## INTRODUCTION

Tobacco smoking is a leading cause of premature death. It can kill 5 million people a year world-wide and will probably kill 8 million people a year between now and 20301.

In Hong Kong, substantial effort has been made to ban cigarette advertisements, promulgate a smoke-free environment and raise taxes on cigarettes, and this has led to a decreasing smoking prevalence in recent years. However, according to the Thematic Household Survey Report No. 59 in 2016, there were 646400 current smokers at the time of the census, accounting for 10.5% of all persons aged 15 years and over in Hong Kong2. Of these current smokers, 93.5 % were daily smokers and 6.5% were non-daily smokers.

With the implementation of the Smoking Public Health Ordinance on 1 January 2007 and the increase in tobacco tax by 50% in February 2009, it is expected that more smokers would be motivated to quit smoking. In 2009, the Tung Wah Group of Hospitals, with funding from the Department of Health of Hong Kong SAR, eight integrated centres on smoking cessation were established to provide clinic-based smoking cessation services to help smokers to quit. This service is free of charge for Hong Kong citizens. The clinics are run by family physicians and social workers who have been trained in smoking cessation. Free NRT or non-NRT medications are provided when necessary3-5. In fact, according to the Stead et al.3 study, the joint effort of clinicians and counsellors has resulted in quit rates at 26 weeks and 52 weeks of 41.9% and 38.9%, respectively.

In our service, the average quit rate from 2009 to 2014 was 37%. In order to boost the quite rate, we implemented a residential treatment program for those who failed to quit smoking. A residential (inpatient) 8-day treatment program for smoking cessation conducted by Mayo Clinic showed that residential treatment for tobacco dependence was associated with significantly greater odds of 6-month abstinence compared with outpatient treatment6.

Mylena et al.7 conducted a study, Five-Day Plan (FDP), for smoking cessation using group behaviour therapy and concluded that the FDP appeared to be an effective smoking cessation therapy.

Residential treatment provides the smoker with a smoke-free and controlled environment that may not be available in a non-residential setting. Residential therapy can provide the opportunity to intensify the counseling intervention with peer experience sharing and support.

In Hong Kong, the working population is very busy and some even have to work on Saturdays, Sundays and even public holidays. It is very difficult to ask them to spare eight days in a residential program for smoking cessation. Therefore, we wished to study the effect and feasibility of a 3-day residential treatment program from Friday to Sunday, in a local setting.

## METHODS

### Recruitment

This is a retrospective case review study. From March 2012 to October 2014, smokers who had a history of failed quit attempts and who came to seek our smoking cessation services just before the residential program dates were asked whether they would like to join our program free of charge. The program was jointly run by family physicians and social workers who had been trained in smoking cessation counseling. It was held twice a year and the residential sites were holiday villages in Hong Kong. Smokers who were under 18 years old, physically handicapped, cognitively impaired or had a hospital admission in the past three months, were excluded for safety and security reasons. Initial assessment included basic sociodemographics, smoking related data, a Fagerström test for nicotine dependence (FTND)8, self-efficacy on quitting and past medical history. Informed consent was obtained. Before admission to the residential site, a family physician would screen each client to assess suitability for admission and a medical treatment plan was formulated.

### Intervention

The 3-day program involved residence overnight on Friday and Saturday, and ended on late afternoon Sunday. The participants were offered medical consultation, exercise prescription, pharmacotherapy and group counseling to enhance their motivation. They were also provided with an opportunity for social support and learning new coping skills through interactive and experiential practice. During the 3-day period, medical rounds by doctors were conducted each morning and medications for smoking cessation, which included nicotine replacement therapy (NRT) and non-NRT, would be prescribed and adjusted if necessary. Their use was determined by medical conditions and the participant’s preference.

Intensive counseling was conducted by experienced social workers who had been trained in smoking cessation, and included motivational interview, acknowledgement of barriers and coping, decisional balance, understanding conditioning and un-conditioning, exploration of emotions, performing relapse prevention activities, and teaching new skills on coping with stress and mood management. Exercise was also taught, and included Qigong: ‘Ba Dua Jin’ (Eight Section Brocade Chi Kung)9.

After the residential treatment, the participants were followed up in the same manner as other nonresidential smokers in our centre, i.e. weekly for the first 2 weeks and every 2 weeks for the rest of the treatment phase, lasting from 8–12 weeks depending on whether NRT or non-NRT was prescribed. During each follow-up visit, a Bedfont Smokerlyzer (carbon monoxide analyzer) was used to ascertain abstinence. At each visit, individual counseling medical treatment would continue until completion of the whole treatment phase, as for our nonresidential smokers. At 26 weeks, a phone follow-up was conducted to get a self-reported smoking status. Table 1 gives the program activities during the 3-day residency.

##### Table 1

Program activities of the 3-day residency

* Participants were asked to write down a summary of the learning points in each session.

### Statistical analysis

Descriptive statistics were used to analyze basic sociodemographics. Chi-squared test was used for statistical analysis among the quitters and non-quitters; and p<0.05 was considered as statistically significant.

The outcome measure was a self-reported 7-day point prevalence abstinence rate at 26 weeks, based on intention-to-treat (ITT) analysis. Participants who were not able to be followed-up or with missing responses on smoking status were considered still smoking.

## RESULTS

We were able to retain all participants during this program. All had received pharmacotherapy with 39 participants (97.5%) on NRT and 1 participant (0.5%) on Varenicline. On the average, we were able to recruit 6 participants for each session. The basic sociodemographics of the participants are shown in Table 2. A total of 40 clients volunteered to join the program, age from 26 to 65 years old with a mean of 44.1 years. There were 25 (62.5%) males and 15 (37.5%) females, and 17 (42.5%) were married. In all, 42.5% had a monthly income of HK$10000 to HK$19999 (US$1250 to US$2500), and 70% had finished Secondary school education. The average cigarette consumption per day was 17 and the mean score for the Fagerström test for nicotine dependence (FTND) was 4.6 (maximum 10). Eleven (27.5%) had a history of mental illness. In general, all participants had rated quitting to be of high importance, with a mean score of 82 out of 100, and had great perceived difficulty in quitting, with a mean score of 70.6 out of 100.

##### Table 2

Basic sociodemographics of participants and smoking related variables (N=40)

Mean ± SDn (%)
Age (years)44.1 ± 9.3
Gender
Male 25 (62.5)
Female 15 (37.5)
Marital status
Single 11 (27.5)
Married/cohabiting 20 (50)
Widow/divorced/separated 9 (22.5)
Education
Primary school or lower 1 (2.5)
F1–F3 11 (27.5)
F4–F7 23 (57.5)
Post-Secondary or Tertiary 5 (12.5)
Monthly income (HKD)
<10000 13 (32.5)
10000–19999 17 (42.5)
20000–29999 5 (12.5)
≥30000 3 (7.5)
Not disclosed 2 (5)
Year of smoking25.2 ± 9.6
Cigarettes per day17.1 ± 10.8
Fagerström score4.6 ± 2.6
Previous quit attempts2.7 ± 2.1
Cohabiting smokers 6 (15)
Perceived importance of quitting (0–100)82 ± 17.1
Perceived difficulty of quitting (0–100)70.6 ± 18.8
Perceived confidence of quitting (0–100)63.4 ± 19.5
History of mental illness* 11 (27.5)

* Both past or current psychiatric illnesses requiring regular follow-up by psychiatrist.

The quit rate at the earlier follow-up at 8 weeks with verification by Smokerlyzer was 70% (n=28) while the self-reported quit rate at 26 weeks was 57.5% (n=23). We were able to trace and contact all participants at this time point. To test each sociodemographic and smoking-related variable that could give the biggest difference in quit rate at 26 weeks, we dichotomized these variables to identify which one could yield a statistically significant quit rate. Table 3 shows that smokers older than 43 years, married or cohabiting, and smoking for more than 28 years were more likely to quit, and this was statistically significant. Having quit more than once was marginally significant.

##### Table 3

Comparison of characteristics of quitters and non-quitters at 26 weeks

Sociodemographic & smoking related variablesQuitter n (%)Non-quitter n (%)χ2Cramer’s Vp
Age (years)  5.0130.3540.02**
≤ 438 (40)12 (60)
> 4315 (75)5 (25)
Gender  1.1530.170.283
Male16 (64)9 (36)
Female7 (47)8 (53)
Marital status  5.0130.3540.025**
Married/cohabiting15 (75)5 (25)
Single/separated/ divorced/ widowed8 (40)12 (60)
Income (HKD)  3.0860.2850.08
< 2000016 (53)14 (47)
≥ 20000*7 (88)1 (12)
Years of smoking  9.5480.4890.002**
≤ 2811 (41)16 (59)
> 2812 (92)1 (8)
Cohabiting smoker  1.6870.2050.194
yes2 (33)4 (67)
no21 (62)13 (38)
Cigarettes per day  0.6280.1250.428
≤ 1912 (52)11 (48)
> 1911 (65)6 (35)
Fagerström score  3.0080.2740.083
≤ 517 (68)8 (32)
> 56 (40)9 (60)
Quit attempts  3.6720.3030.055
19 (82)2 (18)
>114 (48)15 (52)
Importance (0–100)  0.2340.0760.629
≤ 9016 (55)13 (45)
> 907 (64)4 (36)
Difficulty (0–100)  2.2830.2390.131
≤ 7515 (68)7 (32)
> 758 (44)10 (56)
Confidence (0–100)  0.0510.0360.822
≤ 5010 (56)8 (44)
> 5013 (59)9 (41)
History of mental illness  2.7740.2630.096
yes4 (36)7 (64)
no19 (66)10 (34)

** p<0.05 is considered as statistically significant.

* Two missing data (not disclosed).

## DISCUSSION

This report describes the high rate of smoking abstinence from a 3-day residential smoking cessation start-up program. The reasons for the apparent efficacy of this residential treatment are many. One important element of our program is the combined use of group behavioral treatment and pharmacological therapy, of which the efficacy has been well established10,11. In addition, our participants were able to practice the skills, and not just learning the skills, in relapse prevention12. The skill training was done by experienced social workers who had been trained in behavioral counselling; this differs from many other residential programs that are conducted by health professionals. We also incorporated relaxation exercise and a Chinese aerobic stretching exercise (The Eight Section Brocade Chi Kung)13-15 . The outcomes in the residential group per se could also be influenced by the greater expense in terms of time commitment required by the participants. The time-cost of residential treatment undoubtedly results in the selection of more motivated smokers. However, we did have some recruitment difficulty in this program but we are not able to give a number, as many smokers felt that they did not have the time to participate.

A few studies on residential programs have demonstrated their effectiveness. However, these programs vary from 4 to 14 days, with different recruitment methods and sampling subjects6,7,16-18. A 4-day residential program for tobacco dependence by Green et al.17 reported a 6-month abstinence rate of about 26%. Results from the Mayo Clinic 8-day residential smoking cessation program indicate higher self-reported 7-day point prevalence smoking abstinence for the residential (45%) versus outpatient (26%) groups at 6 months19. In contrast to the Mayo Clinic, which is a referral centre, our service is a primary care setting and the smokers are entirely self-referred through our quit-smoking hotline. However, direct comparison between these residential programs cannot be made because of the different characteristics of participants and interventions.

The present study is the first short residential progam for smoking cessation in Hong Kong. Since this is a pilot study, there are several limitations. The main limitations are the small sample size and that the treatment was not randomized. The medication was self-selected after discussion of options with physicians. The participants had previous failed quit attempts and hence could have had past experience with smoking cessation therapy, which might have contributed to a positive response to treatment. The observational nature of the current study also limits our ability to make causal inferences about the higher smoking abstinence rates observed or its generalizability to other smoking populations. Whether residential treatment is truly superior to other methods of treatment can only be determined with a high degree of certainty from a randomized controlled trial. We do not provide biochemical confirmation of 6-month smoking abstinence and yet we are able to yield an abstinence rate of 70% after 8 weeks, with verification by Smokerlyser. Although self-reported abstinence without biochemical confirmation may lead to overestimates of abstinence rates, a previous study has indicated that self-reported abstinence did not differ appreciably from that with biochemical verification20.

As to the comparison of quitters and non-quitters at 26 weeks, there are some commonalities with a study in Hong Kong that showed that older age, being married/cohabiting and years of smoking were independent predictors of quitting21. However, unlike this previous study, our study does not indicate daily cigarette consumption, Fagerström test score for nicotine dependence or having mental illness as independent predictors. This may be due to the fact that this is an intragroup comparison and the different characteristics of the participants are dichotomized to give the biggest difference in quit rate at the 26 weeks. Besides, the subjects are only those who joined the program and the sample size is too small.

## CONCLUSIONS

This pilot study is an observational study. It involves clients who have failed previous quit attempts and are motivated to quit smoking. The quit rate is higher compared with our outpatient setting. It is comparable to or even surpasses some overseas studies with longer duration of residence. It appears that a 3-day residential program from Friday to Sunday as a startup in a local setting is feasible and can save the cost of longer residential programs. It provides an effective alternative to help those addicted smokers who want to seek help to quit smoking.