Association of the DRD2 gene Taq1A polymorphism and
alcoholism: a meta-analysis of case–control studies
and evidence of publication bias
MR Munafo `1, IJ Matheson1and J Flint2
1Department of Experimental Psychology, University of Bristol, Bristol, UK and2Wellcome Trust Centre for Human Genetics,
University of Oxford, Oxford, UK
We investigated the association of the dopamine D2 receptor (DRD2) Taq1A polymorphism and
alcoholism, using meta-analytic techniques, and specifically undertook an investigation of
possible publication bias. Potential publication bias represents a genuine risk to the integrity
of published research, but its impact has rarely been documented. We observed a small effect
of the DRD2 Taq1A polymorphism on risk of alcoholism, indicating increased alcoholism in
individuals possessing the A1 allele of the Taq1A polymorphism (OR=1.21, 95% CI 1.13–1.30,
P<0.001). This association remained significant when data from samples of European and
East Asian ancestry were analyzed separately. We did not find evidence for association in
high-severity alcoholism compared to low-severity alcoholism. Removing the first published
study significantly reduced the magnitude of the pooled effect size estimate, although the
association remained significant. In addition, we observed evidence for possible publication
bias and for the strength of individual study effect size to be inversely related to year of
publication. These results support the association of the DRD2 Taq1A polymorphism with
alcoholism. This conclusion is qualified by the possibility of publication bias in the literature
and the observed between-study heterogeneity, which indicates that the observed association
may differ in strength between populations or may not exist at all in some populations.
Molecular Psychiatry (2007) 12, 454–461. doi:10.1038/sj.mp.4001938; published online 9 January 2007
Keywords: DRD2; ANKK1; genetics; alcoholism; meta-analysis; publication bias
The human central dopamingeric system is widely
considered to play an important role in substance use
and the development of subsequent dependence.
Evidence for a role for this system extends to a range
of psychoactive substances, including opiates, co-
caine, nicotine and alcohol.1–3In consequence, a great
deal of attention has been devoted to determining
whether variation in genes with a dopaminergic
function could account for the heritable variation in
susceptibility to substance abuse. In particular, the
dopamine D2 receptor (DRD2) gene on chromosome
11 (q22–q23) has been widely studied.4
Following a report in 19905that the A1 allele of the
Taq1A polymorphism (rs1800497) of the DRD2 gene, a
C>Tsubstitution located in a noncoding region of the
DRD2 locus, was associated with alcoholism, several
studies have attempted to replicate the finding.
Despite the large number of individual studies,
results have been equivocal. A 1993 report found
that heterogeneity between populations was consid-
erably greater than differences between alcoholics
and controls overall, indicating that the positive
findings could be due to sampling error and popula-
tion stratification.6By contrast, a recent meta-analy-
sis,7surveying 55 studies involving almost 10000
participants, found that the A1 allele was signifi-
cantly more likely to be found in the substance abuse
groups than in controls. Nevertheless, despite the
large number of studies, the effect was only detected
at a 0.05 significance threshold.
It remains unclear whether inconsistency in indivi-
dual study results arises from ancestral variation,
phenotypic variation, sampling variation or some other
source of heterogeneity. Identifying these sources is a
major challenge for the genetic analysis of any
complex trait, not just alcoholism. Fortunately, the
large number of studies of the DRD2 Taq1A1 poly-
morphism makes it possible to investigate heterogene-
ity within the context of a meta-analysis. For example,
in the meta-analysis reported above, the association
was not significant when studied in samples of non-
European ancestry, but significant when studies were
limited to those assessing the association of the A1
allele and severe substance dependence.7
Received 13 July 2006; revised 6 October 2006; accepted 30
October 2006; published online 9 January 2007
Correspondence: Dr MR Munafo `, Department of Experimental
Psychology, University of Bristol, 12a Priory Road, Bristol BS8
Molecular Psychiatry (2007) 12, 454–461
& 2007 Nature Publishing Group All rights reserved 1359-4184/07 $30.00
We undertook an investigation of another source of
heterogeneity: publication bias. Potential publication
bias represents a genuine risk to the integrity of
published research. However, although there are
grounds for believing publication bias to be present
in the psychiatric genetic literature (e.g., because of
the relative difficulty in interpreting nonsignificant
results compared to statistically significant results
within a null hypothesis testing framework), its
impact has rarely been documented. This is because
methods to detect publication bias require relatively
large numbers of published studies.
The methods that may be used to detect possible
publication bias are of two types: graphical and
statistical. The funnel plot is a commonly used
graphical test to assess publication bias in meta-
analytic data sets.8The rationale behind a funnel plot
analysis is that if all studies come from a single
population, then the plot should look like a funnel
with the diameter of the funnel decreasing (i.e., effect
size estimate becoming more accurate) as sample size
increases.9However, there are often insufficient large
studies to form the apex of the predicted funnel.
Formal tests of publication bias are statistical analogs
of the funnel plot.8,10Begg and Mazumdar10proposed
an adjusted rank correlation method to examine the
association between the effect estimates and their
variances, whereas Egger et al.8
approach that tests for asymmetry in the funnel plot
and corresponds to a weighted regression of effect
sizes on their standard errors, where the weights are
inversely proportional to the variance of the effect
size. Asymmetry may be a result of the non-publica-
tion of nonsignificant studies, and this is a formal test
of the null hypothesis that such a bias is not present.
The regression method is more sensitive than the rank
correlation approach, but the sensitivity of both
methods is generally low in meta-analyses based on
less than 20 studies.
Although the Egger method is widely used, it has
been criticized for being intrinsically biased11,12
owing to the correlation between the effect size
estimate and its standard error when using odds
ratios (OR),13which may lead to a high type I error
rate (i.e., falsely identifying possible publication
bias). Alternative methods have been suggested,
including the Macaskill method, which corresponds
to a weighted regression of effect sizes on their sample
sizes,14and a modified Macaskill method, which
employs the inverse of individual study sample
sizes.13These methods have been argued to be more
conservative than the Egger method, and simulation
data suggest that the modified Macaskill method is
The large number of DRD2 Taq1A studies means
that we are able to apply a formal test of publication
bias. We attempted to replicate the findings of a recent
meta-analysis7and include reports not included in
this earlier study (e.g., those published subsequently).
We restricted our analysis to case–control studies.
Although it has recently been shown that the Taq1A
variant alters an amino acid in a protein kinase gene
(ANKK1) near the DRD2 locus,15we refer to the
variant throughout as the DRD2 Taq1A polymorph-
ism, as this is the nomenclature used in the majority
of published studies to date.
Selection of studies for inclusion
Case–control genetic association studies of the DRD2
Taq1A polymorphism in healthy controls and clini-
cally diagnosed alcoholic patients were included.
Studies reporting data on either single-sex or both
male and female participants of any ethnic origin
were included. Studies with data for only alcoholic
patients or only healthy participants were excluded,
as were family-based studies that only reported
transmission disequilibrium to affected offspring.
The principal outcome measure was the allelic OR
for the Taq1A polymorphism and alcoholism case
The search was performed on three databases:
PubMed, PsycInfo and Medline. These databases
were searched from the first date available in each
database up to 30 June 2006, using the search terms
‘alcohol’, ‘alcoholism’, ‘DRD2’, ‘dopamine D2’ and
‘Taq1A’. Once articles had been collected, bibliogra-
phies were then hand-searched for additional refer-
The abstracts of studies identified by these search
strategies were then examined with reference to the
inclusion and exclusion criteria. Duplications were
deleted and the whole text of each reference was then
checked to further establish whether the study met
the study inclusion criteria. Studies that reported
previously published data were excluded.
For each study, the following data were extracted
independently by two authors (MM and IM) using
standard forms: (1) author(s) and year of publication;
(2) methods (country of origin, dominant ancestry of
sample, case and control sample size, diagnostic
criteria for alcoholism case status, candidate gene,
polymorphism, statement of Hardy–Weinberg equili-
brium, method of genotyping); (3) data (number of
participants in control and case groups, mean age and
sex ratio by allele frequency). Genotype frequencies
were used to calculate whether or not these deviated
among controls. Ancestry was coded as European,
East Asian or Other (which included cases where
ancestry was stated as mixed or not stated). Discre-
pancies were resolved by mutual consent.
Analysis of data
Data were analyzed using the Comprehensive Meta-
analysis (v.2) statistical software package. A P-value
of 0.050 was retained throughout.
DRD2 gene and alcoholism
MR Munafo ` et al
Data were initially analyzed within a fixed-effects
framework and OR pooled using inverse variance
methods to generate a summary OR and 95%
confidence interval (CI). A fixed-effects framework
assumes that the effect of allele frequency is constant
across studies and between-study variation is con-
sidered to be due to chance or random variation. The
assumption was checked using a w2test of goodness of
fit for homogeneity. The significance of the pooled OR
was determined using a Z-test.
Where there was evidence of a significant associa-
tion between DRD2 Taq1A1 allele frequency and
alcoholism case status in the presence of significant
between-study heterogeneity, a random-effects frame-
work was employed, with ORs pooled using DerSi-
monianand Laird methods.
framework assumes that between-study variation is
due to both chance or random variation and an
individual study effect. Random-effects models are
more conservative than fixed-effects models and
generate a wider CI. The significance of the pooled
OR was determined using a Z-test.
Stratified analyses by sample ancestry and alcohol-
ism severity were conducted in order to assess
potential moderating effects of these variables. Stu-
dies with samples of predominantly European or East
Asian ancestry were combined separately and the
difference in pooled OR was determined using a Z-
test. For studies that identified low- and high-severity
subgroups and reported data separately for these
subgroups (e.g., alcohol abuse vs alcohol depen-
dence), these subgroups were combined separately
and compared with each other. Such groupings were
made on the basis of the classifications implemented
by individual studies.
The OR of the first published study was compared
to the pooled OR of the remaining studies using a
Z-test, as there is evidence for a substantially greater
estimate of effect size in the first published study.16
Funnel plots were created in order to assess potential
ascertainment bias by plotting individual study log
OR against the standard error of the log OR.
Ascertainment bias was also assessed using the Egger
test8and the modified Macaskill test.13
Description of studies
A total of 40 studies published between 1990 and
2006 were identified by the search strategy, met the
inclusion criteria and contributed to the meta-analy-
sis.5,6,17–54The characteristics of these studies are
described in Table 1.
Twenty-five studies reported data on samples of
predominantly European ancestry, eight on samples
of predominantly East Asian ancestry and seven on
samples of Other ancestry. Three studies reported
DRD2 genotype frequencies for controls that deviated
significantly from Hardy–Weinberg equilibrium (Bo-
los et al., 1990; Comings et al., 1994; Lu et al., 1996).
Four studies used DSM-IV criteria for assessing
schizophrenia case status, whereas 26 used DSM-III-
R criteria, five used ICD-10 criteria and one each used
Feighner, DIGS and RDC criteria, whereas two did not
state the criteria used.
When all studies (k=40) were included, there was
evidence of a significant association between DRD2
Taq1A1 allele frequency and alcoholism case status
(Z=5.50, P<0.001, OR=1.21, 95% CI 1.13–1.30).
There was evidence of significant between-study
heterogeneity (w2=92.87, P<0.001), but when
the analysis was re-run within a random-effects
framework the evidence for association remained
statistically significant (Z=4.39, P<0.001, OR=1.29,
95% CI 1.15–1.45). These results are presented
graphically in Figure 1.
When the first published study5was removed from
the analysis (k=39), there was evidence of a sig-
nificant association between DRD2 Taq1A1 allele
P<0.001, OR=1.20, 95% CI 1.11–1.28). There was
evidence of significant between-study heterogeneity
(w2=67.68, P=0.002), but when the analysis was
re-run within a random-effects framework the evi-
dence for association remained statistically signifi-
cant (Z=4.23, P<0.001, OR=1.23, 95% CI 1.12–1.36).
The comparison of the effect size indicated by the
first published study (OR=8.73) compared to the
pooled effect size for subsequent studies (OR=1.20)
indicated a significant difference within both a fixed-
effects (Z=5.02, P<0.001) and a random-effects
framework (Z=4.92, P<0.001). Meta-regression in-
dicated a significant negative association between
year of publication (corrected for month of publica-
tion) and individual study effect size (Z=?2.15,
P=0.032), with this trend reflecting a decrease in
individual study effect size over time. These data are
presented graphically in Figure 2.
The removal of three studies that reported DRD2
genotype frequencies for controls that deviated
significantly from Hardy–Weinberg equilibrium23,26,47
did not alter these results substantially.
When studies that recruited samples of predomi-
nantly European ancestry were analyzed separately
(k=24), there was evidence for a significant associa-
tion of DRD2 Taq1A1 allele frequency and alcoholism
case status (Z=3.89, P<0.001, OR=1.19, 95% CI
1.09–1.29). There was evidence of significant be-
tween-study heterogeneity (w2=43.44, P=0.009),
but when the analysis was re-run within a random-
effects framework the evidence for association re-
mained statistically significant (Z=3.16, P<0.001,
OR=1.22, 95% CI 1.08–1.38).
When studies that recruited samples of predomi-
nantly East Asian ancestry were analyzed separately
(k=8) there was evidence for a significant association
of DRD2 Taq1A1 allele frequency and alcoholism case
status (Z=2.11, P=0.034, OR=1.17, 95% CI 1.01–
DRD2 gene and alcoholism
MR Munafo ` et al
1.35). There was no evidence of significant between-
study heterogeneity (w2=2.46, P=0.929). When the
analysis was re-run within a random-effects frame-
work, however, there was no evidence for a signifi-
cant association (Z=1.54, P=0.120, OR=1.10, 95%
The comparison of the pooled effect size for studies
of participants of predominantly European ancestry
(OR=1.19) compared to the pooled effect size for
studies of participants of predominantly East Asian
ancestry (OR=1.17) did not indicate a significant
difference (Z=0.17, P=0.865).
When studies that explicitly recorded allele frequen-
cies in cases with severe and mild alcoholism
separately (k=10) were analyzed with mild alcohol-
ism cases as the comparison group, there was no
evidence for a significant difference in DRD2 Taq1A1
allele frequency and cases of severe and mild
alcoholism (Z=1.15, P=0.249, OR=1.13, 95% CI
0.92–1.40). There was evidence of significant be-
tween-study heterogeneity (w2=17.36, P=0.043).
A visual inspection of a funnel plot of 1/S.E. against
effect size estimate suggested evidence of ascertain-
ment bias, due to asymmetry in the plot in the
predicted direction, with a relative lack of low
accuracy (i.e., small) studies indicating no effect or
a direction of effect opposite to that reported in the
first published study. Both the Egger test (t=3.31,
Characteristics of included studies
StudyYear Case n Case N Control nControl NAncestryDiagnosisHWE
Blum et al.5
Bolos et al.23
Blum et al.22
Noble et al.48
Parsian et al.51
Comings et al.26
Gelernter et al.34
Goldman et al.35
Cook et al.28
Arinami et al.19
Goldman et al.6
Amadeo et al.17
Comings et al.27
Geijer et al.32
Higuchi et al.39
Noble et al.49
Sander et al.53
Finckh et al.29
Heinz et al.37
Chen et al.24
Lu et al.47
Chen et al.25
Goldman et al.36
Hietala et al.38
Kono et al.43
Lawford et al.44
Lee et al.45
Ishiguro et al.40
Gelernter and Kranzler33
Sander et al.54
Ovchinnikov et al.50
Bau et al.20
Samochowiec et al.52
Anghelescu et al.18
Limosin et al.46
Foley et al.30
Karaoguz et al.41
Konishi et al.42
Freire et al.31
Berggren et al.21
n, number of T alleles.
N, number of C and T alleles.
DRD2 gene and alcoholism
MR Munafo ` et al
(t=2.90, P=0.006) also indicated the presence of
such bias. These data are presented graphically in
A pooled OR corrected for possible publication bias
was calculated, using Duval and Tweedie’s trim-and-
fill method,55which is an extension of the funnel plot
and themodifiedMacaskill test
method. This removes studies with outlying effect
size values until symmetry is achieved, and then
replaces these along with imputed ‘mirror’ values in
order to retain symmetry. This indicated a reduced
pooled OR (OR=1.17, 95% CI 1.09–1.25), although
this was still statistically significant, which corre-
sponds to 0.2% of phenotypic variance.
indicates significant association between DRD2 Taq1A1 allele frequency and alcoholism case status (P<0.001). Bars
represent individual study 95% CI, with a central block proportional to study size. The summary diamond bar represents the
pooled effect size estimate and 95% CI.
Meta-analysis of case–control studies of DRD2 Taq1A allele frequency and alcoholism case status. Meta-analysis
effect size estimate (log OR). Year of publication (corrected
for month of publication) is negatively associated with
individual study effect size (P=0.032), with this trend
reflecting a decrease in individual study effect size over
Association between year of publication and
estimate (log OR). Asymmetry in the plot in the predicted
direction (i.e., a relative lack of low accuracy or small
studies which indicate no effect or a direction of effect
opposite to that reported in the first published study)
suggests possible publication bias.
Funnel plot of accuracy (1/S.E.) and effect size
DRD2 gene and alcoholism
MR Munafo ` et al
The results of our meta-analysis agree with those
recently published in finding a significant but small
effect of the DRD2 Taq1A polymorphism on risk of
alcoholism. Combining all studies, we found a
statistically significant OR of 1.21, indicating in-
creased alcoholism in individuals possessing the A1
allele of the Taq1A polymorphism. This association
remained significant when data from samples of
European and East Asian ancestry were analyzed
separately. We did not find evidence for association in
high-severity alcoholism compared to low-severity
alcoholism, although the direction of effect was
consistent with this possibility. We found that
removing the first published study significantly
reduced the magnitude of the pooled effect size
estimate, although the association remained signifi-
cant. In all cases except for the analyses of samples of
East Asian ancestry, there was evidence of significant
between-study heterogeneity, although the observed
associations were robust to the application of a
In addition, we observed evidence for possible
publication bias and for the strength of individual
study effect size to be inversely related to year of
publication. This observation is important as the
literature on the association of the DRD2 Taq1A
polymorphism and alcoholism case status is one of
the few in the psychiatric genetics literature where
publication bias can be examined with reasonable
power, given the large number of studies. It is
noteworthy that studies of potential publication bias
do not themselves appear to demonstrate evidence of
publication bias.56Once we had corrected the pooled
effect size estimate for possible publication bias we
found that, if the association between the DRD2
Taq1A polymorphism and alcoholism case status is
real, the single nucleotide polymorphism (SNP) likely
accounts for 0.2% of phenotypic variance. In addi-
tion, given a minor allele frequency of 0.3 and a
prevalence of alcoholism of 5%, for an alpha level of
0.05, in excess of 1500 cases and a similar number of
controls would be required in order to achieve 80%
power to detect significant association.
It should be noted that publication bias is not the
only explanation for an asymmetrical funnel plot, and
the results of formal tests of bias based on the funnel
plot, such as those used in this study, should
therefore be interpreted with caution. Other possibi-
lities include other selection biases (e.g., English
language bias or multiple publication of small
studies), true heterogeneity (e.g., differences in effect
between populations of differing ancestry), data
irregularities (e.g., poor methodological quality in
small studies), artifacts (e.g., differences due to effect
measure employed) or chance8,57There are also
possible explanations for asymmetry which are
specific to genetic studies, such as the violation of
tempted to reduce the impact of these other sources
of bias, for example, by not explicitly excluding
non-English language journals and by attempting to
identify cases of multiple publication.
The between-study heterogeneity observed in the
majority of our analyses may be due to any potentially
relevant differences between the study designs and
methodologies, such as populations from which the
study samples are drawn. In the case of genetic
association studies, possible causes of between-study
heterogeneity include, for example, the possibility
that an association exists in one population but not
another, that different studies did not use comparable
measures of phenotype, or that allelic distributions
deviated from Hardy–Weinberg equilibrium in some
studies. However, we did not observe a difference in
the strength of association between samples of
European and East Asian ancestry, nor did we find
evidence that samples of low-severity alcoholics
differ from those of high-severity alcoholics in
prevalence of the A1 allele, and the between-study
heterogeneity remained significant in the majority of
these analyses, suggesting that the observed hetero-
geneity is not due to these factors. We attempted to
accommodate this heterogeneity in our analyses by
implementing a random-effects framework in the
presence of significant heterogeneity. It should be
noted that fixed-effects and random-effects analyses
address fundamentally different research questions.
The former asks what the best estimate of the true
effect size is in the population studied, whereas the
latter asks what the range and distribution of effect
sizes is in the distribution of populations studied.57
Therefore, the calculation of the mean of the distribu-
tion of population effect sizes (random-effects model)
provides quite different information to the calculation
of the mean of the distribution of sample effect sizes
(fixed-effects model). In other words, our analyses
suggest that there may be populations in which the
DRD2 Taq1A polymorphism is associated with alco-
holism and others where it is not, or the strength of
this association is different (although we were unable
to identify any such populations).
One of the reasons for the interest in the Taq1A
variant is that it may alter the function of the nearby
DRD2 gene. The SNP has been reported to affect
dopamine receptor DRD2 availability in post-mortem
striatal samples48,59and there is also evidence from in
vivo studies for an association between the A1 allele
and lower mean relative glucose metabolic rate in
dopaminergic regions in the human brain.60Positron
emission tomography studies have indicated that this
allele is also associated with low receptor density.61
Evidence that the Taq1A variant alters an amino acid
in the ANKK1 protein kinase gene, near the DRD2
locus,15does not rule out an effect on the DRD2 gene:
data from the HapMap project reveal that the variant
is in linkage disequilibrium with other variants in the
DRD2 gene, but not with variants in the ANKK1 gene.
Thus, it is possible that additional functional variants
in DRD2 are contributing to the observed association
DRD2 gene and alcoholism
MR Munafo ` et al
Another, more speculative, possibility is that
ANKK1 may exert an effect on dopaminergic neuro-
transmission itself. The function of many proteins can
be influenced or regulated by a process of phosphor-
ylation of key amino-acid residues within the protein.
This process can influence factors such as the affinity
of the protein for ligands that bind to it, such as
dopamine to its transporter in this instance. Phos-
phorylation can also influence other aspects of
activity, and kinases catalyze these phosphorylation
processes. ANKK1 might therefore be a kinase that
acts on the transporter to influence its activity. If this
were to be so, it might explain how a polymorphism
in a gene that was not the transporter itself might
relate to dopaminergic activity, so that the poly-
morphism in ANKK1 may influence the activity or
regulation of the kinase, thereby influencing the
activity of the transporter (DJK Balfour, personal
communication, 23 June 2006). Data do not currently
exist to test this possibility directly.
The results of our meta-analysis support the
association of the DRD2 Taq1A polymorphism with
alcoholism, suggesting that possession of the A1
allele confers a modest increase in risk. However,
this conclusion is qualified by the possibility of
publication bias or other bias in the literature and
the observed between-study heterogeneity, which
indicates that the observed association may differ in
strength between populations, or may not exist at all
in some populations. Other sources of heterogeneity
than ancestry and severity of disease are therefore
likely to exist, and further research on adequately
large samples is required to explicitly identify these.
Publication of nonsignificant results in the psychia-
tric genetics literature is important to protect against
the existence of a biased corpus of data in the public
Jonathan Flint is supported by the Wellcome Trust.
1 Koob GF, Le Moal M. Drug addiction, dysregulation of reward, and
allostasis. Neuropsychopharmacology 2001; 24: 97–129.
2 Lingford-Hughes A, Nutt D. Neurobiology of addiction and
implications for treatment. Br J Psychiatry 2003; 182: 97–100.
3 Munafo MR, Johnstone E, Murphy M, Walton R. New directions in
the genetic mechanisms underlying nicotine addiction. Addict
Biol 2001; 6: 109–117.
4 Blum K, Sheridan PJ, Wood RC, Braverman ER, Chen TJ, Comings
DE. Dopamine D2 receptor gene variants: association and linkage
studies in impulsive-addictive-compulsive behaviour. Pharmaco-
genetics 1995; 5: 121–141.
5 Blum K, Noble EP, Sheridan PJ, Montgomery A, Ritchie T,
Jagadeeswaran P et al. Allelic association of human dopamine
D2 receptor gene in alcoholism. J Am Med Assoc 1990; 263:
6 Goldman D, Brown GL, Albaugh B, Robin R, Goodson S, Trunzo M
et al. DRD2 dopamine receptor genotype, linkage disequilibrium,
and alcoholism in American Indians and other populations.
Alcohol Clin Exp Res 1993; 17: 199–204.
7 Young RM, Lawford BR, Nutting A, Noble EP. Advances in
molecular genetics and the prevention and treatment of substance
misuse: implications of association studies of the A1 allele of the
D2 dopamine receptor gene. Addict Behav 2004; 29: 1275–1294.
8 Egger M, Davey Smith G, Schneider M, Minder C. Bias in meta-
analysis detected by a simple, graphical test. Br Med J 1997; 315:
9 Munafo MR, Clark TG, Flint J. Assessing publication bias in
genetic association studies: evidence from a recent meta-analysis.
Psychiatry Res 2004; 129: 39–44.
10 Begg CB, Mazumdar M. Operating characteristics of a rank
11 Seagroatt V, Stratton I. Bias in meta-analysis detected by a simple,
graphical test. Test had 10% false positive rate. Br Med J 1998; 316:
470; author reply 470–471.
12 Irwig L, Macaskill P, Berry G, Glasziou P. Bias in meta-analysis
detected by a simple, graphical test. Graphical test is itself biased.
Br Med J 1998; 316: 470; author reply 470–471.
13 Peters JL, Sutton AJ, Jones DR, Abrams KR, Rushton L. Compar-
ison of two methods to detect publication bias in meta-analysis.
J Am Med Assoc 2006; 295: 676–680.
14 Macaskill P, Walter SD, Irwig L. A comparison of methods to detect
publication bias in meta-analysis. Stat Med 2001; 20: 641–654.
15 Neville MJ, Johnstone EC, Walton RT. Identification and char-
acterization of ANKK1: a novel kinase gene closely linked to
DRD2 on chromosome band 11q23.1. Hum Mutat 2004; 23:
16 Trikalinos TA, Ntzani EE, Contopoulos-Ioannidis DG, Ioannidis JP.
Establishment of genetic associations for complex diseases is
independent of early study findings. Eur J Hum Genet 2004; 12:
17 Amadeo S, Abbar M, Fourcade ML, Waksman G, Leroux MG,
Madec A et al. D2 dopamine receptor gene and alcoholism.
J Psychiatr Res 1993; 27: 173–179.
18 Anghelescu I, Germeyer S, Muller MJ, Klawe C, Singer P, Dahmen
N et al. No association between the dopamine d2 receptor taqi
a1 allele and earlier age of onset of alcohol dependence according
to different specified criteria. Alcohol Clin Exp Res 2001; 25:
19 Arinami T, Itokawa M, Komiyama T, Mitsushio H, Mori H, Mifune
H et al. Association between severity of alcoholism and the A1
allele of the dopamine D2 receptor gene TaqI A RFLP in Japanese.
Biol Psychiatry 1993; 33: 108–114.
20 Bau CH, Almeida S, Hutz MH. The TaqI A1 allele of the dopamine
D2 receptor gene and alcoholism in Brazil: association and
interaction with stress and harm avoidance on severity prediction.
Am J Med Genet 2000; 96: 302–306.
21 Berggren U, Fahlke C, Aronsson E, Karanti A, Eriksson M,
Blennow K et al. The Taqi Drd2 A1 allele is associated with
alcohol-dependence although its effect size is small. Alcohol
Alcohol 2006; 41: 479–485.
22 Blum K, Noble EP, Sheridan PJ, Finley O, Montgomery A, Ritchie
T et al. Association of the A1 allele of the D2 dopamine receptor
gene with severe alcoholism. Alcohol 1991; 8: 409–416.
23 Bolos AM, Dean M, Lucas-Derse S, Ramsburg M, Brown GL,
Goldman D. Population and pedigree studies reveal a lack of
association between the dopamine D2 receptor gene and alcohol-
ism. J Am Med Assoc 1990; 264: 3156–3160.
24 Chen CH, Chien SH, Hwu HG. Lack of association between TaqI
A1 allele of dopamine D2 receptor gene and alcohol-use disorders
in atayal natives of Taiwan. Am J Med Genet 1996; 67: 488–490.
25 Chen WJ, Lu ML, Hsu YP, Chen CC, Yu JM, Cheng AT. Dopamine
D2 receptor gene and alcoholism among four aboriginal groups
and Han in Taiwan. Am J Med Genet 1997; 74: 129–136.
26 Comings DE, Comings BG, Muhleman D, Dietz G, Shahbahrami B,
Tast D et al. The dopamine D2 receptor locus as a modifying
gene in neuropsychiatric disorders. J Am Med Assoc 1991; 266:
27 Comings DE, Muhleman D, Ahn C, Gysin R, Flanagan SD. The
dopamine D2 receptor gene: a genetic risk factor in substance
abuse. Drug Alcohol Depend 1994; 34: 175–180.
28 Cook BL, Wang ZW, Crowe RR, Hauser R, Freimer M. Alcoholism
and the D2 receptor gene. Alcohol Clin Exp Res 1992; 16: 806–809.
DRD2 gene and alcoholism
MR Munafo ` et al
29 Finckh U, von Widdern O, Giraldo-Velasquez M, Podschus J,
Dufeu P, Sander T et al. No association of the structural dopamine
D2 receptor (DRD2) variant 311Cys with alcoholism. Alcohol Clin
Exp Res 1996; 20: 528–532.
30 Foley PF, Loh EW, Innes DJ, Williams SM, Tannenberg AE, Harper
CG et al. Association studies of neurotransmitter gene polymorph-
isms in alcoholic Caucasians. Ann NYAcad Sci 2004; 1025: 39–46.
31 Freire MT, Marques FZ, Hutz MH, Bau CH. Polymorphisms in the
DBH and DRD2 gene regions and smoking behavior. Eur Arch
Psychiatry Clin Neurosci 2006; 256: 93–97.
32 Geijer T, Neiman J, Rydberg U, Gyllander A, Jonsson E, Sedvall G
et al. Dopamine D2-receptor gene polymorphisms in Scandinavian
chronic alcoholics. Eur Arch Psychiatry Clin Neurosci 1994; 244:
33 Gelernter J, Kranzler H. D2 dopamine receptor gene (DRD2) allele
and haplotype frequencies in alcohol dependent and control
subjects: no association with phenotype or severity of phenotype.
Neuropsychopharmacology 1999; 20: 640–649.
34 Gelernter J, O’Malley S, Risch N, Kranzler HR, Krystal J,
Merikangas K et al. No association between an allele at the D2
dopamine receptor gene (DRD2) and alcoholism. J Am Med Assoc
1991; 266: 1801–1807.
35 Goldman D, Dean M, Brown GL, Bolos AM, Tokola R, Virkkunen
M et al. D2 dopamine receptor genotype and cerebrospinal fluid
homovanillic acid, 5-hydroxyindoleacetic acid and 3-methoxy-4-
hydroxyphenylglycol in alcoholics in Finland and the United
States. Acta Psychiatr Scand 1992; 86: 351–357.
36 Goldman D, Urbanek M, Guenther D, Robin R, Long JC. Linkage
and association of a functional DRD2 variant [Ser311Cys] and
DRD2 markers to alcoholism, substance abuse and schizophrenia
in Southwestern American Indians. Am J Med Genet 1997; 74:
37 Heinz A, Sander T, Harms H, Finckh U, Kuhn S, Dufeu P et al.
Lack of allelic association of dopamine D1 and D2 (TaqIA) receptor
gene polymorphisms with reduced dopaminergic sensitivity to
alcoholism. Alcohol Clin Exp Res 1996; 20: 1109–1113.
38 Hietala J, Pohjalainen T, Heikkila-Kallio U, West C, Salaspuro M,
Syvalahti E. Allelic association between D2 but not D1 dopamine
receptor gene and alcoholism in Finland. Psychiatr Genet 1997; 7:
39 Higuchi S, Muramatsu T, Murayama M, Hayashida M. Association
of structural polymorphism of the dopamine D2 receptor gene and
alcoholism. Biochem Biophys Res Commun 1994; 204: 1199–1205.
40 Ishiguro H, Arinami T, Saito T, Akazawa S, Enomoto M, Mitushio
H et al. Association study between the -141C Ins/Del and TaqI A
polymorphisms of the dopamine D2 receptor gene and alcoholism.
Alcohol Clin Exp Res 1998; 22: 845–848.
41 Karaoguz MY, Onen I, Cosar B, Arlkan Z, Sahin FI, Menevse A.
Lack of association between TaqIA RFLP of the dopamine D2
receptor gene and alcoholism in type II Turkish alcoholics. Korean
J Genetic 2004; 26: 149–154.
42 Konishi T, Calvillo M, Leng AS, Lin KM, Wan YJ. Polymorphisms
of the dopamine D2 receptor, serotonin transporter, and GABA(A)
receptor beta(3) subunit genes and alcoholism in Mexican-
Americans. Alcohol 2004; 32: 45–52.
43 Kono Y, Yoneda H, Sakai T, Nonomura Y, Inayama Y, Koh J et al.
Association between early-onset alcoholism and the dopamine D2
receptor gene. Am J Med Genet 1997; 74: 179–182.
44 Lawford BR, Young RM, Rowell JA, Gibson JN, Feeney GF, Ritchie
TL et al. Association of the D2 dopamine receptor A1 allele with
alcoholism: medical severity of alcoholism and type of controls.
Biol Psychiatry 1997; 41: 386–393.
45 Lee MS, Lee KJ, Kwak DI. No association between the dopamine
D2 receptor gene and Korean alcoholism. Psychiatr Genet 1997; 7:
46 Limosin F, Gorwood P, Loze JY, Dubertret C, Gouya L, Deybach JC
et al. Male limited association of the dopamine receptor D2 gene
TaqI a polymorphism and alcohol dependence. Am J Med Genet
2002; 112: 343–346.
47 Lu RB, Ko HC, Chang FM, Castiglione CM, Schoolfield G, Pakstis
AJ et al. No association between alcoholism and multiple
polymorphisms at the dopamine D2 receptor gene (DRD2) in three
distinct Taiwanese populations. Biol Psychiatry 1996; 39: 419–429.
48 Noble EP, Blum K, Ritchie T, Montgomery A, Sheridan PJ. Allelic
association of the D2 dopamine receptor gene with receptor-
binding characteristics in alcoholism. Arch Gen Psychiatry 1991;
49 Noble EP, Syndulko K, Fitch RJ, Ritchie T, Bohlman MC, Guth P et
al. D2 dopamine receptor TaqI A alleles in medically ill alcoholic
and nonalcoholic patients. Alcohol Alcohol 1994; 29: 729–744.
50 Ovchinnikov IV, Druzina E, Ovtchinnikova O, Zagarovskaya T,
Nebarakova T, Anokhina AP. Polymorphism of dopamine D2 and
D4 receptor genes and Slavic-surnamed alcoholic patients. Addict
Biol 1999; 4: 399–404.
51 Parsian A, Todd RD, Devor EJ, O’Malley KL, Suarez BK, Reich T
et al. Alcoholism and alleles of the human D2 dopamine receptor
locus. Studies of association and linkage. Arch Gen Psychiatry
1991; 48: 655–663.
52 Samochowiec J, Ladehoff M, Pelz J, Smolka M, Schmidt LG,
Rommelspacher H et al. Predominant influence of the 3’-region of
dopamine D2 receptor gene (DRD2) on the clinical phenotype in
German alcoholics. Pharmacogenetics 2000; 10: 471–475.
53 Sander T, Harms H, Podschus J, Finckh U, Nickel B, Rolfs A et al.
Dopamine D1, D2 and D3 receptor genes in alcohol dependence.
Psychiatr Genet 1995; 5: 171–176.
54 Sander T, Ladehoff M, Samochowiec J, Finckh U, Rommelspacher
H, Schmidt LG. Lack of an allelic association between polymorph-
isms of the dopamine D2 receptor gene and alcohol dependence in
the German population. Alcohol Clin Exp Res 1999; 23: 578–581.
55 Duval S, Tweedie R. Trim and fill: a simple funnel-plot-based
method of testing and adjusting for publication bias in meta-
analysis. Biometrics 2000; 56: 455–463.
56 Dubben HH, Beck-Bornholdt HP. Systematic review of publication
bias in studies on publication bias. Br Med J 2005; 331: 433–434.
57 Munafo MR, Flint J. Meta-analysis of genetic association studies.
Trends Genet 2004; 20: 439–444.
58 Attia J, Thakkinstian A, D’Este C. Meta-analyses of molecular
association studies: methodologic lessons for genetic epidemiol-
ogy. J Clin Epidemiol 2003; 56: 297–303.
59 Thompson J, Thomas N, Singleton A, Piggott M, Lloyd S, Perry EK
et al. D2 dopamine receptor gene (DRD2) Taq1 A polymorphism:
reduced dopamine D2 receptor binding in the human striatum
associated with the A1 allele. Pharmacogenetics 1997; 7: 479–484.
60 Noble EP, Gottschalk LA, Fallon JH, Ritchie TL, Wu JC. D2
dopamine receptor polymorphism and brain regional glucose
metabolism. Am J Med Genet 1997; 74: 162–166.
61 Jonsson EG, Nothen MM, Grunhage F, Farde L, Nakashima Y,
Propping P et al. Polymorphisms in the dopamine D2 receptor
gene and their relationships to striatal dopamine receptor density
of healthy volunteers. Mol Psychiatry 1999; 4: 290–296.
DRD2 gene and alcoholism
MR Munafo ` et al