Alcohol abusers who want to quit smoking: Implications for clinical treatment

Transcription

1 Drug and Alcohol Dependence 54 (1999) Alcohol abusers who want to quit smoking: Implications for clinical treatment Timothy P. Ellingstad a, Linda C. Sobell a,b, *, Mark B. Sobell a,b, Patricia A. Cleland b, Sangeeta Agrawal b a No a Southeastern Uni ersity, Center for Psychological Studies, 3301 College A enue, Fort Lauderdale, FL 33314, USA b Addiction Research Foundation, 33 Russell St., Toronto, Ont. M5S 2S1, Canada Received 5 November 1997; accepted 8 October 1998 Abstract Although most alcohol abusers are dependent on nicotine, studies of such individuals have been scarce. Consequently, little information is available for advising clients who wish to consider resolving both problems. Clients entering an outpatient alcohol treatment program who were also current smokers were asked about their temporal preferences for changing their alcohol and cigarette use. Over three-quarters of alcohol abusers who were also smokers when asked said they would be willing to consider stopping smoking during or after treatment for an alcohol problem. Individuals who were interested in quitting smoking cigarettes while in treatment for an alcohol problem were different from those who did not want to stop smoking, and such differences may influence their ability to successfully address both problems together. Compared to those who preferred to change their drinking first then address their smoking, those who said they would be willing to address both problems (i.e. smoking and drinking) together in treatment were not only considerably more likely to think that quitting smoking would affect quitting drinking, but also more likely to be planning to quit smoking in the next six months. These results suggest that some individuals whose alcohol problems are not severe and who also smoke cigarettes will be more receptive to a dual recovery approach than others. In the absence of research findings, health care practitioners who encounter individuals who drink heavily and smoke cigarettes should at a minimum explore the option of dual cessation with their clients. The clinical and research implications of the present results are discussed Elsevier Science Ireland Ltd. All rights reserved. Keywords: Alcohol abuse; Alcohol dependence; Nicotine; Smoking; Treatment 1. Introduction * Corresponding author. Tel.: ; fax: ; sobelll@cps.nova.edu It is estimated that 80% to 90% of all people with alcohol problems smoke cigarettes. While this rate is very high compared to the 25% rate in the general population (Hurt et al., 1996), this is not surprising as there is a strong demonstrated link between alcohol and nicotine use (Hurt et al., 1996). For example, individuals who are more severely dependent on alcohol smoke more cigarettes per day and smoke sooner upon awakening than those who are less severely dependent drinkers (Batel et al., 1995). Also, individuals with a history of alcohol problems are less likely to quit smoking than those without such a history (Breslau et al., 1996; Hughes et al., 1996). While the mortality and morbidity rates are very high for both drinking and smoking, when the two drugs are used concurrently the resultant health costs and consequences of using both drugs jointly are staggering. Several years ago Luce and Schweitzer (1978) estimated the combined economic costs (direct and indirect) of alcohol and nicotine use to be about 25% of the total economic costs of all illness in the USA. Today these costs might be even greater as together alcohol and /99/$ - see front matter 1999 Elsevier Science Ireland Ltd. All rights reserved. PII: S (98)00180-X

2 260 T.P. Ellingstad et al. / Drug and Alcohol Dependence 54 (1999) nicotine contribute to over half a million deaths a year and are the leading cause of preventable disease, disability and excessive health care costs (Brady, 1995; Hurt et al., 1996). Further, it has been estimated that the combined health risks of smoking and alcohol use are 50% higher than the sum of their individual risks (Bien and Burge, 1990). Despite the fact that there is a strongly documented association between tobacco and alcohol use, and that an overwhelming majority of alcohol abusers smoke cigarettes, most alcohol treatment programs still ignore their clients smoking behavior (Hurt et al,. 1996; Seidner et al., 1996; Burling et al., 1997). Not only do very few alcohol treatment programs offer dual cessation interventions, most do not even bother to raise the issue of smoking cessation with their clients. This has occurred for several reasons, including assumptions that substance abusers do not want to stop smoking (Bobo, 1992; Sees and Clark, 1993), and that attempting to quit smoking interferes with the resolution of alcohol problems (Hurt et al., 1996; Burling et al., 1997). Several interview studies, however, have shown that alcohol abusers are not only open to discussing stopping smoking (Kozlowski et al., 1989; Seidner et al., 1996; Burling et al., 1997), but one study has even shown that nicotine dependence treatment provided as part of substance abuse treatment enhanced smoking cessation and did not have a substantial adverse effect on abstinence from the non-nicotine drug of dependence (p. 867; Hurt et al., 1994). To date, the authors are not aware of any published randomized trials evaluating temporal ordering of dual cessation (Sobell and Sobell, 1996). Thus, until studies are forthcoming to guide clinical interventions, the only alternative for developing strategies for alcohol abusers who want to stop smoking cigarettes is to identify characteristics associated with clients who indicate an interest in dealing with both problems concurrently or sequentially (Seidner et al., 1996). In this regard, it was felt that archival data (i.e. data that were routinely collected at the clinical assessment when clients entered the treatment program; there were no experimental manipulations) might be able to identify characteristics associated with alcohol abusers in treatment who were interested in dealing with both problems concurrently, sequentially, or who had no interest in stopping smoking. Once characteristics were identified then dual cessation strategies could be developed and tested. The present study which used actuarial data asked clients in an outpatient alcohol treatment program who also smoked cigarettes questions about whether they would like to quit smoking, and if so, when would be the best time to do so, how serious and confident they felt about stopping, and how stopping smoking might affect their ability to quit or reduce their drinking. 2. Methods 2.1. Participants Routine clinical assessment data from clients who voluntarily entered an outpatient alcohol treatment program at the Guided Self Change (GSC) Unit of the Addiction Research Foundation (ARF) in Toronto were evaluated. The GSC program is a brief (four-session) motivational intervention (Sobell and Sobell, 1993). Criteria for eligibility for the treatment program were: (1) primary alcohol problem (American Psychiatric Association, 1994); (2) 18 years of age; (3) an Alcohol Dependence Scale (ADS) (Skinner and Allen, 1982) score of 25 (such scores avoided clients whose alcohol problems are severe); (4) voluntary entrance into treatment (i.e. not mandated by courts or employers); (5) not currently receiving alcohol treatment elsewhere; and (6) not on anti-psychotic medications. At the assessment, each client was interviewed about their alcohol and nicotine use and administered the ADS as a measure of alcohol problem severity (Skinner and Allen, 1982). There were 185 clients who completed the clinical assessment as part of their entrance into the GSC treatment during a 5-month period in 1995/1996. The great majority of clients (84.0%, 152/181; four clients had missing data) met the DSM civ(american Psychiatric Association, 1994) criteria for alcohol dependence, with the remainder meeting the criteria for alcohol abuse Measures Questions on smoking were part of the clinical assessment in an effort to determine if clients who also smoked cigarettes would like to consider stopping smoking. All clients at the time of their clinical assessment who were smoking cigarettes completed a questionnaire about various aspects of their use of alcohol and cigarettes (e.g. number of years smoking cigarettes; number of years problem drinking; average number of cigarettes smoked per day). Because few studies have assessed the smoking behavior of individuals with alcohol problems, the following questions were chosen based on a review of significant issues raised by alcohol and smoking researchers in relation to possible links and cues shared by these two substances (Fertig and Allen, 1995): (l) percentage of time when smoking and alcohol use occurred together; (2) percentage of time had urges/cravings for one substance when using the other substance; (3) how stopping smoking might affect one s ability to quit or reduce drinking; (4) what might be the best way to stop or reduce both substances (i.e. concurrently, sequentially); (5) number of previous serious quit attempts for both substances; (6) how serious they were about stopping smoking and drinking in the

3 T.P. Ellingstad et al. / Drug and Alcohol Dependence 54 (1999) next 6 months as well as the next 30 days; (7) how confident they were that they could successfully stop smoking (0=not at all confident to 100=totally confident); (8) ever advised by a physician to stop smoking or to reduce or quit drinking; (9) ever quit smoking for at least one day in the past year; (10) anticipated side effects upon smoking cessation; and (11) if they stopped smoking today which of several options (e.g. patch, cold turkey, treatment) would be their first choice to help them stop. Although a Fägerstrom Test for Nicotine Dependence (Heatherton et al., 1991) is useful for research purposes, the question about number of minutes upon waking to the first cigarette (which is one question on the Fägerstrom) has been shown to be strongly predictive of nicotine dependence (Kozlowski et al., 1981; Pomerleau et al., 1990), and is often used in studies to describe the severity of clients smoking behavior, e.g. Seidner et al., 1996). Thus, all clients were also asked how many minutes after waking until the first cigarette in the morning? Lastly, everyone was asked which of five different options (i.e. stop smoking cigarettes and quit or reduce drinking at the same time; stop smoking cigarettes first and then reduce or quit drinking; reduce or quit drinking first and then stop smoking cigarettes; it would not matter which was dealt with first; or did not want to quit smoking cigarettes at the current time) they would currently choose if it were recommended that they also stop smoking cigarettes at the same time when they entered a treatment program to deal with their alcohol problem. 3. Results 3.1. Total sample description The smoking status of the 185 respondents at the time of the study was: (a) current cigarette smokers, n=100; (b) never smoked cigarettes, n=36; (c) exsmokers, n=46; and (d) only ever smoked pipes/cigars, n=3. Combining current and ex-smokers resulted in 78.9% (146/185) of all respondents reporting being cigarette smokers at one time. Because the present study was only looking at individuals with alcohol problems who currently smoked cigarettes, these individuals were the focus of this paper Data analysis procedures The data analysis procedures in the present study paralleled those reported by Seidner and her colleagues (1996) who looked at potential differences between dual treatment acceptors and refusors. In the present study, to identify possible differences between clients who preferred different temporal sequences for quitting smoking and drinking, appropriate univariate comparisons were first performed on variables listed in Table 1, and a liberal, non-corrected significance level of 0.1 was adopted for these analysis. To determine the best model for predicting preferred order of treatment, a stepwise logistic regression analysis was conducted using the 12 variables that were significant on the univariate comparisons as shown in Table 1. To avoid collinearity problems, examination of the correlations among 12 significant variables eliminated those that were highly correlated. To generate a parsimonious set of variables, the variable with the greatest number of 0.40 correlations was iteratively identified, and the variables with which it was correlated ( 0.40) were then eliminated. A forward stepwise procedure was used, and both p to enter and p to remain were set at To aid in interpreting the odds ratios, the significant variables entered into the logistic regression analysis were dichotomized (based on their distributions and theoretical considerations) prior to the logistic regression analysis Sample demographics Table 1 displays data for four different types of variables (demographic, smoking, alcohol, and smoking and alcohol) by clients preferred temporal resolution order (alcohol and cigarettes concurrently, ACC, n= 21; alcohol first, cigarettes second, A1C2, n=52; alcohol only, not cigarettes, ANC, n=21). Because too few respondents reported wanting to first stop smoking and then stop or reduce drinking (n=1), or that the order did not matter (n=4), their data are not included in further analyses. One client was also excluded because of missing data. 2 tests were performed on non-parametric variables and independent t-tests were used for parametric variables, as shown in Table Uni ariate comparisons between ACC and A1C2 Table 1 shows that clients who wanted to deal with both alcohol and cigarettes at the same time compared to clients who wanted to tackle their drinking first and then stop smoking were significantly more confident they would be able to quit smoking in the next 30 days as well as the next 6 months, smoked fewer cigarettes per day, and a greater percentage indicated they were planning to quit smoking in the next 30 days and next 6 months. Significantly more clients in the ACC group also reported having urges to drink for a greater percentage of the time when smoking, and significantly more said they believed that quitting smoking would help resolve their drinking problem Uni ariate comparisons between ACC and ANC Table 1 shows that clients who wanted to quit both alcohol and cigarettes at the same time compared to clients who wanted to change their drinking but had no

4 262 T.P. Ellingstad et al. / Drug and Alcohol Dependence 54 (1999) intention of stopping smoking were significantly more confident they would be able to quit smoking in the next 30 days as well as the next 6 months, smoked fewer cigarettes per day, felt they would have fewer side effects upon stopping, a greater percentage indicated they were planning to quit smoking in the next 30 days and next 6 months, a greater percentage reported having quit for at least one day in the past year, and their ADS scores were significantly higher. Significantly more also reported having urges to drink a greater percentage Table 1 Demographic, smoking, and alcohol variables by preferred temporal order of dual resolution Variables Preferred order of treatment t/x 2 (1) value ACC; n=21; A A1C2; n=52; B ANC; n=21; C A vs. B A vs. C Demographic Mean (S.D.) age (years) 36.6 (11.2) 38.8 (8.8) 36.5 (10.4) t (71)=0.88 t (40)=0.03 Mean (S.D.) education (years) 14.5 (2.5) 13.9 (2.0) 14.2 (2.3) t (71)=1.11 t (40)=0.45 % White-collar x 2 =1.32 x 2 =1.53 % Male x 2 =0.00 x 2 =0.87 % Married x 2 =1.32 x 2 =0 % Employed Smoking Mean (S.D.) confident able to quit smoking next (28.0) (25.4) (28.8) x 2 =0.10 t (44)=2.30 f x 2 =0.43 t (20)=3.58 g months a,c Mean (S.D.) confident able to quit cigarettes in next 30 days b,c Mean (S.D.) smoking quit attempts 70.4 (30.4) 11.0 (21.4) 18.1 (23.9) 3.6 (3.9) 7.5 (15.0) 3.9 (9.4) t (25)=4.93 h t (68)=1.53 t (16)=3.94 h t (38)=1.36 Mean (S.D.) [median] minutes to first cigarette upon (261.1) 91.4 (172.6) [15] 76.5 (149.9) [10] t (62)=1.82 t (35)=1.95 waking [60] Mean (S.D.) years of regular smoking 16.9 (12.9) 19.8 (9.4) 19.0 (9.8) t (62)=0.95 t (35)=0.55 Mean (S.D.) cigarettes per day c 14.4 (8.9) 19.9 (7.9) 23.7 (12.3) t (62)=2.31 f t (35)=2.54 f Mean (S.D.) perceived side effects quitting smoking c 2.6 (2.3) 3.6 (1.9) 4.6 (1.8) t (71)=1.89 t (40)=3.09 g % Advised by MD to stop smoking x 2 =0.97 x 2 =0 % Planning to quit smoking in next 6 months x 2 =6.96 g x 2 =20.5 h % Planning to quit smoking next 30 days x 2 =28.53 h x 2 =22.8 h % Quit smoking for 1 day in the past year x 2 =1.72 x 2 =9.72 g % Nicotine patch and/or gum would be first choice to help quit smoking d % Would quit smoking on own d x 2 (2)=4.71 x 2 (2)=0.51 % Would quit smoking using self help, treatment, or other method d Alcohol Mean (S.D.) ADS score (5.2) (4.9) (5.8) t (68)=1.77 t (40)=2.37 f Mean (S.D.) years with alcohol problem 9.1 (6.4) 8.1 (6.6) 6.6 (6.3) t (71)=0.62 t (40)=1.23 Mean (S.D.) alcohol related hospitalizations 0.2 (0.5) 0.2 (0.4) 0.1 (0.3) t (71)=0.54 t (40)=1.06 % Meet DSM criteria for alcohol dependence n.d. e x 2 =1.73 Mean (S.D.) alcohol related arrests 0.5 (0.8) 0.5 (0.9) 0.5 (0.8) t (71)=0.02 t (40)=0.21 Mean (S.D.) alcohol quit attempts 5.2 (6.2) 5.7 (8.1) 5.6 (6.3) t (70)=0.28 t (39)=0.21 % Advise by MD to reduce/stop drinking Alcohol and smoking Mean (S.D.) % time have urges to smoke when (15.2) (21.4) (21.5) x 2 =0.09 t (71)=0.46 x 2 =0.59 t (40)=0.62 drinking Mean (S.D.) % time smoke when drinking 87.1 (21.4) 93.9 (16.9) 89.5 (24.4) t (71)=1.45 t (40)=0.34 Mean (S.D.)% time have urges to drink when 46.7 (40.4) 16.0 (20.6) 10.3 (17.6) t (67)=3.08 g t (35)=3.52 h % Reducing alcohol first would help quit cigarettes x 2 =4.05 x 2 =14.0 h smoking c % Believe quitting cigarettes would help resolve x 2 =14.13 h x 2 =10.1 h drinking a Of those planning to quit in next 6 months. b Of those planning to quit in next 30 days. c n too small to include in logistic regression. d Analyzed using a 2 3 design. e n.d.=x 2 not determinable as 25% of cells have expected frequencies of 5. f P g P h P

5 T.P. Ellingstad et al. / Drug and Alcohol Dependence 54 (1999) Table 2 Stepwise logistic regression: Prediction of order of preferred treatment (concurrent vs. alcohol first) Step entered Wald P Odds ratio (95% confidence interval) Quitting smoking affect drinking ( ) Planning to quit smoking in next 6 months ( ) of the time when smoking, and significantly more not only said that they believed that quitting smoking would help resolve their drinking problem, but that reducing their drinking would help them quit smoking cigarettes Logistic regression analysis Only two groups (ACC and A1C2) were compared in a logistic regression analysis for two reasons. First, the group of interest for comparison purposes was alcohol abusers who wanted to quit at the time they were in treatment (i.e. ACC). Second, the total sample size available for the second comparison of interest (i.e. ACC and ANC) would have been less than that required for reliability in relation to the number of variables that were being examined in a logistic regression analysis. After eliminating those variables that would cause problems with collinearity, seven variables that significantly differentiated (P 0.10) respondents in the ACC group versus the A1C2 group in the univariate analyses were entered into a stepwise logistic regression analysis. Table 2 shows that the belief that quitting smoking would affect drinking, and planning to quit to smoking in the next 6 months combined to produce a multivariate model to predict preferred order of treatment. Compared to those who preferred to change their drinking first and then look at their smoking, those who wanted to address both smoking and drinking concurrently in treatment were about ten times more likely to think that quitting smoking would affect quitting drinking and were 7.5 times more likely to be planning to quit smoking in the next 6 months. This multivariate model, using the intercept plus the identified variables, accounted for a 33% reduction in error variance from using the intercept only. Selecting any cut-off probability level from 0.29 to 0.79 resulted in correct classification of 83.1% of the cases. 4. Discussion A substantial number of respondents (73.5%) in the present study reported being past or present cigarette smokers, a figure that parallels other studies of alcohol abusers who smoke cigarettes (Hurt et al., 1996; Seidner et al., 1996). Also, as reflected by the variables in Table 1, the clients in the present study were very similar in characteristics to data reported for other not severely dependent alcohol abusers in outpatient treatment (Sobell and Sobell, 1993; Sobell et al., 1995; Breslin et al., 1996). Based on their self-reports, those clients in the present study who indicated that they would prefer to quit both alcohol and cigarettes concurrently reported a link between the two substances. For many of these clients, smoking cigarettes appears to be a stimulus for their alcohol use as a significantly larger percentage of this group (62%) compared to the other two groups (17% and 14%) said they believed that quitting smoking would be helpful in resolving an alcohol problem and also reported urges to drink for a greater percentage of time (47%) when smoking than did the other two groups (16% and 10%). These differences may reflect that the clients who wished to resolve both problems felt that there was a strong relationship between their drinking and smoking and based their preference for resolving problems concurrently on that perception. Shiffman et al. (1994) have reported that although actual field test assessments of the timing of smoking and drinking occasions were associated, that association was not significantly correlated with subjects questionnaire reports of the relationship drawing into question the validity of subjects perceptions. However, that study used a community sample of participants in a smoking cessation study rather than problem drinkers. Further research is necessary to evaluate whether self-reported relationships between drinking and smoking have a basis in fact. Whatever the source of the differences between groups, the present study also suggests that the majority (77.7%, 73/94) of individuals with alcohol problems in treatment who were also smokers may be willing to consider stopping smoking during or after treatment for an alcohol problem. These results are consistent with those reported by Sees and Clark (1993) who found that a majority of inpatient substance abusers, and a presumably more severely dependent group, were interested in quitting cigarettes either concurrently (i.e. during their alcohol or drug treatment) or sometime in the future. Unfortunately, few alcohol treatment programs presently offer clients such options (Hurt et al., 1996). Interestingly, respondents in the present study who wanted to stop smoking at the same time as reducing or

6 264 T.P. Ellingstad et al. / Drug and Alcohol Dependence 54 (1999) quitting drinking were more confident that they would quit smoking in the next 30 days and 6 months. If such self-confidence is reflective of readiness to change, then treatment entry would be a good time to present a dual cessation program option to these clients. In fact, several smoking cessation studies have found that stage of change theory predicts that motivation or readiness to change is associated with positive outcomes (Di- Clemente et al., 1991; Velicer et al., 1992). Further support for this group s motivation to stop smoking is reflected in their greater belief that smoking cessation should be concurrent with alcohol treatment. Such beliefs were similar to those reported for substance abusers entering inpatient treatment who were interested in quitting smoking (Irving et al., 1994). While the present sample consisted of not severely dependent individuals with alcohol problems the results parallel those of a study of more severely dependent substance abusers attitudes about the resolution of alcohol and other drugs and their cigarette use (Seidner et al., 1996). The Seidner et al. study, like the present study, found that those who accepted or would consider concurrent treatment for smoking and substance use were more confident in their ability to stop smoking, had more smoking quit attempts, were more likely to believe that quitting smoking would benefit the resolution of their substance abuse problem, and were more likely to believe that the best time to quit smoking was during treatment or in next 6 months. The current findings have multiple implications for the treatment of individuals with both an alcohol and a smoking problem. This study as well as several others suggest that many individuals with alcohol problems who smoke are receptive to smoking cessation during the time they are in treatment for an alcohol problem (Kozlowski et al., 1989; Seidner et al., 1996; Burling et al., 1997). Further, it appears that individuals who are interested in quitting cigarettes when they are in treatment for an alcohol problem are different from those who do not want to stop smoking, and those differences may influence their ability to successfully address both problems together. Lastly, it is encouraging that many individuals with alcohol problems recognized and reported links between their alcohol and cigarette use. Such individuals may be good candidates for a dual cessation program because cessation of one substance should help reduce the potential cues for the second drug. While a few studies have provided data on specific characteristics of alcohol abusers who might do well or should be approached concerning concurrent resolution of their smoking and alcohol use (Burling et al., 1997), data from the several studies, including the present one, suggest that health care practitioners who encounter clients who drink heavily and also smoke cigarettes should routinely assess the relationship between their clients smoking and drinking as well as their clients interest in resolving both problems. For clients who are interested and willing, to work on both problems, dual cessation should be explored as part of a formal treatment plan. Finally, more studies are needed to further delineate the characteristics of alcohol abusers who are interested in and might benefit from receiving smoking cessation interventions. Acknowledgements This study was supported, in part, by a grant (AA08593) from the National Institute on Alcohol Abuse and Alcoholism. References American Psychiatric Association, Diagnostic and statistical manual of mental disorders, 4th ed. American Psychiatric Association, Washington, DC. Batel, P., Pessione, F., Maitre, C., Rueff, B., Relationship between alcohol and tobacco dependencies among alcoholics who smoke. Addiction 90, Bien, T.H., Burge, R., Smoking and drinking: A review of the literature. Int. J. Addict. 25, Bobo, J.K., Nicotine dependence and alcoholism epidemiology and treatment. J. Psychoact. Drugs 24, Brady, K., Prevalence, consequences and costs of tobacco, drug and alcohol use in the United States. In: Lewis, D.C. (Ed.), Training about alcohol and substance abuse for all primary care physicians, Josiah Macey Jr. Foundation, New York. pp Breslau, N., Peterson, E., Schultz, L., Andreski, P., Chilcoat, H., Are smokers with alcohol disorders less likely to quit? Am. J. Pub. Health 86, Breslin, C., Sobell, L.C., Sobell, M.B., Buchan, G., Kwan, E., Aftercare telephone contacts with problem drinkers can serve a clinical and research function. Addiction 91, Burling, T.A., Ramsey, T.G., Seidner, A.L., Kondo, C.S., Issues related to smoking cessation among substance abusers. J. Subst. Abuse 9, DiClemente, C.C., Prochaska, J.O., Fairhurst, S.K., Velicer, W.K., Velasquez, M.M., Rossi, J.S., The process of smoking cessation: An analysis of precontemplation, contemplation, and preparation stages of change. J. Cons. Clin. Psychol. 59, Fertig, J.B., Allen, J.A., Alcohol and tobacco: From basic science to clinical practice (NIAAA Research Monograph No. 30). National Institute on Alcohol Abuse and Alcoholism, Rockville, MD. Heatherton, T.F., Kozlowski, L.T., Frecker, R.C., Fagerström, K.O., The Fagerström test for nicotine dependence: A revision of the Fagerström tolerance questionnaire. Br. J. Addict. 86, Hughes, J.R., Fiester, S., Goldstein, M., et al., Practice guideline for the treatment of patients with nicotine dependence. Am. J. Psychiatry 153, Hurt, R.D., Eberman, K.M., Croghan, I.T., et al., Nicotine dependence treatment during inpatient treatment for other addictions: A prospective intervention trial. Alcohol: Clin. Exp. Res. 18,

7 T.P. Ellingstad et al. / Drug and Alcohol Dependence 54 (1999) Hurt, R.D., Offord, K.P., Croghan, I.T., Gomezdahl, L., Kottke, T.E., Morse, R.M., Melton, L.J., Mortality following inpatient addictions treatment: Role of tobacco use in a communitybased cohort. J. Am. Med. Assoc. 275, Irving, L.M., Seidner, A.L., Burling, T.A., Thomas, R.G., Brenner, G.F., Drug and alcohol abuse inpatients attitudes about smoking cessation. J. Subst. Abuse 6, Kozlowski, L.T., Director, J., Harford, M.A., Tobacco dependence, restraint and time to the first cigarette of the day. Addict. Behav. 6, Kozlowski, L.T., Skinner, W., Kent, C., Pope, M., Prospects for smoking treatment in individuals seeking treatment for alcohol and other drug problems. Addict. Behav. 14, Luce, B.R., Schweitzer, S.O., Smoking and alcohol abuse: A comparison of their economic consequences. N. Engl. J. Med. 298, Pomerleau, C.S., Pomerleau, O.F., Majchrzak, M.J., Kloska, D.D., Malakuti, R., Relationship between nicotine tolerance questionnaire scores and plasma cotinine. Addict. Behav. 15, Sees, K.L., Clark, H.W., When to begin smoking cessation in substance abusers. J. Subst. Abuse Treat. 10, Seidner, A.L., Burling, T.A., Gaither, D.E., Thomas, R.G., Substance-dependent inpatients who accept smoking treatment. J. Subst. Abuse 8, Shiffman, S., Fischer, L.A., Paty, J.A., Gnys, M., Hickcox, M., Kassel, J.A., Drinking and smoking: a field study of their association. Ann Behav. Med. 16, Skinner, H.A., Allen, B.A., Alcohol dependence syndrome: Measurement and validation. J. Abnormal Psychol. 91, Sobell, L.C., Sobell, M.B., Alcohol abuse and smoking: Dual recoveries. Alcohol Health Res. World 20, Sobell, M.B., Sobell, L.C., Problem Drinkers: Guided Self Change Treatment. Guilford Press, New York. Sobell, M.B., Sobell, L.C., Gavin, D.R., Portraying alcohol treatment outcomes: Different yardsticks of success. Behav. Ther. 26, Velicer, W.F., Prochaska, J.O., Rossi, J.S., Snow, M.G., Assessing outcome in smoking cessation studies. Psych. Bull. 111,