James Neuberger and Damian Dowling (liver unit, Queen
Elizabeth Hospital, Birmingham); Mervyn Davies and Helen
Aldersley (liver unit, St James’s Hospital, Leeds); Oliver James,
Martin Prince, and Mark Hudson (liver unit, Freeman Hospital,
Contributors: KH initiated the study and contributed to the
design,interpretation,and reporting.ET coordinated the collec-
tion of the data and contributed to the study design,
interpretation, and reporting. JD conducted the statistical analy-
ses and contributed to the interpretation and reporting. LA and
DG contributed to the design of the study, data collection,
interpretation, and reporting. JC and OB contributed to
database design, data collection, and reporting. KH is guarantor
for the study.
Funding: South East Region NHSE Research and Develop-
ment. KH is also supported by Oxfordshire Mental Healthcare
Competing interests: None declared.
1Hawton K, Fagg J, Simkin S, Bale E, Bond A. Trends in deliberate
self-harm in Oxford,1985-1995.Implications for clinical services and the
prevention of suicide. Br J Psychiatry 1997;171:556-60.
Bialas MC, Reid PG, Beck P, Lazarus JH, Smith P M, Scorer RC. Chang-
ing patterns of self-poisoning in a UK health district. Q J Med
Spooner JB, Harvey JG. Paracetamol overdoses—facts not misconcep-
tions. Pharm J 1993; May 22:706-7.
Bray GP. Liver failure induced by paracetamol. BMJ 1993;306:157-8.
Gunnell D, Hawton K, Murray V, Garnier R, Bismuth C, Fagg J, et al. Use
of paracetamol for suicide and non-fatal poisoning in the UK and
France: are restrictions on availability justified? J Epidemiol Community
O’Grady J. Acute liver failure. Medicine 1999;27:80-2.
Committee on Safety of Medicines, Medicines Control Agency. Paraceta-
mol and aspirin. Curr Probl Pharmacovigilance 1997;23:9.
Hawton K,Ware C,Mistry H,Hewitt J,Kingsbury S,Roberts D, et al.Para-
cetamol self-poisoning. Characteristics, prevention and harm reduction.
Br J Psychiatry 1996;168:43-8.
Hawton K, Ware C, Mistry H, Hewitt J, Kingsbury S, Roberts D, et al. Why
patients choose paracetamol for self poisoning and their knowledge of its
dangers. BMJ 1995;310:164.
10 Charlton J, Kelly S, Dunnell K, Evans B, Jenkins R, Wallis R. Trends in sui-
cide deaths in England and Wales. Popul Trends 1992;69:10-6.
11 Prince MI,Thomas SHL,James OFW,Hudson M.Reduction in incidence
of severe paracetamol poisoning. Lancet 2000;355:2047-8.
12 Poulin C. Prevention of paracetamol poisoning. Lancet 2000;355:2010-1.
13 Atcha Z, Majeed A. Paracetamol related deaths in England and Wales,
1993-97. Health Statistics Q 2000;7:5-9.
14 Harrison PM, O’Grady J, Keays RT, Alexander GJ, Williams R. Serial pro-
thrombin time as a prognostic indicator in paracetamol induced
fulminant hepatic failure. BMJ 1990;301:964-6.
15 Turvill JL, Burroughs AK, Moore KP. Change in occurrence of paraceta-
mol overdose in UK after introduction of blister packs. Lancet
(Accepted 8 February 2001)
Antipsychotic drugs and heart muscle disorder in
international pharmacovigilance: data mining study
David M Coulter, Andrew Bate, Ronald H B Meyboom, Marie Lindquist, I Ralph Edwards
Objectives To examine the relation between
antipsychotic drugs and myocarditis and
Design Data mining using bayesian statistics
implemented in a neural network architecture.
Setting International database on adverse drug
reactions run by the World Health Organization
programme for international drug monitoring.
Main outcome measures Reports mentioning
antipsychotic drugs, cardiomyopathy, or myocarditis.
Results A strong signal existed for an association
between clozapine and cardiomyopathy and
myocarditis. An association was also seen with other
antipsychotics as a group. The association was based
on sufficient cases with adequate documentation and
apparent lack of confounding to constitute a signal.
Associations between myocarditis or cardiomyopathy
and lithium, chlorpromazine, fluphenazine,
haloperidol, and risperidone need further
Conclusions Some antipsychotic drugs seem to be
linked to cardiomyopathy and myocarditis. The study
shows the potential of bayesian neural networks in
analysing data on drug safety.
The antipsychotic drug clozapine has been reported to
cause myocarditis or cardiomyopathy.1 2other drugs in
the same therapeutic class may share similar toxicity.
Data mining of a large database of suspected adverse
reactions can find such new signals. As part of the
World Health Organization’s programme for inter-
national drug monitoring, national pharmacovigilance
centres in 60 countries report adverse reactions to a
central database maintained by the Uppsala Monitor-
ing Centre in Sweden.3
To analyse this large database an approach using
bayesian statistics implemented in a neural network
architecture has been developed. The approach is able
to look for new adverse reactions from combinations
of drugs and also to identify previously unknown
patterns, such as risk factors for adverse events with
specific drugs—for example, patient age, underlying
diseases, and drug interactions. We used the bayesian
approach to look for cardiac effects related to antipsy-
chotic drugs in the WHO database of adverse
We used the bayesian confidence propagation network,
which implements bayesian statistics in a neural network
architecture, in the WHO database. The network was
used to test reports of clozapine and all other
antipsychotic drugs suspected of causing myocarditis or
cardiomyopathy against a background of all reports in
the database. We calculated the strength of dependency
between a drug (or drug group) and adverse reaction
using a logarithmic measure of disproportionality called
the information component.4An association between
the drug and the reaction was considered significant if
the information component minus 2 standard devia-
tions was positive. The value of the information compo-
Details of the
available on the
Centre for Adverse
University of Otago,
David M Coulter
BMJ VOLUME 322 19 MAY 2001bmj.com
nent is based on the number of case reports for drug(s)
“x” (Cx); the number of case reports of adverse
reaction(s) “y”(Cy);the number of reports of the specific
combination (Cxy); and the total number of reports (C).
Further details of the methods are available on the
Myocarditis and cardiomyopathy were reported rarely
as suspected adverse drug reactions, accounting for
less than 0.1% (2121) of almost 2.5 million reports.The
table shows the antipsychotic drugs reported to have
caused either myocarditis or cardiomyopathy on two
or more occasions.Clozapine has a much higher infor-
mation component than other antipsychotics together
and than the general background database. Most
reports predated recent publicity about clozapine. The
statistical associations of clozapine with myocarditis
and cardiomyopathy individually were also significant.
significantly associated with myocarditis and cardiomy-
opathy together (table) and individually compared
with the general database, although these associations
were much weaker than for clozapine.
Chlorpromazine, lithium, and fluphenazine were
significantly associated with myocarditis and cardiomy-
opathy. The 16 cases with risperidone were not more
than expected given the high overall reporting of the
drug in the database. Chlorpromazine was also signifi-
cantly associated with myocarditis and cardiomyopathy
separately. Lithium, fluphenazine, and risperidone
were significantly associated with cardiomyopathy but
not myocarditis.In contrast,haloperidol was associated
with myocarditis but not cardiomyopathy.
Our analysis suggests that antipsychotic drugs other
than clozapine may be associated with myocarditis and
cardiomyopathy. The findings may have three explana-
tions. The conditions for which antipsychotics are used
could be risk factors for myocarditis and cardiomyopa-
thy; the antipsychotic drug could be an innocent
bystander;or there may be a causal association.Despite
patients taking clozapine being intensively monitored
for agranulocytosis, the former two are unlikely expla-
nations for the strong relation between clozapine and
myocarditis and cardiomyopathy.5The association with
clozapine cannot be explained by coprescribed drugs.
In some of the cases in the other antipsychotics group
the patient was also taking clozapine or non-
antipsychotic drugs known to cause myocarditis or car-
diomyopathy. However, standardised clinical evalua-
tion6shows that there were sufficient cases with
confounding to constitute a signal for cardiomyopathy
or myocarditis in the other antipsychotics identified
andapparent lack of
Choice of methods
Our results were obtained by a data mining approach.
A concern had been raised about myocarditis with
clozapine. We then examined the association between
the group of antipsychotics with myocarditis or cardio-
myopathy.Having discovered a quantitative association
between the antipsychotics group and cardiomyopathy
and myocarditis, we investigated individual antipsy-
chotic drugs and then performed a case by case analy-
sis. Our study shows that data mining can be used
successfully to detect signals of adverse reactions in the
Our results could have been shown using a simpler
method. However, the simpler methods rely on some-
one deciding to look for an association.7A data mining
approach that routinely looks for associations between
all possible combinations of drugs and adverse
reactions is computer intensive (hence the use of a
neural network).However,it increases the objectivity of
signal detection by introducing an effective quantitative
filtering step before clinical analysis.8We believe that
this is enormously beneficial.
The summaries of case histories in the database do not
allow us to draw definite conclusions about the
likelihood of the possible causes of the associations we
observed between antipsychotic drugs and myocarditis
and cardiomyopathy. Adverse drug reactions are
Antipsychotic drugs (anatomical, therapeutic, chemical drug classification NO5A) for
which two or more reports of cardiomyopathy or myocarditis have been registered in
No of case
Total No of
reports for drug
*All antipsychotic drugs other than clozapine.
In this table a single case report is counted for more than one drug adverse reaction combination if there
are two or more suspected antipsychotic drugs in that case report.
What is known on this topic
Clozapine has been reported to be associated with
myocarditis and cardiomyopathy
What this study adds
The WHO database shows that clozapine is
significantly more frequently reported in relation
to cardiomyopathy and myocarditis than other
Myocarditis and cardiomyopathy were also
particularly associated with chlorpromazine,
lithium, fluphenazine, risperidone, and
These associations need to be investigated further
to establish whether they are causal
Data mining is a useful tool in pharmacovigilance
Ronald H B
head of research and
I Ralph Edwards
I R Edwards
BMJ VOLUME 32219 MAY 2001 bmj.com
greatly underreported worldwide. Further study is
needed to determine if antipsychotics other than
particularly lithium, chlorpromazine, fluphenazine,
haloperidol, and risperidone, and to consider the com-
parative risks and effectiveness of antipsychotics.This is
especially important given the recent finding that older
and newer drugs have similar efficacy.9Antipsychotic
drugs should also be considered in unexplained
sudden deaths in psychotic patients.
We thank the national centres that contribute data to the WHO
international drug monitoring programme. The opinions and
conclusions, however, are not necessarily those of the various
national centres or of the WHO. Roland Orre was central in
developingthe bayesian confidence
network as a routine tool for signal detection in the WHO data-
base of drug adverse reactions.
Contributors: DMC suggested the study and made a
provisional investigation of the data, AB and IRE planned and
designed the study; AB carried out the study; and IRE, AB, and
ML evaluated the results. RHBM drafted the first report of the
study, AB and IRE wrote the paper, and all authors contributed
to modifying the manuscript and the final editing of the paper.
IRE is the guarantor.
Competing interests: None declared.
1 Jensen VE, Gotzsche O. Allergic myocarditis in clozapine treatment.
Ugeskrift for Laeger 1994;156:4151-2.
Killian JG, Kerr K, Lawrence C, Celermajer DS. Myocarditis and cardio-
myopathy associated with clozapine. Lancet 1999;354:1841-5.
Olsson S.The role of the WHO programme on international drug moni-
toring in coordinating worldwide drug safety efforts. Drug Safety
Bate A, Lindquist M, Edwards IR, Olsson S, Orre R, Lansner A, et al. A
Bayesian neural network method for adverse drug reaction signal
generation. Eur J Clin Pharmacol 1998;54:315-21.
Honigfeld G, Arellano F, Sethi J, Bianchini A, Schein J. Reducing
clozapine-related morbidity and mortality: 5 years of experience with the
clozaril national registry. J Clin Psychiatry 1998;59(suppl 3):3-7.
Edwards IR,Lindquist M,Wiholm B-E,Napke E.Quality criteria for early
signals of possible adverse drug reactions. Lancet 1990;336:156-8.
Hand DJ. Statistics and data mining: intersecting disciplines. SIGKDD
Edwards IR. Adverse drug reactions: finding the needle in the haystack.
Geddes J, Freemantle N, Harrison P, Bebbington P. Atypical antipsychot-
ics in the treatment of schizophrenia: systematic overview and
meta-regression analysis. BMJ 2000;321:1371-6.
(Accepted 20 February 2001)
Effect of improved housing on illness in children under
5 years old in northern Malawi: cross sectional study
Christopher G Wolff, Dirk G Schroeder, Mark W Young
Objective To evaluate the effects of a Habitat for
Humanity housing improvement programme in
northern Malawi on the prevalence of childhood
Design Household based cross sectional study.
Setting Rural communities centred near the small
northern Malawi town of Ekwendeni.
Subjects 318 children under 5 years old.
Main outcome measures Prevalence of respiratory,
gastrointestinal, and malarial infections according to
maternal recall, laboratory, or clinical data.
Results Children living in improved homes were less
likely to have respiratory, gastrointestinal, or malarial
illnesses (odds ratio 0.56, 95% confidence interval 0.35
to 0.91) after confounding factors were controlled for.
The reductions in individual diseases were not
Conclusion Improved housing significantly reduced
the burden of disease among children under 5 years
Poor quality housing is generally accepted to be an
important contributor to ill health.1Rates of disease
have been associated with the quality and specific
attributes of a house as well as the conditions that those
Although the importance of housing for health is
recognised,1 12 13few well designed studies have quanti-
fied this impact,especially in the developing world.The
objective of this study was to assess the impact on chil-
dren’s health of a housing improvement project in
rural Malawi. We examined the effect on illness of
living in improved housing compared with living in
Participants and methods
The study was conducted in collaboration with
Ekwendeni Hospital, Homeless International UK, and
Habitat for Humanity International in the town of
Ekwendeni, Malawi. Traditional houses in the area are
constructed of mud brick walls with thatch roofing,
hard packed mud floors, and possibly a pit latrine.
Houses are usually about 25 m2and consist of two or
three rooms.Houses constructed under the Habitat for
Humanity programme in Ekwendeni have fired mud
bricks, tile roofing, concrete foundation, and a pit
latrine. Habitat houses have a mean size of 30 m2and
three rooms. The cost of a habitat house at the time of
the study was about $550 (£370), offset by a 10 year no
interest loan. Habitat houses were built next to or
replaced the traditional house of the intended owner
and were non-systematically dispersed throughout the
communities among traditional houses.
selected by a village habitat committee. Applicants had
to be unable to provide adequate housing for
themselves because of financial, social, or physical
reasons and to have shown their commitment to the
programme by spending a standardised amount of
time helping to build another applicant’s house.
We used data from two surveys conducted in March
and August 1997. Households for the first survey were
randomly selected from a list of about 300 habitat
School of Public
Health of Emory
Atlanta, GA 30322,
Dirk G Schroeder
Mark W Young
D G Schroeder
BMJ VOLUME 32219 MAY 2001bmj.com