over a 6 week, VNP-use study period. There are no other trials of VNPs for smoking cessation amongst OAT clients published or registered.

Objectives

Hypotheses

The primary hypothesis is that treatment with VNP has a greater 7-day point prevalence of abstinence to tobacco smoking than best practice NRT. Where appropriate, the study will compare outcomes under a superiority hypothesis, however, it may be appropriate to consider non-inferiority hypothesis when considering the outcomes related to safety, cravings and adherence.

Trial design

A two-arm randomised single blind trial comparing a 12-week course of liquid nicotine delivery via VNP’s compared to best practice NRT. The two treatments are randomised in a 1:1 ratio of treatment arm allocation.

Randomisation

The randomisation schedule was developed by an independent statistician and conducted in REDCap. The randomisation was stratified, through blocking, by treatment site and cannabis use.

Sample size

A sample of 200 smokers in each treatment group were needed to detect a difference of 6% between groups (i.e., 1% in NRT group and 7% in VNP group continuous abstinence at 6-month follow-up) with 80% power and a 5% significance level. Based on our previous research in AOD setting with clients from various programs including OAT and assuming 30% attrition rate at 6 months follow-up, we required a sample size of 572 eligible smokers across 6 OAT sites

Framework

This is a superiority trial.

Statistical interim analyses and stopping guidance

No interim analyses are planned.

Timing of final analysis

The primary analysis will occur at the completion of the trial, once all participants have completed the three-month (24 – 28 weeks) follow-up or are determined to be lost to follow-up.

Timing of outcome assessments

Outcome data will be collected from participants at two time-points: (i) at baseline following consent; (ii) at the end of the 12-week treatment and, (iii) 3-months post treatment over the phone by a member of an independent CATI team. See Figure 1 for participant flow through the study.

• Download figure
• Open in new tab

Figure 1.

CONSORT diagram of participant flow through the study.

Screening data

A RO/RN will complete a pre-screening questionnaire with the client either in person or by telephone. Potential eligible persons will be invited to attend the clinic in person or via Telehealth for an informed consent discussion and a formal screening assessment.

Eligibility

Patients will be eligible to participate if the following apply:

• Provide written, informed consent to participate in the study.

• Aged 18 years or older

• Be accessing opioid agonist treatment from a participating service

• Current daily tobacco smokers on self-report

• Want to quit or cut down their smoking

• Be willing and able to comply with requirements of study (including having access to a phone)

1.1. Exclusion Criteria

Patients will be ineligible to take part in the study if the following apply:

• Currently breastfeeding or pregnant, or of childbearing potential and planning/trying to fall pregnant during the study period.

• Current, severe medical disorder assessed by study medical officer (such as but not limited to, unstable cardiovascular/peripheral vascular disease, poorly controlled hypertension).

• Current, severe or unstable psychiatric disorder assessed by study medical officer (such as but not limited to, acute psychosis, severe anxiety and/or mood disorder, intent to harm self or others).

• Current enrolment in a clinical trial involving any investigational drug.

• Regular use (more than one day per week) of Vaporised Nicotine Product (or e-Cigarette) containing nicotine in the last 30 days.

• Not available for follow-up (e.g. likely imprisonment or transfer out of service to another service that is not a trial recruitment site).

Recruitment

The recruitment strategy for this study will incorporate approved advertisements being displayed in participating services OAT waiting rooms and other suitable locations. Due to recent changes in OAT service delivery, advertisements may also be sent via text or email to clinic clients that are no longer visiting clinics daily. Interested clients may contact the researchers directly using contact details contained in the advertisements.

Introductions to the trial shall also be made by the clinicians at each site during appointments or conversations. Clinicians will be asked to identify clients who smoke and enquire about their desire to stop and their interest in participating in a smoking cessation treatment using the script provided. The contact details of any interested clients will be provided to the research team for pre-screening.

Withdrawal/follow-up

Treatment discontinuation does not equate withdrawal from the study and these participants will be invited to continue completing the research component of the study.

Involuntary discontinuation

Participants may be withdrawn involuntarily by the investigator (or delegate) if they meet the following criteria:

• Participant experiences a severe or serious adverse event, thought to be related to the study drug/device, which is not resolving.

• Absence from the protocol monthly visits (Week 4, 8 or 12) for more than four weeks.

Even if the participant has capacity to consent at baseline, the intervention staff will assess capacity to consent at each contact and will cease the intervention if the participant is deemed to lack capacity. In this case the reasoning will be explained and recorded on the CRF.

Voluntary discontinuation

Participants have the option to stop treatment or revoke their consent at any time without giving a reason. The distinction between stopping treatment and revoking consent is shown by the following definitions:

• Discontinuation of treatment: A participant would be considered to have discontinued protocol treatment if they stop using the VNP device/NRT medication study. In this case the participant will continue in all remaining research interviews and assessments.

• Revocation of consent: Total withdrawal from the trial would occur in the circumstance that the participant decides to revoke their consent. Under these circumstances no further information would be collected from the participant for the purpose of the trial.

Participants may at any time elect to revoke consent for study participation without jeopardising their relationship with either their doctor or treatment service.

General Statistical methodology

Statistical inference for assessing differences between groups for all outcome measures will be within a Bayesian framework. This involves for each outcome specifying a likelihood function, and a prior distribution. Together these form the posterior distribution for the parameters of interest.

Analysis populations

The primary analysis modelling will be conducted as intention-to-treat (ITT). The ITT sample will include all participants who are randomised (excluding those who choose to withdraw data from analysis), with their data analysed according to the experimental group to which they were randomised to. Descriptive analysis of study data will be performed for all participants who complete at least one time point.

Baseline patient characteristics

Baseline characteristics will be summarized descriptively using means, standard deviations, medians and interquartile ranges for continuous data, and frequencies and percentages for categorical data. The following baseline characteristics will be presented by treatment arm:

Outcome definitions

All primary, secondary outcomes will be measured at baseline, at the end of the 12-week treatment and 3-months after treatment. An overview of the outcomes to be assessed are shown in Table 2.

Analysis methods

Secondary outcomes

Hierarchical models with appropriate distributional assumptions will be used to analyse secondary outcomes.

Secondary population

The secondary population in the trial is the compliers. The complier average causal effect (CACE) for VPN’s will be used to explore the impact of compliance on the primary outcome. The level of compliance will be defined from retention (secondary outcome 10). The same model to assess the primary outcome will be used, with one difference: X_ijk = 0 (NRT group) or R, where R is defined as the proportion of participants in the VNP group that adhered to treatment. The value of y is the level of compliance observed, and y is then estimated as if all participants were adherent by scaling y by .

Sensitivity analyses

Sensitivity analyses will be used to examine the robustness of results.

1. The impact of the choice of prior on the primary outcome will be examined by running the primary analysis with informative (power) priors with three different weights (0.25, 0.5, 1). The power prior will be based on previous literature [4, 5] that compared VNP and NRT.

2. The primary model will be run again with only complete cases to determine the robustness of the multiple imputation.

3. The primary model will be run with varying MNAR assumptions deemed to be clinically probable.

The sensitivity analysis results will be presented alongside the primary analysis in the format of Table A1 (See Appendix).

Adverse events

All adverse events will be recorded and reported via descriptive statistics.

Statistical software

Analyses will be performed using SAS v9.4 (SAS Institute, Cary, North Carolina, USA) or R (R Core Team, 2022) or Stan (Stan Development Team, 2024).