Subst Use Misuse Downloaded from informahealthcare.com by Freie Universitaet Berlin on 11/28/14 For personal use only.
The International Journal of the Addictions, 26(1 I), 1223-1231, 1991
The Effect of Nicotine, Silver Acetate, and Placebo Chewing Gum on the Cessation of Smoking. The Influence of Smoking Type and Nicotine Dependence E. J. Jensen,* MD E. Schmidt, MD 6. Pedersen, MD
R. Dahl, MD Department of Respiratory Medicine The University Hospital Noerrebrogade, DK-8000 Aarhus C, Denmark
Abstract The effects of nicotine (NI), silver acetate (SA), and ordinary (PL) chewing gum on the cessation of smoking were compared through ratings on six smoking types and physiological nicotine dependence (F-score) in a placebo-controlled and randomized study that included 496 smokers in a group-support setting. ResuIts were evaluated after 26 weeks.Ratings on and 6 and F-score were related to success rates. smoking types 1,3,4,5, The effect of NI was superior to SA in smokers with high type 1 ratings, the effects of NI and SA were almost equal and superior to PL in smokers with high type 4 and 6 ratings, and the effect was independent of the Fscore. Treatment of smokers with NI and SA should be restricted to smok*To whom correspondenceshould be addressed. 1223
Copyright 0 1991 by Marcel Dekker, Inc.
JENSEN ET AL.
1224
ing types in which an effect superior to PL is documented, and research should be performed to find support for smoking types where SA and NI have a clear effect.
Subst Use Misuse Downloaded from informahealthcare.com by Freie Universitaet Berlin on 11/28/14 For personal use only.
INTRODUCTION Smoking cessation programs employing nicotine chewing gum (NI) have varied in outcome after 1 year, and in some studies the effect was dependent on nicotine dependence (Table 1). Silver acetate chewing gum (SA), which gives tobacco smoke a bad taste, has been shown to have a short-term antismoking effect (Malcolm, 1986; Schmidt, 1977). The purpose of the present study was to compare the effects of NI, SA, and ordinary chewing gum (PL) by considering their scores on six smoking types (Ikard et al., 1969) and on the Fagerstrom nicotine tolerance questionnaire (Fagerstrom, 1978).
SUBJECTS AND METHODS Smokers whose daily cigarette consumption had been more than 10 for more than 10 years were included in the study, which was performed in 1986/87. Smokers with drug or alcohol misuse, psychiatric problems, cardiovascular diseases, and pregnant women were excluded. Participants were randomized into 24 main groups, and each group was randomly allocated to treatment with one of the chewing gums. Ten groups (N= 203), 10 groups (N = 21 l), and 4 groups (N = 82) were treated with SA, NI, and PL, respectively. The study could not be blind because the chewing gums had different modes of action and because consumption of SA had to be restricted to a daily maximum of six pieces to avoid the risk of argyria. The participants attended eight sessions of 2 h each at the clinic. They were controlled at each meeting for tobacco abstinence with an Ecolyzer CO-monitor (Energetics Science, Hawthorne, New York) which has been shown to be a reliable method of tobacco abstinence control (Jarvis et al., 1980). Participants presenting a CO concentration greater than 4 pprn were excluded. Reentrance was not allowed. The chewing gums could be used for 3 months. Because relapse rates were expected to be greatest during the first 6 weeks and to stabilize after 26 weeks (Hunt and Bespalec, 1974), sessions were placed at 1, 2, 3, 4, 5, 6, 12, and 26 weeks after the day of quitting. The results evaluated after 26 weeks. Before smoking cessation took place, all participants were guided in filling out a questionnaire concerning age, sex, smoking habits, drug and alcohol intake, living arrangements, and job status, and were then rated into six smoking types. The types included smoking for stimulation (type l), smoking due to a need for hand-
352
Fee and Steward, ScoBand, 1982
Hjallmarson, Sweden, 1984
3.
4.
173
118
9. Tonnesen et at., Denmark, 1988
10. Raw et al., England, 1980
6 + 3 follow-up
6 + 2 follow-up
9 + 3 follow-up
6 + 3 follow-up
8
14
6 during 6 weeks
10 during 5 weeks
Success rates of NI and PL treatments were 63 and 45 % ! @ < .05). NI only superior to PL in participants with F-score > 7. NI recommended for all Success rates of NI and PL treatments were 25 and 9%QJ < .05). Effect of NI was superior to PL only if F-score > 7 After 5 weeks, success rates of NI and PL treatments were 45.6 and 33.1% @ < 0.05); after 1 year, 13 and 9% (N.S.). Substitution time longer than 5 weeks was recommended Success rates of NI and PL treatments were 29 and 13%@ c .05). Daily cigarette consumption was without influence on effect of NI and success rates Success rates of NI and PL treatment were 44 and 21% @ < 01). Effect of NI was superior to PL only if F-scores belonged to the medium interval. No influence of daily cigarette consumption. NI recommended for all smokers In participants with F-score > 6, success rates of NI and PL treatment were 41 and 8% (N..S.), and in participants with F-score I7 success rates were 0 and 302, respectively (N.S.). No conclusion Success rates of NI and PL treatments were 31 and 14%@ < .05). Low p-nicotine was associated with high success rates. Sex, social class, and number of daily cigarettes had influence. NI recommended for well-motivated smokers Treatment with NI not superior to PL. Success rates were 30 and 12%in the NI and PL groups, respectively Effect of NI 2 and 4 mg, and NI 2 mg and PL were compared. Success rates of NI 2 mg and PL in participants with HR scores 2 119 were 44 and 12%@ c .05) and in participants with HR score I119 were 38 and 23%(N.S.).NI 2 mg recommended for light and medium nicotine-depend ent smokers Success rate of NI and PL treatments were 38 and 14%(p < .01). NI recommended for all smokers
Selected variables, results, and implications
nAll studies were double-blind, placebo controlled, and randomized except for study 2, which was not blind, and study 10, which was not blinded or randomized. Studies 2 and 4 used a 2xZ factorial design to evaluate the effect of behavioral treatment and long- vs short-term follow-up in addition to the effect of nicotine chewing gum. In all studies, results from 1 year follow-upare given, except for study 1 in which ir was 6 months. Categories of smokers included in the studies were rarely given. Two-tailed statistical tests were employed except in studies 1 and 7. F-score is the Fagerstrom nicotine-tolerance score, HR score is the Horn-Russell nicotine dependence score (Russell et al., 1974), NI is nicotine chewing gum, and PL is placebo.
60
Schneider et al., USA, 1983
8.
48 1 I6
Murray and Schneider, USA, 1984
139
7. Jarvis et al., England, 1982
6.
5. Hall et al., USA, 1987
151
Fagerstrom, Sweden, 1984
2.
206
8
100
Fagerstrom, Sweden, 1982
1.
2 + 1 call or I S
Sessions (S)
N
Author, country, and year
Smoking Cessation Programs with Nicotine Chewing Gum: Selected Parameters and Studiesa
Table 1
Subst Use Misuse Downloaded from informahealthcare.com by Freie Universitaet Berlin on 11/28/14 For personal use only.
uI
-
3
6
0
2
$
z
55
0
2!
JENSEN ET AL.
1226
Table 2
Subst Use Misuse Downloaded from informahealthcare.com by Freie Universitaet Berlin on 11/28/14 For personal use only.
Demographic Derailsa SA
NI
PL
Overall
N
203
211
82
496
Age ( y e a )
41.8 f 11.8
42.7 f 12.4
41.4 f 12.4
42.1 f 12.2
Males/females
941115
861116
39/41
2191272
+/- living with partner (percent)
78.4121.6
74.3125.7
78.5121.5
77.5122.5
+I- employment (percent)
87.2112.8
82.5117.5
85.0/15.0
84.81152
+/- earlier attempts to quit
83.9116.1
89.6110.4
82.7117.3
86.1113.9
Motivation score
14.6 f 2.2
15.2 f 2.3
15.0 f 2.3
14.8 2 2.3
Initial age of smoking Wears)
14.5 2 2.4
15.2 f 2.3
14.7 f 2.5
14.8 f 2.4
Cigarette consumption (numberlday)
21.7 f 9.9
21.8 f 8.2
21.0 f 9.7
21.7 f 9.1
(percent)
Pack years consumption
25.3 f 17.8
25.8 f 14.3
23.1 f 15.7
24.9 f 16.0
Smoking duration (years)
22.8 f 8.8
23.1 f 8.8
21.6 f 8.3
22.5 f 8.5
F-score
6.3 f 1.9
6.3 f 2.2
6.2 f 2.0
6.3 f 2.0
Type 1
8.4 f 2.3
8.3 f 2.6
8.1 f 2.4
8.3 2 2.5
Type 2
7.0 2 2.7
6.8 f 2.5
6.5 2 2.4
6.8 2 2.6
Type 3
12.4 f 2.2
12.3 f 2.2
12.0 f 2.0
12.3 2 2.2
Type 4
10.5 k 2.7
10.8 f 2.9
10.3 f 3.1
10.6 f 3.0
Type 5
10.6 f 2.7
10.8 f 2.5
10.8 2 2.7
10.8 f 2.6
TYPC: 6
6.8 f 2.9
6.8 2 2.6
6.2 2 2.6
6.7 f 2.8
oSA, NI, and PL are the nicotine, silver acetate, and placebo groups respectively. Means f SD are given if not otherwise indicated. All participants were Caucasians. None had medication for psychiatric disorders or used stimulating drugs. Motivation score was calculated as the sum of scores on a 6-item questionnaire with four gradings on the importance for quitting because of health concerns.
economics, looks, irritation of having a bad habit, wish to have cleaner surroundings, and consideration for others. F-score is the score on the Fagerstrom nicotine-tolerance questionnaire which is a 9-item questionnaire that evaluated physiological nicotine dependence through scores on time from awakening to first cigarette, number of cigarettes per day, nicotine content of the brand usually smoked, difficultieswith abstaining from smoking in public places, cigarette of the day most difficult to spare, smoking if sick, inhalationof the smoke, and time of day with mrst intensive smoking. Maximum score is 11, which indicates high depcndence. Description of the smoking types is given in the text. All demographic details were distributed equally between the treatment groups.
Subst Use Misuse Downloaded from informahealthcare.com by Freie Universitaet Berlin on 11/28/14 For personal use only.
CESSATION OF SMOKING
1227
ling things (type 2), smoking for pleasurable relaxation (type 3), smoking to reduce negative feelings (type 4), smoking due to psychological addiction or craving (type 5), and smoking due to habit (type 6) (Ikard et al., 1969). Each smoking type was related to three questions, each rated 1 to 5.The sum (3-15 points) was the rating for each type. The degree of physiological nicotine dependence (F-score) was assessed through the questionnaire of Fagerstrom (Fagerstrom, 1978). The RATE groups of a smoking type and F-score were defined as participants with, respectively, low and high ratings partitioned by the rating value that optimized the chi-square value calculated from the outcome of participants with rating values lower, equal to, or greater than the rating value used as the cutpoint. The analyses were carried out in the total population and controlled for treatment by using the test of Mantel-Haenszel (Mantel and Haenszel, 1953). In each RATE group the success rates of treatments were compared pairwise. The smokers gave their informed consent to participate and were given 1 to 2 weeks to consider their motivation to quit smoking. The study was approved by the ethical committee of Aarhus County. Statistics
The BMDP statistical program package was used in all calculations (BMDP Srutisticul Software, 1985).To evaluate differences in the success rates, Pearson’s chi-square test was employed. Uniformity in demographic variables between treatment groups was evaluated with Pearson’s chi-square and Student’s t-tests. In all tests a significance limit of 0.05% was used and two-tailed tests were employed.
RESULTS Demographic details are presented in Table 2. The overall success rate after 26 weeks were 42.6, 38.9, and 34.2% in the NI, SA, and PL groups, respectively (N.S.). Correlations between most smoking types and between smoking types and F-score were significant but small (Table 3). High ratings on smoking types 1 (p < .05),4 (p < .Ol), 5 (p < .05),6 @ < .03), and F-score (p < .05)and low ratings on type 3 (p < 0.01) were associated with low success rates (Table 4). In smokers with type 1 ratings greater than 10, the effect of NI was superior to SA (p < .01) and a 15% better outcome of NI treatment compared to PL did not reach significance. The effect of NI (p < .05)and SA @ < .02) was superior to PL in smokers with type 4 ratings greater than 10 and in smokers with type 6 ratings greater than 5 NI (p < .03) and SA (p < .07) tended to be superior to PL (Table 4). NI tended to have a better effect than PL in participants with an F-score greater than 4 (.05 < p < . l ) (Table 4).The effect of NI and SA was independent of the F-score in participants in RATE groups with high type 4 and 6 ratings (data not presented).
JENSEN ET AL.
1228
Table 3
Subst Use Misuse Downloaded from informahealthcare.com by Freie Universitaet Berlin on 11/28/14 For personal use only.
Correlation between Smoking Types and F-scores
Type 2
.19*
TYP 3
.12*
.29**
Type 4
.44**
.16**
.12*
TYP 5
.33**
.15**
.12*
.39**
TYPC 6
.17**
.18**
.05
.23**
.16**
F-score
.14*
.06
.0006
.15**
.33**
.29**
OCorrelation coefficients between ratings on the smoking types and between ratings on the smoking types and ratings on F-score. *p < 0.01. **p < 0.001.
Study limitations were lack of blindness, a too small placebo group considering some RATE groups, intercorrelations between ratings on smoking types, and probably an inaccurate F-score. The lack of blindness probably did not influence the outcome because the overall success rates were equal. Independency between blindness and success rates was found in a study with NI (Hughes and Krahn, 1985). The small intercorrelations between ratings for smoking types 1-6 were confirmed (Ikard et al., 1969; Tobacco Research Council, 1973), and in these studies a high reproduceability was shown. The reliability of the nicotine-tolerance questionnaire has recently been questioned (Lombard0 et al., 1988), and the implications are unclear. Smoking types with negligible intercorrelations and a reliable method of assessing physiological nicotine dependence might improve the classification of smokers. Differences in success rates of 15% between active and inactive treatment in some RATE groups did not reach significance. This may have been caused by the relatively few PL treated in some RATE groups. Sessions in this study were led by two physicians and a nurse. More staff members, including a psychologist, and smaller support-groups would have improved the possibility of individualizing support and increased follow-up. Our results indicate that ratings on most of the smoking types were predictors for success of smoking cessation and of value in identifying smokers who benefited significantly from active treatment. The equality of the effects of NI and SA
CESSATION OF SMOKING
1229 Table 4
Overall and Treatment-Specific Success Rates of RATE Groups
Subst Use Misuse Downloaded from informahealthcare.com by Freie Universitaet Berlin on 11/28/14 For personal use only.
Smoking type 5pe 1
Abstainers Treatment NI SA
PL NI SA
PL 5Pe3
NI SA
PL Type 4
NI SA
PL Type 5
NI SA
PL NI SA
PL
RATE groups
N
RATE S 10 RATE > 10 RATE S 10 RATE > 10 RATE S 10 RATE > 10
171 38 165 36 71
(96)
Pl
P2
N.S.
10
41.5 44.7 44.2 13.9 36.6 30.0
RATE S 7 RATE > 7 RATE 7 RATE > 7 RATE S 7 RATE > 7
127 82 131 70 51 30
39.2 36.6 38.9 38.6 39.2 30.0
N.S.
R A T E S 10 RATE > 10 RATE S 10 RATE > 10 RATE 5 10 RATE > 10
33 176 44 157 19 62
24.2 45.5 25.0 42.7 36.8 35.5
0.01
R A T E S 10 RATE > 10 RATE 5 10 RATE > 10 RATE I 10 RATE > 10
90 119 77 124 37 44
47.8 37.8 42.9 36.3 41.3 22.7
N.S.
RATE 5 10 RATE > 10 RATE I10 RATE > 10 RATE S 10 RATE > 10
97 112 76 125 32 49
46.4 38.4 44.7 35.9 40.6 32.6
N.S.
RATE S 5 RATE > 5 RATE S 5 RATE > 5 RATE S 5 RATE > 5
74 135 72 129 35 46
43.2 41.5 41.7 37.4 51.2 24.9
0.05
0.001 N.S. N.S.
N.S. N.S.
0.01
0.03 N.S. 0.01
N.S. 0.01
0.05
N.S. N.S. N.S.
0.03
N.S. 0.02 ~~
~
(continued)
JENSEN ET AL.
1230 Table 4 (continued) Smoking type
Subst Use Misuse Downloaded from informahealthcare.com by Freie Universitaet Berlin on 11/28/14 For personal use only.
F-score
Abstainers Treatment N1 SA
PL
RATE groups RATE 5 4 RATE > 4 RATE 5 4 RATE > 4 RATE 5 4 RATE > 4
N 33 162 35 157 17 62
(I) 51.5 43.2 48.6 37.6 58.8 30.6
Pl
P2
N.S.
0.03
N.S. 0.03
OAbstainer rates for each treatment within RATE groups of each smoking type 1-6 and F-score. Results are given as percentages. The RATE groups are designated by the rating value used as the cutpoint. The significanceof the difference in outcome between rate groups was tested for each treatment group (pl) and for the tohl population controlled for treatment (p2).
in the smoking types, except type 1, and their independence of the F-score did not indicate any advantage in employing smoking deterrents with different modes of action. It is suggested that pharmacologicalsupport with NI and SA is restricted to smoking types with a documented effect superior to placebo, and research efforts should be made to generate support for smoking types where NI and SA have clear value. Intensive psychological support may be of importance, and individualization of smoking cessation programs may make it easier and more effective to quit. Consideration of smoking types might be a basis for improved support. Studies to improve classification of smokers should be performed and evaluated for further individualization of treatment.
REFERENCES BMDP STATISTICAL SOFTWARE (1985).Los Angeles: University of California Press. FAGERSTROM. K-0.(1978). Measuring degree of physiological dependence to tobacco smoking with reference to individualization of treatment. Addict. Behav. 3: 235-241. FAGERSTROM, K-0.(1982). A cornparkon of psychological and pharmacological treatment in smoking cessation J. Behuv. Med. 5: 343-351. FAGERSTROM, K-0. (1984).Effects of nicotine chewing gum and follow-up appointments in physician-based smoking cessation Prev. Med. 13: 517-527. FEE, W. M.,and STEWARD, M.J. (1982). A controlled trial of nicotine chewing gum in a smoking withdrawal clinic. Practitioner 226 148-151. HALL, S. M., TUNSTALL, C. D., GINSBERG, D., BENOWITZ, N. L.. and JONES, R. T. (1987). Nicotine gum and behavioral treatment: A placebo-controlled trial. J. Consufr. Clin. Psychol. 55: 603-605. HJALMERSON, A. I. M. (1984).Effect of nicotine chewing gum in smoking cessation JAMA 252: 2835-2838. HUGHES,J. R., and KRAHN, D. (1985).Blindness and the validity of the double-blind procedure. J. Clin. Psychopharmacol. 5: 138-142.
Subst Use Misuse Downloaded from informahealthcare.com by Freie Universitaet Berlin on 11/28/14 For personal use only.
CESSATION OF SMOKING
1231
HUNT, A. W., and BESPALEC,D. A. (1974). An evaluation of current methods of modifying smoking behavior. J. Clin. Psychol. 3 0 431-438. . IKARD, F. F., GREEN, D. E., and HORN, D. (1969). A scale to differentiate between types of smoking as related to the management of affect. Int. J. Addict. 4: 649-669. JARVIS, M. E., RAW, M., RUSSELL, M. A. H., and FEYERABEND, C. (1982). Randomized controlled trial of nicotine chewing gum. Br. Med. J. 285: 537-540. JARVIS, M. E., RUSSELL, M. A. H., and SALOOJE, Y. (1980). Expired aircarbon monoxide: A simple breath test of tobacco smoke intake. Br. Med J. 281: 484-485. LOMBARDO, W. L., HUGHES, J. R., and FROSS, J. D. (1988). Failure to support the validity of the Fagerstrom tolerance questionnaire as a measure of physiological tolerance to nicotine. Addict. Behav. 13: 87-90. MALCOLM, R. (1986). Silver acetate gum as a smoking deterrent. Chest 8 9 107-111. MANTEL, N., and HAENSZEL, W. (1953). Statistical aspects of the analysis of data from retrospective studies. J. Natl. Cancer Inst. 22: 719-748. MURRAY, E., and SCHNEIDER,N. G. (1984). Degree of addiction and effectiveness of nicotine gum therapy for smoking. Am. J. Psychol. 141: 790-791. RAW, M., JARVIS, M. J., FEYERABEND, C., and RUSSELL, M. A. H. (1980). Cornparkon of Ncotine chewing gum and psychological .treatment for dependent smokers. Br. Med J. 281: 481-482.
RUSSELL, M. A. H., PETO, J., and PATEL, U. A. (1974). The classification of smoking by factorial structure of motives. J. R Stat. SOC., A 137: 313-364. SCHMIDT, F. (1977). Raucherentwohnungdurch anti-raucher kaugummi dragees im doppelblindversuch. Munch. Med Wochenschr. 119: 1343-1344. SCHNEIDER, N. G., JARVIK, M. E., FORSYTHE, A. B., READ, L. L., ELLIOT, M. L., and SCHWEIGER, A. (1983). Nicotine gum in smoking cessation: A placebo-controlled, doubleblind trial. Addict. Behav. 8: 253-261. TOBACCO RESEARCH COUNCIL (1973). A Cornparison of Two Smoking Typologies (Research paper 12). London. TONNESEN,P.,FRYD, V., HANSEN, M.,HELSTED, J., GUNNERSEN, A. B., FORCHAMMER, H., and STOCKNER, M. (1988). Effect of nicotine gum in combination with group counseling on the cessation of smoking. N. Engl. J. Med. 318: 15-18.
The International Journal of the Addictions, 26(1 I), 1223-1231, 1991
The Effect of Nicotine, Silver Acetate, and Placebo Chewing Gum on the Cessation of Smoking. The Influence of Smoking Type and Nicotine Dependence E. J. Jensen,* MD E. Schmidt, MD 6. Pedersen, MD
R. Dahl, MD Department of Respiratory Medicine The University Hospital Noerrebrogade, DK-8000 Aarhus C, Denmark
Abstract The effects of nicotine (NI), silver acetate (SA), and ordinary (PL) chewing gum on the cessation of smoking were compared through ratings on six smoking types and physiological nicotine dependence (F-score) in a placebo-controlled and randomized study that included 496 smokers in a group-support setting. ResuIts were evaluated after 26 weeks.Ratings on and 6 and F-score were related to success rates. smoking types 1,3,4,5, The effect of NI was superior to SA in smokers with high type 1 ratings, the effects of NI and SA were almost equal and superior to PL in smokers with high type 4 and 6 ratings, and the effect was independent of the Fscore. Treatment of smokers with NI and SA should be restricted to smok*To whom correspondenceshould be addressed. 1223
Copyright 0 1991 by Marcel Dekker, Inc.
JENSEN ET AL.
1224
ing types in which an effect superior to PL is documented, and research should be performed to find support for smoking types where SA and NI have a clear effect.
Subst Use Misuse Downloaded from informahealthcare.com by Freie Universitaet Berlin on 11/28/14 For personal use only.
INTRODUCTION Smoking cessation programs employing nicotine chewing gum (NI) have varied in outcome after 1 year, and in some studies the effect was dependent on nicotine dependence (Table 1). Silver acetate chewing gum (SA), which gives tobacco smoke a bad taste, has been shown to have a short-term antismoking effect (Malcolm, 1986; Schmidt, 1977). The purpose of the present study was to compare the effects of NI, SA, and ordinary chewing gum (PL) by considering their scores on six smoking types (Ikard et al., 1969) and on the Fagerstrom nicotine tolerance questionnaire (Fagerstrom, 1978).
SUBJECTS AND METHODS Smokers whose daily cigarette consumption had been more than 10 for more than 10 years were included in the study, which was performed in 1986/87. Smokers with drug or alcohol misuse, psychiatric problems, cardiovascular diseases, and pregnant women were excluded. Participants were randomized into 24 main groups, and each group was randomly allocated to treatment with one of the chewing gums. Ten groups (N= 203), 10 groups (N = 21 l), and 4 groups (N = 82) were treated with SA, NI, and PL, respectively. The study could not be blind because the chewing gums had different modes of action and because consumption of SA had to be restricted to a daily maximum of six pieces to avoid the risk of argyria. The participants attended eight sessions of 2 h each at the clinic. They were controlled at each meeting for tobacco abstinence with an Ecolyzer CO-monitor (Energetics Science, Hawthorne, New York) which has been shown to be a reliable method of tobacco abstinence control (Jarvis et al., 1980). Participants presenting a CO concentration greater than 4 pprn were excluded. Reentrance was not allowed. The chewing gums could be used for 3 months. Because relapse rates were expected to be greatest during the first 6 weeks and to stabilize after 26 weeks (Hunt and Bespalec, 1974), sessions were placed at 1, 2, 3, 4, 5, 6, 12, and 26 weeks after the day of quitting. The results evaluated after 26 weeks. Before smoking cessation took place, all participants were guided in filling out a questionnaire concerning age, sex, smoking habits, drug and alcohol intake, living arrangements, and job status, and were then rated into six smoking types. The types included smoking for stimulation (type l), smoking due to a need for hand-
352
Fee and Steward, ScoBand, 1982
Hjallmarson, Sweden, 1984
3.
4.
173
118
9. Tonnesen et at., Denmark, 1988
10. Raw et al., England, 1980
6 + 3 follow-up
6 + 2 follow-up
9 + 3 follow-up
6 + 3 follow-up
8
14
6 during 6 weeks
10 during 5 weeks
Success rates of NI and PL treatments were 63 and 45 % ! @ < .05). NI only superior to PL in participants with F-score > 7. NI recommended for all Success rates of NI and PL treatments were 25 and 9%QJ < .05). Effect of NI was superior to PL only if F-score > 7 After 5 weeks, success rates of NI and PL treatments were 45.6 and 33.1% @ < 0.05); after 1 year, 13 and 9% (N.S.). Substitution time longer than 5 weeks was recommended Success rates of NI and PL treatments were 29 and 13%@ c .05). Daily cigarette consumption was without influence on effect of NI and success rates Success rates of NI and PL treatment were 44 and 21% @ < 01). Effect of NI was superior to PL only if F-scores belonged to the medium interval. No influence of daily cigarette consumption. NI recommended for all smokers In participants with F-score > 6, success rates of NI and PL treatment were 41 and 8% (N..S.), and in participants with F-score I7 success rates were 0 and 302, respectively (N.S.). No conclusion Success rates of NI and PL treatments were 31 and 14%@ < .05). Low p-nicotine was associated with high success rates. Sex, social class, and number of daily cigarettes had influence. NI recommended for well-motivated smokers Treatment with NI not superior to PL. Success rates were 30 and 12%in the NI and PL groups, respectively Effect of NI 2 and 4 mg, and NI 2 mg and PL were compared. Success rates of NI 2 mg and PL in participants with HR scores 2 119 were 44 and 12%@ c .05) and in participants with HR score I119 were 38 and 23%(N.S.).NI 2 mg recommended for light and medium nicotine-depend ent smokers Success rate of NI and PL treatments were 38 and 14%(p < .01). NI recommended for all smokers
Selected variables, results, and implications
nAll studies were double-blind, placebo controlled, and randomized except for study 2, which was not blind, and study 10, which was not blinded or randomized. Studies 2 and 4 used a 2xZ factorial design to evaluate the effect of behavioral treatment and long- vs short-term follow-up in addition to the effect of nicotine chewing gum. In all studies, results from 1 year follow-upare given, except for study 1 in which ir was 6 months. Categories of smokers included in the studies were rarely given. Two-tailed statistical tests were employed except in studies 1 and 7. F-score is the Fagerstrom nicotine-tolerance score, HR score is the Horn-Russell nicotine dependence score (Russell et al., 1974), NI is nicotine chewing gum, and PL is placebo.
60
Schneider et al., USA, 1983
8.
48 1 I6
Murray and Schneider, USA, 1984
139
7. Jarvis et al., England, 1982
6.
5. Hall et al., USA, 1987
151
Fagerstrom, Sweden, 1984
2.
206
8
100
Fagerstrom, Sweden, 1982
1.
2 + 1 call or I S
Sessions (S)
N
Author, country, and year
Smoking Cessation Programs with Nicotine Chewing Gum: Selected Parameters and Studiesa
Table 1
Subst Use Misuse Downloaded from informahealthcare.com by Freie Universitaet Berlin on 11/28/14 For personal use only.
uI
-
3
6
0
2
$
z
55
0
2!
JENSEN ET AL.
1226
Table 2
Subst Use Misuse Downloaded from informahealthcare.com by Freie Universitaet Berlin on 11/28/14 For personal use only.
Demographic Derailsa SA
NI
PL
Overall
N
203
211
82
496
Age ( y e a )
41.8 f 11.8
42.7 f 12.4
41.4 f 12.4
42.1 f 12.2
Males/females
941115
861116
39/41
2191272
+/- living with partner (percent)
78.4121.6
74.3125.7
78.5121.5
77.5122.5
+I- employment (percent)
87.2112.8
82.5117.5
85.0/15.0
84.81152
+/- earlier attempts to quit
83.9116.1
89.6110.4
82.7117.3
86.1113.9
Motivation score
14.6 f 2.2
15.2 f 2.3
15.0 f 2.3
14.8 2 2.3
Initial age of smoking Wears)
14.5 2 2.4
15.2 f 2.3
14.7 f 2.5
14.8 f 2.4
Cigarette consumption (numberlday)
21.7 f 9.9
21.8 f 8.2
21.0 f 9.7
21.7 f 9.1
(percent)
Pack years consumption
25.3 f 17.8
25.8 f 14.3
23.1 f 15.7
24.9 f 16.0
Smoking duration (years)
22.8 f 8.8
23.1 f 8.8
21.6 f 8.3
22.5 f 8.5
F-score
6.3 f 1.9
6.3 f 2.2
6.2 f 2.0
6.3 f 2.0
Type 1
8.4 f 2.3
8.3 f 2.6
8.1 f 2.4
8.3 2 2.5
Type 2
7.0 2 2.7
6.8 f 2.5
6.5 2 2.4
6.8 2 2.6
Type 3
12.4 f 2.2
12.3 f 2.2
12.0 f 2.0
12.3 2 2.2
Type 4
10.5 k 2.7
10.8 f 2.9
10.3 f 3.1
10.6 f 3.0
Type 5
10.6 f 2.7
10.8 f 2.5
10.8 2 2.7
10.8 f 2.6
TYPC: 6
6.8 f 2.9
6.8 2 2.6
6.2 2 2.6
6.7 f 2.8
oSA, NI, and PL are the nicotine, silver acetate, and placebo groups respectively. Means f SD are given if not otherwise indicated. All participants were Caucasians. None had medication for psychiatric disorders or used stimulating drugs. Motivation score was calculated as the sum of scores on a 6-item questionnaire with four gradings on the importance for quitting because of health concerns.
economics, looks, irritation of having a bad habit, wish to have cleaner surroundings, and consideration for others. F-score is the score on the Fagerstrom nicotine-tolerance questionnaire which is a 9-item questionnaire that evaluated physiological nicotine dependence through scores on time from awakening to first cigarette, number of cigarettes per day, nicotine content of the brand usually smoked, difficultieswith abstaining from smoking in public places, cigarette of the day most difficult to spare, smoking if sick, inhalationof the smoke, and time of day with mrst intensive smoking. Maximum score is 11, which indicates high depcndence. Description of the smoking types is given in the text. All demographic details were distributed equally between the treatment groups.
Subst Use Misuse Downloaded from informahealthcare.com by Freie Universitaet Berlin on 11/28/14 For personal use only.
CESSATION OF SMOKING
1227
ling things (type 2), smoking for pleasurable relaxation (type 3), smoking to reduce negative feelings (type 4), smoking due to psychological addiction or craving (type 5), and smoking due to habit (type 6) (Ikard et al., 1969). Each smoking type was related to three questions, each rated 1 to 5.The sum (3-15 points) was the rating for each type. The degree of physiological nicotine dependence (F-score) was assessed through the questionnaire of Fagerstrom (Fagerstrom, 1978). The RATE groups of a smoking type and F-score were defined as participants with, respectively, low and high ratings partitioned by the rating value that optimized the chi-square value calculated from the outcome of participants with rating values lower, equal to, or greater than the rating value used as the cutpoint. The analyses were carried out in the total population and controlled for treatment by using the test of Mantel-Haenszel (Mantel and Haenszel, 1953). In each RATE group the success rates of treatments were compared pairwise. The smokers gave their informed consent to participate and were given 1 to 2 weeks to consider their motivation to quit smoking. The study was approved by the ethical committee of Aarhus County. Statistics
The BMDP statistical program package was used in all calculations (BMDP Srutisticul Software, 1985).To evaluate differences in the success rates, Pearson’s chi-square test was employed. Uniformity in demographic variables between treatment groups was evaluated with Pearson’s chi-square and Student’s t-tests. In all tests a significance limit of 0.05% was used and two-tailed tests were employed.
RESULTS Demographic details are presented in Table 2. The overall success rate after 26 weeks were 42.6, 38.9, and 34.2% in the NI, SA, and PL groups, respectively (N.S.). Correlations between most smoking types and between smoking types and F-score were significant but small (Table 3). High ratings on smoking types 1 (p < .05),4 (p < .Ol), 5 (p < .05),6 @ < .03), and F-score (p < .05)and low ratings on type 3 (p < 0.01) were associated with low success rates (Table 4). In smokers with type 1 ratings greater than 10, the effect of NI was superior to SA (p < .01) and a 15% better outcome of NI treatment compared to PL did not reach significance. The effect of NI (p < .05)and SA @ < .02) was superior to PL in smokers with type 4 ratings greater than 10 and in smokers with type 6 ratings greater than 5 NI (p < .03) and SA (p < .07) tended to be superior to PL (Table 4). NI tended to have a better effect than PL in participants with an F-score greater than 4 (.05 < p < . l ) (Table 4).The effect of NI and SA was independent of the F-score in participants in RATE groups with high type 4 and 6 ratings (data not presented).
JENSEN ET AL.
1228
Table 3
Subst Use Misuse Downloaded from informahealthcare.com by Freie Universitaet Berlin on 11/28/14 For personal use only.
Correlation between Smoking Types and F-scores
Type 2
.19*
TYP 3
.12*
.29**
Type 4
.44**
.16**
.12*
TYP 5
.33**
.15**
.12*
.39**
TYPC 6
.17**
.18**
.05
.23**
.16**
F-score
.14*
.06
.0006
.15**
.33**
.29**
OCorrelation coefficients between ratings on the smoking types and between ratings on the smoking types and ratings on F-score. *p < 0.01. **p < 0.001.
Study limitations were lack of blindness, a too small placebo group considering some RATE groups, intercorrelations between ratings on smoking types, and probably an inaccurate F-score. The lack of blindness probably did not influence the outcome because the overall success rates were equal. Independency between blindness and success rates was found in a study with NI (Hughes and Krahn, 1985). The small intercorrelations between ratings for smoking types 1-6 were confirmed (Ikard et al., 1969; Tobacco Research Council, 1973), and in these studies a high reproduceability was shown. The reliability of the nicotine-tolerance questionnaire has recently been questioned (Lombard0 et al., 1988), and the implications are unclear. Smoking types with negligible intercorrelations and a reliable method of assessing physiological nicotine dependence might improve the classification of smokers. Differences in success rates of 15% between active and inactive treatment in some RATE groups did not reach significance. This may have been caused by the relatively few PL treated in some RATE groups. Sessions in this study were led by two physicians and a nurse. More staff members, including a psychologist, and smaller support-groups would have improved the possibility of individualizing support and increased follow-up. Our results indicate that ratings on most of the smoking types were predictors for success of smoking cessation and of value in identifying smokers who benefited significantly from active treatment. The equality of the effects of NI and SA
CESSATION OF SMOKING
1229 Table 4
Overall and Treatment-Specific Success Rates of RATE Groups
Subst Use Misuse Downloaded from informahealthcare.com by Freie Universitaet Berlin on 11/28/14 For personal use only.
Smoking type 5pe 1
Abstainers Treatment NI SA
PL NI SA
PL 5Pe3
NI SA
PL Type 4
NI SA
PL Type 5
NI SA
PL NI SA
PL
RATE groups
N
RATE S 10 RATE > 10 RATE S 10 RATE > 10 RATE S 10 RATE > 10
171 38 165 36 71
(96)
Pl
P2
N.S.
10
41.5 44.7 44.2 13.9 36.6 30.0
RATE S 7 RATE > 7 RATE 7 RATE > 7 RATE S 7 RATE > 7
127 82 131 70 51 30
39.2 36.6 38.9 38.6 39.2 30.0
N.S.
R A T E S 10 RATE > 10 RATE S 10 RATE > 10 RATE 5 10 RATE > 10
33 176 44 157 19 62
24.2 45.5 25.0 42.7 36.8 35.5
0.01
R A T E S 10 RATE > 10 RATE 5 10 RATE > 10 RATE I 10 RATE > 10
90 119 77 124 37 44
47.8 37.8 42.9 36.3 41.3 22.7
N.S.
RATE 5 10 RATE > 10 RATE I10 RATE > 10 RATE S 10 RATE > 10
97 112 76 125 32 49
46.4 38.4 44.7 35.9 40.6 32.6
N.S.
RATE S 5 RATE > 5 RATE S 5 RATE > 5 RATE S 5 RATE > 5
74 135 72 129 35 46
43.2 41.5 41.7 37.4 51.2 24.9
0.05
0.001 N.S. N.S.
N.S. N.S.
0.01
0.03 N.S. 0.01
N.S. 0.01
0.05
N.S. N.S. N.S.
0.03
N.S. 0.02 ~~
~
(continued)
JENSEN ET AL.
1230 Table 4 (continued) Smoking type
Subst Use Misuse Downloaded from informahealthcare.com by Freie Universitaet Berlin on 11/28/14 For personal use only.
F-score
Abstainers Treatment N1 SA
PL
RATE groups RATE 5 4 RATE > 4 RATE 5 4 RATE > 4 RATE 5 4 RATE > 4
N 33 162 35 157 17 62
(I) 51.5 43.2 48.6 37.6 58.8 30.6
Pl
P2
N.S.
0.03
N.S. 0.03
OAbstainer rates for each treatment within RATE groups of each smoking type 1-6 and F-score. Results are given as percentages. The RATE groups are designated by the rating value used as the cutpoint. The significanceof the difference in outcome between rate groups was tested for each treatment group (pl) and for the tohl population controlled for treatment (p2).
in the smoking types, except type 1, and their independence of the F-score did not indicate any advantage in employing smoking deterrents with different modes of action. It is suggested that pharmacologicalsupport with NI and SA is restricted to smoking types with a documented effect superior to placebo, and research efforts should be made to generate support for smoking types where NI and SA have clear value. Intensive psychological support may be of importance, and individualization of smoking cessation programs may make it easier and more effective to quit. Consideration of smoking types might be a basis for improved support. Studies to improve classification of smokers should be performed and evaluated for further individualization of treatment.
REFERENCES BMDP STATISTICAL SOFTWARE (1985).Los Angeles: University of California Press. FAGERSTROM. K-0.(1978). Measuring degree of physiological dependence to tobacco smoking with reference to individualization of treatment. Addict. Behav. 3: 235-241. FAGERSTROM, K-0.(1982). A cornparkon of psychological and pharmacological treatment in smoking cessation J. Behuv. Med. 5: 343-351. FAGERSTROM, K-0. (1984).Effects of nicotine chewing gum and follow-up appointments in physician-based smoking cessation Prev. Med. 13: 517-527. FEE, W. M.,and STEWARD, M.J. (1982). A controlled trial of nicotine chewing gum in a smoking withdrawal clinic. Practitioner 226 148-151. HALL, S. M., TUNSTALL, C. D., GINSBERG, D., BENOWITZ, N. L.. and JONES, R. T. (1987). Nicotine gum and behavioral treatment: A placebo-controlled trial. J. Consufr. Clin. Psychol. 55: 603-605. HJALMERSON, A. I. M. (1984).Effect of nicotine chewing gum in smoking cessation JAMA 252: 2835-2838. HUGHES,J. R., and KRAHN, D. (1985).Blindness and the validity of the double-blind procedure. J. Clin. Psychopharmacol. 5: 138-142.
Subst Use Misuse Downloaded from informahealthcare.com by Freie Universitaet Berlin on 11/28/14 For personal use only.
CESSATION OF SMOKING
1231
HUNT, A. W., and BESPALEC,D. A. (1974). An evaluation of current methods of modifying smoking behavior. J. Clin. Psychol. 3 0 431-438. . IKARD, F. F., GREEN, D. E., and HORN, D. (1969). A scale to differentiate between types of smoking as related to the management of affect. Int. J. Addict. 4: 649-669. JARVIS, M. E., RAW, M., RUSSELL, M. A. H., and FEYERABEND, C. (1982). Randomized controlled trial of nicotine chewing gum. Br. Med. J. 285: 537-540. JARVIS, M. E., RUSSELL, M. A. H., and SALOOJE, Y. (1980). Expired aircarbon monoxide: A simple breath test of tobacco smoke intake. Br. Med J. 281: 484-485. LOMBARDO, W. L., HUGHES, J. R., and FROSS, J. D. (1988). Failure to support the validity of the Fagerstrom tolerance questionnaire as a measure of physiological tolerance to nicotine. Addict. Behav. 13: 87-90. MALCOLM, R. (1986). Silver acetate gum as a smoking deterrent. Chest 8 9 107-111. MANTEL, N., and HAENSZEL, W. (1953). Statistical aspects of the analysis of data from retrospective studies. J. Natl. Cancer Inst. 22: 719-748. MURRAY, E., and SCHNEIDER,N. G. (1984). Degree of addiction and effectiveness of nicotine gum therapy for smoking. Am. J. Psychol. 141: 790-791. RAW, M., JARVIS, M. J., FEYERABEND, C., and RUSSELL, M. A. H. (1980). Cornparkon of Ncotine chewing gum and psychological .treatment for dependent smokers. Br. Med J. 281: 481-482.
RUSSELL, M. A. H., PETO, J., and PATEL, U. A. (1974). The classification of smoking by factorial structure of motives. J. R Stat. SOC., A 137: 313-364. SCHMIDT, F. (1977). Raucherentwohnungdurch anti-raucher kaugummi dragees im doppelblindversuch. Munch. Med Wochenschr. 119: 1343-1344. SCHNEIDER, N. G., JARVIK, M. E., FORSYTHE, A. B., READ, L. L., ELLIOT, M. L., and SCHWEIGER, A. (1983). Nicotine gum in smoking cessation: A placebo-controlled, doubleblind trial. Addict. Behav. 8: 253-261. TOBACCO RESEARCH COUNCIL (1973). A Cornparison of Two Smoking Typologies (Research paper 12). London. TONNESEN,P.,FRYD, V., HANSEN, M.,HELSTED, J., GUNNERSEN, A. B., FORCHAMMER, H., and STOCKNER, M. (1988). Effect of nicotine gum in combination with group counseling on the cessation of smoking. N. Engl. J. Med. 318: 15-18.
