The number needed to treat (NNT) and cost per responder for behavioural interventions for smoking cessation

This additional information describes in detail the effectiveness evidence and cost information on behavioural interventions for smoking cessation. The selection of the effectiveness evidence had following steps: the literature search, selection of effectiveness evidence, and quality appraisal. The effectiveness evidence is gathered from Stead et al. (2013), Stead et al. (2017), Rice et al. (2017), Matkin et al. (2019) and Tzelepis et al. (2019). Effectiveness information for individual intervention will be updated when meta-analysis is completed. Table 3 describes studies included in those meta-analyses. «Stead LF, Buitrago D, Preciado N, ym. Physician ad...»1, «Stead LF, Carroll AJ, Lancaster T. Group behaviour...»2, «Rice VH, Heath L, Livingstone-Banks J, ym. Nursing...»3, «Matkin W, Ordóñez-Mena JM, Hartmann-Boyce J. Telep...»4, «Tzelepis F, Paul CL, Williams CM, ym. Real-time vi...»5

Source of the effectiveness evidence:

• Stead LF, Buitrago D, Preciado N, Sanchez G, Hartmann-Boyce J, Lancaster T. Physician advice for smoking cessation. Cochrane Database Syst Rev. 2013 May 31;2013(5):CD000165. doi: 10.1002/14651858.CD000165.pub4.
• Stead LF, Carroll AJ, Lancaster T. Group behaviour therapy programmes for smoking cessation. Cochrane Database Syst Rev. 2017 Mar 31;3(3):CD001007. doi: 10.1002/14651858.CD001007.pub3.
• Rice VH, Heath L, Livingstone-Banks J, Hartmann-Boyce J. Nursing interventions for smoking cessation. Cochrane Database Syst Rev. 2017 Dec 15;12(12):CD001188. doi: 10.1002/14651858.CD001188.pub5. PMID: 29243221; PMCID: PMC6486227.
• Matkin W, Ordóñez-Mena JM, Hartmann-Boyce J. Telephone counselling for smoking cessation. Cochrane Database Syst Rev. 2019 May 2;5(5):CD002850. doi: 10.1002/14651858.CD002850.pub4. PMID: 31045250; PMCID: PMC6496404.
• Tzelepis F, Paul CL, Williams CM, Gilligan C, Regan T, Daly J, Hodder RK, Byrnes E, Byaruhanga J, McFadyen T, Wiggers J. Real-time video counselling for smoking cessation. Cochrane Database Syst Rev. 2019 Oct 29;2019(10):CD012659. doi: 10.1002/14651858.CD012659.pub2. PMID: 31684699; PMCID: PMC6818086.

Literature search and selection

The literature search included network meta-analyses, meta-analyses and systematic reviews of behavioural interventions for smoking cessation. Information specialist conducted the literature search in October 2022. The timeframe for the search was years 2018–2022. The search found 47 references. We searched for the most recent evidence by viewing only network meta-analyses published in 2022 and 2021, if satisfactory number of comprehensive studies were found. Only one network meta-analysis (Hartmann-Boyce et al. 2021) «Hartmann-Boyce J, Livingstone-Banks J, Ordóñez-Men...»6 was published in 2022 and 2021 «Description of the network meta-analysis by Hartmann-Boyce et al. (2021) of behavioural interventions for smoking cessation and the suitable meta-analyses included the network meta-analysis. ...»1. The Hartmann-Boyce et al. (2021) «Hartmann-Boyce J, Livingstone-Banks J, Ordóñez-Men...»6 searched for Cochrane Reviews of behavioural interventions for smoking cessations but compared only the effectiveness of different components of behavioural interventions. Thus, the results did not represent effectiveness of the interventions. We used Hartmann-Boyce et al. (2021) «Hartmann-Boyce J, Livingstone-Banks J, Ordóñez-Men...»6 to find suitable meta-analyses of the behavioural interventions. The publication included 33 reviews and the primary outcome measure was smoking cessation at six months or longer from baseline.

We selected five meta-analyses (Matkin et al. 2019; Rice et al. 2017; Stead et al. 2013; 2017; Tzelepis et al.) for further examination of the effectiveness of behavioural interventions as they evaluated the Current Care working group's selected interventions. Stead et al. (2013) «Stead LF, Buitrago D, Preciado N, ym. Physician ad...»1 evaluated physician brief advice compared to no intervention or usual care. Stead et al. (2017) «Stead LF, Buitrago D, Preciado N, ym. Physician ad...»1 evaluated group counselling programme compared to no intervention and group counselling programme compared to face-to-face individual intervention. Rice et al. (2017) «Rice VH, Heath L, Livingstone-Banks J, ym. Nursing...»3 evaluated nurse-delivered cessation interventions compared to no interventions. Matkin et al. (2019) «Matkin W, Ordóñez-Mena JM, Hartmann-Boyce J. Telep...»4 evaluated telephone counselling as an adjunct to self-help materials or to a minimal intervention compared to self-help materials or minimal intervention. Tzelepis et al. (2019) «Tzelepis F, Paul CL, Williams CM, ym. Real-time vi...»5 evaluated real time video counselling compared to telephone counselling.

Quality of the selected meta-analyses

We evaluated the quality of the Stead et al. (2013) «Stead LF, Buitrago D, Preciado N, ym. Physician ad...»1 and (2017) «Stead LF, Carroll AJ, Lancaster T. Group behaviour...»2, Rice et al. (2017) «Rice VH, Heath L, Livingstone-Banks J, ym. Nursing...»3, Matkin et al. 2019 «Matkin W, Ordóñez-Mena JM, Hartmann-Boyce J. Telep...»4, Tzelepis et al. (2019) «Tzelepis F, Paul CL, Williams CM, ym. Real-time vi...»5 meta-analyses by using AMSTAR 2 «Shea BJ, Reeves BC, Wells G, ym. AMSTAR 2: a criti...»7. The results of the quality appraisal are shown in table «Critical appraisal of selected meta-analyses by using AMSTAR 2 ...»2.

The chosen meta-analyses were suitable base of our effectiveness evidence as they were the most comprehensive and included interventions that was selected by the Current Care working group. All but Tzelepis et al. (2019) «Tzelepis F, Paul CL, Williams CM, ym. Real-time vi...»5 had some weaknesses. Most of the trials in Stead et al. (2017) «Stead LF, Carroll AJ, Lancaster T. Group behaviour...»2 had unclear risk of bias and the meta-analysis did not evaluate how it impacted the summary estimates of effect. Rice et al. (2017) «Rice VH, Heath L, Livingstone-Banks J, ym. Nursing...»3 included all RCTs regardless of the RoB and did not discuss the likely impact of the RoB on the results. In Stead et al. (2013) «Stead LF, Buitrago D, Preciado N, ym. Physician ad...»1 and (2017) «Stead LF, Carroll AJ, Lancaster T. Group behaviour...»2 study selection was not performed in duplicate for all review versions. In both of the most recent update, the selection was done in duplicate. In Matkin et al. (2019) «Matkin W, Ordóñez-Mena JM, Hartmann-Boyce J. Telep...»4 there was no mention that study selection was performed in duplicate. Rice et al. 2017 «Rice VH, Heath L, Livingstone-Banks J, ym. Nursing...»3 and Matkin et al. (2019) «Matkin W, Ordóñez-Mena JM, Hartmann-Boyce J. Telep...»4 did not describe timeframe for follow up.

RCT= randomized controlled trial, RoB = risk of bias

The effectiveness evidence

Stead et al. (2013) «Stead LF, Buitrago D, Preciado N, ym. Physician ad...»1 included 17 RCTs comparing minimal intervention delivered by physician to no intervention. Minimal intervention was defined as advice which was provided during a single consultation lasting less than 20 minutes plus up to one follow-up visit. Minimal interventions did not include additional pharmacotherapies. Stead et al. (2017) «Stead LF, Carroll AJ, Lancaster T. Group behaviour...»2 included 9 RCTs comparing Group intervention vs. no intervention. The interventions included scheduled meetings and some form of behavioural intervention, such as information, advice and encouragement or cognitive behavioural therapy (CBT) delivered over at least two sessions. Rice et al. (2017) «Rice VH, Heath L, Livingstone-Banks J, ym. Nursing...»3 included 58 studies in their review. A total of 44 studies compared a nursing intervention to a control or usual care. Nursing intervention was defined as offering advice, counseling and/or strategies to aid people quit smoking. This analysis will be updated in fall 2023 and the results will be used in the table.

Matkin et al. (2019) «Matkin W, Ordóñez-Mena JM, Hartmann-Boyce J. Telep...»4 included 104 RCT studies in their review. Of these, 35 RCTs were used to examine the effectiveness of telephone counseling as an in addition to self-help or a minimum intervention. Telephone counselling was defined as a behavioral intervention to help people stop smoking. It can be used either alone or in combination with other therapies. Tzelepis et al. (2019) «Tzelepis F, Paul CL, Williams CM, ym. Real-time vi...»5 included 2 RCTs comparing real-time video counselling compared with telephone counselling. Real-time video counselling consists of a video camera connected to a computer or mobile device, to securely transmit live video and audio of the counsellor and client to one another over the Internet. Interventions were delivered by therapist or counsellor. The effectiveness of real-time video counseling for quitting smoking is not well supported by the available research. According to the available studies, there is no difference between telephone and video counseling. However, the results should be interpreted cautiously.

Three meta-analyses [Rice et al. 2017; Stead et al. 2013; 2017] included tobacco smokers but excluded trials of pregnant women. Additionally, Stead et al. (2017) «Stead LF, Carroll AJ, Lancaster T. Group behaviour...»2 excluded studies recruiting only adolescent smokers. Matkin et al. (2019) «Matkin W, Ordóñez-Mena JM, Hartmann-Boyce J. Telep...»4 included studies in which participants were mostly adult smokers from the general population, but some studies included also teenagers, pregnant women, and people with long term or mental health conditions. Tzelepsis et al. (2019) «Tzelepis F, Paul CL, Williams CM, ym. Real-time vi...»5 had no age, gender, nicotine dependency, or comorbidities limitations. All meta-analyses required the follow-up for smoking cessation outcome at least six months after the start of treatment.

The selected reviews reported RRs and 95%CIs for the Current Care working group's selected interventions. We converted the RRs to NNTs «Schünemann HJ, Vist GE, Higgins JPT, Santesso N, D...»8. Assumed comparator risk (5%) was the pooled result of the 58 RCT's minimal intervention or no intervention groups' smoking cessation in 6-months «Rice VH, Heath L, Livingstone-Banks J, ym. Nursing...»3.

The resource use and costs of interventions

We searched for information about the resource use and cost of intervention for five different interventions: minimal intervention (short intervention), group intervention, individual intervention, telephone counselling and real time video counselling. Information about the use of resources is based on meta-analyses [Matkin et al. (2019); Rice et al. (2017); Stead et al. 2013; 2017; Tzelepis et al. (2019)]. If the information was lacking, we searched for it from original studies. The resource use of the individual intervention will be gathered from updated meta-analyses (Rice et al. 2017). We looked for information on the number of appointments, the time use, the provider of the intervention and the possible size of the group if the intervention was delivered in group format [table 3]. Table 3 describes the average number of sessions, time use per session, total time use of the sessions and the most usual provider of the interventions in the original studies. Also, the table describes corresponding numbers used in cost calculation assumed by the Current Care working group to represent Finnish practice. We used the Finnish unit costs for physician and public health nurse (Mäklin & Kokko, 2021) «Mäklin, S., & Kokko, P. (2021). Terveyden- ja sosi...»9 converted to 2021 value using the Statistics Finland's public expenditure price index (2023) «Official Statistics of Finland (OSF): Price index ...»10. Unit prices for professionals were converted to represent cost per minute by assuming 30-minute appointments. Cost of interventions represent the intervention costs per patient.

*Individual intervention: meta-analysis is in process.