Effectiveness of pharmacologic therapy for smoking cessation in ...

clinical Report Smoking cessation

CLINICAL report

Effectiveness of pharmacologic therapy for smoking cessation in adolescent smokers: Meta-analysis of randomized controlled trials Yeol Kim, Seung-Kwon Myung, Young-Jee Jeon, Eun-Hyun Lee, Chang-Hae Park, Hong Gwan Seo, and Bong Yul Huh

S

moking is the leading cause of preventable death and is responsible for the death of approximately 5 million people every year worldwide.1 Smoking cessation is very difficult for most people because of nicotine’s addictive effects.2 Most smokers started smoking before the age of 18 years,3 and 75% of teenage smokers continue to smoke into adulthood.4 Thus, smoking prevention in adolescents is the most important and effective method for controlling adulthood smoking.3 However, more-effective programs are required to help current adolescent smokers quit smoking, because the successful quit rate for adolescent smokers is low. The Centers for Disease Control and Prevention (CDC) reported that in 2007, about 61% of adolescents who ever smoked cigarettes daily tried to quit smoking but only about 12% succeeded. 5 Most programs currently in place rely on counsel-

Purpose. The effectiveness of pharmacologic therapy for smoking cessation in adolescent smokers was evaluated. Methods. In this meta-analysis, the medical literature was searched for randomized controlled trials (RCTs) investigating the effect of pharmacologic therapy for smoking cessation in smokers age 20 years or younger. The overall effect of pharmacologic therapy was based on the longest follow-up data available in each study. The effects of pharmacologic therapy by follow-up period, type of pharmacologic therapy, and type of strategy analysis were also compared among RCTs. Secondary outcome measures were adverse events reported from each study. Results. Six RCTs involving 816 smokers age 12–20 years were included in the final analysis. No significant increase in abstinence rates was detected with pharmacologic therapy (relative risk [RR], 1.38; 95% confidence interval [CI], 0.92–2.07; I2 = 0.0%) in a fixed-effects meta-analysis. Simi-

ing interventions, such as cognitive behavioral therapy, but the overall

Yeol Kim, M.D., M.P.H., is Staff Physician, Smoking Cessation Clinic, Family Medicine Clinic, and Center for Cancer Prevention and Detection; and Seung-Kwon Myung, M.D., M.S., is Research Scientist, Cancer Epidemiology Branch, Research Institute, and Staff Physician, Smoking Cessation Clinic, Family Medicine Clinic, and Center for Cancer Prevention and Detection, National Cancer Center, Goyang, Republic of Korea. Young-Jee Jeon, M.D., is Clinical Fellow, Department of Family Medicine, Inje University, Haeundae Paik Hospital, Busan, Republic of Korea. Eun-Hyun Lee, M.D., is Clinical Physician, Health Promotion Center, Chung-Ang University Hospital, Seoul, Republic of Korea. Chang-Hae Park, M.D., is Clinical Fellow, Center for Cancer Prevention and Detection; and Hong Gwan Seo, M.D.,

larly, no significant increase in abstinence rates was found in subgroup meta-analyses of studies with both short-term (≤12 weeks) (RR, 1.23; 95% CI, 0.92–1.65) and mid-term (26 weeks) follow-up periods (RR, 1.60; 95% CI, 0.90–2.82). Although few serious adverse events were reported, there was no evidence directly linking these effects to the pharmacologic therapy used. Conclusion. A meta-analysis found that pharmacologic therapy for smoking cessation among adolescent smokers did not have a significant effect on abstinence rates at short-term and mid-term followup times of 50% was considered to indicate substantial heterogeneity. The summary relative risk (RR) was estimated with 95% confidence intervals (CIs) on the basis of the fixed-effects or randomeffects models. Publication bias was evaluated using Begg’s funnel plot and Egger’s test.18 If publication bias exists, the funnel plot is asymmetrical or the p-value is 3 were considered high quality.

Am J Health-Syst Pharm—Vol 68 Feb 1, 2011

223


Figure 3. Effect of pharmacologic therapy for smoking cessation in adolescent smokers in a meta-analysis of randomized controlled trials based on the longest follow-up data in each study, excluding a low-quality study38 (n = 6, fixed-effects model). RR = relative risk, CI = confidence interval. The CI for each study is represented by a horizontal line, and the point estimate is represented by a square. The size of the square corresponds to the weight of the study. The dotted vertical line represents the RR of the combined total, and the diamond represents the CI for the combined total. Reference 12 was divided into two individual trials because it included groups receiving nicotine patches or gums.

Ref.

RR (95% CI)

Weight (%)

11

1.17 (0.60–2.27)

37.72

12 (patch)

4.12 (0.92–18.52)

5.78

12 (gum)

1.74 (0.34–9.00)

6.72

13

1.05 (0.41–2.69)

24.55

14

1.49 (0.55–4.02)

18.95

37

2.50 (0.51–12.28)

6.29

Combined total (fixed)

1.49 (0.98–2.26)

0.1

1

100.00

10

Figure 4. Effect of pharmacologic therapy for smoking cessation in adolescent smokers in a meta-analysis of randomized controlled trials by follow-up period: short-term (A) or mid-term (B) (fixed-effects model). RR = relative risk, CI = confidence interval. The CI for each study is represented by a horizontal line, and the point estimate is represented by a square. The size of the square corresponds to the weight of the study. The dotted vertical line represents the RR of the combined total, and the diamond represents the CI for the combined total. Reference 12 was divided into two individual trials because it included groups receiving nicotine patches or gums.

A. Ref.

RR (95% CI)

Weight (%)

11

1.17 (0.60–2.27)

18.69

12 (patch)

4.12 (0.92–18.52)

2.86

12 (gum)

1.74 (0.34–9.00)

3.33

13

0.84 (0.53–1.33)

45.63

14

1.70 (0.93–3.09)

21.92

37

2.50 (0.51–12.28)

3.12

38

0.15 (0.01–2.94)

4.45

1.23 (0.92–1.65)

100.00


0.1

224

1

10


B. Ref.

RR (95% CI)

Weight (%)

12 (patch)

4.12 (0.92–18.52)

10.32

12 (gum)

1.74 (0.34–9.00)

12.01

13

1.05 (0.41–2.69)

43.84

14

1.49 (0.55–4.02)

33.84

1.60 (0.90–2.82)

100.00


0.1

1

10


Other minor adverse events in the studies analyzed were sleep problems or abnormal dreams, joint or muscle aches, lightheadedness or dizziness, headache, stomachache, nausea, constipation, dyspepsia, sweating, pruritus, anxiety, cough, and sore throat. Discussion Although nicotine replacement therapy is known to be safe for use in adolescent smokers, evidence that these medications and bupropion are effective in increasing long-term abstinence rates is lacking.39 Unlike pharmacologic therapy, psychological counseling has been associated with a twofold increase in the longterm abstinence rate when compared with the standard care or no treatment,39 and relatively higher quit rates were found in programs that involved a motivation-enhancement component, cognitive–behavioral techniques, and social influence approaches.40 The current meta-analysis also found that pharmacologic therapy had no significant effect on smoking cessation among adolescent smokers in either a meta-analysis of all included trials or subgroup metaanalyses by type of pharmacologic therapy and analysis strategy. These findings might be associated with the low statistical power (i.e., small sample size), as only six studies, involving a total of 883 participants, were included in the analysis due to the paucity of published data. In the current study, when data based on the longest follow-up period in each study were pooled, successful abstinence rates were 11.5% (95% CI, 8.4–14.6%) among 409 intervention participants and 8.4% (95% CI, 5.7–11.0%) among 407 placebo-receiving or untreated participants. Based on a sample-size calculation, at least 2920 participants (1460 for each group) are required to show a statistically significant difference between groups with a power of 80%, an a of 0.05, and the above

Table 3.

Effects of Pharmacologic Therapy for Smoking Cessation in Adolescent Smokers by Type of Therapy and Analysis Strategya Variable Type of pharmacologic therapy Nicotine replacement therapy Nicotine patch Nicotine gum Nicotine nasal spray Bupropion Analysis strategy Intent-to-treat analysis Per-protocol analysis

No. Trials

Summary RR (95% CI)

Heterogeneity, I2 (%)

5 3 1 1 2

1.47 (0.89–2.44) 1.68 (0.96–2.93) 1.74 (0.34–9.00) 0.15 (0.01–2.94) 1.24 (0.63–2.45)

20.0 27.6 NA NA 0.0

7 7

1.35 (0.90–2.03) 1.31 (0.88–1.94)

0.0 0.0

RR = relative risk, CI = confidence interval, NA = not applicable.

a

corresponding smoking abstinence rates in the intervention and placebo groups. Therefore, if further trials with a larger sample size are conducted, it might be possible to determine the effect of pharmacologic therapy on smoking cessation in adolescents. In addition to the small sample size, other explanations for the nonsignificant effect seen with pharmacologic therapy in these RCTs may include low abstinence rates and high attrition rates (i.e., low compliance rates) in adolescent smokers. Abstinence rates and overall attrition rates were 0–28.0% and 32.5–91.8%, respectively (data from longest follow-up periods). CDC reported that in 2003, the rates of failed quitting attempts in adolescent smokers were higher (58%) than those in adult smokers (43%) and that the smoking cessation rate at 1 year among smokers age 12–19 years was only about 4%.41 Also, with low abstinence rates, high attrition would contribute to the nonsignificant effect, since study participants lost to follow-up are regarded as nonquitters in an intent-to-treat analysis. Regarding the safety of pharmacologic therapy for adolescent smokers, only one of the six studies reported serious adverse events.14 There was no evidence, however, that these two events were directly related to

the pharmacologic therapy used in the trial. Although various minor adverse events were reported, both bupropion and nicotine replacement therapies appeared to be safe and well tolerated. This study had several limitations. First, the long-term effects of pharmacologic therapy could not be evaluated because there have been no published RCTs reporting long-term (i.e., >26 weeks) smoking abstinence rates among adolescent smokers. Further research is needed to investigate the long-term effects of pharmacologic therapy for smoking cessation among adolescent smokers. Another limitation was the small number of trials examined and their small sample sizes; thus, the analysis had inadequate power to differentiate the effectiveness of the interventions. Further, the RCTs examined in this meta-analysis were too heterogeneous, particularly with respect to follow-up period and intervention types. Lastly, smokingcessation rates or attrition rates might have been overestimated or underestimated in trials whose participants received several hundred dollars as compensation. Conclusion A meta-analysis found that pharmacologic therapy for smoking


225


cessation among adolescent smokers did not have a significant effect on abstinence rates at short-term and mid-term follow-up times of