- Email: [email protected]
Combined use of alcohol and nicotine gum
Addictive Behaviors, Vol. 16, pp. l-10, Printed in the USA. All rights reserved.
COMBINED
1991 Copyright
USE OF ALCOHOL
JIM MINTZ, * CATHERINE Department
SHIRLEY
AND NICOTINE
GUM
C. PHIPPS , * and MARY JANE ARRUDA
of Psychiatry
and Biobehavioral
Sciences,
UCLA
M. GLYNN and NINA G. SCHNEIDER
West Los Angeles VA Medical Center, Brentwood
MURRAY Department
0306-4603/91 $3.00 + .OO o 1991 Pergamon Press plc
of Psychiatry
Division,
UCLA
E. JARVIK
and Biobehavioral
Sciences,
UCLA
Abstract - This study evaluated the effects of chewing nicotine gum immediately before and just after drinking a moderate amount of alcohol. Four research questions were addressed. First, does chewing nicotine gum prior to drinking alcohol attenuate the increased craving to smoke that is typically associated with alcohol use? Second, does drinking prior to chewing reduce the gum’s effectiveness? Third, are significant side effects observed with nicotine gum, and is their severity affected by alcohol use? Finally, can we identify subjects who are more likely to respond well to the gum on the basis of smoking history or pattern or other descriptive-demographic, psychologic, or historical variables? Smokers who had abstained for at least 12 h were studied in a fully crossrandomized experimental design that contrasted nicotine gum (before or after drinking) versus sugarless gum, and alcohol versus a no-alcohol comparison condition. Nicotine gum use was associated with significantly greater immediate reduction in craving to smoke, regardless of whether it preceded or followed alcohol, but the effects were weak and short-lived in either case. Moderate use of alcohol after chewing the gum eliminated virtually all of its beneficial effects. Mild side effects were common with nicotine gum, but equally so regardless of alcohol use. A small battery of demographic and historical variables failed to identify those subjects who responded well to nicotine gum.
One of the most promising strategies in the psychologic treatment of smoking in recent years has been relapse prevention (Marlatt & Gordon, 1980). Within this framework, smokers are asked to identify “high-risk” situations, in which they perceive they have a strong probability of relapse, and to plan coping strategies as part of their cessation efforts. Alcohol consumption is clearly an important high-risk factor for the recent ex-smoker. Drinking alcohol appears to result in increased urges to smoke and/or decreased self-control. Interviews from 264 ex-smokers calling a relapse hotline indicated that 19% of the callers experienced smoking urges directly associated with alcohol consumption. Being at a social gathering where alcohol was available was especially problematic (Shiffman, 1986). In a laboratory setting, Mintz, Boyd, Rose, Charuvastra, and Jarvik (1985) found that drinking alcohol was associated with increased smoking. Given the association between drinking alcohol, smoking, and relapse, ex-smokers could benefit from interventions that could help them avoid smoking when they use alcohol. Nicotine chewing gum, used in conjunction with some type of group counseling or support, has been demonstrated to help individuals quit smoking (Christen, McDonald, Olson, Drook, & Stookey, 1984; Jarvis, Raw, Russell, & Feyerabend, 1982; Schneider et al.,
*Also at West Los Angeles VA Medical Center, Brentwood Division. The authors thank Bob Benson, B.A., for his work on the project, Stephen Miller and Phillip Ackerman of Merrill Dow, and acknowledge the technical support provided by our colleague, Jed Rose, Ph.D. This work was supported by Protocol 675, Study 719, and Grant No. P860501 from Merrill Dow Pharmaceuticals, Inc. Reprint requests should be sent to Jim Mintz, Ph.D., Professor, UCLA Department of Psychiatry and Biobehavioral Sciences, Brentwood VA Hospital (691/B] 17). 11301 Wilshire Blvd., Los Angeles, CA 90073.
2
JIM MINTZ et al
1983). A recent meta-analysis of 14 randomized trials comparing the use of nicotine gum to no gum or placebo during cessation clinics reported that gum improved l-year abstinence rates by 77% (Lam, Sze, Sacks, & Chalmers, 1987; see also Fagerstrom, 1988). However, a number of factors may bear on the effectiveness of the gum, including the degree of nicotine dependence (Fagerstrom & Schneider, in press; Jarvik & Schneider, 1984), lack of understanding of how the gum should be used (Schneider, 1988), and aversive side effects (Hughes & Miller, 1984). Given these difficulties, it is important to investigate for whom and in what ways nicotine chewing gum can be used most effectively. Recent ex-smokers who are social drinkers might be particularly good candidates for nicotine gum, both before and after drinking alcohol. Craving is likely to be maximized when alcohol is used, and both judgment and firmness of resolve may suffer. By satisfying the craving for nicotine, gum might have a preventive or therapeutic effect if used before drinking. The loss of taste sensitivity after drinking might render the gum more palatable at a time when craving is likely to be highest. On the other hand, there are several reasons to suspect that nicotine gum and alcohol might interact badly, particularly if the gum use followed drinking. For example, the careful systematic method of chewing the gum required for maximum effectiveness could be disrupted by mild intoxication. Improper technique could also lead to swallowing the nicotine, a problem that could be compounded if drinking was done while chewing. The nicotine in the gum can only be absorbed by holding it in the mouth, and any swallowed is lost. Furthermore, the combination of two gastric irritants-alcohol and nicotine-might lead to increased side effects. The current study explored some of these issues. It investigated the following research hypotheses: 1. Chewing nicotine gum immediately before drinking moderate amounts of alcohol will attenuate the increased craving typically associated with alcohol use. 2. Drinking alcohol immediately prior to chewing nicotine gum will reduce the effectiveness of nicotine gum. 3. Combined use of nicotine gum and alcohol will result in increased reports of unpleasant side effects. 4. A smoking history that suggests addiction to nicotine, other descriptive-demographic characteristics, and psychologic factors can identify subjects who are good gum responders. The hypotheses are stated above as directional, but in each case the statistical analyses were performed with two-tailed tests to allow the possibility that the results would be contrary to these expectations. METHOD
Subjects
Seventy-two subjects were recruited from community volunteers through newspaper advertisements and word of mouth. Initial screening was done by telephone with a structured interview form designed to rule out obviously inappropriate subjects such as ex-smokers. individuals with medical contraindications, impossible scheduling conflicts that would preclude participation, or individuals with drug or alcohol problems. Potential subjects were informed that the study involved the effectiveness of nicotine gum in altering psychologic and behavioral aspects of smoking for people who had been drinking alcohol at moderate levels. All subjects were active smokers who agreed to abstain for 12 h prior to the actual
Combined
Table Age Mean (? Range
use of alcohol and nicotine gum
1, Demographic
information
Gender (% male)
31 (2 21-60 68
Marital status(%) Single Married Widowed Divorced/separated
68 14 0 18
Educarion
SD)
7.87)
(9’~)
Some high school High-school graduate Some college/technical College graduate Postgraduate work
school
7 16 43 21 13
experimental session. Subjects were accepted into the study if they were competent to give informed consent, were above the legal drinking age of 21, and stated that they drank socially and smoked regularly. Demographic characteristics of the subjects are summarized in Table 1. Information obtained from a Smoking History Form is presented in Table 2. Subjects agreed to stop smoking for at least 12 h prior to participation in each study session but also agreed not to stop smoking entirely. Subjects who gave evidence of drug or alcohol abuse, or of medical problems that contraindicated cigarette, nicotine, or alcohol use in the judgment of the study’s responsible physician, were excluded. Subjects who regularly used any drugs that could potentially interact negatively with alcohol were also excluded from participation unless the subject received clearance from the research project’s physician for discontinuing the medications and agreed to refrain from taking them 24 h prior to each study session. Subjects who wished to stop smoking were invited to participate in a research smoking cessation program upon completion of the study. Experimental design and conditions One-third of the subjects chewed nicotine gum prior to drinking, one-third chewed nicotine gum after drinking, and the remaining subjects chewed sugarless gum. Half of each of these groups were given an alcoholic drink, and the other half a nonalcoholic drink. Twelve subjects were assigned randomly to each of the design’s six cells. Each subject participated first in an orientation session, followed by a week of practice with the nicotine gum at home. Then subjects returned to the laboratory for the experimental session, usually 1 week after the orientation. Nicotine gum versus sugarless gum. The experimental dose of nicotine gum was a 2-mg piece. Comparison subjects chewed sugarless gum. The use of sugarless gum as a comparison condition was a considered decision based on the judgment that the lack of a truly credible placebo precluded effective blinding of subjects. The comparison condition, which was included to control for physiologic effects of oral activity, does not control for expectancy effects resulting from the belief in the efficacy of the active gum. Thus, the results reported below can not unambiguously be attributed to pharmacologic versus psychologic factors. Alcohol versus no-alcohol. Half of the subjects in each gum condition drank an alcoholic drink containing 1.5 mL/kg alcohol. The amount is roughly equivalent to three drinks. Grain
JIM MINTZ et al.
Table 2. Smoking profile Cigaretres smoked daily PIT) -C I Pack 1 Pack Up to 2 packs > 2 Packs Other smokers in household (% yes) Nicotine content of preferred brand (mg) Mean (? SD) Range Yews smoked Mean (-c SD) Range Times quit smoking Mean ( ir SD) Range Lorqest time away from cigarettes ( % ) No time Hours Days Weeks Months Years
24 24 49 3 53 .96 (-t .27) .08-l .8
14(+ 7.23) 3-35 2 (? o-12
2.61)
4 6 19 17 37 17
neutral spirits mixed with an equivalent amount of orange juice provided the alcohol, a quantity chosen to produce blood alcohol levels in the range of social drinking (greater than 0.05 but less than the legal drunk limit of 0.10). Subjects in the nonalcohol condition received only orange juice, with a trace (5 mL) of alcohol floated on the top of the drink, a technique that provides some degree of blindness on the part of subjects regarding the experimental condition. Measures
After a telephone screening, potential subjects came to the laboratory for a longer screening, using standard questionnaires. These included a Smoking and Alcohol History form, Eysenck’s Personality Questionnaire (Extraversion and Neuroticism factors and a validity score), the Fagerstrom Tolerance Questionnaire (a total score reflecting dependence), the Russell Smoking Motivation Questionnaire (seven cluster scores representing Stimulation, Handling, Relaxation, Tension Reduction, Addictive, Automatic, and Psychosocial motivational bases for smoking), and the Shiffman-Jarvik Smoking Withdrawal Scale (four cluster scores representing Craving, Physiologic Discomfort, Sedation, and Tension). Subjects also completed the Cornell Medical Index, a detailed medical problems questionnaire, to screen out any individuals with medical problems that would contraindicate participation. Immediate consultation was obtained with the project physician for all questionable cases. All potential problems were brought to the attention of the principal investigator or the responsible physician, as appropriate, for clearance. Orientation
session
The purpose of the orientation session was to familiarize the subjects with the study procedures, obtain consent, instruct them in the use of nicotine gum, and take baseline measures. During this orientation session, the subjects were informed that the study was designed to determine the effectiveness of the gum when combined with alcohol and to identify any problems that might arise from the combination of alcohol and nicotine gum. Several special measures were obtained. A smoke mixer designed in our laboratory (Rose, Lafer, & Jarvik, 1982) allows subjects to manipulate nicotine level in inhaled smoke. This
Combined
use of alcohol and nicotine gum
5
device was used to determine whether nicotine gum had any effects on the preferred nicotine level. The mixer blends the smoke from two cigarettes-one high in nicotine (1.87 mg per cigarette) and the other low (0.2 mg per cigarette). The initial puff is entirely from the low nicotine cigarette. Then, the ratio is successively changed through six steps. At each step, an increasing proportion of smoke is drawn from the high nicotine cigarette, from 0% to 100% Subjects rated each puff from 0 to 10 for its strength, desirability, and harshness (0 = low or not at all, 10 = high or as much as possible). Subjects waited 1 minute between puffs. During the orientation, subjects received instruction in use of nicotine gum. They then practiced for a 2-min period by chewing a piece of 2-mg gum (the most commonly used dose) to insure that instructions provided by the manufacturer were correctly followed and that no immediate adverse consequences resulted. The instructions were to chew for 10 s (or 10 chews), and then hold the gum in the mouth for the remainder of a minute (50 s), and repeat this procedure for the 20-30-min period needed to get almost all of the nicotine out of the gum. Subjects were then instructed to use the gum at home for a week of practice, chewing two to three pieces daily when desired, with the exception that gum was not to be used within 2 h of alcohol use. All pieces of gum chewed at home were saved by subjects and returned to the laboratory when subjects came for the experimental session. Thus, it was clear by inspection that all the pieces of gum had been chewed. Subjects were aware that they could withdraw from participation because of problems with gum use during the orientation session or during the following orientation week. Experimental session After one week of practice with the gum, subjects returned for the experimental session. All subjects had abstained from smoking for at least 12 h prior to the experimental session. To confirm this, carbon monoxide (CO) levels were measured at two time points. Subjects took a breath sample at home the night before the experimental session after their last cigarette, and then provided another in the laboratory at the start of the experimental session. Compliance was excellent. The average subject dropped 65% in CO level, and more than 95% of the subjects had at least a 20% decrease. Subjects were interviewed by the experimenter to assess their general experience with the gum over the l-week practice period, and ratings of liking for the gum and reports of problems and side effects during the practice week were obtained. These interviews provided further evidence that subjects had complied with the request to use the gum on their own during the practice week. Subjects completed the side effects and Shiffman-Jarvik Scales to describe their current status, and rated their current degree of craving for a cigarette on a lOO-point rating scale. The experimental session was divided into three periods-a gum-chewing period lasting about 20 min, a drinking period lasting about 10 min, and a second gum-chewing period. During the first period, one-third of the subjects (n = 24) chewed one piece of 2-mg nicotine gum for 20 min, while the remaining 48 subjects chewed sugarless gum. The same chewing procedures were used by all subjects regardless of whether they were chewing nicotine or sugarless gum, with the rationale that this would help provide experimental control. Chewing was observed by a research assistant. The test battery, which included the side effects scale, the loo-point craving rating, and the Shiffman-Jarvik Smoking Withdrawal Scale, was then repeated to assess the effects of chewing the gum. During the second period, subjects were given the experimental drink to consume over a lo-min period. The subjects, who were run in small groups of approximately two to six, then read magazines or chatted. At the start of the session, blood alcohol levels (BALs) were checked with a breathalyzer to insure that they were zero. Beginning at 15 min after the drink, BALs were measured in breath samples taken at roughly IO-min intervals in all
JIM MINT2 et al.
Table 3. Mean changes in craving associated
with gum use before drinking
Experimental
Group
Rating scale
Nicotine Gum (N = 24)
Craving scale (loo-point)
-22.6 (2 16.4) r = 5.63
- 1.4 (2 24.6) t = .s
df = 1.69
p = .OcMll
n.6.
p = .OOOl
_ .? (2 .8) f = 1.62 n.s.
F = 13.61 df = 1.65 p = .0005
Shiffman-Jarvik Craving cluster (7.point)
-1.2
(?
1.3)
t = 5.65 p = .OOOl
Sugarless Gum (Pi = 48)
Between-Groups ANCOVA F =
17.93
Note. Covariance adjusted means (i SD), within-group f tests,and p levels are presented in each cell. Results of between-groups ANCOVAs, adjusting for baseline, are listed in the last column. Alcohol and no-alcohol groups were combined because all alcohol main and interaction effects were nonsignificant.
subjects using a breathalyzer until a level of at least 0.05 had been achieved in subjects who had received alcohol. at which point measurement was suspended. One subject whose BAL began to decline without having reached 0.05 was given a booster drink half the size of the original, following which a level exceeding 0.05 was obtained. At the end of this experimental period, subjects again rated their current craving for a cigarette on the lOOpoint scale. During the third period, half of the subjects who had chewed sugarless gum during the first period were now given one piece of 2-mg nicotine gum, following the same procedures used in the first period. The remaining subjects chewed sugarless gum. After 20 min, the measures of side effects and craving were repeated. Finally, the smoke mixer measurement of nicotine preferences was obtained, and subjects were offered a cigarette and smoked. After the cigarette, a final rating of craving was obtained. Subjects then remained in the laboratory until BAL levels fell below 0.03. RESULTS
Nicotine gum prior to alcohol First, the effect of gum on craving prior to drinking was analyzed using an analysis of covariance (ANCOVA), with change in the rating of craving on a lOO-point scale as the dependent measure, and baseline craving level as the covariate. This analysis, summarized in Table 3, indicated that the degree of craving for a cigarette decreased significantly more in subjects who chewed nicotine gum (F = 17.93, df = 1, 69, p = .OOOl). Their craving dropped an average of 22.6 (SD = 16.4) points on a loo-point scale (all means reported from covariance analyses are adjusted for covariates), a highly significant change (t = 5.63, df = 22, p = .OOOl). In contrast, the craving of those who chewed sugarless gum dropped only 1.4 points (SD = 24.6), a nonsignificant change (t = .53. df = 46). This finding was echoed by the Shiffman-Jarvik Craving cluster, which also showed a significantly greater decrease in the nicotine gum group (F = 13.61, df = 1, 65, p = .OOOS). The cluster score, which ranges from one to seven, decreased significantly in the nicotine gum group (mean change = 1.2, t = 5.65. p = .OOOl) versus a nonsignificant change of .2 in the comparison group. Because of the highly significant effect of nicotine gum prior to the experimental drink, the groups were no longer comparable at the point the drink was offered. This, in addition to significant heterogeneity of within-cell regression coefficients, precluded use of AN-
Combined
use of alcohol and nicotine gum
Table 4. Mean changes in craving associated
with gum use after drinking
Experimental Rating Scale
Nicotine Gum (N = 24)
Craving Scale (100 point)
10.1 (2 20.0) t = 3.12 p = .0027
Shiffman-Jarvik Craving Cluster (7 point)
- .42 (-+ 1.2) t = 2.03 p = .0465
Group
Sugarless Gum (N = 48) 1.6 (i-
13.6)
F = 8.68 1.67 p = .OO44 df =
t = .72 n.s. .30 (k
Between Groups ANCOVA
.9)
t = 2.03 p = .0461
F = 7.92 df = 1.63 p = ,006s
Note. Covariance adjusted means (+ SD), within-group t tests, andp levels are presented in each cell. Results of between-groups ANCOVAs, adjusting for baseline, are listed in the last column. Alcohol and no-alcohol groups were combined because all alcohol main and interaction effects were nonsignificant.
COVA. Instead, simple change scores during this period were analyzed using simple one-way analysis of variance (ANOVA). Alcohol had virtually no effect on craving in those subjects who had not chewed nicotine gum during Period 1. High craving remained constant during Period 2 whether subjects drank alcohol (mean change = 0.0; ns.) or not (mean change = 1.88; n.s.). In contrast, alcohol effects were possible for the subjects whose craving had been lowered by nicotine gum during Period 1. During Period 2, their self-rated craving for a cigarette increased dramatically and very significantly (mean increase with alcohol = 21.7 (SD = 27.3; t = 4.30, p = .OOOl). Subjects who had chewed nicotine gum followed by a nonalcoholic drink showed a small and nonsignificant increase in self-rated craving (mean change = 6.25, n.s.). The overall ANOVA main effect was significant (F = 4.58; d! = 3; p = .006), and Fisher’s Least Significant Difference indicated (at alpha = .OS) that craving had increased significantly more in the group that had chewed nicotine gum followed by alcohol than in any of the other three groups. Nicotine
gum following alcohol The results of a two (nicotine gum postalcohol vs. sugarless) x two (alcohol vs. no alcohol) factorial ANCOVA on change scores during the postdrink smoking period, using the baseline craving rating as the covariate, are summarized in Table 4. Nicotine gum significantly decreased craving relative to sugarless gum (F = 8.68, df = 1,67, p = .0044). Alcohol had no effects whatsoever (F ratios for the alcohol main effect and its interaction with gum condition were both
JIM MINTZ
8
et al
of the items were rated as moderate or more, and fewer than 1% were rated as severe. Nevertheless, chewing nicotine gum did result in a significant increase in complaints of side effects during both chewing periods. In the predrink session, the total side effects score increased .12 in the nicotine gum group (SD = .16, t = 3.69, p = .OOOS), while the control subjects showed no change (M = 0). a highly significant difference (F = 9.18, df = 1, 68, p = .0035). Although the absolute numbers of ratings of “severe” were very small, the side effect data were reanalyzed using a count based only on symptoms rated as severe. This analysis also showed a highly significant increase in the nicotine gum group @ = .0006), with no change in the group that chewed sugarless gum. Again, this difference was highly significant (F = 8.21, p = .0056). Increased side effects were again seen in subjects who chewed nicotine gum after drinking (mean change = .15, t = 3.61, p = .0006). This was a significantly larger change than the small drop in complaints (M = - .09, ns) seen in subjects who chewed sugarless gum during this period (F = 21.57, df = 1,63, p = .OOOl). No interaction with alcohol was observed, however, and increased complaints with nicotine gum use were no more serious when alcohol preceded the nicotine gum use (mean change = .15, p = .013) than when it did not (mean change = .14, p = .013). Reanalysis based on a count of symptoms rated as severe also showed a significantly greater increase in the nicotine gum group (F = 4.33, df = 1, 63, p = .042), but the increase was extremely small, averaging under 1% of the rated items. In all, 10 of the 48 subjects who chewed nicotine gum ever rated a side effect as severe, 5 in the predrink group and 5 in the postdrink condition. Only two complaints were given by more than one subject, excess sulivation (6 subjects) and unpleasnnt taste (5 subjects). Other complaints, given by only 1 subject each, were stomach discomfort, loss of appetite, and soreness of mouth. It should be noted, however, that in no case were side effects experienced severe enough for a subject to drop out of the study. Nicotine preferences
and effects of a real cigarette
After completing the postdrink chewing period, all subjects were tested on the smoke mixer to evaluate their nicotine preference levels. A repeated measures analysis of variance evaluating changes from the orientation session at each of the six mixer setting levels revealed no significant differences. Thus, nicotine preference was not related to nicotine gum use or alcohol consumption. After smoking a cigarette at the end of this period, dramatic and highly significant @ < 0.0001) drops in self-rated craving were seen in all groups. The effects of a cigarette on craving were roughly three to four times as great as those of nicotine gum. Changes in self-rated craving on the loo-point scale associated with cigarette smoking were comparable in all groups, averaging 40 points (SD = 24.8). Predictors
of gum effectiveness
Several sets of variables were examined in an exploratory manner to determine whether individuals who responded well to nicotine gum during the experimental session could have been identified on the basis of preexisting characteristics. These sets included the motivational factors from the Russell scale, the personality factors from the Eysenck Personality Inventory, the tolerance total score from the Fagerstrom instrument, and a number of historical and descriptive-demographic variables. A series of stepwise and multiple regression analyses were done, using variables taken in these sets separately. In each analysis, the dependent variable was the craving rating after smoking, and the contribution of craving immediately before the gum chewing session was forced into the model first. Thus, all correlations examined were partial correlations, adjusting for baseline craving. All subjects who chewed nicotine gum during the study, whether before or after drinking, were included in these analyses, and subjects in the alcohol and no-alcohol conditions were pooled, because
Combined
use of alcohol and nicotine gum
9
alcohol use had no effect on craving during either of the gum chewing experimental periods (Periods 1 and 3). In sum, these results were negative. All of the multiple correlations were nonsignificant. None of the separate motivational clusters from the Russell scale correlated significantly with change in craving after chewing nicotine gum in the laboratory session. Nor did any of the personality factors from the Eysenck Personality Inventory, or the total tolerance score from the Fagerstrom instrument. A battery of 15 descriptive-demographic and historical variables including age, gender, education, number of smokers in the household, previous attempts to quit smoking, use of caffeinated and alcoholic beverages, and self-perception of the degree of addiction to cigarettes yielded a nonsignificant multivariate test (F = 1.08,
.045)-were significance confirmation
p = significant correlates of change in craving. Of course, in the context of multiple testing, these isolated results could well be adventitious and would require in another sample. DISCUSSION
Nicotine gum appeared to be moderately effective in reducing self-reported craving for a cigarette, even when used after consumption of moderate amounts of alcohol. The absence of a true blind control condition means that any pharmacologic effects of the gum were possibly confounded with and augmented by psychologic expectancy effects associated with awareness of the experimental condition. Of course, those expectancy effects would presumably also operate in a real-life application of the gum to the problem of quitting. Despite any such augmentation, however, the gum had no prophylactic benefits with regard to increased craving caused by alcohol, and its effects on craving were extremely short-lived. Using nicotine gum immediately before drinking alcohol did not prevent alcohol from increasing the craving for a cigarette. In fact, drinking alcohol immediately after chewing nicotine gum essentially eliminated all of the gum’s beneficial effects, resulting in dramatic and immediate returns of craving to pregum levels. Thus, individuals who are using nicotine gum as an adjunct to quitting smoking should not view themselves as less vulnerable to the increased craving they are likely to experience if they drink. On the other hand, the data provide no evidence that nicotine gum should be avoided after alcohol. In fact, chewing after drinking alcohol reduced craving about as much as usual. Some increased side effects were typically reported with nicotine gum, but these were usually relatively mild and were no worse whether or not alcohol use preceded gum use. These findings suggest that social drinkers may find it useful to use nicotine gum when confronted with urges to smoke after drinking alcohol. The data provided relatively little evidence that any particular kind of smoker is more likely to find the gum effective. We were unable to find statistically compelling correlates of the response to the gum in a variety of measures of smoking motives or pattern, psychologic factors, or a number of descriptive and demographic variables. The failure to see alcohol-induced increases in craving in subjects who had not chewed nicotine gum prior to drinking was somewhat surprising. This may be at least partly attributable to the very high craving levels many of these subjects were already experiencing prior to drinking. Most of them (29/48 = 60%) rated their craving for a cigarette at the start of the drinking period at or above 85 on the loo-point rating scale. For all of its experimentally demonstrated effectiveness in attenuating the craving to smoke in recently deprived smokers, it must be acknowledged that nicotine gum’s benefits in this study were relatively fleeting and weak, particularly when compared to the effects of
10
JIM MINTZ et al.
smoking a cigarette. Although objective measures of chewing behavior were not taken, we do not think this was due to failure to chew the gum properly. Chewing during the experimental session was observed in the small experimental groups by the project’s research assistant, and all subjects had careful training in the laboratory, a week of documented practice, and an opportunity to consult about problems in nicotine gum use with research staff prior to the experiment. In our view, these results are more consistent with recent clinical reports on cessation programs that suggest that the 2-mg dose used in this study may be too low for many smokers. The new availability of gum with higher doses of nicotine may result in more effective cessation efforts (Tonnesen et al., 1988).
REFERENCES Christen, A., McDonald, J.L., Olson, B.L., Drook, C.A., & Stookey, G.K. (1984). Efficacy of nicotine chewing-gum in facilitating smoking cessation. Journal of rhe American Denrul Association, 108, 594-597. Fagerstrom, K.O. (1988). Efficacy of nicotine chewing-gum: A review. In O.F. Pomerleau & C.S. Pomerleau (Eds.), Nicotine replacement A cririccd evaluation (pp. 109-128). New York: Alan R. Liss. Fagerstrom, K.O., & Schneider, N.G. (in press). Measuring nicotine dependence: A review of the Fagerstrom Tolerance Questionnaire. Journal of Behavioral Medicines. Hughes, J.R., & Miller, S.A. (1984). Nicotine gum to help stop smoking. Journal of rheAmerican Medical Association, 252, 2855-2858. Jarvik, M.E., & Schneider, N.G. (1984). Degree of addiction and effectiveness of nicotine gum therapy for smoking. American Journal of Psychiqv, 141, 790-791. Jarvis, M.J., Raw, M., Russell, M.A.H., & Feyerabend, C. (1982). Randomized controlled trial of nicotine chewing-gum. British Medical Journal, 285, 537-540. Lam, W., Sze, P.C., Sacks, H.S., & Chalmers, T.C. (1987). Meta-analysis of randomized controlled trials of nicotine chewing-gum. Lancer, II, 27-30. Marlatt, G.A., & Gordon, J.R. (1980). Determinants of relapse: Implications for the maintenance of behavioral change. In P. Davidson & S. Davidson (Eds.), Behavioral medicine: Changing health lifesryles (pp. 410452). New York: Brunner/Mazel. Mintz. J., Boyd, G., Rose, J.E., Charuvastra, V.C., & Jarvik, M.E. (1985). Alcohol increases cigarette smoking: A laboratory demonstration. Addictive Behaviors, 10. 203-207. Rose, J.E., Lafer, R.L., & Jarvik, M.E. (1982). A smoke-mixing device for measuring nicotine preference. Behavior
Research
Merhods and Insrrumenrarion,
14, 501-503.
Schneider, N.G. (1988). How to use nicotine gum and orher srruregies ro quit smoking. New York: Pocket Books. Schneider, N.G., Jarvik, M.E., Forsythe, A.B., Read, L.L., Elliott, M.C., & Schweiger, A. (1983). Nicotine gum in smoking cessation: A placebo-controlled, double-blind trial. Addictive Behaviors, 8, 253-261, Shiffman, S. (1986). A cluster-analytic classification of smoking relapse episodes. Addictive Behaviors, 11, 295-307. Tonnesen, P., Fryd, V., Hansen, M., Helsted, J. 1Gunnersen, A.B., Forchammer, H., & Stockner, M. (1988). Effect of nicotine chewing gum in combination with group counselling on the cessation of smoking. Neiv England
Journal
of Medicine,
318,
15-18.
COMBINED
1991 Copyright
USE OF ALCOHOL
JIM MINTZ, * CATHERINE Department
SHIRLEY
AND NICOTINE
GUM
C. PHIPPS , * and MARY JANE ARRUDA
of Psychiatry
and Biobehavioral
Sciences,
UCLA
M. GLYNN and NINA G. SCHNEIDER
West Los Angeles VA Medical Center, Brentwood
MURRAY Department
0306-4603/91 $3.00 + .OO o 1991 Pergamon Press plc
of Psychiatry
Division,
UCLA
E. JARVIK
and Biobehavioral
Sciences,
UCLA
Abstract - This study evaluated the effects of chewing nicotine gum immediately before and just after drinking a moderate amount of alcohol. Four research questions were addressed. First, does chewing nicotine gum prior to drinking alcohol attenuate the increased craving to smoke that is typically associated with alcohol use? Second, does drinking prior to chewing reduce the gum’s effectiveness? Third, are significant side effects observed with nicotine gum, and is their severity affected by alcohol use? Finally, can we identify subjects who are more likely to respond well to the gum on the basis of smoking history or pattern or other descriptive-demographic, psychologic, or historical variables? Smokers who had abstained for at least 12 h were studied in a fully crossrandomized experimental design that contrasted nicotine gum (before or after drinking) versus sugarless gum, and alcohol versus a no-alcohol comparison condition. Nicotine gum use was associated with significantly greater immediate reduction in craving to smoke, regardless of whether it preceded or followed alcohol, but the effects were weak and short-lived in either case. Moderate use of alcohol after chewing the gum eliminated virtually all of its beneficial effects. Mild side effects were common with nicotine gum, but equally so regardless of alcohol use. A small battery of demographic and historical variables failed to identify those subjects who responded well to nicotine gum.
One of the most promising strategies in the psychologic treatment of smoking in recent years has been relapse prevention (Marlatt & Gordon, 1980). Within this framework, smokers are asked to identify “high-risk” situations, in which they perceive they have a strong probability of relapse, and to plan coping strategies as part of their cessation efforts. Alcohol consumption is clearly an important high-risk factor for the recent ex-smoker. Drinking alcohol appears to result in increased urges to smoke and/or decreased self-control. Interviews from 264 ex-smokers calling a relapse hotline indicated that 19% of the callers experienced smoking urges directly associated with alcohol consumption. Being at a social gathering where alcohol was available was especially problematic (Shiffman, 1986). In a laboratory setting, Mintz, Boyd, Rose, Charuvastra, and Jarvik (1985) found that drinking alcohol was associated with increased smoking. Given the association between drinking alcohol, smoking, and relapse, ex-smokers could benefit from interventions that could help them avoid smoking when they use alcohol. Nicotine chewing gum, used in conjunction with some type of group counseling or support, has been demonstrated to help individuals quit smoking (Christen, McDonald, Olson, Drook, & Stookey, 1984; Jarvis, Raw, Russell, & Feyerabend, 1982; Schneider et al.,
*Also at West Los Angeles VA Medical Center, Brentwood Division. The authors thank Bob Benson, B.A., for his work on the project, Stephen Miller and Phillip Ackerman of Merrill Dow, and acknowledge the technical support provided by our colleague, Jed Rose, Ph.D. This work was supported by Protocol 675, Study 719, and Grant No. P860501 from Merrill Dow Pharmaceuticals, Inc. Reprint requests should be sent to Jim Mintz, Ph.D., Professor, UCLA Department of Psychiatry and Biobehavioral Sciences, Brentwood VA Hospital (691/B] 17). 11301 Wilshire Blvd., Los Angeles, CA 90073.
2
JIM MINTZ et al
1983). A recent meta-analysis of 14 randomized trials comparing the use of nicotine gum to no gum or placebo during cessation clinics reported that gum improved l-year abstinence rates by 77% (Lam, Sze, Sacks, & Chalmers, 1987; see also Fagerstrom, 1988). However, a number of factors may bear on the effectiveness of the gum, including the degree of nicotine dependence (Fagerstrom & Schneider, in press; Jarvik & Schneider, 1984), lack of understanding of how the gum should be used (Schneider, 1988), and aversive side effects (Hughes & Miller, 1984). Given these difficulties, it is important to investigate for whom and in what ways nicotine chewing gum can be used most effectively. Recent ex-smokers who are social drinkers might be particularly good candidates for nicotine gum, both before and after drinking alcohol. Craving is likely to be maximized when alcohol is used, and both judgment and firmness of resolve may suffer. By satisfying the craving for nicotine, gum might have a preventive or therapeutic effect if used before drinking. The loss of taste sensitivity after drinking might render the gum more palatable at a time when craving is likely to be highest. On the other hand, there are several reasons to suspect that nicotine gum and alcohol might interact badly, particularly if the gum use followed drinking. For example, the careful systematic method of chewing the gum required for maximum effectiveness could be disrupted by mild intoxication. Improper technique could also lead to swallowing the nicotine, a problem that could be compounded if drinking was done while chewing. The nicotine in the gum can only be absorbed by holding it in the mouth, and any swallowed is lost. Furthermore, the combination of two gastric irritants-alcohol and nicotine-might lead to increased side effects. The current study explored some of these issues. It investigated the following research hypotheses: 1. Chewing nicotine gum immediately before drinking moderate amounts of alcohol will attenuate the increased craving typically associated with alcohol use. 2. Drinking alcohol immediately prior to chewing nicotine gum will reduce the effectiveness of nicotine gum. 3. Combined use of nicotine gum and alcohol will result in increased reports of unpleasant side effects. 4. A smoking history that suggests addiction to nicotine, other descriptive-demographic characteristics, and psychologic factors can identify subjects who are good gum responders. The hypotheses are stated above as directional, but in each case the statistical analyses were performed with two-tailed tests to allow the possibility that the results would be contrary to these expectations. METHOD
Subjects
Seventy-two subjects were recruited from community volunteers through newspaper advertisements and word of mouth. Initial screening was done by telephone with a structured interview form designed to rule out obviously inappropriate subjects such as ex-smokers. individuals with medical contraindications, impossible scheduling conflicts that would preclude participation, or individuals with drug or alcohol problems. Potential subjects were informed that the study involved the effectiveness of nicotine gum in altering psychologic and behavioral aspects of smoking for people who had been drinking alcohol at moderate levels. All subjects were active smokers who agreed to abstain for 12 h prior to the actual
Combined
Table Age Mean (? Range
use of alcohol and nicotine gum
1, Demographic
information
Gender (% male)
31 (2 21-60 68
Marital status(%) Single Married Widowed Divorced/separated
68 14 0 18
Educarion
SD)
7.87)
(9’~)
Some high school High-school graduate Some college/technical College graduate Postgraduate work
school
7 16 43 21 13
experimental session. Subjects were accepted into the study if they were competent to give informed consent, were above the legal drinking age of 21, and stated that they drank socially and smoked regularly. Demographic characteristics of the subjects are summarized in Table 1. Information obtained from a Smoking History Form is presented in Table 2. Subjects agreed to stop smoking for at least 12 h prior to participation in each study session but also agreed not to stop smoking entirely. Subjects who gave evidence of drug or alcohol abuse, or of medical problems that contraindicated cigarette, nicotine, or alcohol use in the judgment of the study’s responsible physician, were excluded. Subjects who regularly used any drugs that could potentially interact negatively with alcohol were also excluded from participation unless the subject received clearance from the research project’s physician for discontinuing the medications and agreed to refrain from taking them 24 h prior to each study session. Subjects who wished to stop smoking were invited to participate in a research smoking cessation program upon completion of the study. Experimental design and conditions One-third of the subjects chewed nicotine gum prior to drinking, one-third chewed nicotine gum after drinking, and the remaining subjects chewed sugarless gum. Half of each of these groups were given an alcoholic drink, and the other half a nonalcoholic drink. Twelve subjects were assigned randomly to each of the design’s six cells. Each subject participated first in an orientation session, followed by a week of practice with the nicotine gum at home. Then subjects returned to the laboratory for the experimental session, usually 1 week after the orientation. Nicotine gum versus sugarless gum. The experimental dose of nicotine gum was a 2-mg piece. Comparison subjects chewed sugarless gum. The use of sugarless gum as a comparison condition was a considered decision based on the judgment that the lack of a truly credible placebo precluded effective blinding of subjects. The comparison condition, which was included to control for physiologic effects of oral activity, does not control for expectancy effects resulting from the belief in the efficacy of the active gum. Thus, the results reported below can not unambiguously be attributed to pharmacologic versus psychologic factors. Alcohol versus no-alcohol. Half of the subjects in each gum condition drank an alcoholic drink containing 1.5 mL/kg alcohol. The amount is roughly equivalent to three drinks. Grain
JIM MINTZ et al.
Table 2. Smoking profile Cigaretres smoked daily PIT) -C I Pack 1 Pack Up to 2 packs > 2 Packs Other smokers in household (% yes) Nicotine content of preferred brand (mg) Mean (? SD) Range Yews smoked Mean (-c SD) Range Times quit smoking Mean ( ir SD) Range Lorqest time away from cigarettes ( % ) No time Hours Days Weeks Months Years
24 24 49 3 53 .96 (-t .27) .08-l .8
14(+ 7.23) 3-35 2 (? o-12
2.61)
4 6 19 17 37 17
neutral spirits mixed with an equivalent amount of orange juice provided the alcohol, a quantity chosen to produce blood alcohol levels in the range of social drinking (greater than 0.05 but less than the legal drunk limit of 0.10). Subjects in the nonalcohol condition received only orange juice, with a trace (5 mL) of alcohol floated on the top of the drink, a technique that provides some degree of blindness on the part of subjects regarding the experimental condition. Measures
After a telephone screening, potential subjects came to the laboratory for a longer screening, using standard questionnaires. These included a Smoking and Alcohol History form, Eysenck’s Personality Questionnaire (Extraversion and Neuroticism factors and a validity score), the Fagerstrom Tolerance Questionnaire (a total score reflecting dependence), the Russell Smoking Motivation Questionnaire (seven cluster scores representing Stimulation, Handling, Relaxation, Tension Reduction, Addictive, Automatic, and Psychosocial motivational bases for smoking), and the Shiffman-Jarvik Smoking Withdrawal Scale (four cluster scores representing Craving, Physiologic Discomfort, Sedation, and Tension). Subjects also completed the Cornell Medical Index, a detailed medical problems questionnaire, to screen out any individuals with medical problems that would contraindicate participation. Immediate consultation was obtained with the project physician for all questionable cases. All potential problems were brought to the attention of the principal investigator or the responsible physician, as appropriate, for clearance. Orientation
session
The purpose of the orientation session was to familiarize the subjects with the study procedures, obtain consent, instruct them in the use of nicotine gum, and take baseline measures. During this orientation session, the subjects were informed that the study was designed to determine the effectiveness of the gum when combined with alcohol and to identify any problems that might arise from the combination of alcohol and nicotine gum. Several special measures were obtained. A smoke mixer designed in our laboratory (Rose, Lafer, & Jarvik, 1982) allows subjects to manipulate nicotine level in inhaled smoke. This
Combined
use of alcohol and nicotine gum
5
device was used to determine whether nicotine gum had any effects on the preferred nicotine level. The mixer blends the smoke from two cigarettes-one high in nicotine (1.87 mg per cigarette) and the other low (0.2 mg per cigarette). The initial puff is entirely from the low nicotine cigarette. Then, the ratio is successively changed through six steps. At each step, an increasing proportion of smoke is drawn from the high nicotine cigarette, from 0% to 100% Subjects rated each puff from 0 to 10 for its strength, desirability, and harshness (0 = low or not at all, 10 = high or as much as possible). Subjects waited 1 minute between puffs. During the orientation, subjects received instruction in use of nicotine gum. They then practiced for a 2-min period by chewing a piece of 2-mg gum (the most commonly used dose) to insure that instructions provided by the manufacturer were correctly followed and that no immediate adverse consequences resulted. The instructions were to chew for 10 s (or 10 chews), and then hold the gum in the mouth for the remainder of a minute (50 s), and repeat this procedure for the 20-30-min period needed to get almost all of the nicotine out of the gum. Subjects were then instructed to use the gum at home for a week of practice, chewing two to three pieces daily when desired, with the exception that gum was not to be used within 2 h of alcohol use. All pieces of gum chewed at home were saved by subjects and returned to the laboratory when subjects came for the experimental session. Thus, it was clear by inspection that all the pieces of gum had been chewed. Subjects were aware that they could withdraw from participation because of problems with gum use during the orientation session or during the following orientation week. Experimental session After one week of practice with the gum, subjects returned for the experimental session. All subjects had abstained from smoking for at least 12 h prior to the experimental session. To confirm this, carbon monoxide (CO) levels were measured at two time points. Subjects took a breath sample at home the night before the experimental session after their last cigarette, and then provided another in the laboratory at the start of the experimental session. Compliance was excellent. The average subject dropped 65% in CO level, and more than 95% of the subjects had at least a 20% decrease. Subjects were interviewed by the experimenter to assess their general experience with the gum over the l-week practice period, and ratings of liking for the gum and reports of problems and side effects during the practice week were obtained. These interviews provided further evidence that subjects had complied with the request to use the gum on their own during the practice week. Subjects completed the side effects and Shiffman-Jarvik Scales to describe their current status, and rated their current degree of craving for a cigarette on a lOO-point rating scale. The experimental session was divided into three periods-a gum-chewing period lasting about 20 min, a drinking period lasting about 10 min, and a second gum-chewing period. During the first period, one-third of the subjects (n = 24) chewed one piece of 2-mg nicotine gum for 20 min, while the remaining 48 subjects chewed sugarless gum. The same chewing procedures were used by all subjects regardless of whether they were chewing nicotine or sugarless gum, with the rationale that this would help provide experimental control. Chewing was observed by a research assistant. The test battery, which included the side effects scale, the loo-point craving rating, and the Shiffman-Jarvik Smoking Withdrawal Scale, was then repeated to assess the effects of chewing the gum. During the second period, subjects were given the experimental drink to consume over a lo-min period. The subjects, who were run in small groups of approximately two to six, then read magazines or chatted. At the start of the session, blood alcohol levels (BALs) were checked with a breathalyzer to insure that they were zero. Beginning at 15 min after the drink, BALs were measured in breath samples taken at roughly IO-min intervals in all
JIM MINT2 et al.
Table 3. Mean changes in craving associated
with gum use before drinking
Experimental
Group
Rating scale
Nicotine Gum (N = 24)
Craving scale (loo-point)
-22.6 (2 16.4) r = 5.63
- 1.4 (2 24.6) t = .s
df = 1.69
p = .OcMll
n.6.
p = .OOOl
_ .? (2 .8) f = 1.62 n.s.
F = 13.61 df = 1.65 p = .0005
Shiffman-Jarvik Craving cluster (7.point)
-1.2
(?
1.3)
t = 5.65 p = .OOOl
Sugarless Gum (Pi = 48)
Between-Groups ANCOVA F =
17.93
Note. Covariance adjusted means (i SD), within-group f tests,and p levels are presented in each cell. Results of between-groups ANCOVAs, adjusting for baseline, are listed in the last column. Alcohol and no-alcohol groups were combined because all alcohol main and interaction effects were nonsignificant.
subjects using a breathalyzer until a level of at least 0.05 had been achieved in subjects who had received alcohol. at which point measurement was suspended. One subject whose BAL began to decline without having reached 0.05 was given a booster drink half the size of the original, following which a level exceeding 0.05 was obtained. At the end of this experimental period, subjects again rated their current craving for a cigarette on the lOOpoint scale. During the third period, half of the subjects who had chewed sugarless gum during the first period were now given one piece of 2-mg nicotine gum, following the same procedures used in the first period. The remaining subjects chewed sugarless gum. After 20 min, the measures of side effects and craving were repeated. Finally, the smoke mixer measurement of nicotine preferences was obtained, and subjects were offered a cigarette and smoked. After the cigarette, a final rating of craving was obtained. Subjects then remained in the laboratory until BAL levels fell below 0.03. RESULTS
Nicotine gum prior to alcohol First, the effect of gum on craving prior to drinking was analyzed using an analysis of covariance (ANCOVA), with change in the rating of craving on a lOO-point scale as the dependent measure, and baseline craving level as the covariate. This analysis, summarized in Table 3, indicated that the degree of craving for a cigarette decreased significantly more in subjects who chewed nicotine gum (F = 17.93, df = 1, 69, p = .OOOl). Their craving dropped an average of 22.6 (SD = 16.4) points on a loo-point scale (all means reported from covariance analyses are adjusted for covariates), a highly significant change (t = 5.63, df = 22, p = .OOOl). In contrast, the craving of those who chewed sugarless gum dropped only 1.4 points (SD = 24.6), a nonsignificant change (t = .53. df = 46). This finding was echoed by the Shiffman-Jarvik Craving cluster, which also showed a significantly greater decrease in the nicotine gum group (F = 13.61, df = 1, 65, p = .OOOS). The cluster score, which ranges from one to seven, decreased significantly in the nicotine gum group (mean change = 1.2, t = 5.65. p = .OOOl) versus a nonsignificant change of .2 in the comparison group. Because of the highly significant effect of nicotine gum prior to the experimental drink, the groups were no longer comparable at the point the drink was offered. This, in addition to significant heterogeneity of within-cell regression coefficients, precluded use of AN-
Combined
use of alcohol and nicotine gum
Table 4. Mean changes in craving associated
with gum use after drinking
Experimental Rating Scale
Nicotine Gum (N = 24)
Craving Scale (100 point)
10.1 (2 20.0) t = 3.12 p = .0027
Shiffman-Jarvik Craving Cluster (7 point)
- .42 (-+ 1.2) t = 2.03 p = .0465
Group
Sugarless Gum (N = 48) 1.6 (i-
13.6)
F = 8.68 1.67 p = .OO44 df =
t = .72 n.s. .30 (k
Between Groups ANCOVA
.9)
t = 2.03 p = .0461
F = 7.92 df = 1.63 p = ,006s
Note. Covariance adjusted means (+ SD), within-group t tests, andp levels are presented in each cell. Results of between-groups ANCOVAs, adjusting for baseline, are listed in the last column. Alcohol and no-alcohol groups were combined because all alcohol main and interaction effects were nonsignificant.
COVA. Instead, simple change scores during this period were analyzed using simple one-way analysis of variance (ANOVA). Alcohol had virtually no effect on craving in those subjects who had not chewed nicotine gum during Period 1. High craving remained constant during Period 2 whether subjects drank alcohol (mean change = 0.0; ns.) or not (mean change = 1.88; n.s.). In contrast, alcohol effects were possible for the subjects whose craving had been lowered by nicotine gum during Period 1. During Period 2, their self-rated craving for a cigarette increased dramatically and very significantly (mean increase with alcohol = 21.7 (SD = 27.3; t = 4.30, p = .OOOl). Subjects who had chewed nicotine gum followed by a nonalcoholic drink showed a small and nonsignificant increase in self-rated craving (mean change = 6.25, n.s.). The overall ANOVA main effect was significant (F = 4.58; d! = 3; p = .006), and Fisher’s Least Significant Difference indicated (at alpha = .OS) that craving had increased significantly more in the group that had chewed nicotine gum followed by alcohol than in any of the other three groups. Nicotine
gum following alcohol The results of a two (nicotine gum postalcohol vs. sugarless) x two (alcohol vs. no alcohol) factorial ANCOVA on change scores during the postdrink smoking period, using the baseline craving rating as the covariate, are summarized in Table 4. Nicotine gum significantly decreased craving relative to sugarless gum (F = 8.68, df = 1,67, p = .0044). Alcohol had no effects whatsoever (F ratios for the alcohol main effect and its interaction with gum condition were both
JIM MINTZ
8
et al
of the items were rated as moderate or more, and fewer than 1% were rated as severe. Nevertheless, chewing nicotine gum did result in a significant increase in complaints of side effects during both chewing periods. In the predrink session, the total side effects score increased .12 in the nicotine gum group (SD = .16, t = 3.69, p = .OOOS), while the control subjects showed no change (M = 0). a highly significant difference (F = 9.18, df = 1, 68, p = .0035). Although the absolute numbers of ratings of “severe” were very small, the side effect data were reanalyzed using a count based only on symptoms rated as severe. This analysis also showed a highly significant increase in the nicotine gum group @ = .0006), with no change in the group that chewed sugarless gum. Again, this difference was highly significant (F = 8.21, p = .0056). Increased side effects were again seen in subjects who chewed nicotine gum after drinking (mean change = .15, t = 3.61, p = .0006). This was a significantly larger change than the small drop in complaints (M = - .09, ns) seen in subjects who chewed sugarless gum during this period (F = 21.57, df = 1,63, p = .OOOl). No interaction with alcohol was observed, however, and increased complaints with nicotine gum use were no more serious when alcohol preceded the nicotine gum use (mean change = .15, p = .013) than when it did not (mean change = .14, p = .013). Reanalysis based on a count of symptoms rated as severe also showed a significantly greater increase in the nicotine gum group (F = 4.33, df = 1, 63, p = .042), but the increase was extremely small, averaging under 1% of the rated items. In all, 10 of the 48 subjects who chewed nicotine gum ever rated a side effect as severe, 5 in the predrink group and 5 in the postdrink condition. Only two complaints were given by more than one subject, excess sulivation (6 subjects) and unpleasnnt taste (5 subjects). Other complaints, given by only 1 subject each, were stomach discomfort, loss of appetite, and soreness of mouth. It should be noted, however, that in no case were side effects experienced severe enough for a subject to drop out of the study. Nicotine preferences
and effects of a real cigarette
After completing the postdrink chewing period, all subjects were tested on the smoke mixer to evaluate their nicotine preference levels. A repeated measures analysis of variance evaluating changes from the orientation session at each of the six mixer setting levels revealed no significant differences. Thus, nicotine preference was not related to nicotine gum use or alcohol consumption. After smoking a cigarette at the end of this period, dramatic and highly significant @ < 0.0001) drops in self-rated craving were seen in all groups. The effects of a cigarette on craving were roughly three to four times as great as those of nicotine gum. Changes in self-rated craving on the loo-point scale associated with cigarette smoking were comparable in all groups, averaging 40 points (SD = 24.8). Predictors
of gum effectiveness
Several sets of variables were examined in an exploratory manner to determine whether individuals who responded well to nicotine gum during the experimental session could have been identified on the basis of preexisting characteristics. These sets included the motivational factors from the Russell scale, the personality factors from the Eysenck Personality Inventory, the tolerance total score from the Fagerstrom instrument, and a number of historical and descriptive-demographic variables. A series of stepwise and multiple regression analyses were done, using variables taken in these sets separately. In each analysis, the dependent variable was the craving rating after smoking, and the contribution of craving immediately before the gum chewing session was forced into the model first. Thus, all correlations examined were partial correlations, adjusting for baseline craving. All subjects who chewed nicotine gum during the study, whether before or after drinking, were included in these analyses, and subjects in the alcohol and no-alcohol conditions were pooled, because
Combined
use of alcohol and nicotine gum
9
alcohol use had no effect on craving during either of the gum chewing experimental periods (Periods 1 and 3). In sum, these results were negative. All of the multiple correlations were nonsignificant. None of the separate motivational clusters from the Russell scale correlated significantly with change in craving after chewing nicotine gum in the laboratory session. Nor did any of the personality factors from the Eysenck Personality Inventory, or the total tolerance score from the Fagerstrom instrument. A battery of 15 descriptive-demographic and historical variables including age, gender, education, number of smokers in the household, previous attempts to quit smoking, use of caffeinated and alcoholic beverages, and self-perception of the degree of addiction to cigarettes yielded a nonsignificant multivariate test (F = 1.08,
.045)-were significance confirmation
p = significant correlates of change in craving. Of course, in the context of multiple testing, these isolated results could well be adventitious and would require in another sample. DISCUSSION
Nicotine gum appeared to be moderately effective in reducing self-reported craving for a cigarette, even when used after consumption of moderate amounts of alcohol. The absence of a true blind control condition means that any pharmacologic effects of the gum were possibly confounded with and augmented by psychologic expectancy effects associated with awareness of the experimental condition. Of course, those expectancy effects would presumably also operate in a real-life application of the gum to the problem of quitting. Despite any such augmentation, however, the gum had no prophylactic benefits with regard to increased craving caused by alcohol, and its effects on craving were extremely short-lived. Using nicotine gum immediately before drinking alcohol did not prevent alcohol from increasing the craving for a cigarette. In fact, drinking alcohol immediately after chewing nicotine gum essentially eliminated all of the gum’s beneficial effects, resulting in dramatic and immediate returns of craving to pregum levels. Thus, individuals who are using nicotine gum as an adjunct to quitting smoking should not view themselves as less vulnerable to the increased craving they are likely to experience if they drink. On the other hand, the data provide no evidence that nicotine gum should be avoided after alcohol. In fact, chewing after drinking alcohol reduced craving about as much as usual. Some increased side effects were typically reported with nicotine gum, but these were usually relatively mild and were no worse whether or not alcohol use preceded gum use. These findings suggest that social drinkers may find it useful to use nicotine gum when confronted with urges to smoke after drinking alcohol. The data provided relatively little evidence that any particular kind of smoker is more likely to find the gum effective. We were unable to find statistically compelling correlates of the response to the gum in a variety of measures of smoking motives or pattern, psychologic factors, or a number of descriptive and demographic variables. The failure to see alcohol-induced increases in craving in subjects who had not chewed nicotine gum prior to drinking was somewhat surprising. This may be at least partly attributable to the very high craving levels many of these subjects were already experiencing prior to drinking. Most of them (29/48 = 60%) rated their craving for a cigarette at the start of the drinking period at or above 85 on the loo-point rating scale. For all of its experimentally demonstrated effectiveness in attenuating the craving to smoke in recently deprived smokers, it must be acknowledged that nicotine gum’s benefits in this study were relatively fleeting and weak, particularly when compared to the effects of
10
JIM MINTZ et al.
smoking a cigarette. Although objective measures of chewing behavior were not taken, we do not think this was due to failure to chew the gum properly. Chewing during the experimental session was observed in the small experimental groups by the project’s research assistant, and all subjects had careful training in the laboratory, a week of documented practice, and an opportunity to consult about problems in nicotine gum use with research staff prior to the experiment. In our view, these results are more consistent with recent clinical reports on cessation programs that suggest that the 2-mg dose used in this study may be too low for many smokers. The new availability of gum with higher doses of nicotine may result in more effective cessation efforts (Tonnesen et al., 1988).
REFERENCES Christen, A., McDonald, J.L., Olson, B.L., Drook, C.A., & Stookey, G.K. (1984). Efficacy of nicotine chewing-gum in facilitating smoking cessation. Journal of rhe American Denrul Association, 108, 594-597. Fagerstrom, K.O. (1988). Efficacy of nicotine chewing-gum: A review. In O.F. Pomerleau & C.S. Pomerleau (Eds.), Nicotine replacement A cririccd evaluation (pp. 109-128). New York: Alan R. Liss. Fagerstrom, K.O., & Schneider, N.G. (in press). Measuring nicotine dependence: A review of the Fagerstrom Tolerance Questionnaire. Journal of Behavioral Medicines. Hughes, J.R., & Miller, S.A. (1984). Nicotine gum to help stop smoking. Journal of rheAmerican Medical Association, 252, 2855-2858. Jarvik, M.E., & Schneider, N.G. (1984). Degree of addiction and effectiveness of nicotine gum therapy for smoking. American Journal of Psychiqv, 141, 790-791. Jarvis, M.J., Raw, M., Russell, M.A.H., & Feyerabend, C. (1982). Randomized controlled trial of nicotine chewing-gum. British Medical Journal, 285, 537-540. Lam, W., Sze, P.C., Sacks, H.S., & Chalmers, T.C. (1987). Meta-analysis of randomized controlled trials of nicotine chewing-gum. Lancer, II, 27-30. Marlatt, G.A., & Gordon, J.R. (1980). Determinants of relapse: Implications for the maintenance of behavioral change. In P. Davidson & S. Davidson (Eds.), Behavioral medicine: Changing health lifesryles (pp. 410452). New York: Brunner/Mazel. Mintz. J., Boyd, G., Rose, J.E., Charuvastra, V.C., & Jarvik, M.E. (1985). Alcohol increases cigarette smoking: A laboratory demonstration. Addictive Behaviors, 10. 203-207. Rose, J.E., Lafer, R.L., & Jarvik, M.E. (1982). A smoke-mixing device for measuring nicotine preference. Behavior
Research
Merhods and Insrrumenrarion,
14, 501-503.
Schneider, N.G. (1988). How to use nicotine gum and orher srruregies ro quit smoking. New York: Pocket Books. Schneider, N.G., Jarvik, M.E., Forsythe, A.B., Read, L.L., Elliott, M.C., & Schweiger, A. (1983). Nicotine gum in smoking cessation: A placebo-controlled, double-blind trial. Addictive Behaviors, 8, 253-261, Shiffman, S. (1986). A cluster-analytic classification of smoking relapse episodes. Addictive Behaviors, 11, 295-307. Tonnesen, P., Fryd, V., Hansen, M., Helsted, J. 1Gunnersen, A.B., Forchammer, H., & Stockner, M. (1988). Effect of nicotine chewing gum in combination with group counselling on the cessation of smoking. Neiv England
Journal
of Medicine,
318,
15-18.