Abstract
This retrospective study aims to determine whether
there is a difference in the additional consumption of
alcohol between addicts treated with methadone or dihydrocodeine
(DHC) and untreated addicts injecting heroin.
1,685 patients admitted for opioid withdrawal between
1991 and 1997 were reviewed. Cross-reference tables
and multiple logistic regression analyses were carried
out. 28% of patients take more than 40 g of alcohol daily
(on average 176 g). We found that patients who are
treated with methadone or DHC drink alcohol significantly
more often daily than the heroin-dependent patients
(p ! 0.01). Using multiple regression analyses, the results
were confirmed. Additionally, we found that co-abuse of
alcohol was predicted by male gender, longer duration
of drug use, additional daily consumption of tetrahydrocannabinol
and daily consumption of benzodiazepines.

Alcohol consumption by opioid-addicted patients
treated with methadone or DHC presents a serious medical
problem. Co-abuse of alcohol will receive more attention.

Introduction
Consumption of other psychotropic substances during
substitution treatment of opioid addicts with methadone
and dihydrocodeine (DHC) may have a substantial impact
on morbidity, mortality and clinical course. While
consumption of illegal drugs is usually reduced during
substitution treatment, additional consumption of legal
psychotropic substances, especially of alcohol, has not
been examined in such detail. Chronic alcohol consumption
leads to a variety of somatic effects and diseases.
Therefore, where a high prevalence of regular or severe
alcohol consumption exists among patients in substitution
programs, the advantages and risks should be carefully
weighed up. Further, to the consumption of respirantdepressive
opioids, additional complications and dangers
are involved when taking other psychotropic substances
with respiration-depressive effects, such as alcohol, benzodiazepines
or barbiturates. 62–72% of the patients, who
had been treated for overdosing, had consumed various
substances [1, 2], of which 23–35% comprised alcohol. In
cases of fatal overdosing, abuse with multiple substances
was reported in 71–92% of patients [3–7]. In 41–51% of
these deaths, alcohol was identified. In addition to the

acute risk of an overdose from combined consumption of

a number of psychotropic substances, the consumption of

alcohol, in contrast to opioids, leads to chronic permanent
damage and disease concerning almost all aspects of medicine
[8–11].
On the one hand, a number of studies have ascertained
that the supplementary consumption of illegal drugs such
as heroin and cocaine may be reduced by methadone
maintenance treatment programs [12–15] and codeine
maintenance programs [16]. On the other hand, it has
often been reported that about 30% of the patients in
methadone maintenance treatment programs have an
alcohol problem or are even alcoholics [17–19]. It is not
clear, however, whether the patients already had alcohol
problems before their entry into the methadone maintenance
treatment. The North Rhine Westphalia study on
the efficacy of outpatient medical rehabilitation with
methadone maintenance indicated that over time the
number of patients being abstinent of alcohol increased
during treatment [20].
In Germany, the critical dose is often stated as 60 g for
men and 40 g for women [21]. More than 8,000 deaths of
people aged 15–29 years in Europe in 1999 were attributable
to alcohol [22].
In addition to the risk of overdosing, alcoholism for
example leads to an increased risk of long-term secondary
physical sicknesses [8, 9]. With regard to narcotics fatalities,
toxicological data from southern Bavaria – the same
area as in our study – indicate that alcohol is a frequent
covariant in drug-related deaths and in patients treated
with codeine (27% each) and that it is less frequent in
methadone patients (16%) [23].
Those studies concerned with supplementary consumption
within methadone and codeine maintenance
treatment programs have mainly focused on the supplementary
consumption of illegal drugs. We further consider
that in Germany, and especially Bavaria, where alcohol
is everywhere easily available and where alcohol consumption
is well established (e.g. the ‘Oktoberfest’), heroin
addicts consider alcohol consumption analogous to a
‘normal’ individual’s regular consumption of alcohol, and
more so when heroin becomes more difficult to obtain.
We thus present our hypothesis that those patients treated
with methadone or DHC drink less alcohol than the
untreated patients injecting illegal heroin.
Subjects and Methods
Sample
All drug-addicted patients voluntarily admitted to inpatient detoxification
treatment between April 1991 and December 1996, in
whom, according to ICD-10 criteria, an opioid or multiple-substance

addiction had been diagnosed, were included in the study. Of all
patients enrolled in the treatment program between April 1991 and
December 1996, those subjects coming for a second or third detoxification
visit within this time period had to be excluded, except for the
first visit, otherwise the assumption of independent observations
would have been violated. Patients could come in of their own volition,
and previous contact with a counsellor or physician was not
necessary. The treatment strategy called ‘qualified detoxification’ has
been described elsewhere [24].
Measures
On the day of admission, the daily intake of psychotropic substances
over the preceding 6 months was established. The patients
were questioned individually regarding their intake of opioids, especially
heroin, codeine/DHC and methadone (D, L-methadone, levomethadone),
and of other psychotropic substances such as nicotine,
benzodiazepines, barbiturates, cocaine, cannabis, amphetamines/
amphetamine derivatives and alcohol. The alcohol intake
was noted in grams of alcohol per day. 500 ml beer was calculated as
20 g alcohol, 500 ml cognac (40%) or vodka (40%) as 160 g alcohol
[21]. The declarations were verified by an immunoassay urine test
(Triage®) and a KIMS test (kinetic interaction of microparticles in a
solution).
The patients were divided into 3 groups according to the preferred
opioid of each individual. Those who daily received methadone
or levomethadone (summarized as methadone) were included
in group 1, those who took codeine or DHC daily (summarized as
DHC) in group 2 and those daily consuming heroin in group 3. If
primarily two opioids were taken daily, then the priority sequence of
heroin before methadone before DHC was decisive. If none of the
opioids were taken daily, but rather several of them alternatively,
then these patients were not introduced into any of the 3 groups.
Data concerning the daily intake of other psychotropic substances
were collected for each group, as well as the gender of the patient,
duration of addiction, age at first opioid use, age, marital status, history
of imprisonment, history of suicide attempts and employment.
Alcohol was separated and selected as a dependent variable (co-abuse
of alcohol). Co-abuse of alcohol was defined as consumption of more
than 40 g alcohol per day. The 40-gram value was chosen since it is
the stated critical dose of alcohol per day in most other published
reports [21, 22, 25, 26].
Statistics
After cross-tabulation and bivariate analysis, a logistic regression
model was established. Bivariate analyses were performed for the
variable of interest and all further potentially relevant variables.
Since preliminary analysis indicated nonlinear associations involving
type of opioid dependency, age, duration of drug use, age at first
opioid use, marital status and history of imprisonment, these variables
were transformed from ordinal to categorical variables. Results
are summarized by reporting a ‘full model’ that includes all investigated
variables regardless of their statistical significance. Tests for
interactions were used as a check on the uniformity assumption
under which multiple regression estimates are derived.

Results
During the observation period from April 1991 to
December 1996, 1,656 patients were voluntarily admitted
to stationary qualified withdrawal treatment, 36% women
and 64% men. 537 of these were patients readmitted
within the given time frame. Of these patients only the
first admission was included in the study in order not to
violate the assumption of independent observations. 49
patients were not included since they had daily consumed
several different opioids. 137 patients, 36% women and
64% men, who daily received methadone, were included
in group 1. 658 patients, 34% women and 66% men, due
to their daily intake of codeine/DHC were placed into
group 2. 275 patients, 39% women and 61% men, daily
consuming heroin formed group 3. The average age was
30.7 years in group 1, 28.9 years in group 2 and 28.3 years
in group 3. The average duration of addiction to opioids
was substantiated as 10.5 years in group 1, 8.9 years in
group 2 and 8.2 years in group 3.
301 patients consumed more than 40 g alcohol per day.
These were evaluated as positive for the dependent variable
‘co-abuse of alcohol’. On average, alcohol consumption
was 176 g/day (table 1).
Contrary to our hypothesis, the bivariate analysis indicated
that patients who are in a methadone or DHC maintenance
treatment program daily drink alcohol significantly
more often than the heroin-dependent patients (p !
0.01). Fewer patients who were treated with DHC
(31.3%) drink alcohol than patients who were treated with
methadone (36.5%). Table 2 presents bivariate analyses
on the key variable and possible confounding variables
predicting co-abuse of alcohol. As shown, co-abuse of
alcohol was predicted by male gender, older age, longer
duration of drug use, additional daily consumption of
tetrahydrocannabinol (THC), daily consumption of barbiturates
and daily consumption of benzodiazepines.
Table 3 shows the results of a multiple logistic regression
analysis. The key finding that patients consuming
heroin drink less alcohol than patients who were treated
with DHC or methadone persists even when all investigated
variables regardless of their statistical significance
were included in the model. Age and daily consumption
of barbiturates turned out not to be a significant predictor
of co-abuse of alcohol. None of the other previously
entered significant variables like gender (odds ratio, OR,
0.61, 95% confidence interval, CI, 0.44–0.84, p ! 0.01),
duration of drug use (OR 0.19, 95% CI 0.05–0.74, p !
0.05), daily consumption of benzodiazepines (OR 0.52,
95% CI 0.38–0.71, p ! 0.001) and daily consumption of
Table 1. Consumption of alcohol by opioid patients consuming
more than 40 g of alcohol per day (defining ‘co-abuse of alcohol’)
Quantity of alcohol
g/day
Patients (n = 301)
%
40–60 12.0
61–120 23.9
121–180 27.2
181–240 13.3
More than 240 23.6
Total 100
THC (OR 0.67, 95% CI 0.47–0.95, p ! 0.05) became
insignificant, even though our variable of interest ‘type of
opioid dependency’ remained only marginally significant
(OR 2.38, 95% CI 1.42–4.00, p ! 0.001; table 3).
Discussion
The aim of this study was to investigate the association
between the consumption of alcohol and the daily preferred
opioid (methadone, DHC or heroin). The results
indicate that patients consuming heroin drink less alcohol
on a regular basis. In comparison to the patients who consume
DHC, patients in a methadone maintenance program
drink alcohol more often (36.5% of the methadone
patients vs. 31.3% of the DHC patients). The hypothesis
had therefore to be rejected. One explanation might be
due to the effect of the substance, i.e. that heroin is the one
substance with which the desired effect according to the
addicts may best be attained. Other opioids such as methadone
and DHC do not completely produce the desired
effect so that other additional substances, e.g. alcohol, are
consumed. In published reports, problematic alcohol consumption
by opiate users is given as varying between 12%
[27], 32% [18, 28] and 42% [27]. In this spectrum, the
individual subgroups of our study are found, with 16%
problem drinkers in the heroin group, 31.3% in the
codeine group and 36.5% in the methadone group. It is
becoming increasingly possible in Europe to use various
opioids as substitutes. This is understandable when one
considers the high mortality rate of drug addicts, which,
in a meta-analysis, was found to be more than 13 times
greater than in the average populations’ equivalent age
group [29]. Researchers and practitioners are therefore
required to re-evaluate indication positions and to annotate
advantages and disadvantages of each substance. An

important target criterion should be the parallel consumption
of further psychotropic substances, especially alcohol.
It is known that alcohol has a toxic effect on all
organic systems and that therefore chronic consumption
of alcohol causes many severe illnesses. If the patients
now begin to drink alcohol during the methadone or DHC
substitution treatment, then a serious responsibility falls
upon the physicians to recognize, and where necessary to
prevent, the injurious alcohol consumption being caused
perhaps partly by the substitution treatment itself. Further
studies had to be conducted to prove the association
between methadone or other substitutes and drinking
alcohol, and to investigate how long patients can be
treated with methadone before the risk of excessive alcohol
consumption increases significantly. A previous study
showed that 29% of the patients initially began to drink
during the course of methadone maintenance treatment
programs and that they are younger in comparison to nonopioid-
addicted alcoholics [30]. Numerous studies continue
to substantiate that patients during substitution
treatment with methadone consume fewer illegal drugs,
that the risk of HIV infection is reduced and that drugrelated
crime rates fall [20, 31–33]. DHC, which is frequently
prescribed in Germany, appears to improve the
addicts’ situation just as well as methadone [16]. In all
these studies, the illicit drug co-abuse but not alcohol was
a target criterion. In a new study, it was reported that only
one third of the patients in methadone maintenance treatment
abstained from alcohol at the intake and follow-up
[34]. It is known that with a lower dosage of methadone,
co-abuse is higher [35]. Apparently, patients who discover
they are hepatitis C positive, reduce their alcohol con

sumption considerably [28]. In principle it should be taken
into consideration that in Germany alcohol is culturally
accepted. Possibly the results would be somewhat different
in a more abstinence-oriented culture.
Independently of the preferred daily opioid the following
variables were associated with co-abuse of alcohol:
(1) in the total population, males more often have an alcohol
problem than females [36–38] – so it was not surprising
that male opioid addicts more often had drunk alcohol
daily than female opioid addicts; (2) that older patients,
consuming drugs over a longer period, drink more alcohol
seems realistic since the drug alone no longer appears to
have the desired effect; (3) this would be an analogous
explanation of the significantly higher benzodiazepine
and cannabis consumption. Barbiturates are consumed
(in total) in such small quantities that its use is difficult to
assess with certainty. The subgroup of cocaine users is
also so small that an assessment here appears to be of little
practical value.
The multiple logistic regression analysis result proves
that patients who have been treated with methadone or
have taken DHC drink quite significantly more than
patients who have injected heroin intravenously. To our
knowledge this has not yet been described thus in any other
literature. It is nevertheless necessary to consider that
this concerns a retrospective study. Being male, additional
consumption of benzodiazepines and THC was also
confirmed as independent predictor for alcohol co-abuse.
Increased age was not asserted as an independent predictor,
but duration of drug use was. This may be explained
by the fact that increased age implies a longer period of
drug use, since drug addicts in this study on average first
began injecting heroin at 20 years of age. In the first year
of opioid consumption, it appeared that the opioid itself
achieved the desired effect. Not quite every sixteenth
patient drinks alcohol additionally. Already from the second
year onward every fourth patient drinks alcohol daily.
From the sixth to the tenth year, the proportion was
30%, from the eleventh to the fifteenth year 40%. The
effect of the opioid itself no longer appears to be sufficient.
With an opioid consumption of more than 15 years,
slightly fewer patients drank alcohol daily (32.4%). This is
comparable to a British study in which a 32% harmful
daily alcohol consumption has been diagnosed, where the
average duration of opioid dependence was 17 years and
the average age was 39 [28].
The results of this study suggest that co-abuse of alcohol
should receive more attention in further studies of the
effectiveness of substitution treatment programs with various
substitution drugs. Particularly, a comparison between
heroin and methadone should prove very interesting.
More longitudinal studies are still needed.
Acknowledgements
This study was supported by the ‘Modellprogramm Kompakttherapie
im Verbund der Drogenhilfe 1990–1995’ from the Department
of Health of the Federal Republic of Germany. The authors
would like to thank the staff of the detoxification unit ‘villa’ for their
help with data collection and entry.

Read the whole Material here: alcoholConsumption.in.methadone