Article Text
PDF
Abstract
Background The point-of-sale (POS) is adapting to marketing restrictions, societal changes and the inclusion of new products, such as heated tobacco products (eg, Philip Morris International’s (PMI) IQOS device and HEETS sticks). We aimed to assess (1) PMI’s influences on IQOS/HEETS POS marketing and (2) the implications of the new legislation (POS display ban and plain packaging) for retailers.
Methods A cross-sectional survey of 43 IQOS/HEETS POS owners/managers in five Israeli cities assessed POS and participant characteristics, marketing strategies, attitudes towards IQOS, and POS implications of the legislation and COVID-19, including industry reactions. Bivariate analysis explored differences between POS selling of the IQOS device versus POS selling of HEETS only.
Results A higher proportion of those carrying IQOS (n=15) (vs HEETS only) had special displays (100% vs 17.9%, p<0.001) and interacted with specific IQOS salespersons (73.3% vs 28.6%, p=0.013). Common promotions were financial incentives based on HEETS sales for retailers (37.5%) and price discounts on HEETS for customers (48.7%). Most indicated positive attitudes towards IQOS (72.1%; eg, ‘less harmful’), opposition to the legislation (62.7%), limited government assistance to implement the legislation (62.8%), and industry provision of display cases and/or signage to comply with the legislation (67.4%).
Conclusion PMI uses similar tactics to promote IQOS at POS as they previously used for combustible products, including direct promotional activities with retailers, and circumvented legislation by using special displays and signage. Governments need to ban these measures and support retailers with clear practical guidance regarding the implementation of marketing restrictions at POS.
- advertising and promotion
- tobacco industry
- non-cigarette tobacco products
Data availability statement
Data are available upon reasonable request.
Statistics from Altmetric.com
Introduction
The point-of-sale (POS) environment is an important venue for tobacco product marketing and advertisement.1 2 POS advertisement exposure may contribute to higher smoking initiation and use among youth,1 2 hinder smoking cessation and trigger relapse among ex-smokers.3–5 Tobacco companies use various POS marketing techniques, including paying retailers for designated shelf spaces and/or special displays, discounts and price reductions.6
Tobacco advertisement and POS display bans are effective tobacco control strategies; they reduce exposure to displays, brand awareness, impulse purchases and smoking norms.1 7–11 Such bans achieve high compliance; nonetheless, tobacco companies continue to directly incentivise retailers and use different promotional strategies.12 13 For example, tobacco companies used direct communication with retailers (eg, provide retailers with materials to help explain changes to consumers) to circumvent plain packaging in the UK.14
Aside from the implementation of new tobacco control legislation, societal changes, such as those imposed by the COVID-19 pandemic, may also alter POS.15 16 For example, a study assessing US vape shop activities during COVID-19 found that many offered pick-up or home delivery and some began selling food items to qualify as ‘essential’ businesses.17 US retail tobacco purchases increased by 13% in the early months of the pandemic compared with the year before, with increases across all geographical areas and demographic categories.18
As the POS is evolving to adapt to new tobacco and nicotine products, marketing restrictions, and other societal changes such as the COVID-19 pandemic, it is important to comprehensively assess tobacco industry influences on retailers and POS. Most previous studies have used either POS audits or cross-sectional consumer studies, and fewer have directly engaged merchants. Moreover, direct research on retailers’ experiences with tobacco companies’ influence on POS strategies has focused on combustible tobacco products. Thus, little is known about whether or how the same tactics are being used to promote non-combustible tobacco products such as heated tobacco products (HTPs).
Philip Morris International’s (PMI) IQOS is the leading HTP globally.19 In July 2020, IQOS was the first HTP to receive Food and Drug Administration’s (FDA) modified risk tobacco product authorisation, allowing PMI to market the product as ‘reduced exposure’ (ie, reduced exposure to harmful chemicals).20 However, FDA did not authorise IQOS as a ‘reduced risk’ product, stating that current evidence does not support that its use reduces harm or risk.20
IQOS, with its corresponding HEETS tobacco sticks, entered Israel in December 2016; since April 2017 IQOS is regulated as all other tobacco products and is currently (2021) the only HTP available in Israel.21 In March 2019, an advertisement ban (including at the POS) went into effect, and in January 2020 plain packaging and a POS display ban were implemented.21 Nonetheless, tobacco control legislation in Israel does not include any reference to promotions or incentives provided to retailers.21 Regarding consumers, promotions such as providing free samples, paraphernalia and/or any other gifts were banned previously (April 2017), and the new advertisement ban (effective March 2019) extended the ban to include also price promotions. However, price discounts to customers are still allowed.21
In 2019, 1.8% of Israeli adults reported using either IQOS or electronic cigarettes (e-cigarettes).22 Youth IQOS experimentation has increased from 1.0% in 2019 to 5.6% in 2021.23 Globally, awareness and use of IQOS have been steadily increasing, particularly among smokers, but also among non-smokers.24–31
A prior study assessing IQOS POS marketing in Israel after the advertisement ban but before plain packaging and the POS display ban, indicated that although 95% of POS complied with the advertisement ban,32 there were special efforts directed towards promoting IQOS, including those exposing youth.32 However, this study used concealed audits, preventing assessment of marketing strategies directed at retailers themselves and was conducted prior to the pandemic.
This literature underscores the need to better understand the role of tobacco companies in influencing retailers’ POS strategies in the context of restricted POS marketing and COVID-19, specifically for emerging tobacco products. This study used a direct survey of retailers in Israel, conducted after the implementation of plain packaging and a POS display ban, which occurred during the COVID-19 pandemic. This study assessed (1) PMI’s influences on IQOS/HEETS marketing at POS, in general and before and after implementation of plain packaging and a POS display ban; and (2) the implications of societal circumstances, particularly the new legislation and COVID-19, for tobacco retailers.
Methods
Study setting, sample selection and procedure
We used publicly available data on the IQOS Israel website33 to identify IQOS/HEETS POS locations in five large cities (Haifa, Nazareth, Tel-Aviv, Jerusalem, Beer-Sheva) in November 2020 (n=713) (figure 1). Then, we searched online for each POS phone number. From 14 December 2020 to 18 February 2021, research assistants (RAs) contacted each POS, determined if the store sold the IQOS device and/or HEETS (currently or in the past) and asked to talk to the manager or owner of the POS. If the store did sell IQOS/HEETS and had a manager or owner available, the RAs then explained the purpose of the study and requested they complete the survey. Owners/managers (one per POS) were offered the choice of completing the survey through an online link or completing it via phone with the RA. Those who completed the survey were compensated with a 100 New Israeli Shekel voucher (~US$30). Three attempts were made to contact each POS. Overall, out of 171 eligible POS (working phone numbers, able to be contacted and sold IQOS/HEETS), 43 completed the survey (25.1% response rate) (figure 1).
Flow chart of POS participation. PMI, Philip Morris International; POS, point-of-sale; RA, research assistant.
Data collection
Survey
The survey assessed (1) POS characteristics (type of store, belonging to a chain); (2) participant characteristics (gender, age, cigarette smoking status, IQOS use status); (3) IQOS/HEETS products sold (when they began selling the IQOS device and/or the HEETS, number of HEETS colours sold); (4) marketing strategies, including promotions (eg, price discounts, incentives for sales) and advertisements (eg, presence of a special display, advertisements to customers)—for each question, respondents could expand with free text; and (5) attitudes towards IQOS via nine checkbox questions assessing how they would communicate with consumers about IQOS and/or HEETS (eg, ‘IQOS is as satisfying as regular cigarettes’) and one open-ended question assessing their own personal opinion about IQOS and/or HEETS (online supplemental file 1).
Supplemental material
We also assessed implications of the new legislation for POS, including (1) opinion on the new legislation, from 1=very much in support to 5=very much in opposition; (2) difficulty in understanding and in implementing the new legislation, from 1=not at all to 4=very much for each; (3) impact on sales for cigarettes, IQOS/HEETS and e-cigarettes, from 1=far less sales to 5=far more sales for each item; (4) industry reactions to the tobacco legislation (ie, provided education on the new legislation; provided free cabinets, display cases and/or signage to address the tobacco legislation; checkbox for each item with room for elaborating); and (5) open-ended items asking about any difficulties in understanding/implementing the legislation and what the government/local municipality has done to help with its implementation and enforcement.
As the implementation of the POS display ban and plain packaging occurred just prior to the COVID-19 pandemic, we assessed the implications of the pandemic, including its impact on sales of cigarettes, IQOS/HEETS and e-cigarettes, from 1=far less sales to 5=far more sales for each. This section also included open-ended items asking about tobacco industry response to COVID-19 and the impact of COVID-19 on their ability to comply with and/or implement the new legislation.
Neighbourhood data
Data describing the population characteristics for each POS location were provided by Points Location Intelligence, a private Israeli mapping and data company, using the Israel Central Bureau of Statistics 2017 National Population Statistics Survey. The bureau divides Israel into small, homogeneous statistical areas with an average of 3000 people per statistical area,34 and characterises the population by (1) Socioeconomic Status (SES) ranking, calculated based on 14 factors (eg, monthly income) to yield scores on a 0–10 ordinal scale (eg, industrial unpopulated area (0), low SES (1–3), medium SES (4–7) and high SES (8–10))34; and (2) population group (>50% Israeli Arabs or >50% Ultra-Orthodox Jews or general population).
Data analysis
Analysis was performed using SPSS V.26. Descriptive analysis was conducted using frequencies (%) for categorical variables. Bivariate analysis was conducted to explore differences between POS that sell both the IQOS device and HEETS versus POS that sell only HEETS using χ2 and Fisher’s exact tests. Significance was set at p<0.05. Open field questions were coded into categorical responses by one researcher (YB-Z).
Results
POS characteristics
The final sample included 43 POS (Beer-Sheva n=4, Haifa n=11, Jerusalem n=13, Tel-Aviv n=14, Nazareth n=1), of which 39 currently sell HEETS and 4 sold them previously; 15 currently sell both HEETS and the IQOS device and 5 sold the device in the past (of these, 3 continued to sell HEETS and 2 stopped selling both IQOS and HEETS). POS were largely convenience stores with or without gas stations (51.2%, n=22) or grocery stores (30.2%, n=13); the remaining eight stores were liquor stores (n=4) or tobacco shops (n=4). Most of the POS were not part of a chain (86.0%, n=37) and were mostly in medium or high SES neighbourhoods (46.5% (n=20) and 23.3% (n=10), respectively) and non-minority populations (ie, only 11.6% (n=5) in Arab or Ultra-Orthodox Jewish neighbourhoods, respectively). Majority of the participants were current smokers or ex-smokers (57.1%, n=24), with a smaller proportion being current users or ex-users of IQOS (23.8%, n=10), which was more common among retailers who carried the IQOS device versus HEETS only (46.7% vs 11.1%, p=0.020).
Most POS started selling HEETS more than 2 years ago (n=29, 67.4%). Of the POS currently selling the IQOS device, 60% (n=9) started selling it in the last 2 years. Majority of the stores (69.8%, n=30; missing n=1) sold at least four different HEETS colours, more so among POS also carrying the IQOS device versus only HEETS (100% vs 55.6%, p=0.003). Half (50%, n=21; missing n=1) also sold e-cigarettes.
POS marketing strategies
A higher proportion of those carrying the IQOS device (vs HEETS only) had special displays (100% vs 17.9%, p<0.001) and interacted with a specific IQOS/HEETS salesperson (73.3% vs 28.6%, p=0.013) (table 1).
POS marketing strategies of IQOS/HEETS, total and across those selling both the IQOS device and HEETS versus those selling HEETS only
The most common form of promotions offered to stores were financial incentives based on HEETS sales (37.5%, n=15, missing n=3). Some participants mentioned they received a discount based on sales, while others specified receiving incentives proportional to sales, with additional incentive for each IQOS device sold. The most common forms of promotion offered to customers were price discounts for HEETS (48.7%, n=19, missing n=4). Most described this as a 2 NIS (~US$0.6) discount on HEETS price that they received back from the company. Three participants also described a specific discount on the IQOS device or for a bundle including IQOS plus HEETS. These financial incentives marketing strategies (both to the retailer and to the customers) increased after the POS display ban and plain packaging went into effect (table 1).
More than half of the sample (67.6%, n=25, missing n=6) stated that they advertised IQOS/HEETS inside the POS; of these, 12 (48%) reported currently doing so, most in a way that complied with the legislation, through flags or signs that highlight that they sell HTPs (without mentioning the brand IQOS or HEETS). Two participants mentioned having female sales staff that came to the store to promote these products: one mentioned this was done as a single event at the launch of the product sales and one mentioned such events being done weekly. Social and print media advertisements were mentioned scarcely and only in the past, prior to the new legislation.
Almost half (56.8%, n=21, missing n=6) mentioned receiving some sort of directions/instructions from the PMI salesperson regarding IQOS or HEETS in comparison with other tobacco products in terms of their product characteristics, placement within the POS, targeted consumers and strategies for communication with customers; more retailers reported this was done after the POS display ban and plain packaging went into effect (table 1). Others mentioned that they could access information on the products through a designated app. One participant stated that the PMI salesperson “only provide[s] information to those who sell the device. Those who sell only the HEETS, it’s just like any other cigarette in the store.”
Retailer attitudes towards IQOS
Overall, 31 (72.1%) participants expressed at least one positive attitude towards IQOS, with statements regarding it being “less harmful for smokers” and producing “less smell” (table 2). One participant stated that “the fact that the FDA has approved the product proves that it is a quality product and not harmful to health.” Several participants suggested that it suits only some people and not everyone will find it “satisfying,” with some expressing negative views stating that IQOS produces an “awful distinctive smell,” is “more expensive” and “requires cleaning.”
Attitudes towards IQOS, total and across POS selling both the IQOS device and HEETS versus those selling HEETS only
Implications of the new legislation for POS
Almost two-thirds of the participants (62.7%, n=27) were opposed to the new legislation (41.9%, n=18, very much opposed; 20.9%, n=9, somewhat opposed). Most (69.8%, n=30) stated no (or only a little) trouble understanding the legislation. However, almost half (46.5%, n=20) stated that it was difficult to implement, mentioning “confusion between brands” and the “additional time that is needed to get the right order to the customer.” While most participants stated that the legislation did not impact their cigarette, IQOS/HEETS or e-cigarette sales (69.8% (n=30), 62.8% (n=27), 57.1% (n=12/21 selling e-cigarettes), respectively), a small proportion stated that the legislation reduced cigarette and IQOS/HEETS sales (16.3% (n=7) for each) and 38.1% (n=8/21) that it reduced their e-cigarette sales. Two-thirds (62.8%, n=27) indicated limited government or local municipality assistance in implementing the legislation. Less than a quarter (23.3%, n=10) mentioned that if they were non-compliant, they might be fined, with 9.3% (n=4) stating limited/no knowledge of the consequences of non-compliance. One participant stated that “It is something you can play with. You can say what it covered and what is not, it’s all up to your understanding.”
Tobacco industry’s response to the new legislation
More than half (53.5%, n=23) stated that the tobacco industry provided free cabinets/display cases and/or signage to comply with the legislation; an additional 13.9% (n=6) mentioned that they sold them these items. One participant mentioned that “the company was willing to provide the cabinet for free on condition that a minimum number of products will be bought each week.” More than a quarter (27.9%, n=12) mentioned that tobacco companies provided education on the new legislation, and five (11.6%) stated that they advised on how to work around the legislation or minimised the importance of compliance.
Regarding other notable responses, one participant stated that “their company (PMI) is domineering; they demanded that only their products will be displayed in the cabinet and that we sell a lot, and put the other cigarettes on the side.” Another mentioned having a specific app and loyalty programme for certain POS: “There’s a program that send us updates every week, and they tell us about new products and they give us points when we solve puzzles on this app, it’s only for special shops that have all the products in their line.”
Several participants stated limited comprehension of the new legislation’s purpose since “those who want to smoke will continue and it will not affect them, they don’t care about the colour” and that it makes their job harder. Only one participant acknowledged that the legislation might impact future generations: “those that are 12, they won’t see the products, and therefore will smoke less.”
Implications of the COVID-19 pandemic
Approximately 40% stated that the pandemic increased their cigarettes and HEETS sales (41.9%, n=18, for cigarettes; 39.5%, n=17, for IQOS/HEETS sales), while only one stated that e-cigarette sales increased. However, a proportion stated that the pandemic decreased their cigarette, IQOS/HEETS and e-cigarette sales (30.2% (n=13), 25.6% (n=11) and 18.6% (n=8), respectively). Participants mentioned that this was highly variable throughout the pandemic, with people’s behaviour changing according to lockdowns or other restrictions. While a third (32.6%, n=14) stated that the industry did nothing in response to the pandemic, five (11.6%) stated that the industry lowered prices or provided discounts due to COVID-19. None of the participants mentioned that the pandemic impacted their ability to comply with the legislation; one participant mentioned that “it was excellent, because they’re focused on giving fines to those who don’t wear masks,” implying that it allowed non-compliance.
Discussion
Tobacco companies exert specific efforts directed at tobacco retailers to (1) promote IQOS and HEETS and differentiate them from combustible tobacco products; and (2) proactively address or respond to societal circumstances such as new legislation and COVID-19. As restrictions to POS marketing expanded to include a POS display ban and plain packaging, PMI increased marketing strategies directed at the retailers themselves and price discounts to the customers, both legal under the current tobacco control legislation in Israel. These findings underscore the importance of using multiple strategies (eg, POS audits,10 32 35 36 merchant surveys/interviews6 12 13 37–40 and mystery shopper assessments41 42) to better understand the retail environment as new products emerge, changes to tobacco legislation and regulation are implemented, and societal circumstances alter the retail environment.17 43 44
Findings from this study, as well as a previous study in Israel using POS audits, suggest that PMI is circumventing the POS display ban through special displays, flags and signs to continue promoting IQOS and HEETS at POS.32 45 This is also in line with similar research on IQOS marketing at POS from Canada46 and with previous research exploring PMI’s strategies to adapt to tobacco control legislation for combustible tobacco products.47–49 PMI is also influencing the IQOS and/or HEETS retail environment through direct communication with retailers, as was also previously shown in other studies assessing tobacco POS retailers.12–14 Marketing strategies that were reported in this study closely resemble previously noted past marketing for combustible products, including special displays, discounts and incentives, and product launches with female salespersons.6 12 13
The findings suggest that PMI is targeting and investing in specific POS that also carry the IQOS device, which receive more instructions and promotions from the company to promote IQOS and/or HEETS. Geodemographic targeting approaches (ie, developing algorithms to target specific areas that take into account various neighbourhood characteristics such as age, ethnicity, income and education) are a known strategy employed by the tobacco industry.50 Despite the commonality of this practice, our limited neighbourhood data did not find differences based on SES rank or population group. Nearly half of the merchants in this sample of IQOS device retailers were current or ex-IQOS users, compared with 11% among retailers selling only HEETS and with 1.8% of the general adult population who report IQOS or e-cigarette use.22 This study’s cross-sectional nature does not allow us to determine whether using the product leads to higher chances of selling the device or vice versa. Future research should examine whether PMI specifically targets retailers for using the product, given the unusually high use rate.
The high proportion of retailers who were opposed to the new legislation and found it difficult to implement should raise a ‘red flag’ to the ‘real-world’ effectiveness of these strategies due to barriers impeding compliance. A study conducted in South Carolina, USA, assessing compliance with the Tobacco Control Act (which included various marketing restrictions at the POS level) across retailers (n=252), found that reporting less support for restrictions and more barriers correlated with greater non-compliance.40 Similar to our study, most retailers mentioned not receiving any guidance from the government, as opposed to receiving information from the tobacco industry.40
Study strengths and limitations
To the best of our knowledge, this study is the first globally to assess IQOS marketing strategies at POS through retailer surveys, particularly at a notable time—immediately following the new legislation and during the pandemic—providing additional important data on tobacco industry involvement in the POS environment. Our study sample was small and covered only five cities, with a 25% response rate; therefore, caution should be used when interpreting these results. Specifically, we had few Arab POS and only one from an Arab city. As the Arab population is the largest minority group in Israel (20%), with the highest smoking rate,22 PMI might be using other strategies for this population. A larger study comparing Arab with Jewish POS might provide additional insight. We did not assess PMI’s influences regarding POS online or direct product delivery, which might be more prevalent due to COVID-19.16 17 The survey used mostly closed-ended questions, with most respondents providing minimal elaboration, for example regarding the use of a dedicated app. However, in a previous study including POS owner/manager interviews, few provided full answers, or agreed to recording, and most mentioned they would prefer a survey.
Implications for future research and policy
As long as the sale of tobacco products continues, making the POS environment devoid of any marketing and advertisement of tobacco products, including HTPs, and restricting tobacco product sales to designated licensed tobacco shops are the only ways to reduce the impact of POS marketing. In the mean time, governments must anticipate that the industry will circumvent such legislation, which may require bans on (1) signs/flags or any other means to alert customers that the POS sells these products without customers specifically asking; and (2) special displays/cabinets specifying that these products need to be kept in areas outside customers’ view (eg, under the counter, in separate back rooms). As restrictions on advertisement are increasingly implemented, retailers are becoming PMI’s ‘representatives’ incentivised to spread the company’s messages to the public. This further supports the suggestion that governments must ban direct promotional activities between tobacco companies and retailers.12 Governments and local municipalities need to take active roles in providing rational assistance and practical guidance to retailers when implementing tobacco control measures that affect the POS, as well as enforcement afterwards with clear instructions on how to comply. Public health authorities and non-governmental organisations that work in tobacco control need to be proactive, not only in advocating for these measures in the prelegislation period, but also subsequently working with retailers as partners to ensure compliance.
Conclusions
This study suggests that tobacco companies use similar strategies to promote IQOS at POS as previously used for combustible products.12–14 In addition, they invest efforts to differentiate IQOS and HEETS from combustible tobacco products. Tobacco control measures should tackle industry-selective promotion strategies; specifically, governments must ban direct promotional activities between the industry and retailers and the use of any means to alert customers to tobacco products at POS, including designated displays. To enhance their ability to protect public health, retailers must be supported by governments and other agencies with clear practical guidance and assistance in order to increase support and compliance.
What this paper adds
What is already known on this subject
Tobacco companies have used various marketing techniques at the point-of-sale (POS), including direct promotions to retailers, to influence product placement, advertisement and price.
Israel has implemented progressive tobacco control legislation in recent years, comprising an advertisement ban, plain packaging and a POS display ban for all tobacco products, including heated tobacco products (eg, IQOS device with HEETS sticks).
What important gaps in knowledge exist on this topic
It is not known whether or how the same marketing strategies are being used to promote non-combustible tobacco products such as IQOS/HEETS at POS and whether marketing restrictions have impacted tobacco companies’ influence on IQOS/HEETS POS in Israel.
What this paper adds
The tobacco industry is using similar approaches in Israel to promote IQOS at POS as previously used for combustible products, including direct promotional activities with retailers and circumvention of legislation by using special displays and signage.
Retailers need to be supported by governments and other agencies with clear practical guidance and assistance in order to increase support and compliance with new tobacco control legislation.
Data availability statement
Data are available upon reasonable request.
Ethics statements
Patient consent for publication
Ethics approval
This study involves human participants and was approved by Hebrew University, Faculty of Health Ethics Committee (approval #30112020). Participants gave informed consent to participate in the study before taking part.
Acknowledgments
The authors acknowledge and thank the members of the External Advisory Board for their input to the study’s design and methodology and interpretation of results (by alphabetical order): Joanna Cohen PhD, Nadav Davidovitch MD, MPH, PhD, Milka Donchin MD, MPH, Anat Gesser-Edelsburg PhD, Michael Eriksen ScD, Haim Geva Haspil, Lisa Henriksen PhD, Elad Godinger LLM, Shira Kislev, and James Thrasher PhD.
References
Supplementary materials
Supplementary Data
This web only file has been produced by the BMJ Publishing Group from an electronic file supplied by the author(s) and has not been edited for content.
Footnotes
Twitter @yaelbarzeev, @AmalKhayat5, @RommKatelyn
Contributors YB-Z, CJB, HL and LA designed this study. KFR, CW and AK contributed to study design and methodology. AK and DE contributed to data collection. YB-Z analysed the data and wrote the draft manuscript. All coauthors reviewed the final manuscript. YB-Z accepts full responsibility for the finished work and/or the conduct of the study, had access to the data, and controlled the decision to publish.
Funding This research was supported by the National Cancer Institute (R01CA239178-01A1; MPIs: CJB, HL) and by the Israel Lung and Tuberculosis Association (MPIs: YB-Z, HL).
Competing interests YB-Z has received fees for lectures from Pfizer, Novartis NCH and GSK Consumer Health (distributors of smoking cessation pharmacotherapy in Israel) in the past (2012–July 2019). HL had received fees for lectures from Pfizer Israel (distributor of a smoking cessation pharmacotherapy in Israel) in 2017. LA receives royalties for the sale of Text2Quit and is a shareholder in Welltok.
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.