Article Text

PDF

Original article
Self-reported quit rates and quit attempts among subscribers of a mobile text messaging-based tobacco cessation programme in India
  1. Preetha Gopinathan1,
  2. Jagdish Kaur2,
  3. Surabhi Joshi3,
  4. Vinayak Mohan Prasad3,
  5. Sameer Pujari3,
  6. Pradeep Panda4,
  7. Pratima Murthy5
  1. 1International Institute of Health Management Research, New Delhi, Delhi, India
  2. 2Tobacco Free Initiative, World Health Organization South East Asia Regional Office, New Delhi, Delhi, India
  3. 3Be He@lthy, Be Mobile, World Health Organization, Geneva, Switzerland
  4. 4International Institute of Health Management Research, New Delhi, Delhi, India
  5. 5Centre for Addiction Medicine, National Institute of Mental Health and Neuro Sciences, Bangalore, Karnataka, India
  1. Correspondence to Dr Preetha Gopinathan, International Institute of Health Management Research, New Delhi 110075, India; preetha{at}iihmr.org

Abstract

Introduction In 2015, as part of the WHO and International Telecommunication Union’s ‘Be Healthy Be Mobile’ initiative using mobile technology to combat non-communicable diseases, the Ministry of Health and Family Welfare and the Ministry of Communication and Information Technology in India developed a short text message-based mobile health programme (the ‘mCessation’ programme) to support tobacco users to quit tobacco use.

Objectives To evaluate the effectiveness of the mCessation programme by estimating quit rates and quit attempts among registered subscribers of the programme and to understand subscriber perceptions of the programme.

Methods Subscribers to the mCessation (QuitNow) programme were telephonically interviewed 4–6 months after registration. A total of 12 502 calls were made, and completed responses recorded from 3362 ever tobacco users. A total of 6978 respondents either gave very few responses or refused to participate in the telephonic survey. Never tobacco users (1935) and subscribers to the mDiabetes programme (227) were excluded from the sample.

Results A large proportion of registrants (1935 out of 12 502 respondents) were found to be never users. The quit rate (estimated as those who stated they had not had any tobacco in the past 30 days at 4–6 months after registering with the programme from the total sample (excluding never smokers and mDiabetes registrants)) was 19%. Sixty-six per cent of registered subscribers who were current tobacco users had made quit attempts in the period between registration and survey. Seventy-seven per cent of respondents reported that the programme was helpful/very helpful to quit tobacco.

Conclusion The mCessation programme has successfully helped tobacco users in India to quit tobacco by motivating and supporting registered participants through mobile text messages.

  • mHealth
  • quit rate
  • quit attempt
  • tobacco use
  • mCessation

Statistics from Altmetric.com

Introduction

Globally, tobacco use has killed 100 million people in the 20th century.1 More than 7 million people die each year due to tobacco use and this number is estimated to increase to 8 million by 2025 if unchecked.2 3 For every smoking-related death, at least 30 people live with a serious smoking-related illness.3 Globally, there are 879 million tobacco users (721 million men and 158 million women).4 India is the second largest consumer of tobacco in the world after China. In 2009–2010, 34.6% of the adult population in India used tobacco, of whom 14% were smokers and 25.9% users of smokeless tobacco.5

In India, tobacco causes nearly 1 million deaths annually, which is approximately one-sixth of all tobacco-related deaths worldwide.6 The total economic burden attributable to tobacco use from all diseases in the country in 2011 among adults aged 35–69 years amounted to Indian rupee 104.500 crore (US$22.4 billion), which was approximately 1.16% of the gross domestic product and 12% more than the combined health expenditure of the national and state governments.7

India enacted the tobacco control law, Cigarette and Other Tobacco Products Act (COTPA) in 2003 and became party to the WHO Framework Convention on Tobacco Control (FCTC) in 2004. India is also implementing a National Tobacco Control Programme since 2007.6 8 Article 14 of the WHO FCTC and the WHO MPOWER advocates offering help to tobacco users for quitting.9 Nearly half of the adult tobacco users, both smokers and smokeless tobacco users interviewed in the Global Adult Tobacco Survey (GATS India 2009–2010), expressed a desire to quit tobacco use.5 This demonstrated demand is not matched by capacity to provide support for tobacco cessation, which is restricted to limited and scattered tobacco cessation services in the country.8

Text messaging (or short message service (SMS)) is a popular mode of communication worldwide and has great potential to deliver healthcare and promote behaviour change. Recently, multiple systematic reviews have demonstrated the efficacy of text messaging as a tool for behaviour change for an array of targeted behaviours including smoking cessation.10–12

Mobile phone use in India has grown exponentially and is expected to increase from 525 million in 2013 to 813 million in 2019.13 This proliferation in mobile use, along with one of the cheapest data and SMS plans14 in the world, makes India an attractive market for mobile intervention programmes. The mCessation programmes can potentially leverage this unique opportunity to reach out to a larger population of tobacco users through mobile texting. Given the scarcity of tobacco cessation programmes in India relative to the demand for support to quit, mCessation programmes can significantly expand the reach of cessation services to tobacco users. A recent meta-analysis of text-based tobacco cessation programmes has shown them to be cost-effective compared with other programmes.15

About the tobacco mCessation programme

The WHO and the International Telecommunication Union, through a partnership called the Be He@lthy, Be Mobile initiative, support member states to scale up mHealth programmes nationally. mCessation, in particular, provides support for users to quit tobacco use through a two-way text message programme, developed with support of global mHealth and tobacco cessation experts. The Ministry of Health and Family Welfare (MOHFW) and the Ministry of Communication and Information Technology of India partnered with the Be He@lthy, Be Mobile initiative to introduce innovative technologies for strengthening tobacco cessation services in the country. The MOHFW, with support from the National Institute of Mental Health and Neuro Sciences and an Expert Group, adapted the content for the messaging from a publicly available message library.16 The messages were adapted both for smoking and smokeless tobacco cessation, in English and Hindi. Web-based culturally appropriate information on the reasons to quit, simple strategies to deal with craving, sleep, hygiene and alternative techniques to manage moods and deal with stress were developed. The Centre for Health Informatics at the National Institute of Health and Family Welfare under the MOHFW implemented the technology platform to create a two-way text message delivery system. The Government of India’s MyGov platform and National Informatics Centre provided technical support for the text messaging (SMS) short code and coordinated with the telecom companies. Telecom Regulatory Authority of India negotiated the cost of text message delivery for the mCessation programme with telecom companies. Through this multistakeholder collaboration, a nationwide mCessation programme (QuitNow) was launched in January 2016, as part of the Prime Minister’s Digital India initiative and approximately 180 000 people registered in the first month of the programme. As of November 2016, the programme had about 2 million users.17–19

Registration with the QuitNow programme follows the steps below17:

  • Missed call (call from a mobile and hang up within two rings) to toll free number 011 22 901 701, or registration at the website: http://www.nhp.gov.in/quit-tobacco.

  • Text message sent to the subscriber confirming the registration.

  • Series of text messages sent to the subscriber providing information on:

    • The importance of quitting and encouragement to quit.

    • Setting up a quit date.

    • Tips to support the quit attempt.

    • Follow-up support.

The programme sends about 150 text messages according to a defined algorithm to the registered user over a period of 6 months. The text messages are delivered to the registered users, encouraging them to set and achieve quit goals, promoting self-management skills at the same time.18 19 The cost of the programme (sending ‘push’ text messages) is entirely borne by the Government and the user does not incur any cost for the received messages. The ‘pull’ messages (sent by the users in response to questions asked) are charged and the cost incurred on the responses to all the questions is nearly half a US$. However, sending these text message responses is entirely optional as the push messages continue, even if a response is not sent by the user.

This paper describes an evaluation of the mCessation programme with two specific objectives: to estimate quit rates and quit attempts among the registered subscribers, and to understand subscriber perceptions of the programme.

Materials and methods

Study setting and study population

We contacted subscribers of the mCessation programme who had registered between 21 June and 20 September 2016.

Sample size

A feasible sample for this evaluation was determined in collaboration with an external expert group to be approximately 3500 subscribers. The estimated baseline quit rate was considered to be around 5%5 from population surveys. A sampling frame was constructed of 453 990 people who subscribed to the mCessation programme between 21 June and 20 September 2016, so that they would have completed approximately 4–6 months of the programme.

Data collection tools and techniques

A telecalling team of 10 staff conducted telephonic interviews of subscribers using a semistructured interview schedule. Information collected included background demographic characteristics, current tobacco use status, type and frequency of tobacco use, quit attempts, 30-day quit status as well as impressions about the effectiveness of the programme.

Data collection process

The team made 12 502 connected calls and completed interviews with 3362 registered ever users of tobacco who had registered for the mCessation programme. Subscribers to the mDiabetes programme (227) and respondents who did not complete the interviews (6978) were excluded from the analysis.

Figure 1 summarises the telecall details and the sample obtained.

Figure 1

Details of the telecalls to registered subscribers.

Quality assurance of data collection

A 2-day training programme was conducted for the telecalling team. The team was provided exposure to the mCessation programme, the study objectives, research ethics, tools and soft skills for communication. The second day of the training was devoted to practice and pretesting of the tools. A team of two research staff from the evaluation team closely monitored the calls by the trained investigators during the initial stage and provided feedback. Ten per cent of the interviews were checked for the veracity of the collected data by the evaluation team through independent telephonic calls with respondents.

Measures

Ever use of tobacco was measured by asking the respondents if they had ever in their life used any form of tobacco (smoked or smokeless forms).

Current use of tobacco was measured by asking respondents if they had a puff or pinch of tobacco in the past 30 days.

Recall of messages was assessed by asking if respondents could recall seeing and reading messages received through the programme.

Quit attempt (among current users) was defined as having tried to quit tobacco use on the quit day/s provided by the programme.

The quit rate was calculated as percentage of:

Embedded Image

To assess subscriber perceptions of helpfulness to quit tobacco use, the respondents were requested to rate the programme on a scale of 1 to 5, with 1 being ‘not at all helpful to quit’ and 5 being ‘very helpful to quit’, and to provide reasons for their rating. These responses were grouped according to broad themes including the information provided, motivational content, language, technological issues and additional suggestions.

Data analysis

Data entry was done in IBM SPSS V.22 and were analysed as per objectives. For further statistical analysis, only completed interviews of ever tobacco users were considered. Multivariate logistic regression was performed to draw associations of quit status with selected independent variables.

Results

A total of 12 502 telecalls were made. After excluding the subscribers to mDiabetes (227) and never users of tobacco (1935), there were 10 340 ever users of tobacco who were contacted, of whom 3362 users provided completed interviews. Therefore, the telephonic survey for estimating quit rates among registered users yielded 3362 completed interviews of those who had ever used tobacco and had subscribed to the programme 4–6 months before the survey.

Demographics

The sociodemographic characteristics of these respondents are summarised in table 1. The participants of the telephonic survey were predominantly men, under 45 years of age, had secondary-level education or above and were employed.

Table 1

Sociodemographic characteristics of tobacco ever user respondents who provided completed interviews

Current tobacco use

More than 40% of the 3362 ever tobacco users used tobacco in the last 30 days preceding the call, as presented in table 2. Of the 3362 ever users, 15% were current smokers, 29% current users of smokeless tobacco and 14% used both smoking and smokeless forms.

Table 2

Current tobacco use among respondents who provided completed interviews

Quit attempts

Among the registered subscribers of the QuitNow programme who had at least a puff or pinch of tobacco in the 30 days preceding the survey, there were many who made attempts to quit tobacco use between the period of registration and the survey. Figure 2 depicts the percentage of the subscribers who attempted to quit on the quit day provided by the mCessation programme. Sixty-six per cent of the respondents who were current tobacco users at the time of the survey had made attempts to quit tobacco use.

Figure 2

Quit attempts after registration among respondents who used tobacco in the last 30 days.

Quit rates

Overall past 30-day quit rate at 4–6 months of follow-up

The overall quit rate, estimated as the percentage of respondents stating they did not use any tobacco in the past 30 days (at 4–6 months after enrolling in the mCessation programme) from the total sample of ever users, was 19.06% (1971/10 340).

This rate assumes that all those who did not complete the interview were still using tobacco. However, on conducting the interviews, it became apparent that some subscribers had very little understanding of what they were subscribing to when they registered with the programme. For this reason, those who stated in the interview that they had never used tobacco were excluded from the denominator for this calculated rate (as well as those who were registering with the diabetes programme which had the same registration programme as the mCessation programme).

However, if we did not exclude never users of tobacco the quit rate would be 1971/12 275 or 16.06%.

Table 3 represents only those respondents with a completed interview who stated they remembered seeing some of the QuitNow programme messages (n=1345), and then whether they had used any tobacco in the past 30 days or not.

Table 3

Tobacco use among those who could recall reading messages

A multivariate logistic regression analysis was performed with 30-day quit status (of those who read the messages) as dependent variable and the sociodemographic variables and mobile phone possession characteristics as independent variables. Being male, of younger age (<45 years), higher level of education as well as possession of smartphone had a positive predictive effect on the quit rates (table 4).

Table 4

Effect of selected variables on 30-day tobacco quit status

Helpfulness of the programme to quit tobacco use

The respondents were requested to rank the programme on a scale of 1 to 5, with 1 being not at all helpful and 5 very helpful (figure 3). Seventy-seven per cent of the respondents reported that they found the programme helpful or very helpful to help quit tobacco use.

Figure 3

Feedback from respondents regarding the utility of the programme.

Opinions of the respondents about the mCessation programme

Respondents were asked to substantiate their ranking through verbatim opinions, which were then categorised under the following broad themes:

  1. Providing useful information: Many respondents expressed that the programme was useful and provided information about tobacco use-related harm, including cancer.

  2. Motivation to quit tobacco use: While some respondents reported that they felt motivated to quit after reading the messages, some felt they needed further motivation to quit and maintain the quit status.

  3. Content of the messages: The respondents were generally satisfied with the content of the messages, but some felt that that more information should be provided and language should be simple and easy to understand. Some respondents also felt that messaging should be in regional languages.

  4. Technical problems: Some respondents highlighted technical issues such as not receiving messages regularly.

  5. Other methods: Many respondents felt that the mCessation programme could be further improved with other supportive methods/techniques such as interactive voice recording, audiovisuals, counselling centres and clinical support.

Discussion

Our paper discusses the evaluation results of the largest ever national scale implementation of an mCessation programme. The 30-day point prevalence rate of 19.1% in the overall sample and a quit attempt rate of 66% in our evaluation study find resonance with existing literature. Many studies11 12 15 20–23 which used text messaging as the tool for tobacco cessation, alone or in combination with other methods, report comparable quit rates and quit attempts to this study. In one of the studies, 56% of the smokers who participated in a mobile text-based programme maintained a 30-day quit status after the quit day in a mobile text-based cessation programme and 17% remained smoke free during the programme.23

The efficacy of some text messaging programmes may be accentuated by combining with other effective strategies such as visual and audio media, clinic-based support, counselling (including proactive counselling) through telephonic helplines24 and communication campaigns.12 20 The respondents of our evaluation study also articulated the need for text messages being supported by other appropriate strategies.

The QuitNow programme of India is different from other published mCessation studies in many ways. It used the ‘missed call’ functionality and the capacity of the Indian government to send mass emails and texts promoting the programme. This may be the first published evaluation of a national mCessation programme in a low/middle-income country (LMIC). All studies to date showing the effectiveness and cost-effectiveness of mCessation have been conducted in relatively high-income countries with good tobacco control policies and mass media/education campaigns. Our evaluation study, although not a rigorous randomised controlled trial, has found very similar quit rates to other evaluations and studies in developed countries. This is an important finding and bodes well for other LMICs initiating this type of programme.

It is noteworthy to mention that the national prevalence of tobacco use among the adult Indian population has reduced by about 6% from 34.6% in 2009–2010 to 28.6% in 2016–2017.5 25 Several initiatives, including the National Tobacco Programme, have contributed to this.

Many valuable lessons can be learnt from the mCessation programme. A large proportion of registrants (1935 out of 12 502 respondents) were found to be never tobacco users and a few subscribers were not sure what they were signing up for. Their feedback during the telephone interviews facilitated in making the registration process more user friendly. The unanticipated presence of never users and subscribers who registered inadvertently made the determination of appropriate numerators and denominators for the calculation quit rate more complex than intended. Thus, while this may not be the strictest interpretation of quit rates, it was decided that these were the most useful calculations for the Indian Government on which to base decisions about enhancing features of the current programme and its expansion. This evaluation has also increased the understanding of the needs of the population and appropriate methods of implementing a national scale mHealth programme. Several new features are being added to this programme based on learnings from this evaluation. This includes modification of content algorithm, translation of the mCessation content into regional languages and introduction of integrated voice-based response system, as a considerable proportion of tobacco users in the country fell in the category of those with no or little formal schooling.5

A limitation of our study is that because of survey data limitations, we were unable to separate in the analysis ‘ever users’ from ‘recent users’. As a result, people who had quit tobacco before subscribing to the programme were counted in the quit rate analysis. This may have artificially inflated the quit rate as these successes are being attributed to the programme. In this paper, we have not analysed the quit rate in terms of the type of tobacco used before quitting. Another limitation is that the self-reported tobacco quit status was not backed by any cotinine assays. The chance of inadvertent inclusion of some false positives, as is known from earlier studies,26 cannot be ruled out. Our study sample included predominantly men employed in government or private sector. This may in part be explained by tobacco users being predominantly male (48% male vs 20% female tobacco users in GATS 1) and also because the programme was initially promoted through emails in the government system. This may somewhat limit the generalisability of the findings to the entire population.

Notwithstanding these limitations, the mCessation experience from India can provide several learnings to other countries implementing national mHealth programmes. The core of the programme rested on successful cross-sectoral and multigovernmental organisational collaboration and provided impetus to establish new programmes within the wider umbrella of mHealth—a few months after the launch of the mCessation programme, the Government of India launched the mDiabetes programme using the same technology base and cross-sector collaboration.

Conclusion

The Government of India has demonstrated the feasibility and success of an extremely large-scale mHealth intervention through the mCessation programme. The quit rates achieved are comparable to global standards. The mCessation programme appears to have motivated a large number of subscribers to attempt to quit tobacco use and for many to achieve 30-day quit status. The programme is being further improved by translating to local languages, improving the registration process, tailoring the content to further increase the motivational quotient by appropriate use of audio and visual media and converging with other clinical and counselling programmes of tobacco cessation.

The cessation effectiveness combined with low costs and the wide reach of mobile phones all warrant serious consideration of such interventions in promoting other public health interventions, especially addressing non-communicable diseases and potential risk factors in low-resource settings.

Acknowledgments

We express gratitude to Dr Sanjiv Kumar, Director of IIHMR Delhi and Dr Dipanjan Roy, Former Director of IIHMR Delhi, for their insight and guidance during the evaluation. We also express gratitude to Mr Jitendra Arora, Ministry of Health and Family Welfare, India and Dr Supten Sarbadhikari, National Health Portal, India, for their guidance, support and insights to drafting this paper. Dr Lorien Abroms and Dr Robyn Whittaker designed the evaluation protocol and gave valuable comments for finalising this paper.

References

View Abstract

Footnotes

  • Contributors PG: coordinator for the implementation of the evaluation survey; prepared the first draft of the manuscript, collated inputs from the team and finalised the manuscript. SJ: cowrote the first draft, provided inputs and made revisions in the manuscript. JK: planned the study; reviewed draft manuscripts and provided inputs for revision. VMP: reviewed draft manuscripts and provided overall guidance to the manuscript. SP: reviewed draft manuscripts. PP: co-coordinator for the implementation of the evaluation survey; provided inputs for revision of draft. PM: reviewed the manuscripts and provided inputs for revision of design and content of the paper.

  • Funding Funding for the study was provided by the World Health Organization.

  • Competing interests None delared.

  • Patient consent Consent of respondents obtained.

  • Ethics approval Institute Review Board, Ethics Committee of International Institute of Health Management Research (IIHMR) Delhi.

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

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.