Factors Associated with Smoking Cessation Among Quit Smart™ Participants
Karen M Polizzi, MPH; Douglas W Roblin, PhD; Adrienne D Mims, MD, MPH; Dianne Harris, BS, CHES; Donna D Tolsma, MPH
Perm J 2004 Spring; 8(2):28-33
Objectives: To evaluate social and program factors associated with the one-year smoking cessation rate among participants of a smoking cessation program at a managed care organization (MCO).
Methods: As implemented at this MCO, the Quit Smart program incorporated group sessions taught by health educators, discount vouchers for nicotine replacement patches, self-help manuals, and a relaxation audiotape. A survey of 97 patients who participated in the program during 1999 or 2000 or both was administered one year after these participants completed the program.
Results: Of the 97 participants, 58 responded to the survey. Nineteen (33%) reported not smoking at one year after completing the program; and 11 (19%) reported that they were smoking-abstinent for 12 months after completing the program. Compared with patients who did not use the nicotine patch, respondents who used the nicotine patch were significantly more likely (OR = 4.42 [1.12, 17.35]) to report not smoking at 12 months after completing the program and to be smoking-abstinent for 12 months after completing the program (OR = 8.31 [1.15-60.22]). Respondents who were exposed to smoking in two or three settings (ie, at home, with friends, at work) were significantly less likely to report smoking cessation at 12 months (OR = 0.12 [0.02, 0.70]) and to have abstained from smoking for 12 months (OR = 0.04 [0.01, 0.42]) than were respondents who were not exposed to smoking in these settings.
Conclusions: The Quit Smart program achieved 12-month smoking cessation and abstinence rates comparable with those achieved by other multifactorial programs to promote smoking cessation. Subsidized therapy using the nicotine patch was effective for promoting smoking cessation. However, program success was inhibited by exposure to smoking in domestic and social situations.
Population-based studies of smoking cessation programs indicate that, although initial quit rates are high, quit rates decline to approximately 15%-25% at one year.1,2 Community- and workplace-based interventions generally report quit rates of a similar magnitude,3,4 and some report rates as high as 36%.5 For comparison, the background rate of unassisted smoking cessation is estimated at approximately 7%-8%.2,6 Physician interventions that use nicotine gum as an aid to smoking cessation produce one-year quit rates of about 10%.7,8 Within managed care organizations, one-year smoking cessation rates as high as 30%-40% have been reported.9-11 One of several interventions used in multifactorial health education programs to promote smoking cessation, nicotine replacement therapy is efficacious for promoting and sustaining smoking cessation7 and is also a cost-effective method of treatment.9,12 Controlled studies have shown that quit rates for users of the nicotine patch are approximately double the quit rates for users of placebo.13,14
The Quit Smart smoking cessation program, developed by Robert H Shipley, PhD (founding director of the Duke Medical Center Stop Smoking Clinic), is a multifactorial health education program designed to promote smoking cessation among tobacco users. Quit Smart was implemented in 1998 as part of the health education program of the Kaiser Permanente Georgia Region (KPG). A pilot evaluation of the Quit Smart program was conducted for KPG's quality improvement initiatives and addressed three questions:
This article presents results of the pilot evaluation.
Intervention Used in the Quit Smart Program
Instrument items and scales were developed through an iterative process. We initially reviewed the smoking cessation literature for sample items and for factors associated with promoting or inhibiting smoking cessation. The survey instrument included items about the following topics:
The survey instrument was designed to be completed within 10-15 minutes. A draft instrument was administered to a small convenience sample of colleagues (smokers and former smokers) for assessing flow and clarity of the instrument. The final survey instrument included revisions suggested by the preliminary survey results. The final survey instrument and the protocol for its administration were reviewed, approved, and monitored by the KPG Institutional Review Board.
For the Fall 1998 group, the survey was administered during December 1999; for the Spring 1999 group, the survey was administered during May 2000. Approximately two weeks before receiving the initial telephone call, each potential respondent was mailed a letter containing information about the survey. As many as five attempts were made to contact each potential respondent. A total of 58 participants completed most of the survey (response rate of 60%).
The study had three principal independent variables: use of aids to quit smoking, cumulative number of settings with smoking exposure, and level of physical activity. Use of aids to quit smoking was assessed among all respondents by asking, "What techniques did you use to quit smoking?" Responses included: "Cold turkey, will power" and "Nicotine patch." Both variables were coded as binary (1 = used the technique). Smoking exposure at home, among friends, and at work was ascertained. Exposure at home was measured by asking if the respondent lived in a house with others and whether or not any of these persons smoked. Exposure among friends was assessed by asking how many of the respondent's five closest friends smoked. Exposure at work was ascertained by asking if the respondent was employed and whether or not any of the respondent's five closest colleagues smoked. Each of these three variables was coded as binary (1 = exposed). A cumulative measure of smoking exposure was also computed as the sum of the settings with exposure (0, 1, 2, or 3). Level of physical activity was ascertained from a 5-level response ("Rarely or not at all" through "Every day") to the question "How often do you exercise?" We recoded this item into a binary variable of "Every day" versus "Less than every day."
Patient demographic and socioeconomic measures included age (below median age 48 years vs at or above median age); gender; race/ethnicity (white or African American); level of education; and household income.
Association of the independent variables with respondent status as a 12-month quitter or with respondent status as a 12-month abstainer or not was evaluated by using a c2 test of significance (a = 0.05). Because the sample size was small, we considered any association with an a-level of 0.15 to be marginally significant.
Logistic regression for each of the two dependent variables was estimated to assess competing effects of factors that help smoking cessation and factors that inhibit smoking cessation.
Analyses were performed using SAS (Statistical Analysis Software) Version 6.12 (SAS Institute, Cary NC).
Personal choice was indicated by 71% of respondents as the principal reason for enrolling in the Quit Smart program (data not shown in tables). Physician recommendation to enroll was the principal reason given by 21% of respondents. Only 7% indicated that availability of the nicotine patch was their principal reason for enrollment. Neither the 12-month quit rate nor the 12-month abstinence rate was significantly associated with respondents' reasons for enrolling in the Quit Smart program.
Of the 39 respondents who indicated that they were smoking at 12 months after last attending the Quit Smart program, 67% indicated that they had quit smoking for a limited time after completing the Quit Smart program (data not shown in tables). At the time of survey, current smokers were, on average, smoking 13 cigarettes (half a pack) per day. Mean duration of abstaining from smoking was 2.6 months.
Factors Promoting or Inhibiting Smoking Cessation
The 12-month abstinence rate was significantly associated with exercise frequency, number of settings exposed to smoking, and exposure to smoking among friends (p < 0.950 (Table 2). The 12-month abstinence rate among respondents who reported exercising daily (38.5%) was higher than the abstinence rate among respondents who exercised less frequently (13.3%) (p = 0.04). For respondents who were exposed to smoking in at least two settings, the 12-month abstinence rate (9.1%) was lower than the abstinence rate for respondents who were not exposed to smoking in any setting (40.0%) (p = 0.03). The 12-month abstinence rate was most adversely associated with exposure to smoking among friends (6.5%) of any setting in which respondents were exposed to smoking (33.3%) (p = 0.01).
For exercise frequency, use of the nicotine patch, and settings in which respondents were exposed to smoking, we obtained adjusted odds ratios for 12-month smoking cessation status (Table 3). Compared with respondents who did not use the nicotine patch, respondents who used the nicotine patch were significantly more likely (OR = 4.42 [1.12, 17.35]) to report not smoking at 12 months and to abstain from smoking for 12 months (OR = 8.31 [1.15-60.22]). Compared with respondents who were not exposed to smoking at home, among friends, or at work, respondents who were exposed to smoking in two or three settings were significantly less likely to report smoking cessation at 12 months (OR = 0.12 [0.02, 0.70]). Similarly, respondents who were exposed to smoking in either one, two, or three settings were significantly less likely (OR = 0.09 [0.01, 0.42] and 0.04 [0.01, 0.42], respectively) to abstain from smoking for 12 months than were participants who were not exposed to smoking in these three settings.
As implemented at KPG, the Quit Smart program yielded a 12-month quit rate of 33% and a 12-month abstinence rate of 19%. These rates resemble those achieved in other multifactorial health education programs promoting smoking cessation at other MCOs. Use of the nicotine patch promoted both smoking cessation and smoking abstinence at 12 months, whereas continued exposure to smoking--whether at home, among friends, or at work--inhibited both smoking cessation and smoking abstinence at 12 months. This importance of the nicotine patch (and other forms of nicotine replacement) for facilitating smoking cessation is consistent with results reported for clinical trials as well as for other observational studies of smoking cessation techniques.7,9,12-15 Other studies have affirmed the association between exposure to smoking and temptation to smoke, failure to quit smoking, and smoking relapse among former smokers.16-22
The main strength of the Quit Smart smoking cessation program is its combination of proven methods for aiding smoking cessation. Comments solicited from survey respondents indicated that the program was well received by those who attended it. Even respondents who continued to smoke indicated that they were very satisfied with the program overall.
That smoking cessation programs are cost-effective--both in general and with regard to specific strategies--is widely accepted.9,12,23-25 The cost of an entire smoking cessation program may be justified even if only a low percentage of program participants achieve abstinence.24 Of KPG participants in the Quit Smart program, 19% abstained from tobacco use for 12 months after completing the program. Although we did not calculate a final cost-benefit analysis, the quit rate as calculated would suggest that the Quit Smart program is a success from a cost-benefit standpoint as well as from a health education standpoint.
Although encouraging, the results of our evaluation of the Quit Smart program should be interpreted as preliminary. Although the response rate to the survey was relatively high (60%), the number of respondents was small. This small sample size limited power to detect statistically significant differences (for p < 0.05) in factors promoting or inhibiting smoking cessation and resulted in wide confidence intervals even when a difference was significant (p < 0.05). Moreover, the study sample included only KPG members who completed the Quit Smart program and remained KPG members at 12 months after completing the program. If smoking cessation or abstinence rates differ between survey respondents and nonrespondents, between study participants who remained KP members and study participants who disenrolled from KPG, or between participants who completed the Quit Smart program and those who did not, then our current estimates of the Quit Smart program could overestimate or underestimate the true intervention effects. In addition, we used patient-reported measures for estimating 12-month quit and abstinence rates. Although self-reported measures are generally consistent with biochemical measures of smoking status, self-reported measures may tend to overstate the socially desirable response (ie, smoking cessation).26-30 Because this study was conducted as part of a quality improvement initiative, we did not include a control group (eg, patients randomly assigned at entry to the Quit Smart program or no intervention).
In summary, the Quit Smart program was easily incorporated into the prevention and health promotion objectives of the Kaiser Permanente Georgia Region. Of program participants responding to a survey at 12 months after completing the program, 33% had quit smoking; and 19% reported that they had abstained from smoking for the entire 12 months. Use of the nicotine patch significantly promoted smoking cessation, whereas exposure to smokers in multiple settings significantly inhibited smoking cessation.
1. Bains N, Pickett W, Hoey J. The use and impact of incentives in population-based smoking cessation programs: a review. Am J Health Promot 1998 May-Jun;12(5):307-20.