Tobacco Documents Online

Eur. J. Psychiat. Voi.11, N.° 3, (179-188) 1997 Key words: Smoking, Alcohol, Depression, Psy- chiatry, Smoking Among Psychiatric Patients Antti Tanskanen, M.D. Heimo Viinamiiki, M.D. Heli-Tuulie Koivumaa-Honkanen, M.D. Juha J~.~skel~iinen, M.D. Johannes Lehtonen, M.D. Department of Psychiatry Reseamh and Development Unit, Kuopio University Hospital Kuopio FINLAND ABSTRACT. -The present study examined the prevalence and determinants of smoking among patients receiving different psychiatric diagno~s (DSM-III-R), The study sample of 1,217 in- and outpatients, 13 to 88 years of age, tma~d at the Department of Psychiatry, Kuopio University Hospital, in eastern Finland, was interviewed with two sets of ques- tions during May 1993. The prevalence of smoking was significantly higher in male (62 % vs. 34 %) and female patients (40 % vs. 22 %) compared to the general population. Among men the highest rate was observed in schizophrenia and among women in perso- nality disorder. In persons with major depression or pcrsonaiity disorder smoking was related to the level of depression. The best determinants of smoking in multiple logistic models were alcohol drinking, poor financial situation and male gender. Education, or marital status were not associated with smoking after adjustment for confounders. Because psychiatric patients smoke so much, prevention approaches are clearly indicated. Introduction Recently the American Psychiatric Asso- ciation encouraged psychiatrists to partici- pate in advocacy, prevention and research concerning tobacco use among other things, (APA 1995). In addition, an advisory panel of the US Food and Drug Administration declared that "the amount of nicotine deli- vered by currently marketed cigarettes is likely to lead to addiction in the typical smoker" (Roberts 1994). The notion that psychiatric symptoms are related to smoking is not new. Equally well known is the fact that smoking is very com- mon among psychiatric patients (Hughes et al. 1986). As a matter of fact, it is also a habit frequently found in psychiatrists themselves compared with other medical spe.cialities (Hughes et al. 1992). The rela- tionship between cigarette smoking and psychiatric disorders has been thoroughly reviewed and discussed recently (Glassman 1993). This art~cte is tbr individual use only and may not be further reproduced or stored electronically without written permission fi'om the copyright b~lder. Unautflor/~,eff reprotguctfon may result fn ffnancia~ and o~cr panacftt'c~, fc) EUROf'EAN JOURNAL OF" PSYCHIATRY

180 ANTTI TANSKANEN ETALS. Among those with alcohol/drug abuse, 80 to 95 % smoke (DiFranza & Guerrera 1990, Glassman et aL 1990). Schizophrenic pa- tients also smoke heavily. According to dif- ferent surveys rates of smoking between 50 % and 90 % have been demonstrated (Goff et al. 1992, de Leon et aL 1995). Another study showed that psychiatric patients with a current major depression were more likely to smoke than the general population (Hughes etal. 1986). To their surprise, Glassman and co-wor- kers found that 60 % of the smokers who wished to stop smoking had a previous his- tory of major depression (Glassman et aL 1988). They were also able to show that psychiatric patients with a history of major depression developed depressive symptoms when they tried to quit and were more likely to fail in their smoking cessation efforts (Covey eta/. 1990). Later, examination of a population-based data set revealed that ma- jor depression was significantly more com- mon among smokers than among non smo- kers (Glassman et al. 1990). These findings have now been replicated in other clinical and epidemiological studies (Anda et aL 1990, Breslau et al. 1991, Breslau et aL 1993a, Kendler et al. 1993). Anxiety disorders are also associated with cigarette smoking, but the evidence for this relationship is weaker and less consis- tent (Glassman 1993, Breslau et al. 1992). In addition, smoking has been found to be associated with neuroticism and cynicism (Almada et aL 1991) and with Type A beha- viour (Forgays et aL 1993). In another study, nicotine dependence, but not nonde- pendent smoking was related to four measu- res (neuroticism, negaiive affect, hopeless- ness, general emotional distress) of vulnerability to psychopathology (Breslau ,t al. 1993b). In this report, we examine the association of smoking with sociodemographic and cli- nical factors in a large sample of psychiatric parents. The following questions are addressed: 1) What are the rates of smoking in patients from different psychiatric diag- nostic groups? 2) Does the rate of smoking vary by different factors in those psychiatric illnesses? 3) Does level of depression vary between smokers and non-smokers in those disorders? 4) What are the factors that best explain smoking in this Finnish sample of psychiatric inpatients and outpatients? Methods A cross-sectional survey of all patients at the Department of Psychiatry, Kuopio University Hospital, in eastern Finland, was carried out in May 1993. The total popula- tion of the area is approximately 200,000. The target sample included all the outpa- tients receiving specialist psychiatric care during May 1993, who had visited the out- patient units at least three times during January-April 1993, and all the inpatients during the first week of May 1993. A double set of 1,744 study questionnai- res was distributed. The patients returned 1,290 and the staff 1,683 questionnaires, so the response rams were 74 % and 97 %, res- pectively. Further analyses included only those patients for whom an assessment had been made by both the patients themselves and the staff (n = 1,217). This group com- prised 70 % of the original sample. The patient questionnaire contained 42 questions and covered sociodemographic factors such as sex, age, marital status, edu- cation, and financial situation. Use of alco- hol was determined with the question: "How This article is tbr individual use" only and may not be thnher reproduced or stored electronically without written permission fi'om the copyright holder. Un,a~Ihot~2e.d reproduction may result fn fi'nane~af anti other penalitms. (c) EUROPEAN JOURNAL OF P3YCHIATRY

i I I ! ! ! ! ! ! ! ! ! ! ! ! o/ten, on an verage, have you been drinking alcoholic beverages during the last six months? Subjects who had not been drin- king alcohol at all were classified as non- drinkers. The smoking status was defined by the question: "Have you been smoking daily during the last six months? The patients who answered with "yes" were considered to be current smokers. The level of depression was evaluated by means of the short (13-item) Beck Depression Inventory (BDI), in which the total BDI- score ranges from 0 to 39, with higher sco- res indicating more severe depression (Beck & Beck 1972). The staff questionnaire (23 questions) covered the place of treatment (outpatient clinic, hospital), and the main psychiatric diagnosis based on the DSM-III-R system determined by psychiatrists. Inquiry was also made into the number of previous psy- chiatric hospitalisations, the age of the appearance of psychiatric symptoms, and the start of psychiatric treatment. The psy- chopharmaceuticals used were registered in terms of different drug groups (yes-no). The amount of depressive symptoms at the time of the study was assessed with three op- tions: not at all, mild, severe. The staff had no knowledge patient's answers filling up the staff questionnaires. The nonrespondents (n = 393, 26%) did not differ from the respondents in terms of sex, age, the main psychiatric diagnosis, psychopharmaceuticals used, or the age at which the symptoms started or the treatment was begun. However, the nonrespondents had been in psychiatric inpatient treatment more often than the respondents (mean hos- pitalisations: 4.9 vs 3.9, p < 0.01). The general population we used as a refe- rence group, was a random sample of 5,000 persons, aged 15-64 years, drawn from the SMOKING AMONG PSYCHIATRIC PATIENTS 181 National Population Register of Finland. This postal survey on health behaviour among the Finnish adult population has been carried out annually since 1978 and was also done in spring 1993 (Berg et al. 1993). Its primary purpose is to acquire information specifically on smoking and changes in smoking. The sample covered eastern Finland, too. Associations between categorical varia- bles were estimated by the Chi-square test and Fisher's Exact Test. Relationships bet- ween the continuous variables were asses- sed by the Student's test or by analyses of variance (ANOVA), when appropriate. All tests of significance were two-tailed. Multiple logistic regression analysis (met- hod: enter) was used to control confounding factors when analysing smoking as the de- pendent variable. The computer processing was carded out by the SPSS for Windows. Results Of the 1,217 respondents, 55% were wo- men, 34% were marred or cohabiting, 26% had a low level of basic education, and the financial situation was poor in 39% of the subjects. The mean age was 4t.0 years (range, 13 to 88 years). Of the sample, 79% were outpatients, schizophrenia was the main psychiatric diagnosis in 33%, major depression (includes bipolar disorder) in 29%, neurosis (includes anxiety disorder, panic disorder, dissociative disorder, pho- bia, obsessive-compulsive disorder) in I I %, personality disorder in 11%, and other psy- chiatric disorders 16% of the cases. Of the patients, 53% used neuroleptics, 45% anti- depressants, 37% benzodiazepines, 5% car- bamazepine, 4% anticholinergics, and 14% other psychotropics (incl. hypnotics). The This article is tbr individual use only and may not be further reproduced or stored electronically w~thout written permission tlorn the copyright holder,

182 ANTTI TANSKANEN ETALS. mean duration of psychiatric treatment was 9.9 years (range, 0 m 49). In this sample of Finnish psychiatric pa- tients, 50% had been smoking daily during the last six months before the study in the age group of 15-64 years. This prevalence is significantly (p < 0.001) higher than 27% among the general population of Finland (Berg et al. 1993). Both men (62% vs. 34%, p < 0.001) and women (40% vs. 22%, p < 0.001) in our sample were more likely to be smokers than in the general population sam- ple. The prevalence of smoking was'signifi- cantly higher (p<0.001, expect for neurosis p<0.01) among all the main categories of psychiatric diagnoses compared to the gene- ral population in both sexes. Figure I. shows that the highest rate of smoking among men was reported by schizophrenics (67%) and among women in patients with personality disorder (48%). The lowest pre- valences were observed in neurotic patients both among men (51%) and women (34%). Because the group of subjects with other psychiatric disorders (n = 197) was conside- red to be too heterogeneous, it was excluded from the final analyses. This left us 1,020 subjects, 58% of the original sample to be analysed further. The prevalence of smo- king between inpatients and outpatients did not differ from each other, so these groups were analysed together. Among schizophrenic men, those who used two or more psychotropic drugs or who were alcohol drinkers were more likely to be smokers (table I). Male patients with major depression smoked at a higher rate, if they were under 40 yrs., with poor financial situation, or had mild to severe depressive symptoms. Among neurotic men the preva- lence of smoking was increased by poor financial status, two or more drugs used, or being an alcohol drinker. Men with persona- lity disorder were more likely to be smokers if they were classified as drinkers. Table I also shows that female schizoph- renics had higher prevalences of smoking with a poor financial situation and with a habit of drinking alcohol. Women with major depression were more likely to be smokers if their financial status was poor, or if they had depressive symptoms, or were reported to be drinkers. Neurotic women smoked at a higher rate if they had a poor financial situation' or took alcohol. Among women with personality disorder the diffe- rences in rates of smoking were influenced only by the habit of drinking. Analyses of variance showed that the mean scores of BDI, estimating the level of depression, tended to be higher in smokers than in non-smokers among all diagnostic groups except schizophrenia (table II). This was true in both non-drinkers and drinkers. The association was the most significant in the category of major depression (p<0.001). Alcohol status was not related to the level of depression in any of the disease categories. In table III we present the results of the multiple logistic regressions used to estima- te adjusted odds ratios (AOR) for smoking, controlling for the most potential confoun- ding factors found in bivariate analyses. Alcohol drinkers were three times as likely to be smokers than nondrinkers among schi- zophrenic patients when all other important factors were Controlled for. Male gender and the use of two or more psychotropic drugs were also highly related to smoking in these subjects. In persons with major depression the AOR's ~br alcohol drinking, for poor finan- cial situation, and for level of depression (estimated with BDI-score) were signifi- cantly different from unity (table III). Age This article is tbr individual use only and may not be further reproduced or stored electronically without written permission from the copyright holder Unautflor Cze~" repr~fuc~fon may rew~¢t in ffna~c~'al anff other pcna/(ticn. (~ ~UROf'~AN ,~OURN~f.., OF PSY'CHIP~TRY

OI 1 I I ! ! I I 1 I I I ! I I ! SMOKINGIAMONG PSYCHIATRIC PATIENTS 183 Table I Prevalence of smoking (%) according to sociodemographic factors in different categories of the main psychiatric diagnosis by sex Schizophrenia Major depresssion Neurosis Personality disorder men women men women men women men women % (n=198) (n=205) (n=lS1) (n=198) (n=49) (a=90) (n--69) (n=60) Age group S40 y~. 64 49 68 41 52 35 67 51 >40 yrs. 70 39 47 29 50 26 52 33 p<0.05 Madtal status marded/cohab. 69 single/div/wid 68 Education low 70 average/high 68 Financial st. good 64 poor 74 No of drugs 0-1 2+ 44 53 26 47 29 63 ' 43 44 57 39 52 32 62 48 51 53 23 46 33 75 67 42 55 37 50 30 60 44 38 39 24 25 23 52 37 58 73 48 85 44 70 63 p<O.05 p<0.001 p<0.001 p<0.001 p<0.05 46 43 58 42 38 32 61 42 80 45 53 28 75 29 62 52 p<0.001 p'0.05 Depressive symptoms not at all 7 l mild to severe 68 51 23 0 40 46 38 23 38 58 38 51 29 66 53 p<0.05 p<0.001 Alcohol status non-drinker drinker 58 32 44 8 14 14 38 I1 79 62 59 55 56 39 70 76 p<O.01 p<0.001 p<0.001 1:~0.05 p<0.05 p<0.05 p<O.001 (Significance was tested with Chi-square and Fisher's Exact Test, two-tailed) and male gender also contributed to the model. For neurotic patients, the odds for being a smoker with poor financial status was appr. four-fold and with alcohol drin- king the corresponding figure was about five-fold. Among alcohol drinkers with per- sonality disorder the AOR for being a smo- ker was the highest figure found in these analyses (13. I, 95% CI's 4.4-39.4, p<0.001). In these patients age and BDl-score were also related to smoking after adjustment for other misleading factors. Discussion The major findings of our study indicate the following. 1) The prevalence of smo- king was clearly higher among psychiatric patients compared to the general population in both genders. The highest rate among men was observed in schizophrenia and among women in personality disorder. 2) According to Beck Depression Inventory, in all groups of psychiatric illnesses, except in This article is tbr individual use only and may not be i~rther reproduced or stored electronically wthout written permission from the copyright holder.

184 AN'FFI TANSKANEN ETALS. Prevalence of smoking (%) SCH MD NEU PD 0TH GP (SCH = schizophre~a, MD = major depression, NEU = neurosis, PD = personality disorder. OTH = ot~r psychiatric diagnosis, GP = general population) Figure 1. Prevalence of smoking C',+) among p.~y,:hmu'ic patients (aged I5-64 years) compared to general population in men and women schizophrenia, both male and female smo- kers tended to be more depressed than nons- mokers irrespective of alcohol use. 3) In multivariate analyses alcohol drinking was significantly associated with smoking in all categories of disorders. Poor financial situa- tion was also related to smoking, mainly in patients with major depression and neuro- sis. The male/female ratio was the same bet- ween our material and the control sample (0.80 vs. 0.86, a nonsignificant difference). The age range was wider in the present study (13-88 yrs.), but in the analysis of figure 1 we selected only the patients 15 to 64 years old. The most significant determi- nant of smoking among our patients was the use of alcohol. The prevalence of non-drin- This article is for individual use only and may not be lhrthcr reproduced or stored electronically',mthout'written permission fi'om the cop)right holder. Unauthorized reproduction may resulI in financial and other panalities. (e) EUROPEAN JOURNAL OF PSYCHIATRY

I I ! I I ! SMOKING AMONG PSYCHIATRIC PATIENTS 185 Table 11 Mean (÷sd) BDl.scores in different categories of the main psychiatric diagnosis according to smoking and alcohol status (men and women combined) Schizophrenia Major depression Neurosis Personality disorder BDl-scores smok- smok+ smok- smok+ smok- smok+ smok- smok+ (mean ÷ sd) (n=152) (n=187) (n=174) (n=139) (n=78) (n--49) (n=54) (n=67) Alcohol status non.drinkers 9.3±7.3 8,5±7.1 10.5±8,3 16.0±7.1 9.4±7.3 16.5~-9.0 9.9_-~5.8 15.4±10.6 drinkers 9.5±6.7 9.5±7.5 I 1.3±8.4 14.3±7.8 10.5±5.6 13.1±8.3 10.9"4-7.1 14.3±7.7 ANOVA Main effects: - alcohol n.s. n,s. n.s. n.s. - smoking n.s. p<0.001 p<0.05 p<0.01 2-way n.s. n,s. n.s. n.s. interactions I I I I I I I I Table Ill Multiple logistic regressions of smoking the main psychiatric diagnosis Schir~ophrcnia (n=314) status on potential confounding factors in different categories of Major Depression Neurosis Personality disorder (n=299) (n= 124) (n= 113) FACTOR AOR 95% CI AOR 95 % CI AOR 95 % CI AOR 95% CI Age 0.99 0.96;1.01 0.97" 0.95; 1.00 0.97 0.93;1.01 0,94* 0,89;1.00 Alcohol- 3.36*** 2.00;5.65 4.37*** 2.36;8.09 4.77* 1.33;17.1 13.1*'~* 4 35;39.4 13 DI-score 0.96 0 93; 1.00 1.05"* 1.01; 1.09 1.06 0.99; 1.14 1.07* 1.00; I. 15 Education 0.58 0,33;1.03 0.77 0 42;1.43 0.85 0.27;2.68 0.30 0.05;1.75 Finane.st. 1.54 0,87;2.72 2.77*** 1.58;4.86 3.76** 1.55;9.15 1.48 0.56;3.92 Marital st. 1.32 0.68;2.57 1.20 0.69;2.10 0.60 0.23;1.56 0.59 0,20;1.79 No. nfdrugs 2.75*** I 59;4 74 0.88 0.50;1.57 2.39 0.84;6.84 0.91 0.36;2.35 Sex 0.37**~' 0.22;0.63 0,55* 0.32;0.95 0.63 0.25;1.59 0.90 0.34;2.38 AOR indicates adjusted odds ratio, CI indicates confidence interval. (* = 1)<0.05, ** = p<0.01, *** = p<0.001) Smoking (non-smoke~ 0, smoker = 1) Age (yrs) Alcohol (non-drinker = 0, drinker = 1) BDI-scorc (points, BDI indicates Beck Depression Inventory) Education (low = 0, average/high = 1) Financial status (good = 0, poor = I ) Marital status (married or cohabiting = 0, other = I) No. of psychotropic drugs (0-I = 0, 2 or more = 1) Sex (male = 0, female = 1) This article is tbr individual use only and rn~y not be thrther reproduced or stored electronically gathout written permission fcom the copyright holder, Unautlaorfze~ repmc~uc~fon may result fn ffnancfat and otf~cr pan~ftfc~, f~ ~UROf'~AN JOURNAL OF P~Y(~HIATRY

186 ANTT| TANSKANEN ETALS. kers in our study was 39%, while in the ran- dom sample it was 16% (a significant diffe- rence, p<0.001). This latter observation in- dicates that the high prevalence of smoking in these psychiatric patients seems to be the result of many different factors, and is not determined only by alcohol drinking. On the whole, the rates of smoking in our schizophrenic men were somewhat lower than in previous studies (Ziedonis et al. 1994). This cannot reflect the general ten- dency in the area, because the prevalence of smoking in Finland is higher than in the United States at the population level (Berg et aL 1993, Dalack & Glassmart 1992). Moreover, in Finland the patients are allo- wed to smoke in the psychiatric hospitals, in special rooms reserved only for smokers, while the trend is totally different elsewhere (Hughes 1993), In addition, to date there have been neither smoking cessation nor health education programmes aimed at the psychiatric population in Finland. In schizophrenic patients the number of psychotropic drugs used was associated with smoking. In this group a high rate of smoking is possibly an effort to reduce drug-induced side effects (Jarvik 1991). Nicotine "also increases the release of dopa- mine in the nucleus accumbens which has been shown to be associated with reward or incentive mechanisms in the brain, so this could be a potential basis for the high preva- lence of smoking in these patients (Koob & Bloom 1988). The positive relationship between smo- king and depression especially among patients with major depression was also supported by our study. It is understandable that a poor financial situation could lead to more severe depression and after that to smoking. There are some hypotheses and findings that several potential biological links exist between smoking and depression (Hall et aL 1993). For example, cortical arousal produced by nicotine administration has been hypothesized to refiect activation of central nicotinic receptors in the mid- brain reticular formation and limbic sys- tems. There is also evidence that vulnerabi- lity to nicotine dependence is related to high initial (genetic) sensitivity to nicotine (Pomerleau et aL 1993), and that people who are destined to become smokers are also genetically determined to become depressive, so there is probably a common predisposing factor to both smoking and depression (Kendler et aL 1993). The response rate was moderate. Accor- ding to the drop-out analysis, the main study variables did not differ between the respon- dents and the nonrespondents. A cross-sec- tional study like this has certain limitations, but, however, can offer valuable data on independent variables related to smoking and its relationship with psychiatric disor- ders. In addition, this survey could be the first one to provide information on smoking in all psychiatric disorders as well as in both outpatients and inpatients in a geographi- cally defined area where the treatment units are the only public providers of psychiatric care. The sample size is also much larger than in previous studies (Hughes et al. 1986) and the study population consists of both genders with a wide age range. The present study has the disadvantage that the smoking status was inquired into with only one simple question (yes-no). So we could not assess the relationship of the quantity of smoking with other possible fac- tors. Rowever, the validity of reported smo- king status has been found to be re~onably good in face-to-face interviews (Slattery et aL 1989). Because the individual variation of the use of alcohol in Finland is large, we asked for the average consumption of alto- This article is tbr individual use on12,, and may not be ~rther reproduced or stored electronically without written permission from the copyright holder. Unauthorize~l reproduction may result fn ffnancfat" and other pen~Ift[e~. (c) EUROt'EAN JOURNAL OI= 1";'3YCHIATRY

I I I I I I I I I I I I I I hoL which gives a sufficient estimate of the actual drinking patterns (O'Hare 1991). Disorders of memory caused by illnesses do not seem to affect the credibility of the self- reports (Brown et al. 1992), either. In this survey each patient completed the question- naire anonymously by him/herself which decreases the patients' tendency to adapt their responses to please the staff. Mo- reover, the self-reported prevalence of smo- king is usually an underestimate of the true rate (Hughes et aL 1986). The. validity and generalizability of our results are fairly good, because the response rote, the sample size, the range of disorders, and the representativeness of this patient population are adequate, in terms of tl~ sen- sible interpretation of data. One of the major implications of our study is the need for smoking cessation programmes in psychia- tric hospitals and in outpatient clinics, firstly with volunteers. As social attitudes and public policy act to diminish the rate of smoking at the population level, the remai- ning core of current smokers will increa- singly include those who are most severely addicted (Covey et al. 1990). As the preva- lence of'smoking continues to diminish, an increasing percentage of those who remain smokers will be psychiatric patients (Glassman 1993). The characteristics of this "refractory" group must be defined and the need and efficacy of specially tailored the- rapies for smoking cessation studied in cli- nical trials, even if we know that nicotine is the addicting drug with the poorest success rate (O'Brien & McLellan 1996). Infor. mation about the deleterious effects of ciga- rette smoking should be provided to psy- chiatric personnel and patients. Prevention approaches are clearly indicated and their outcome should be tested in prospective, fo- llow-up studies. SMOKING AMONG PSYCHIATRIC PATIENTS 187 References ALMADA SJ., ZONDERMAN AB., SHEKELLE RB, et aL Neuroticism and cynicism and risk of death in midd- le-aged men: The Western Electric Study. P~'chosomatic Medicine, 53. 165-t75, 1991. AMERICAN PSYCHIATRIC ASSOCIATION. Posi- tion statement on nicotine dependence. American Journal of Psychiat~.,; 152, 481-482. 1995. ANDA RE, WILLIAMSON DF., ESCOBEDO LG, et at. Depression and the dynamics of smoking. A national perspective. Journal of ilte Americtul Medictd Association, 264, 1541-1545, 1990. BECK AT., BECK RW. Screening depressed patients in family practice: A rapid technique. Postgruduate Me- dicine. 52, 81-85. 1972. BERG MA., HELAKORPI S., PUSKA P. Health beha- vigor among Finnish Adult Population, spring 1993. Pu- blications of the Notional Public Health Institute, B 10. Helsinki, Finland, 1993. BRESLAU N,, KILBEY MM., ANDRESKI P. Nicotine dependene~ and major depression. Now evidence from a prospective investigntion, Archives of General Psyebiat~. , 50, 31-35, 1993.(a) BRESLAU N., KILBE~Y MM.. ANDRE, SKI P. Nicotine dependence, major depression, ~nd anxiety in young adults. Archirex of General P.~. cbiato; 4.8. 1069-1074, 1991. BRESLAU N., KILBEY MM., ANDRESKI P. Nicotine withdrawal symptoms and psychiatric disorder;~ findings from an epidemiologie study of young adults. American Journal ofP.Lvchiatry. 149, 464-469, 1992. BRESLAU N,, KILBEY MM.o ANDRESKI P. Vulneo rability to psychopathology in nicotinc-depenflcnt smo- kers: an epidcmiologic study of young adults. Antet~can Journal of P.~. chiatry, 150, 941946, 1993. Co) BROWN J., KRANZLER H., DEL BOCA E Self-re- ports by alcohol and drug abuse inpalienls: factors affec- ting reliability and validity'. British Journal of Addiction.r, 87, 1013-1024-, 1992. COVEY LS., GLASSMAN AH,, STETNER F. Dcpm- ssion and depressive symptoms in smoking cessation. Comprehensive P~,'¢hiatr3~ 31,350-354, 1990. DALACK GW, GLASSMAN AH. A clinicoJ approach to help psychiatric patients with smoking cessation. P.~.whiatric Quarterly, 63, 27-39, 1992. This article is tbr individual use only and may not be thnher reproduced or stored electronically ~thout written permission t~om lhc copyright holder.

188 ANTTI TANSKANEN ETALS. DE LEON J., DADVAND M., CANUSO C. et al. Sclfi- zophrcnia and smoking: An epidemiological survey in a state hospital. American Jouranl of Psychiatry, 152, 453- 455, 1995. DIFRANZA JR., GIJERRERA MP. Alcoholism and smoking. Journal of Studies on Alcohol, 51, 130-135, 1990. FORGAYS DG, BONAIUTO P. WRZESNIEWSKI K, et ul. Personality and cigarette smoking in Italy, Poland, and the United States. lntermztianal Jounzal of Addictions, 28, 399-413, 1993. GLASSMAN AH, HEI..ZER JE, COVEY LS, et al. Smoking, smoking cessation, and major depression. Jour- nal of the American Medical Assoeiatinn, 264, 1546-1549. 1990. GLASSMAN AH, STETNER F, WALSH BT, et al. Heavy smokers, smoking ces~tion, and clonidinc. Results of a double-blind, randomized trial. Journal of the Ame. rtcan Medical Association, 259. 2863-2866, 1988. GLASSMAN AH. Cigarette smoking: implications for psychiatric illness. American Journal of Psychiatry. 150, 546-553, 1993. GOFF DC, HENDERSON DC, AMICO E. Cigarette smoking in schizophrenia: Relationship to psychopatho- logy and medication side effects. American Journal of Psychiatry, 1~.9, 1189ol 194, 1992 HALL SM, MUNOZ RF, REUS VI, et al. Nicotine, negative affect, and depression. Journal of Cansulting and Clinical Psychology, 61,761-767, 1993. HUGHES JR. HATSUKAMI DK, MITCHELL JE. et a/. Prevalence of smoking among psychiatric outpatients. American Journal Psychiatry. 143, 993,-997, 1986. HUGHES JR. Possible effect~ of smoke-free inpatiant units on psychiatric diagnosis and treatmem. Journal of Clinical Psychiatry, 54, 109-I 14, 1993. HUGHES PH, BALDWIN DWC, SHEEHAN DV, et al. Resident physician substance use, by specialty. Ame- ricanJounml of Psychiatry. |49, 1348-i354, 1992. JARVIK ME. Beneficial effects of nicotine. British Journal ofAddi¢'tion~, 86, 571-575, 199l. KENDLER KS, NEALE MC, MACLEAN C J, et aL Smoking and major depression. A causal analysis. Ar- chives of General P=ychiatry, 50, 36-43. 1993. KOOB GF, BLOOM FE. Cellular and molecular mechanisms of drug dependence. Science, 242, 715-723, 1988. O'BRIEN CP0 MCLEI,LAN AT. Myths about the treat- ment of addiction. Lattcet, 347, 237-240, 1996. O'HARE T. Measuring alcohol consumption: a compa- rison of the retrospective diary and the quantity-frequency methods in a college survey. Journal of Steadies on Alcohol, 52, 500-502, 1991. POMERLEAU OF, COLLINS AC, SHIFFMAN S, el a/. Why some people smoke and others do not: New pres° pecfives. Journal of Consulting and Clinical Psychology. 61,723-731, 1993. ROBERTS J. Nicotine is addictive, says advisory panel in US. British Medical Journal, 309, 429. 1994. SLATFERY ML, HUNT SC, FRENCH TK, et aL Validity of clgarette smoking habits in three epidemiotogic studies in Utah. Preventive Medicine, 18, I l - 19,1989. ZIEDONIS DM, KOSTEN TR, GLAZER WM, et Nicotine dependence mad schizophrenia. Hospital and Community psychiatry. 45, 204-206, 1994. Address of author:. Antti Tanskanen Department of Psychiat~ Kuopio University Hospital P.O. Box 1777 Fin-7021 i Kuopio FINLAND This article is tbr individual use only and may not be 0arther reproduced or stored electronically without written permission from the copyright holder. Unauthorized reproduction may result in financial and other penalities. (e) EUROPEAN JOURNAL OF PSYCHIATRY