OUP user menu

DEPENDENCE ON LEGAL PSYCHOTROPIC DRUGS AMONG ALCOHOLICS

Björn Axel Johansson, Mats Berglund, Maria Hanson, Christina Pöhlén, Ingrid Persson
DOI: http://dx.doi.org/10.1093/alcalc/agg123 613-618 First published online: 21 November 2003

Abstract

Aims: Dependence on legal psychotropic drugs (PTD) has been reported to have increased in alcoholics, but previous studies report conflicting results concerning the rate of increase and clinical characteristics. The aim of the present study was first, to assess the dependence rate of PTD among alcoholics in open and institutionalized care, and to compare these populations with the general population, and second, to assess rates and doses of high- and low-dose PTD-dependence among alcoholics. Methods: In 1997, alcoholics in open and institutionalized care were asked to anonymously fill in a questionnaire on their drug use and dependence. Healthy controls were included. The number of attending subjects was 130 open-care alcoholics at the Department of Alcohol and Drug Diseases in Malmö, Sweden; 23 alcoholics in institutionalized care at Karlsvik Rehabilitation Centre in Höör, Sweden; and 120 healthy controls at Vårdcentralen Kirseberg, a primary health care centre located in a Malmö area. The approximate attendance rate was 75, 70 and 95%, respectively. The questionnaire was based on DSM-IV criteria for dependence. Results: The total rate of PTD-dependent alcoholics was higher in the institutionalized group (35%) than in the open-care setting (14%): difference in proportions (p1–p2 21%; 95% CI: 1%, 41%). Alcoholics were more often PTD-dependent (17%) than were healthy controls (2%), (p1–p2 15%; 95% CI: 9%, 21%). Benzodiazepines (BZD) were the most common PTD. Only four out of a total of 23 BZD-dependent alcoholics developed high-dose BZD-dependence. Those subjects were also misusing other drugs, including cannabis. Conclusions: We conclude that alcoholism is associated with legal PTD-dependence and illegal drug misuse. High-dose BZD-dependence is infrequent among BZD-dependent alcoholics.

(Received 8 April 2002; first review notified 14 May 2002; in revised form 4 July 2003; accepted 5 August 2003)

INTRODUCTION

Epidemiological studies

Alcohol dependence and its comorbidity with psychotropic drug (PTD) dependence (both legal and illegal drugs) has been addressed in epidemiological studies. From the literature it is not possible to differentiate between dependence on legal or illegal drugs. In the Epidemiologic Catchment Area (ECA) study, Regier et al. (1990) concluded that the lifetime comorbidity rate of drug disorders in alcoholism was 47.3%. In the National Comorbidity Survey (NCS), Warner et al. (1995) assessed the prevalence of drug dependence in the general population. In the total group (n = 6673) 1.8% had been dependent during the past 12 months according to DSM-III-R. The lifetime prevalence was 7.5%. In the NCS, Kessler et al. (1997) found the lifetime prevalence for drug dependence among alcohol-dependent subjects to be 30% (men 29.5%; women 34.7%). In the National Longitudinal Alcohol Epidemiological Survey (NLAES), Hanna and Grant (1999) reported that the lifetime prevalence to qualify for DSM-IV diagnoses of drug dependence was 5.9%.

Clinical studies

Alcohol dependence and its comorbidity with legal PTD dependence has also been addressed in clinical studies, where the patients in general were using both prescribed and self-administrated PTD. Kryspin-Exner (1966) found that of 302 open-care alcoholics treated with diazepam, 14% were dependent. Ashley et al. (1978) found that of 1001 hospitalized alcoholics, 10% were misusing tranquilizers. Busto et al. (1983) found that of 261 outpatient alcoholics, 18% misused benzodiazepines (BZD). Ciraulo et al. (1988) reviewed the literature concerning liability for BZD misuse among alcoholics and concluded that the prevalence of BZD misuse among alcoholics was higher than in the general population. Ross (1993) included 427 patients with alcohol misuse or dependence according to DSM-III: 20% of the alcoholics had misused or had been dependent on anxiolytics, including BZD, during their lifetime. In summary, these clinical studies showed that 10–20% of the alcoholics developed a concomitant BZD misuse or dependence. Sallmén et al. (1997) assessed 104 alcohol-dependent subjects, 58 coerced and 46 voluntarily admitted for inpatient treatment at Karlsvik Rehabilitation Centre in Höör. 28% were BZD-dependent BZD according to DSM-III-R during the past 12 month period. The correspondent total lifetime prevalence was also 28%.

High-versus low-dose BZD-dependence syndrome

Smith and Wesson (1994) divided BZD dependence into high- and low-dose BZD withdrawal syndromes. The high-dose BZD withdrawal syndrome is characterized by withdrawal symptoms emerging after discontinuation of high doses of BZD, taken for 1 month or more. The syndrome is qualitatively similar to that for all sedative hypnotics. The time course of symptoms depends on the particular drug. Short-acting BZD withdrawal symptoms typically peak between 24–72 h after the last dose. The low-dose BZD withdrawal syndrome is characterized by withdrawal symptoms emerging after discontinuation of therapeutic doses of BZD, taken for at least 4 months. This syndrome is also associated with protracted withdrawal symptoms. The present study addresses the question of dependence on psychotropic drugs by a sample of alcoholics, which will be divided into high- and low-dose BZD dependence, according to the average daily dose taken by the individual patient, regardless of withdrawal symptoms.

Other authors have used this concept with different definitions of doses in high- and low-dose BZD dependence. Busto et al. (1989), Schneider-Helmert (1988) and Bernik et al. (1998) define low-dose BZD (diazepam equivalents) dependence as less than 4.0, less than or equal to 2.0 and 0.5–2.0 defined daily dose (DDD), respectively. Diazepam 10 mg = 1 DDD (WHO Collaborating Centre, 2002). To our knowledge, the definition of high- and low-dose dependence on zolpidem, zopiclone, codeine and dextropropoxyphene is not described in the literature.

SUBJECTS AND METHODS

Subjects and setting

During four consecutive weeks in 1997, consecutive patients at the Department of Alcohol and Drug Diseases (DAD) in Malmö, Karlsvik Rehabilitation Centre (KRC) in Höör, and healthy controls at VC Kirseberg (VCK) in Malmö, were asked to fill in an anonymous self-report concerning their possible use and dependence on legal PTD.

The number of attending subjects was 130 open-care alcoholics at DAD, 23 long-term institutionalized alcoholics at KRC (17 coerced and six voluntarily admitted) and 120 healthy controls attending a general health examination at VCK. The approximate attendance rate was 75, 70 and 95%, respectively. The setting at DAD offers an extensive outpatient treatment of alcoholics, well described by Österling et al. (1994). The setting at KRC offers both coercive and voluntary inpatient treatment of alcohol-dependent subjects. The mean treatment period for coercive and voluntarily treated patients at KRC is 5.5 and 4 months, respectively. The number of beds at KRC is 24. The setting is well described by Sallmén et al. (1997). VCK is a primary health care centre located in the Malmö area with 10 000 inhabitants. Data on psychiatric comorbidity were not collected.

Power analysis

We included a total of 250 subjects in the large groups. This sample size would identify drugs with a frequency of dependence of 10% in the main clinical group and 1% in the general population, with a significant level of 5% and a power of 80%, according to Altman (1994). The small group with severe alcoholics was mainly used as a contrast group and was not included in the power analysis.

Methods

We developed a questionnaire based on DSM-IV (American Psychiatric Association, 1994) to assess the dependence on PTD during the past 12 months (see Fig. 1). The misuse of illegal drugs was also assessed. All patients were asked if they had used one or more of the listed substances during the past 12 months. The questionnaire covered fourteen trademarks (tm) of benzodiazepines registered in Sweden in 1998 (diazepam, four tm; nitrazepam, three tm; flunitrazepam, two tm; oxazepam, lorazepam, triazolam, two tm, and alprazolam), zolpidem and zopiclone, four codeine-containing drugs, five dextropropoxyphene (DPX)-containing drugs, and three muscle relaxants. The patients were asked for length of use and the daily doses of each substance. Sex, age and occupation were also assessed. High-dose PTD dependence was defined as DDD ≥4.0, the definition of low-dose PTD dependence was set as DDD <4.0. The DDD were: diazepam 10 mg, oxazepam 50 mg, alprazolam 1 mg, nitrazepam 5 mg, flunitrazepam 1 mg, zolpidem 10 mg, zopiclone 7.5 mg, codeine 100 mg and dextropropoxyphene 200 mg (WHO Collaborating Centre, 2002).

Fig. 1.

Modified version of DSM-IV used in a study of dependence on legal psychotropic drugs among alcoholics in Sweden.

Ethics

All subjects were given both oral and written information from registered nurses at the three centres. Oral informed consent was given by all participants. The study was accepted by the Ethics committee at Lund University.

Statistical analyses

To determine the degree of significance, ‘chi-squared with Yates’ correction was used for comparing differences between proportions, and Student's t-test was used for interval data. A P-value of <0.05 was considered significant. Analyses were carried out with SPSS for Windows 7.5 (1997). For the main results 95% confidence intervals (CI) for the difference in proportions (p1–p2) were calculated, according to Altman (1994).

RESULTS

Demographic characteristics

Demographic features differed considerably between alcoholics and controls. The employment rate at DAD, KRC and VCK was 55, 41 and 92%, respectively. Alcoholics were unemployed more often than were healthy controls, (P < 0.001). Mean (±SD) age was 45 ± 10, 46 ± 10 and 37 ± 7 years at DAD, KRC and VCK, respectively. The alcoholics were older than the healthy controls, (P < 0.001). The number of men was 76, 96 and 51% at DAD, KRC and VCK, respectively. Alcoholics were more often male than were the healthy controls, (P < 0.001).

Dependence on legal PTD among alcoholics

The rates of dependence on different legal PTD in the three groups are presented in Table 1. The total rate of PTD-dependent alcoholics was higher in the institutionalized group than in the open-care setting (p1–p2 21%; 95% CI: 1%, 41%). Alcoholics were more often dependent on total PTD, BZD and zopiclone than were healthy controls (PTD: p1–p2 15%; 95% CI: 9%, 21%; BZD: p1–p2 14%; 95% CI: 8%; 20%; and zopiclone: p1–p2 5%; 95% CI: 1%; 8%).

View this table:
Table 1.

Dependence on legal psychotropic drugs among alcoholics in open and institutionalized care, respectively, and among healthy controls during the past 12 months, in a study of dependence on legal psychotropic drugs among alcoholics in Sweden

Legal psychotropic drugsAlcoholics in open care (%) (n = 130)Alcoholics in institutionalized care (%) (n = 23)p1–p2(95% CI)Alcoholics total group (%) (n = 153)Healthy controls (%) (n = 120)p1–p2(95% CI)
Benzodiazepines123018 (−2; 38)15114 (8; 20)
Zolpidem341 (−8; 10)303 (0; 6)
Zopiclone495 (−7; 17)505 (1; 9)
Codeine31310 (−4; 24)514 (0; 8)
Dextropropoxyphene297 (−5; 19)312 (−1; 5)
Legal psychotropic drugs143521 (1; 41)17215 (9; 21)

Seven per cent of the alcoholics in open care and 17% of the alcoholics in institutionalized care were dependent on more than one PTD, including different types of BZD, the most commonly used PTD (not significant (NS)). Among the BZD-dependent subjects in open care, nine reported dependence on diazepam, eight on oxazepam, four on nitrazepam, one on alprazolam and one on flunitrazepam. Among the BZD-dependent subjects in institutionalized care, three reported dependence on diazepam and three on oxazepam; two subjects were dependent on nitrazepam and two on alprazolam. Among the healthy controls, 1% was dependent on more than one drug.

Illegal drugs

The misuse rates of different illegal drugs in the three groups are presented in Table 2. Illegal drugs were used more frequently among the institutionalized alcoholics than among the alcoholics in open care (p1–p2 27%; 95% CI: 7%; 47%). Alcoholics had a higher rate of illegal drug abuse than did the healthy controls (p1–p2 12%; 95% CI: 7%; 17%).

View this table:
Table 2.

Use of illegal psychotropic drugs among alcoholics in open and institutionalized care, respectively, and among healthy controls during the past 12 months, in a study of dependence on legal psychotropic drugs among alcoholics in Sweden

Illegal psychotropic drugsAlcoholics in open care (%) (n = 130)Alcoholics in institutionalized care (%) (n = 23)p1–p2(95% CI)Alcoholics total group (%) (n = 153)Healthy controls (n = 120)p1–p2(95% CI)
Cannabis52621 (3; 39)808 (4; 12)
Amphetamine62216 (−1; 33)909 (4; 13)
Cocaine297 (−5; 19)303 (0; 6)
Heroin143 (−5; 11)202 (−0; 4)
Other01313 (−1; 27)202 (−0; 4)
Illegal psychotropic drugs83527 (7; 47)12012 (7; 17)

PTD-dependent alcoholics and misuse of illegal drugs

In open care, 33% of the PTD-dependent alcoholics and 4% of the non-PTD-dependent alcoholics had misused illegal drugs during the past 12 months (P < 0.001). In institutionalized care, the figures were 62 and 20%, respectively (NS).

Background data and PTD-dependence

The rate of PTD dependence was related to age, sex and employment. 20% of the female and 10% of the male alcoholics in open care were PTD-dependent (NS). There was no other relationship between PTD dependence and background data.

High- and low-dose PTD dependence versus length of dependence

BZD and related drugs. The DDD and the duration of dependence on BZD and related drugs among alcoholics in open and institutionalized care are presented in Fig. 2. Only four out of a total of 23 BZD-dependent alcoholics (2/16 in open care and 2/7 in institutionalized care) had developed high-dose BZD dependence. None of the five zolpidem- or the seven zopiclone-dependent alcoholics had developed high-dose dependence.

Fig. 2.

Dependence on benzodiazepines and related drugs among alcoholics in open and institutionalized care, respectively, related to reported length of dependence, in a study of dependence on legal psychotropic drugs among alcoholics in Sweden. Only the highest defined daily dose (DDD) for each patient was selected.

In open care, mean (±SD) DDD for low-dose BZD dependence was 1.1 ± 0.5 and for high-dose BZD dependence 11.5 ± 9.2. Among the institutionalized alcoholics, the mean DDD for low-dose BZD dependence was 1.7 ± 1.0 and for high-dose BZD dependence 5.5 ± 0.7. The mean DDD for zolpidem- and zopiclone-dependent patients in open care were 1.4 ± 0.7 and 1.1 ± 0.5, respectively. In institutionalized care, the figures were 1.0 and 1.5 ± 0.7, respectively. There was no positive correlation between DDD and length of use.

High-dose BZD-dependent alcoholics. All four high-dose BZD-dependent alcoholics were male, mean (±SD) age 34 ± 7 years. Two subjects were employed. In open care, one subject was dependent on three BZD (diazepam, DDD 18.0; nitrazepam, DDD 5.0 and flunitrazepam, DDD 5.0) as well as DPX (DDD 1.0). The second patient was dependent on diazepam only (DDD 5.0). In institutionalized care, one patient was dependent on three BZD (alprazolam, DDD 6.0 and diazepam, DDD 1.5) and oxazepam (DDD 0.8) as well as codeine (DDD 2.4) and DPX (DDD 2.0). The second patient was dependent on diazepam only (DDD 5.0; not plotted due to missing duration data). All four subjects had used cannabis, three amphetamine and one morphine during the past 12 months.

Legal opioids

The DDD and the duration of dependence on codeine and DPX among alcoholics in open and institutionalized care are presented in Fig. 3. Two codeine-dependent alcoholics (both in open care), out of a total of seven, had developed high-dose codeine dependence. None of the five DPX-dependent alcoholics had developed high-dose DPX dependence. In open care, the mean DDD for low-dose codeine dependency was 0.8 ± 0.5 and for high-dose dependency 8.8 ± 6.6. The mean DDD for codeine dependence in the institutionalized setting was 1.8 ± 0.8. The mean DDD for DPX-dependent alcoholics was 1.0 ± 0.1 in the open-care group and 2.5 ± 0.7 in the institutionalized group. There was no correlation between DDD and time.

Fig. 3.

Dependence on legal opioid derivates among alcoholics in open and institutionalized care, respectively, related to reported length of dependence, in a study of dependence on legal psychotropic drugs among alcoholics in Sweden. Only the highest defined daily dose (DDD) for each patient was selected.

High-dose codeine-dependent alcoholics

Both of the high-dose codeine-dependent alcoholics were male, mean age 48 ± 7 years and unemployed. Each was dependent on one additional PTD: zolpidem (DDD 1.0) and diazepam (DDD 2.0), respectively. Neither had used illegal drugs during the past 12 months.

DISCUSSION

The major findings were that (1) 14% of the alcoholics in open care and 35% in institutionalized care were dependent on PTD. BZD was the most misused drug. (2) Institutionalized alcoholism was associated with illegal drug misuse. (3) PTD dependence was associated with a higher frequency of illegal drug use. (4) Only 17% of the BZD-dependent alcoholics and 29% of the codeine-dependent alcoholics had developed high-dose dependence. (5) There was no positive correlation between DDD and length of use.

Limitations of the study

Differences in demographic baseline data could perhaps influence the main results but would hardly make them invalid. The absence of baseline psychiatric comorbidity data makes it difficult to analyse the importance of psychiatric comorbidity for the development of dependence. The results were based on anonymous, self-reported data without objective verification. However, Baillie and Mattic (1996) and Kan et al. (1997) have shown that self-reported data are reliable in the field of dependence research.

BZD and related drugs

Overall, the results are similar to previously reported clinical data, which have shown that 10–20% of alcoholics develop concomitant BZD misuse or dependence. 30% of the alcoholics in long-term institutionalized care were BZD-dependent, a figure corresponding well with Sallmén et al. (1997). In the present study, we used the same rehabilitation centre (KRC) as Sallmén et al. did for the inquiry of the long-term institutionalized alcoholics.

There were several cases of zolpidem and zopiclone dependence, supporting Soyka et al. (2000), who reported that substance misuse was considered a risk for later misuse of zolpidem. The use/dependence quotient was almost similar to BZD: 33, 25 and 38%, respectively, which contradicts the findings of Lader (1997), who reported that zopiclone had substantial advantages over BZD in terms of dependence and misuse potential.

Opioid-containing analgesics

A small number of alcoholics were dependent on codeine and DPX. More than half of these subjects were also BZD-dependent. Sproule et al. (1999) and Ng and Alvear (1993) reported that codeine and DPX were drugs of primary misuse and that problems with alcohol and sedatives were common. Johansson et al. (1997) showed that codeine dependence makes BZD taper more difficult.

High-versus low-dose BZD-dependent alcoholics

We evaluated high- and low-dose PTD dependence among alcoholics which, to our knowledge, has not been investigated before in a systematically sampled clinical population. The categorization of high-dose BZD dependence versus low-dose BZD dependence divided our sample into one large group with no increase of DDD and one small group with a considerable increase of DDD. There was a clear dividing line between the groups.

The small group of high-dose BZD-dependent alcoholics (4/23) consisted of men under 40 years of age. Two were dependent on more than one legal PTD. All four had used cannabis, three had used amphetamines and two opioids during the past 12 months.

This type of addiction has been previously described in different ways. Seivewright and Dougal (1993) described a population of 33 poly-drug, high-dose BZD (diazepam)-dependent subjects with a median DDD of 14.0. Kaufman (1976) distinguished between poly- and multidrug misuse. Polydrug misuse indicates the misuse of more than one psychoactive substance (not including opioids), whereas multidrug misuse involves two psychoactive substances other than alcohol, nicotine, caffeine or prescribed medication. Schuckit (1989) discussed a classification according to primary alcoholics, primary multidrug misusers or primary antisocial personalities. Our sample was too small to address the usefulness of the different concepts to high-dose BZD dependence.

Subjects with low-dose BZD dependence had different characteristics, with less poly- and multidrug misuse (32 and 26%, respectively). The subjects corresponded well to the description by Mueller et al. (1996), who prospectively studied 343 BZD-using subjects with anxiety disorders. 99 subjects had, and 244 subjects did not have, a history of alcohol misuse or dependence (DSM-III-R). The BZD-consumption was equal in the two cohorts at the 6-month follow-up. Six months later, the alcoholics used more BZD (chlordiazepoxide equivalents) than did the non-alcoholics (median 4.0 vs. 3.0 DDD; P < 0.02). Mueller also identified subjects with high-dose BZD tolerance, both alcoholics and non-alcoholics, but did not report actual numbers.

CONCLUSIONS

The prevalence of PTD dependence and the distribution of high- and low PTD dependence among alcoholics was systematically studied in two clinical populations. The total rate of PTD-dependent alcoholics was higher in the institutionalized group (35%) than in the open-care setting (14%) (p1–p2 21%; 95% CI: 1%, 41%). Alcoholics were more often dependent on PTD (17%) than were the healthy controls (2%) (p1–p2 15%; 95% CI: 9%; 21%). BZD were the most common PTD. Only four out of a total of 23 BZD-dependent alcoholics had developed high-dose BZD dependence.

REFERENCES

View Abstract