The adaptation process included the following three major steps: (1) preliminary translation and adaptation design; (2) adaptation test and formative evaluation by users; and (3) linguistic and cultural adaptation and refinement. The design process of the adaptation was dynamic, taking into consideration the user experience as an essential part of the adaptation outcome. As such, the adaptation was responsive to the sociocultural context of Samoa and Samoan smokers. Researchers from the University of Auckland collaborated closely with the Samoa Ministry of Health and the National University of Samoa to provide feedback and finalize the translation to the local language. Ethical approval for this study was granted by the University of Auckland Ethics Committee and the Samoa Ministry of Health Research Committee (reference 016631).

Initially, 14 SMS text messages from the STOMP program were selected for pretesting. SMS text messages were extracted from each of the STOMP mCessation program phases, based on the Theory of Behavior Change model [39]; these included messages based on the countdown to quit, quit day, intensive phase, and relapse prevention phase [15]. The messages included a combination of motivational messages to persevere with the quit attempt, reminders of the benefits of quitting smoking, and provided information about the consequences of smoking, how to quit and stay quit. These messages were translated verbatim to Samoan. Following the translation, SMS text messages that exceeded the 160-character limit were paraphrased, but the overall content and tone were retained.

The recruitment of focus group participants was conducted in the capital of Samoa, Apia. Participants, including smokers and ex-smokers, were recruited through posters placed in high people-traffic areas. The participant inclusion criteria were as follows: being a smoker or an ex-smoker (defined as having quit in the last ≥6 months), being aged ≥16 years, having access to a mobile phone, being able to receive SMS text messages for 1 week, being able to provide written informed consent, and participating in two 1-hour group interviews (before and after receiving the SMS text messages). Although >40 people expressed interest in participating, several were excluded because of the lack of mobile phone access or because they did not arrive at the focus group venue. Others were unable to be available to receive the messages for the 1 week required for participation. Notably, 36 people directly contacted the research assistant to participate in the study and fulfilled the eligibility to participate. Participants were grouped according to the demographic homogeneity (age and gender) [40] and smoking status (Table 1). This strategy was used in recognition of traditional Samoan-speaking customs [41] and to leverage on participants shared experiences.

All 6 focus groups were moderated by a Samoan researcher (HT). Participants completed a brief demographic and tobacco use questionnaire at the beginning of each focus group. During the focus groups, the following topics were included in the discussion: beliefs about tobacco use, mobile phone use and preferences, including phone sharing, typical daily use, perceived benefits, and problems with mobile for messages to support quitting. All participants received 20 tala (approximately US $8) following their participation in the interviews.

Following the in-depth discussion on the participants’ views of tobacco use, smoking initiation, and quitting strategies awareness [42], 36 participants agreed to received SMS text messages for 1 week (2 SMS text messages per day) on their mobile phones. A sample of 14 SMS text messages was sent to participants by the research assistant on a prepaid mobile phone. This process aimed to provide qualitative information on the potential users’ attitude toward the mobile phone use and explore their views on the mCessation program within their local context.

After 1 week, participants reconvened for a follow-up group discussion to provide feedback. The follow-up interviews aimed to draw participants’ general feedback on the SMS text messages they received, with particular attention to the content, tone, comprehension, and acceptability. The follow-up interviews covered the following topics: preferences for the program, technical issues, phone and message sharing, preferences for the timing of messages, message engagement, and suggestions for improvements in the messages and the timing of delivery.

The focus groups were digitally recorded, transcribed, and subsequently translated into English for non-Samoan researchers. The results were analyzed using a general inductive thematic analysis method [43]. Two researchers (HT and EU) independently performed an initial coding, which was later categorized and organized into themes by the research team.

An initial linguistic translation process required all messages to be translated to create a full suite of messages. The translation process was iterative, drawing upon both the content of interviews and consultation with participants regarding the message content and tone. Rigorous consultations with stakeholders were undertaken to ensure the appropriateness and suitability of messages to local setting or preferences.

Messages were delivered at various transition and maintenance stages of the program—countdown to quit day, quit day, 4-week intensive phase, and relapse prevention. Table 2 presents examples of messages created for each of the 4 stages of behavioral change. The SMS text messages were available in both English and Samoan and were then developed into a single, predefined quit day for the pilot.

The themes that informed the development of the adaptation design were described accordingly: (1) initial program perception; (2) message tone and content; (3) message delivery; and (4) perception of the value of mCessation program in general. The results below reflect participants’ perceptions of the messages after receiving them for a week.

Before receiving an actual program of SMS text messages, participants were shown examples of 3 SMS text messages that they were to receive and then asked about their thoughts regarding the concept of a SMS text messaging program to help them quit smoking. Overall, participants were generally positive toward the idea of receiving supportive messages and concurred the program would offer support to Samoans who smoke.

However, a few expressed doubts about the SMS text messages’ effects, comparing them with antismoking advertisements previously viewed on television or billboards.

These participants suggested that messages should be personalized and tailored to Samoans for greater impact on users.

During the follow-up focus group discussions, many participants preferred a positive tone of the SMS text messages—encouraging messages that reflected the physiological challenges when quitting were perceived to be helpful. Practical suggestions on what to do to avoid smoking were appreciated (eg, “Get lots of rest to keep your energy up to fight the cravings. If you get tired, it’s easy to have cravings for a smoke.”), as well as those that highlight the benefits of quitting smoking to their family.

Some participants preferred SMS text messages that contained information about the negative effects of smoking—the shock effect perceived to be a necessary motivator for attracting attention and potentially nudging behavioral change. Some requested photo evidence to be included with the SMS text message. Notably, these suggestions were more likely to come from older males in groups. Furthermore, visual representations of the message were considered important, partly for validation; seeing is believing.

Some SMS text messages stood out among participants and elicited discussion. For instance, many participants disagreed with the SMS text message that encouraged going to public places to avoid smoking, saying that public places are where smokers congregate and many start smoking.

Another message aroused humor and curiosity; that cigarettes contain chemicals found in toilet cleaner. Although some questioned the validity of that specific message, others suggested that these were the most impactful—clear linkages between smoking, the chemicals, and the risk to body organs.

Participants preferred SMS text messages that did not use abbreviated slang language, with some commenting that these SMS text messages will be difficult to understand among older smokers. Furthermore, some participants suggested providing the option for smokers to choose between receiving English and Samoan SMS text messages.

The timing of message delivery was also carefully considered within the groups. When asked about the time of the day they prefer to receive the SMS text message, participants broadly agreed that sending messages at midday (after a meal, which is a trigger to smoke) and in the evening (when family is around and after dinner) were the most beneficial times.

Generally, participants suggested avoiding times when the messages would be disruptive to important events (Bingo night, church services) or rituals (during a meal). Most participants reported being satisfied with the timing of messages and the frequency. Some even suggested they would like to receive more messages to assist them to quit smoking.

After 1 week of receiving messages, many participants reported that they had reduced the number of cigarettes they smoked. A few said nothing changed, but expressed that the SMS text messages increased their intention to quit.

The majority of participants who received the messages were positive about the potential of the program in assisting them to quit smoking. Many felt that, in the absence of any alternative tobacco cessation support (eg, nicotine replacement therapy), messages were likely to be helpful in prompting thoughts about quitting.

Phone sharing in Samoa and the close connectivity with family was believed to help in spreading the message. A number of participants reported that they shared the SMS text messages with family or friends, through either actual forwarding the SMS text message or relaying the content of the SMS text messages to family and friends.

The cultural translation and adaptation of the STOMP program was a necessary first step to implementing an mCessation program in Samoa. The value of undertaking a cultural translation cannot be underestimated, with strong evidence of its marked contribution to the acceptability and effectiveness of mHealth interventions [24,44,45]. The current literature provides some guidance on how to adapt mobile phone-based interventions (for use in countries wherein it was not specifically designed), acknowledging the essential role of the end user in this process [24,46,47].

A sample of Samoan smokers and ex-smokers were involved in the process of cultural translation and adaptation of a mCessation tool designed for use in Samoa. Smokers were essential as key informants on design-related decisions such as which messages to include or revise and the timing of message and themes highlighted in the SMS text messages (eg, benefits to the family, negative effects of smoking, practical suggestions to avoid smoking). Previous qualitative analyses affirm the critical role of the social environment, familial connections, including smokers, and perceived barriers and motivations for quitting; this led to minor revisions of the original STOMP messages to include more reference to family (eg, messages about the benefits of quitting smoking to family), and changes to suggested activities and snacks (reduce craving to smoke). Participants identified when cravings to smoke were the highest and when SMS text messages were most useful in providing timely encouragement. The positive response of participants to the program, including the increased focus on quitting or reducing the consumption of tobacco products, was consistent and provided modest evidence that Samoans smokers are likely to use the mCessation program if the program is scaled up and made available as an option for smoking cessation.

One challenge during the initial translation was to retain the integrity of the message while ensuring that the messages were kept within 160 characters. As Samoan words are typically lengthier than their English equivalent, alternative words were selected.

The involvement of key stakeholders and potential implementers of the mCessation program was critical for the formative linguistic translation and subsequent refinement of the SMS text messages. Colantonio et al [36] conducted similar consultations, mostly with tobacco cessation experts, to validate their translated messages for suitability and relevance to the local setting. The process of translation required considerable time and effort, a step that could potentially be underestimated. Our consultations with experts improved the suitability of messages to the local setting, as well as the accuracy and integrity of translation and language use. The translation quality was determined according to the integrity of the translator, with the primary challenge being the identification of the definitive authority in translation [48]. In our experience, New Zealand-based Samoan translators were the least preferred over local officially appointed translators.

An important aspect of our cultural adaptation process and one of the strengths of this study is integrating both a “bottom-up” and a “top-down” approach; this process underscores the vital role of the end user of mHealth tools, in this case smokers, to provide inputs about the shape and design of the adapted program. It also recognizes the involvement of key stakeholders who can determine the local acceptability and appropriateness of the intervention. Another strength of this study is partnership with the Ministry of Health Samoa to undertake the process of adaptation and the fact that this work reflects the reality of adapting an SMS text message program for a limited-resource environment.

A local coinvestigator, fluent in Samoan, conducted the interviews. Although this adds strength to the integrity of the data collection, it might have introduced a social desirability response, owing to the tendency to offer affirmative opinions when researching within socially connected group settings.

The aim of this study reflects an appreciation that culturally adapted messages will increase engagement, a predictor of the positive impact of SMS text message programs on health [45,49]. Similarly, engagement with SMS text message interventions is recognized as important to uptake and behavioral change [27]. Behavioral change is a complex process, which despite efforts to define a predictable, universal model remains largely elusive. However, what our work has shown is that careful attention to the nuances of the content and delivery timing of SMS text messages for behavioral change is important.

The current provision under the Samoa Tobacco Control Act (2013) covers smoke-free environments, tobacco advertising, and plain packaging [50]. However, most tobacco control initiatives in Samoa have been focused primarily on regulating the sale and distribution of tobacco products and tobacco marketing, with some enforcement at the local level [51]. New Zealand, under its bilateral agreement with Samoa, offers support to Samoa to achieve its health goals and targets [52]. Within this scope, we worked with the Samoa Ministry of Health to adapt a mobile phone-based smoking cessation intervention to support Samoan smokers to quit. Lessons learned and observations made throughout the process of adapting the intervention provide value for others attempting similar efforts. Adapting mobile phone-based interventions requires considerable time and resource and keen attention to the nuances of language and meaning—as well as partnerships. Hence, this study suggests that this investment is essential for meaningful improvements to the integrity of technology-based health interventions.