Article Text

Original research
Does offering small financial incentives to smokers at the time of being wait-listed for surgery increase smoking cessation by the day of surgery? A randomised feasibility trial
  1. Ashley Webb1,2,
  2. Lisa Coward2,
  3. Michael Yousef2,
  4. Alexandra Karamesinis2,
  5. Samuel Leong1
  1. 1Department of Anaesthesia, Monash University Central and Eastern Clinical School, Melbourne, Victoria, Australia
  2. 2Department of Anaesthesia, Peninsula Health, Frankston, Victoria, Australia
  1. Correspondence to Dr Ashley Webb; AWebb{at}phcn.vic.gov.au

Abstract

Objective This study aims to assess whether offering small financial incentives to smokers on elective surgery wait-lists is feasible and increases quitting before surgery.

Design Randomised controlled trial, prospective, double-blinded.

Setting Single-centre, Australian metropolitan public hospital.

Participants 620 adult smokers (≥10 cigarettes per day) were randomised on being wait-listed for elective surgery and 404 underwent operations (28 January 2021–31 July 2022) at the hospital (65.2%) by trial’s end.

Intervention Intervention participants were offered at wait-listing an $A70 supermarket voucher for verified abstinence on the day of surgery, provided they registered an intention to quit before surgery. Registrants intending to quit were also referred to Quitline. Neither intervention was offered to control participants (usual care). Smokers wait-listed from 17 May 2021 were offered an increased incentive of $A140.

Main outcome measures Primary outcome, quitting at least 24 hours before surgery, verified by exhaled carbon monoxide testing. Feasibility outcomes were the proportion taking up offers, ease of patient contact and disputes about quit status.

Results Of 620 randomised participants (control 312, intervention 308), 404 had surgery at the hospital during the trial (control 214, intervention 190), which was lower than expected (for COVID-19 reasons). Offering $A70 resulted in 21.9% registering to quit, increasing to 32.6% with $A140. Telephone calls were the most effective means to gain registrations. The proportion of intervention group patients verified quit at least 24 hours before surgery was similar to controls (9.5% vs 8.9%, OR 1.1, 95% CI 0.5 to 2.2). Quitline contact was higher in the intervention group (13.2% vs 2.3%, OR 6.3, 95% CI 2.3 to 21.6). Disputes over test results did not occur, but 17.4% of intervention participants claiming quit failed verification.

Conclusion A single offer of financial rewards for perioperative cessation was feasible, without achieving clinically important quit differences.

Trial registration number ACTRN12620000130965.

  • adult anaesthesia
  • behavior
  • clinical trial
  • feasibility studies
  • hospitals, public
  • public health

Data availability statement

Data, including individual deidentified participant data, study protocols, ethical approvals and consent forms are available to researchers providing a reasonable request and HREC approved proposal. This will be available from the date of publication with no end-date determined.

http://creativecommons.org/licenses/by-nc/4.0/

This is an open access article distributed in accordance with the Creative Commons Attribution Non Commercial (CC BY-NC 4.0) license, which permits others to distribute, remix, adapt, build upon this work non-commercially, and license their derivative works on different terms, provided the original work is properly cited, appropriate credit is given, any changes made indicated, and the use is non-commercial. See: http://creativecommons.org/licenses/by-nc/4.0/.

Statistics from Altmetric.com

Request Permissions

If you wish to reuse any or all of this article please use the link below which will take you to the Copyright Clearance Center’s RightsLink service. You will be able to get a quick price and instant permission to reuse the content in many different ways.

STRENGTHS AND LIMITATIONS OF THIS STUDY

  • A strength is that this is a large smoking financial incentives study and the largest ever in a perioperative population.

  • A further strength is inclusion of all eligible smokers, regardless of interest in quitting before surgery, so results are likely to represent the ‘real-world’ effect of a single offer of a small financial incentive before surgery.

  • A limitation is that many (34.8%) randomised participants did not have an operation at the hospital during the trial, due to longer than usual elective-surgery wait-times, which was mainly due to COVID-19 elective surgical disruption.

  • Another limitation was changing the protocol mid-trial to double the reward to $A140, which was done due to lower-than-expected registrations to quit for $A70 (although this allowed researchers to study whether there was a ‘dose-effect’ response for quit registrations and quit outcomes with higher rewards).

  • However, another limitation was the small numbers of participants offered $A140 who had surgery at the hospital while the study was operational.

Introduction

Smokers have increased perioperative respiratory, cardiac and wound-complication risks that are significantly reduced by preoperative cessation; particularly for 4 weeks or more.1 2 Preoperative interventions can increase cessation, reduce postoperative complications and increase long-term abstinence.3 4 However, such studies show great heterogeneity and further studies are needed to determine cessation strategies that are acceptable, effective and easily incorporated into elective surgical pathways.4 Because longer cessation confers greater benefits, we previously trialled whether offering free mailed nicotine replacement therapy (NRT) plus telephone Quitline referral at time of wait-listing for elective surgery increased perioperative abstinence.5 6 When offered this help, between 32%6 and 39%5 of smokers requested it (costing approximately $A70 (£37, €42, US$45) per acceptance), resulting in approximately doubling verified abstinence compared with standard care.6

In 2019, a report on financial incentives to quit smoking showed that in mixed non-surgical populations, cessation (6 months) increased 49% compared with control groups, with smaller incentives working as effectively as larger ones.7 Subsequently, a small trial in smokers having cancer surgery were offered payments of up to US$645 ($A996, £522, €603) for preoperative cessation, showing tripling of confirmed smoking abstinence (RR 3.2, 95% CI 1.1 to 9.3).8

Given this evidence, it may have been simpler to redirect the $A70 (per accepting patient) we previously spent on mailed NRT; instead offering this amount as a cash inducement for cessation. Given the perioperative period may be a ‘teachable moment’ for smokers, where positive health behaviour changes are more likely,9 this small amount may be enough to motivate patients to seek out existing community-based cessation programmes and quit using the cessation supports they choose.

In this randomised controlled trial, we hypothesised it was feasible to offer smokers (≥10 cigarettes/day) being wait-listed for elective surgery a $A70 supermarket voucher for proven abstinence on the day of surgery (DOS), and referral to Quitline for those smokers who stated an intention to quit. We further hypothesised that compared with standard care, the group receiving offers would have increased quitting before surgery.

Methods

Study design

This randomised, controlled, blinded trial occurred at Frankston Hospital (Melbourne, Australia), a public tertiary hospital offering most surgical specialties (not cardiac/neurosurgery) between 28 January 2021 and 31 October 2022. In 2017, 18.4% of adults in the Frankston local government area were daily smokers; higher than the Victorian state average (14.7%).10 The trial was prospectively registered with the Australian New Zealand Clinical Trials Registry (ACTRN12620000130965) (11 February 2020).

Research assistants (RAs) identified smokers by screening responses to tobacco use questions in the standard hospital preoperative health questionnaire (HQ) that patients must complete to be wait-listed. All adult patients having elective operations were randomised if they self-reported smoking ≥10 cigarettes daily. Exclusion criteria were minor diagnostic procedures (eg, endoscopy, cystoscopy, interventional radiology) or dates for surgery within 2 weeks of screening the HQ, (usually category 1 (urgent) surgery), due to insufficient time for patient contact/Quitline intervention. Most patients did not have known surgical dates at the time of wait-listing/screening/randomisation and would become aware of their date several weeks/months/years later. Researchers monitored daily theatre lists during the study in order to approach participants and collect their DOS data. Our major analyses were restricted to people undergoing surgery by 31 July 2022, as study funding ended then.

Intervention and study procedures

Eligible smokers were assigned sequential study numbers in order of receipt of their HQ’s, then randomised by AW (1:1) to control/intervention groups using simple computer-generated sequences (www.randomizer.org). Control group participants received the standard hospital smoking cessation care; a previously evaluated brochure on smoking and surgery that was posted to smokers added to the surgical wait-list.11 Intervention group participants also received this, but were posted an additional invitation asking for participation in a ‘cash challenge’ to quit smoking before surgery (online supplemental file 1). This invitation stated that quitting would be an opportunity to improve health as well as earn a financial reward (a $A70 supermarket voucher) on the DOS. It further stated that to obtain the reward, patients were required to notify us (preregister) of an intention to quit (by telephone/email) plus providing DOS cessation verification by carbon monoxide (CO) breath testing. Preregistration was an opportunity to answer participant’s questions and registrants were advised they would be electronically referred to Quitline for free telephone-based behavioural support (unless opting-out). Registrants were sent study-specific printed information outlining options and costs for obtaining cessation pharmacotherapy in the community (online supplemental file 2). No other cessation support was provided to participants. Seventy dollars was chosen as the incentive because it was similar to mailed NRT costs in our previous studies5 6 and the Cochrane report concluded that incentive size was not important.7

Intervention group participants who had not initiated contact within 1 week of posting invitations were telephoned (one call attempt) to discuss participation. For unanswered calls, a voicemail or text message was left, advising our contact details. Participants were deemed to have declined the offer by telling us so, or through failure to ever make contact. Given the Human Research Ethics Committee (HREC) had waived consent requirements at randomisation, smokers added to the wait-list were not informed they were trial participants, and that some patients were offered incentives for quitting.

Whenever surgery eventually occurred, RAs blinded to allocation approached patients in the preoperative ward. Verbal consent for a questionnaire on quitting behaviour since being added to the wait-list was requested (online supplemental file 3). Questionnaire refusal resulted in current smoker classification. People claiming to have quit at least 24 hours before surgery gave written, informed consent for a confirmatory exhaled CO test using a calibrated piCO+Smokerlyzer (Bedfont Scientific, UK). While expired CO values of less than 10 parts per million (ppm) are generally used to confirm smoking cessation,12 we chose a higher level of confirmation (<8 ppm); consistent with other perioperative cessation trials.5 6 Confirmed quit claims resulted in a concealed study allocation envelope being opened. Intervention group participants received the supermarket voucher, but only if marked as having preregistered an intention to quit. Refusal of verification requests or CO levels≥8 ppm resulted in current smoker classification. In case of disputed readings, a second calibrated Smokerlyzer was available, with the lowest reading used for determination. Only verified quit patients were telephoned at 3 months postoperatively for self-reported abstinence during the preceding 7 days. Anyone uncontactable after three call attempts was assumed to have relapsed.

From 17 May 2021, the HREC approved a protocol amendment allowing all new intervention group patients to be offered double the original incentive; that is, $A140 (£74, €84, US$90). After conducting the study for several months, registrations to quit were lower than expected, elective operating had been seriously disrupted by the SARS-COV2 pandemic and only four supermarket vouchers had been given away. This unplanned change provided an opportunity to explore offer uptake and outcomes with a higher amount.

Outcomes

The primary outcome was the proportion of smokers who quit at least 24 hours before surgery, confirmed by CO analysis. Feasibility outcomes included success of patient contact, offer uptake and disputes over non-smoker classification. Secondary outcomes were self-reported quitting four or more weeks prior to surgery (confirmed by CO testing on DOS), quitting activity during waiting period (successful attempts and unsuccessful attempts, sustained for at least 24 hours), Quitline contacts, cessation pharmacotherapy use (including use of e-cigarettes when used as a cessation aid) and 3-month relapse rates (only if quit on DOS).

Statistical analysis

Our a priori power analysis was based on our most recent cessation trial of mailed dual NRT,6 which had shown a control group abstinence rate of approximately 10% and estimated 199 people per group needing to undergo surgery to detect a 10% increase in participants quit before surgery (power, 80%; α=0.05). Our protocol was originally based on an earlier pilot mailed NRT study showing a 4.8% group quit difference, requiring a study with 1590 participants,5 although for the purpose of determining feasibility of the intervention, a study of 400 patients undergoing surgery was planned. We aimed to randomise more participants than this at listing, because long wait-times would mean that some randomised participants would never have surgery during the study. One trial of mailed NRT at wait-listing had 31% of participants not having surgery during the study’s data collection period.6 We undertook an intention-to-treat analysis by assigned group; associations between intervention and quitting are reported as ORs with 95% CIs. The intention-to-treat population was the randomised population of adult smokers (≥10 cigarettes/day) undergoing surgery at the hospital for the duration of the study. Statistical analyses were conducted in Stata .13.0.

Patient and public involvement

There was no involvement during the planning or conduct of the trial. Participants were offered the opportunity to receive study results.

Results

We identified 620 eligible participants from wait-list HQs, of whom 404 (65%) had undergone elective surgery at Frankston Hospital by 31 July 2022 (intervention group, 190; control group, 214, p=0.4) (figure 1). The characteristics of the control/intervention groups at randomisation were similar; typically, middle-aged heavy smokers with high rates of economic disadvantage (concession cardholders) (table 1). The local impacts of the COVID-19 pandemic were significant as elective operating ceased completely for several periods during the study to allow for care of respiratory patients. Closed theatres and surgical cancellations were frequent, but at other times when COVID numbers fell and lockdowns paused, special government initiatives addressing lengthening wait-lists resulted in transfers of public–patient surgery to local private hospitals. Impacts included a loss of participants (figure 1) and the median time from listing to surgery was double that expected (control group; 177 days, intervention group; 215 days, mean difference 38 days, p=0.45) (table 2). Less urgent surgery (category 3) was significantly less likely to be operated on, as 170 of the 620 randomised wait-listed participants were category 3 (27.4%, table 1), but such patients were just 81 of the 404 completed surgeries (17.6%, table 2) (p=0.01). Surgery performed after 31 July 2022 was not included for funding reasons and the final 3-month postoperative data were collected on 31 October 2022.

Figure 1

Flow chart of randomised participants.

Table 1

Participants and surgery characteristics at wait-listing

Table 2

Participants and surgery characteristics at day of surgery

Cessation of smoking before the DOS

The characteristics of 404 randomised participants who had their operation at the hospital are shown (table 2). Refusal of DOS data collection was 3.2% overall, resulting in classification as current smoker (table 2). Nobody claiming to have quit refused CO testing although 7 of the 45 quit claims exceeded 7ppm (15.6%), one of whom had preregistered to quit for a $A140 reward (tables 3 and 4). This claim was rechecked with a second calibrated device and the decision to withhold payment accepted by the patient. One participant claiming quit was not tested in error (tables 3 and 4).

Table 3

Outcomes by group/amount offered

Table 4

Major‡ outcomes by group

Verified quitting (>24 hours) occurred in 19 of 214 control patients (8.9%) and 18 of 190 (9.5%) intervention group patients (OR 1.1, 95% CI 0.5 to 2.2), with median quit durations of 21 days (table 4). There were no significant group differences in other important quitting outcomes, such as quitting≥4 weeks before surgery, or any quitting activity (successful quits or attempts lasting ≥24 hours) (table 4). As expected, Quitline use was much higher for intervention participants, (OR 6.3, 95% CI 2.3 to 21.6), because those registering quit intentions were referred, unless opting-out. Recall of cessation pharmacotherapy use while on the wait-list was 24.5% overall; mostly pharmaceutical NRT, vaping or varenicline, although some patients were using combinations of these (table 4). To determine if there was a change in amount smoked between being listed for surgery and having an operation, we compared DOS smoking data with that self-reported smoking on the preoperative HQ. Smoking reduction just under four cigarettes/day was found for both groups, with no significant group differences (table 4). Overall, 16.2% of those presenting for surgery were smoking more than previously stated (by median 5 /day), 32.8% the same and 51% were smoking less (by median 5 /day).

Of the intervention group’s 18 verified quitters, only 7 had preregistered for the offer and received rewards (5×$A70 and 2×$A140 rewards) (table 3). The other 11 were ineligible due to non-registration; mostly from non-response to contact attempts (table 3). Within intervention group participants, cessation outcomes were better for those registering intentions to quit compared with declining participants (table 5). Increased quitting activity (successful plus sustained attempts ending in relapse), more cessation pharmacotherapy use and higher Quitline use were seen in acceptors, but a trend for higher overall quitting was not significant, (OR 2.3, 95% CI 0.7 to 7.1).

Table 5

Comparison of outcomes within the intervention group

Although the numbers were small, higher rewards were not associated with more quitting (table 3) proved with higher financial rewards. The verified quit rate in the $70 offer group was A70 offer group was 11.4% compared with just 5.2% in the $A140 offer group (table 3).

Cessation 3 months after surgery

One quit intervention group participant withheld consent for 3-month telephone follow-up, saying she planned to resume smoking after surgery. Contact was successful with 12 of the remaining 18 intervention group quitters (66.7%) and 15 of the 19 control group quitters (78.9%), who were asked about any smoking in the last 7 days. Assuming anyone uncontactable had relapsed, 8 of 19 control group quitters remained abstinent (42.1%) and 5 of the 19 intervention group quitters (26.3%). Median time to relapse after surgery was 21-days (IQR 7–33 days)

Feasibility of intervention group contact

Intentions to quit were registered by 77 of the 308 (25%) of intervention group patients at wait-listing (table 1). Participant engagement was higher at $A140 as 48 of 219 140 as 48 of 219 $A70 offers were accepted (21.9%) compared with 29 of 89 (32.6%) at $A140 (p=0.06). Posting the information bundle was ineffective at gaining registrations, with only 2.2% contacting us after receiving the flyer, so hundreds of phone-calls were made to the remaining intervention participants. In 51.6% of these cases, calls were answered immediately, otherwise, messages (voicemail or text) were left. From these messages, just 15.7% would contact us back, so ultimately 60.6% of intervention participants were spoken to before surgery. Of the 231 intervention participants declining the quit incentive, 104 (45%) directly told us so and 127 (55%) declined by non-response. Offer acceptors were older than decliners by a mean 5.1 years (95% CI 1.2 to 8.9 years) with point estimates suggesting a trend for males to be more likely to accept, though this was not quite significant at the 5% level (table 1).

Discussion

We found no clinically important cessation benefits from offering small financial incentives at time of wait-listing to a mixed-surgical public hospital population. Unusual in the financial incentives literature, the HREC exemption for consent at randomisation meant that data collection occurred from all smokers (≥10 /day), regardless of their interest in quitting or taking part in incentive programmes. Consequently, these results are likely to accurately represent the real impact of a single, small financial offer within the Australian public hospital setting. Patients were seeking surgical treatment, not tobacco treatment and although research suggests surgery is a ‘teachable moment’ where quitting is more likely,9 the offer did little to motivate further cessation, despite this population having increased postoperative risks through tobacco use.

In contrast to our findings, the 2019 Cochrane meta-analysis of 33 financial incentive trials found that in mixed populations of smokers, abstinence at 6-month follow-up points was increased if incentives were offered.7 Unlike many of the included studies in the Cochrane report, ours was a one-time offer of a relatively small incentive in a high-income country, and there were few other components to our intervention (Quitline referral and a pamphlet, but no cessation pharmacotherapy or multisession counselling). The included Cochrane studies took place at mostly workplace or community settings in North America, Europe and Asia, with many studies having multiple components (NRT in one-third of the studies) and often with quit motivated participants. For example, of the six included studies that contributed more than 5% wt to the meta-analysis, two involved participants motivated enough to commit their own money as saving deposits that were forfeited if they returned to smoking,13 14 while another recruited 50% of participants from callers to Quitline.15

The Cochrane report on financial incentives included just one study involving hospital outpatients, and these were patients with head/neck cancer undergoing surgery and/or chemoradiation but still smoking.16 Despite this cohort’s risk of cancer recurrence and the opportunity to receive US$50 for attendance at multiple stop-smoking classes (all participants) and earn US$150 at multiple time points for cessation (incentive arm participants), only 21% of eligible smokers enrolled in the study with 33% of these not even presenting for the initial stop-smoking class.16 At 6 months, only two of six intervention participants quit (nil of eight control group) in a study underpowered to conclude effectiveness due to low enrolments. Our enrolment rate of 25% in the intervention group was similar, as was the lack of significant effect from offering incentives, and although we did not have a fixed 6-month time point to verify abstinence, the 215 days (7 months) median time between wait-listing and surgery was close.

The only other presurgical incentives study was a pilot trial in the USA, where financial rewards were studied in smokers 10-days to 5 weeks before undergoing cancer surgery.8 This was an intensive intervention involving multiple hospital-based counselling sessions, NRT and three times weekly exhaled CO breath testing where incentive payments were made for negative results. Participants could earn a maximum of US$645 ($A996, £522, €603), depending on several factors, with those in incentive groups having three times the verified abstinence on DOS (RR 3.2, 95% CI 1.1 to 9.3).8 Nicotine’s addictive quality is highlighted by the fact that for these (often) tobacco-related surgeries, and despite significant money on offer, only 52% of participants randomised to incentives could quit (16% if offered the same help without incentives). As for most incentive research, only data from those consenting for the trial were included, and presumably the picture would be grimmer if those with no interest in participation were included.

Our main comparison was between the control and intervention group on success or failure to quit smoking by surgery. As expected, participants who registered quit intentions (and received potential rewards) had better cessation outcomes in several areas than those who did not (table 5). Incentives may work, but only for those open to the possibility of receiving them. It should also be noted that the majority of intervention group quitting occurred in smokers who had ‘declined’ the offer (table 3). Most of these quit smokers had not responded to our contact efforts, so declining was the result of inaction or indifference rather than actively refusing the incentive programme. Making more telephone call attempts or exploring other contact options may have yielded different results and further research is needed on how to best promote cessation interventions.

The COVID-19 pandemic had significant impacts, including prolongation of waiting periods (median 215 days) that might have allowed more time for quit attempts to end in relapse before surgery. By contrast, this figure was 99 days in our (positive) study of offering mailed NRT on wait-listing, which partly preceded the pandemic.6 Acceptors of cash incentives had significantly more Quitline use, cessation pharmacotherapy use and overall quitting activity, but relapse during the long periods before surgery was frequent. Long waits may have reduced effectiveness in other ways, as studies show smokers display higher delayed discounting rates.17 18 This means smokers tend to value frequent, smaller immediate rewards over larger, later rewards, so offering a gift card, once only, for collection at an unknown future surgical date may have been suboptimal. Offer reminders during the wait-list period, or reoffering once a date for surgery was confirmed, may have yielded better outcomes, even without increasing the amount offered.

Patient and surgical characteristics varied between listing/randomisation and DOS (tables 1 and 2), and this was likely another COVID-19 pandemic effect. Proportionally, more randomised category 1 patients were operated on as the hospital prioritised urgent cases, while category 3 patients were less likely to have operations by study closure. Category 3 patients include major joint arthroplasty patients, who tend to be elderly. Category 1 patients include cancer-related general surgery, who may be younger, possibly more likely to accept cessation support and receive surgery faster. As such factors may not be entirely independent of the propensity to quit, there was potential to introduce bias. Similarly, only 65.2% of randomised participants had their operation at our hospital during the study period, with others having surgery at private hospitals, being removed from the wait-list or dying while waiting (figure 1). Although both groups were similarly affected making this unlikely to introduce bias, such postrandomisation factors may not be entirely independent. For example, if urgent cancer surgeries were more likely to receive transfers to private hospitals (lost data), this could distort study findings if such patients were also more likely to be acceptors. The overall proportion of patients not having surgery at our hospital (34.8%) was higher than our mailed combination NRT study (31%, study period partly during pandemic)6 or our mailed NRT patch study (9.2%, preceded pandemic).5 Similarly, the 1.5% rate for participants either dying or requiring surgery as an emergency was not seen in the previous studies and points to the marked effects of the pandemic on elective surgery.

The $A70 (£37, €42, US$45) incentive was low compared with many financial incentives studies and was chosen because it was similarly priced to our successful mailed NRT trial.6 Furthermore, our thinking to offer lower amounts was influenced by the Cochrane meta-analysis that reported no significant improvements between trials paying less than US$100 compared with those paying over US$700,7 and a subsequent 2020 review finding that higher incentives did not induce more quitting.19 If the $A70 reward triggered successful quitting, not buying cigarettes would confer the greatest financial benefit, given the recommended retail price (2021) of a pack of 25 cigarettes (Winfield) was $A48.70 (£25.50, €29.40, US$31.30).20 As over half the randomised participants had incomes low enough to qualify for government social security benefits (concession card holders), we anticipated offer engagement would be high, but the 21.9% offer acceptance at $A70 was low compared with 32% acceptance for free mailed NRT previously.6 Paradoxically (because they need money less), monetary rewards incentivise quitting in smokers with higher socioeconomic status better than lower.21 Lower socioeconomic status smokers tend to quit less because of higher nicotine dependency, lower cessation self-efficacy and reduced social supports for quitting, with these factors and others interacting to make them less responsive to monetary rewards.21 Increasing incentives to $A140 could not be recommended because although engagement increased to 32.6%, quit outcomes did not.

Although pandemic-related mail disruptions may have relevance, posting study invitations cannot be recommended as just 2.2% contacted us after 1 week. Email contact could have been better, but a recent study found it worse, with just 25% even opening the email and even fewer clicking on contained links.22 Telephone contact was somewhat more efficient as we connected with over half the participants following one attempt, which was slightly higher than 42% in the largely prepandemic mailed NRT study.6 Anecdotally, many we spoke to after they responded to our voicemails stated reluctance to pick up unknown numbers due to high volumes of marketing calls. Difficulty with telephone research recruitment was identified almost 20 years ago23 but scams and telemarketers are likely making matters worse. Other study issues, such as non-quit smokers falsely claiming rewards were not problems. The outcomes of CO testing were universally accepted. Unblinding of research personnel by quit smokers who had preregistered to quit occasionally occurred, as these patients anticipated the test and reward and could divulge this to staff. This was unlikely to bias results given the objective nature of CO testing.

Limitations

The preplanned DOS data collection (199 per group) was not achieved within the available time by 9 patients in the intervention group. Extending the study was not possible and would probably not yield different results given the lack of signal for increased cessation. Doubling the incentive part-way through the study was a protocol amendment addressing low response rates, but was not a preplanned change. There were low numbers of patients in the $A140 arm and the study was not powered to compare outcomes from offering the different amounts.

The mixed-surgical population of an Australian public hospital may not be generalisable to others. Besides factors such as national income, underlying smoking rates and surgical mix, smoking cessation aids vary significantly between countries. For example, nicotine-containing e-cigarettes are illegal in Australia but widely used in the UK and elsewhere.24

Strengths include this being among the largest of all perioperative cessation trials.4 Inclusion of all eligible smoker is another as it showed the real effect of the intervention on a population. Trials randomising only after consenting comprise the vast majority of studies,7 19 yet may overestimate the incentive population effect by including subsets with better prognostic features. Evidence for this was clear in our study, comparing the offer acceptors with those declining.

Conclusions

A single offer of a small financial incentive and Quitline referral at wait-listing for cessation by surgery was feasible but did not result in clinically important improvements in quit outcomes. Differently designed incentive programmes may perform better, but offering pharmacotherapy and behavioural support on wait-listing is already a proven strategy that can be recommended.6 25 Future perioperative incentives research could focus its role increasing smoker engagement and compliance with NRT and/or counselling.

Data availability statement

Data, including individual deidentified participant data, study protocols, ethical approvals and consent forms are available to researchers providing a reasonable request and HREC approved proposal. This will be available from the date of publication with no end-date determined.

Ethics statements

Patient consent for publication

Ethics approval

This study involves human participants and was approved by Peninsula Health Human Research Ethics Committee HREC/61015/PH-2020. Participants gave informed consent to participate in the study before taking part.

Acknowledgments

We thank the following final year Monash University medical students for their assistance in data collection: Sam Le Lievre, Eldon Lee, Ruben Jakob, Elynn Ong, Amanda Goh, Anika Hettipathirana, Anna Langford, Beatrice Yip, Evina Ling and Clara Chua.

References

Supplementary materials

Footnotes

  • Contributors AW: conceived and designed the project, obtained funding and approvals, participated in data collection, data and manuscript preparation. LC: contributed to project design, managed staff/student training and participated in data collection and manuscript preparation. AK: project design, data collection and manuscript preparation. MY: data collection, data analysis, manuscript preparation. SL: data analysis, statistical support and manuscript revisions and final approval. AW is the author responsible as guarantor.

  • Funding This work was supported by a 2019 Australian & NZ College of Anaesthetists, Clinical Trials Network Pilot Grant, $A10 000 (no grant number, unencumbered grant).

  • Competing interests None declared.

  • Patient and public involvement Patients and/or the public were not involved in the design, or conduct, or reporting, or dissemination plans of this research.

  • Provenance and peer review Not commissioned; externally peer reviewed.

  • Supplemental material This content has been supplied by the author(s). It has not been vetted by BMJ Publishing Group Limited (BMJ) and may not have been peer-reviewed. Any opinions or recommendations discussed are solely those of the author(s) and are not endorsed by BMJ. BMJ disclaims all liability and responsibility arising from any reliance placed on the content. Where the content includes any translated material, BMJ does not warrant the accuracy and reliability of the translations (including but not limited to local regulations, clinical guidelines, terminology, drug names and drug dosages), and is not responsible for any error and/or omissions arising from translation and adaptation or otherwise.