Participants were recruited from a government smoking cessation clinic located in Kuala Lumpur. This clinic was chosen based on the large number of walk-in participants. It was the most active clinic in Malaysia with a regular smoking cessation program. In this study, only male participants were included due to the demographics of smokers in Malaysia where close to half adult males smoke compared to less than 2% of females (Institute for Public Health 2015). Participants were screened to ensure that they met our inclusion criteria: males, over 18 years of age, smoked 10 or more cigarettes per day, the first cigarette smoked within 30 min of waking, and were not medically dependent or intellectually impaired. The cut-off of 10 cigarettes was to ensure that smokers were moderate or heavy smokers (Clair et al. 2011) and smoking within 30 min of waking is a reliable indicator of nicotine dependence (Zwar et al. 2014).
Upon recruitment, participants completed a Demographic Form and four questionnaires in the Malay language.
The demographic form covered information on age, occupation, smoking history, current smoking habits, previous smoking cessation attempts, reasons for quitting smoking, and participation in PA.
Shiffman–Jarvik Withdrawal Scale (SJWS; Shiffman and Jarvik 1976)
The SJWS is a 25-item questionnaire that measures participants’ withdrawal symptoms and desire to smoke. The SJWS consists of five subscales: craving, psychological symptoms, physical symptoms, sedation, and appetite. The translated Malay version of this questionnaire (SJWS-M) had an overall Cronbach’s alpha, α = .66, with three of its subscales (cravings, physical, psychological symptoms) showing alpha values greater than .70 (Teo et al. 2015). It also showed good test–retest reliability (r = .76) (Teo et al. 2015).
Cessation Self-efficacy Questionnaire (CSEQ; DiClemente 1981)
The CSEQ measures participants’ beliefs in their capability to avoid smoking in various situations. Pearson item-scale correlations were reported at an average of .68 ranging from .58 to .76 (DiClemente 1981). Internal consistency was good with an α range of .87–.96 (DiClemente et al. 1991; Prapavessis et al. 2007). The translated Malay version of this questionnaire (CSEQ-M) also demonstrated good internal consistency with Cronbach’s alpha of .90 and good test–retest reliability (r = .80) over 2 weeks (Teo et al. 2015).
International Physical Activity Questionnaire (IPAQ; Craig et al. 2003)
In this study, we used the 7-item short form IPAQ. Craig et al. (2003) reported good reliability for the IPAQ with r > .7. This questionnaire was translated to Malay language by Chu and Moy (2015) and was tested for reliability and validity among the Malaysian population. The Malay version of the IPAQ (IPAQ-M) showed good reliability with intra-class correlation coefficients (ICC) of .84–.92, except for walking (ICC = .54), which was still considered moderately correlated. The IPAQ-M also showed good validity with Cohen’s coefficient, κ = .89 (Chu and Moy 2015). The IPAQ measures and records the amount of PA of participants in four intensity levels: vigorous intensity, moderate intensity, walking, and sitting. The amount of PA is based on the number of days and minutes per day of doing each level of PA in the past 7 days. Raw scores are then converted into estimated metabolic equivalents (METs). METs are used to calculate MET-minutes/week scores.
Brunel Mood Scale (BRUMS; Terry et al. 1999)
The BRUMS records participants’ moods states. The six mood states measured are: tension, depression, anger, fatigue, vigour, and confusion. The BRUMS is a valid and reliable measure with α values for all subscales above the recommended value of .70 (Lan et al. 2012; Terry et al. 1999, 2003). The BRUMS was translated to Malay language by Hashim et al. (2010) and was tested for reliability and validity among the Malaysian population. The BRUMS-M showed good factorial validity and satisfactory reliability. Alpha coefficients were reported to range from .58 to .73 (Hashim et al. 2010).
Two interviews were conducted, a post-test interview at Week 9, after the PAC intervention ended, and a follow-up interview at Week 21, after 12 weeks with no intervention support. At post-test (Week 9), participants discussed their smoking history, smoking cessation experience, smoking cessation status, opinions on smoking before and after the cessation experience, PA involvement and opinions on PA before and after the intervention. Topics of discussion also focused on thoughts and opinions regarding the relationship between PA and smoking cessation. Questions were conversational and flexible based on the information provided by participants. At follow-up (Week 21), we interviewed participants about their experiences and feelings since the intervention ended. We also asked them about progress of their smoking cessation and PA after the intervention.
The intervention comprised a combination of the standard Malaysian smoking cessation treatment and PAC to promote PA.
Standard smoking cessation treatment
Smoking cessation consultation sessions were provided by a nurse with a counselling licence approved by the Malaysian Board of Counsellors. The nurse gave the participants behavioral counselling. Nicotine Replacement Therapy (NRT) was provided as well in the form of varenicline (Champix). Participants attended the clinic after 1 week for follow-up. Subsequent visits were scheduled by the clinic nurse depending on participants’ needs. These consultation sessions were conducted as often as needed by the participants until they were successfully abstinent. We arranged for the clinic nurse to organise one visit specifically at Week 9 and one at Week 21 for the purpose of the interviews.
Physical activity consultation (PAC)
Motivational interviewing (Miller and Rollnick 2002) is the primary consultation style used in the PAC. It is often described as a way of being with people and promoting a safe and encouraging environment for therapists to help clients’ growth and change (Westra and Aviram 2013). Motivational interviewing focuses on freedom and autonomy of clients in their change process. Another core component of the PAC is matching support to individuals’ stage of change, based on the Transtheoretical Model of Change (Prochaska and DiClemente 1982). This model proposes five stages of change: pre-contemplation, contemplation, preparation, action and maintenance (Marcus et al. 1992). The PAC also uses mastery goal setting, which focuses on Bandura’s (1977) concept of self-efficacy and involves setting step-by-step, achievable goals in order to increase experiences of mastery that enhance self-efficacy. Lastly, the PAC incorporates relapse prevention (Marlatt and Gordon 1985) into the intervention. Participants are prepared for the chance of relapse and guided on effective courses of action in the event of such relapse. Participants are also advised that relapse does not mean failure, but could be a trigger for success.
PAC was conducted by a facilitator who maintained contact with participants for 21 weeks. PAC consisted of one face-to-face consultation session and two follow-up phone sessions. In the face-to-face consultation, we sought to understand participants’ level of PA, general view on PA, provide motivation and help them shape their own path for change, at the same time providing them with a sense of autonomy. We assessed participants for their current stage of change and advised on PA to suit their current fitness levels. Advice also covered realistic goal setting and relapse prevention. Phone calls at Weeks 3 and 6 functioned as follow-ups with participants on their progress, as well as providing encouragement and motivation. In Weeks 0, 3 and 6 when the facilitator communicated with the participants, the intervention included all the behavior change techniques of PAC, described in the previous paragraph. These techniques were administered as appropriate to address the presentation of progress and concerns by each participant (Additional file 1).
Recruitment was conducted from 15 April 2013 to 15 January 2014. Smokers who attended the clinic for the first time to seek smoking cessation treatment were invited to participate in the study. During the first visit, the nurse provided smoking cessation consultation and scheduled a follow-up meeting a week later. After the smoking cessation consultation during the follow-up meeting, the nurse explained the study to participants and those who were interested to participate met the first author. The first author, who is a Psychology graduate with training and experience conducting questionnaire and interview studies, then explained the study in more detail and those who agreed to participate provided written consent. Acting as facilitator, the first author administered all the questionnaires, carried out all the interviews and executed all aspects of the intervention, following training in PAC delivered by an author with considerable experience of conducting PAC interventions.
At Week 0, we briefed participants about the nature of the study and informed them that their participation in the study was voluntary and they could withdraw at any time. Participants completed the demographic form, the SJWS-M, IPAQ-M, BMS-M, and CSEQ-M. The facilitator then conducted a face-to-face PAC session that lasted between 30 and 40 min. The face-to-face intervention was conducted in a quiet room in the clinic. We advised participants to commence their PA as agreed with the facilitator. During Weeks 1 to 2 participants implemented the PA that they had agreed in the face-to-face intervention without contact with the facilitator. At Week 3 the facilitator followed up participants with a phone interview at a time convenient to the participants. The phone interview lasted between 15 and 20 min and focused on any concerns participants had about their PA, reinforcement of goals they had set for themselves and general encouragement. During Weeks 4 and 5 participants continued to perform PA. At Week 6 the facilitator contacted participants by telephone, conducting an interview similar in format to the Week 3 interview. During Weeks 7 and 8 participants continued their self-selected PA and the intervention terminated at the end of Week 8.
Post-intervention data collection
At Week 9, participants returned to the clinic for another smoking cessation consultation session. Following this consultation, the facilitator administered the same set of questionnaires (SJWS-M, IPAQ-M, BMS-M, CSEQ-M). Then she conducted a post-test interview. During Weeks 10–20 participants continued to perform their PA with no contact from the facilitator. At follow-up in Week 21, the facilitator administered the four questionnaires (SJWS-M, IPAQ-M, BMS-M, CSEQ-M) and interviewed participants about their experiences and feelings since the intervention ended. We then debriefed and thanked participants for their involvement. Figure 1 displays the flow of the study.
Data from the questionnaires were analyzed using SPSS (Version 20.0). We used QSR NVivo (Version 10.0) to analyze qualitative information from the interviews. We sought an essentialist or realist type of knowledge, which reports the experiences, meaning and reality of the participants. We performed an inductive thematic analysis to identify and report patterns of themes within the data. According to Braun and Clarke (2006), thematic analysis is not only an appropriate technique to identify patterns within data, but also efficient in helping to organize and describe a data set in rich detail by using an inductive approach. Only general patterns are reported here.