50904989 MT5003

Analysis of the strength and weakness of the

different methods for detection of adverse
drug reactions
Abstract
13 % of the patients suffered adverse drug reaction in UK, 17 % in Canada
and 20% in Australia and US in 2009 and it costs NHS £ 2 billion for
1,040,000 patients. Commonly affect people with adverse drug reactions are
those who have been prescribed wrong medicine by their GP’s or those who
have responded adversely to the medicine. Most of the adverse drug
reactions are predictable and mostly due to polypharmacy. Children and old
aged are prone to the risk of the adverse drug reaction.

Several methods have been adopted to reduce the incidences of adverse

drug reaction which are- anecdotal reporting by doctors and individual
patients, voluntary reporting, Yellow cards, Post marketing surveillance.
Each method has its own advantages as well disadvantages.

Key words –

Adverse Drug reactions, signal, drug monitoring.

Abbreviation-

Adverse Drug Reaction: ADR, General Practitioners : GP’s, FDA:

Food and drug administration,

Introduction

Any undesirable effect which could be linked directly to a drug and is harmful
for the patient occurring at normally administered doses is called as adverse
drug reaction. Undesirable effects may include any abnormal symptom or lab
test (WHO guidelines, 2009).

An adverse drug experience cannot be termed as adverse drug reaction until

it is directly linked to a drug and is harmful to the patient else it’s just an
adverse drug experience which are coincidental with the drug intake.

1
50904989 MT5003

In a research conducted on 3000 patients by researchers of University of

Liverpool in February, 2009 indicated that 15 % of the admitted patients
suffered from adverse drug reaction which were constipation, confusion,
renal problems and bleeding. Drugs that were commonly linked with ADR
were anticoagulants, analgesics, antineoplastics and diuretics (University of
Liverpool, 2009). Adr’s were found to be the fifth leading cause of death
globally (WHO, 2002).

Classification of Adverse drug reactions

Type A: It is dose related (Augmented). Quite common, predictable.

Associated with the pharmacological effect of the drugs and is non fatal. E.g.
Digoxin toxicity (Aronson & Ferner, 2000).

Type B: It is non dose related (Bizarre). These are very rare and cannot be
predicted. No association with the pharmacological effect of the drug and is
fatal. E.g. immune reactions (Hypersensitivity), Pseudoallergic reactions
(Aronson & Ferner, 2000).

2
50904989 MT5003

Type C: It is dose and time dependent reactions (Chronic). These are rare
reactions and accumulation of drug occurs in the body over time. E.g.
hypothalamic-pituitary-adrenal axis suppression by corticosteroids (Aronson
& Ferner, 2000).

Type D: It is time dependent reaction (Delayed). This reaction is extremely

rare and related to the dose administered. There exists a time gap between
the administration of the drugs and the reaction to occur. E.g.
Teratogenesis, Carcinogenesis, Tardive dyskinesia with neuroleptics
(Aronson & Ferner, 2003).

Type E: It is the withdrawal type reactions (end of use). This reaction is rare
and occurs soon after use of a particular drug is stopped. E.g. opiate
withdrawal syndrome, myocardial ischemia associated with withdrawal of β-
blockers, withdrawal reactions with benzodiazepines (Aronson & Ferner,
2003).

Type F: It is the unexpected failure of therapy. This type of reaction is

common and is associated with the administered dose and is due to drug-
drug interactions. E.g. dose of oral contraceptives have to be increased when
administered with enzyme inducing drugs (Aronson & Edward, 2000).

ADR’s are caused by prescribing wrong medicines or high doses to the

patients by their GP’s or due to polypharmacy (which can cause drug – drug
interaction). Elderly people are more susceptible to drug –drug interactions
as they are on many medications due to a number of health conditions they
are suffering from.

ADR’s are costing NHS exorbitantly and have not only prolonged the duration
of the patient stay in hospital but also exaggerates the patient’s condition
and indirectly affecting the efficiency of the system as a whole.

Methods and processes have been developed to protect the society by early
detection and reporting, valid verification and quantification of adverse drug
reactions which includes anecdotal reporting, voluntary reporting by yellow
card scheme, post marketing epidemiological (surveillance) studies, in silico
search methods.

Strength and weakness of the methods used to detect adverse drug

reactions

●Anecdotal reporting by individuals and doctors

3
50904989 MT5003

It is the record of personal experiences by patient or by doctor during the

treatment. It is different from data obtained from any scientific analysis
(Aronson & Edward, 2000). This kind of report is often appealing and
inspiring. The only advantage with this kind of reporting method is its very
simple and cheap (Aronson & Edward, 2000).

The drawback of this method is that it is depend totally on the vigilance and
judgment of the individual and they usually report common effects (Aronson
& Edward, 2000). But the biggest drawback is that they are easily available
to the media and can be created into huge political or corporate controversy
rather using it for improvement in healthcare system.

●Voluntary but organized reporting

It is the responsibility of doctors, dentists, carers, pharmacists, nurses,

pharmaceutical companies and all health care professionals to report any
suspected case of adverse drug reaction to Medicines and Healthcare
products Regulatory Agency (MHRA), even when it is a well known adverse
drug reaction or when its is uncertain or unknown (British national formulary,
2009). ADR’s could be reported using either electronic form of yellow card or
by reporting at regional monitoring centers (Aronson & Edward, 2000).
Information regarding the reaction and their diagnosis is sent to National
centers which shares the information with the WHO worldwide database for
International drug monitoring (Aronson & Edward, 2000).

In addition to yellow card method of ADR reporting green form are used by
Drug safety and research unit to monitor the safety of the drugs (British
national formulary, 2009). Patient’s prescribed with selected new drugs are
identified and they are monitored (drug safety and research unit, 2009). But
the submission of clinical data is not obligatory on the part of general
practitioner (drug safety and research unit, 2009).

The newly licensed drugs which have new delivery system, new route of
administration, contains new active substances, active substances in
combination with other active substance or drug which indicate altered risk
to benefit ratio are closely monitored by MHRA and are denoted by black
triangle ( ) (British national formulary, 2009). The product bears the black
triangle until the safety data has been reviewed (British national formulary,
2009).

MHRA urges all the healthcare professionals to spontaneously report all the
suspected reaction by yellow cards associated with the medical products

4
50904989 MT5003

bearing black triangle, even when they are not serious or occurring when
other drug is administered at the same time (British national formulary,
2009).

The advantage of this kind of reporting scheme is that they are simple and
economical (Aronson & Edward, 2000).

The drawbacks of this scheme is that as the reporting methods are not
obligatory on healthcare professionals to report so many ADR’s are often
under reported and high probability of biased reporting exists (Aronson &
Edward, 2000).

●Intensive event monitoring

This is monitoring methods usually aimed to collect drug prescription and

adverse events data. This type of study is short observational study and the
objective of this study to collect data about the frequency and type of
adverse drug reaction and susceptibility to risk factors (Mazzeo, et al.,2005).

Safety of newly licensed drug is studied using prescription event monitoring

(PEP) by DSRU (WHO guidelines,2009). PEP is used to generate the
hypothesis to study the safety profile by pharmacoepidemelogy (WHO
guidelines,2009).

E.g. Intensive monitoring of inpatients for antibiotic induced adverse drug

reaction in hospitals (Mazzeo, et al.,2005).

Advantage of this method is that they are easy to organize and is short term
monitoring (Aronson & Edward, 2000).

Main drawback of this method is that monitoring is done on relatively smaller

non-randomly selected population for short duration which reduces the
probability of detecting rare adverse reactions (Aronson & Edward, 2000).

●Post marketing surveillance studies

Post marketing pharmacovigilance is important as it can detect the ADR’s

which went undetected in clinical trials.

Clinical trials have limitations because, study is conducted on relative

healthy patients and often elderly, pregnant women and children are
excluded (WHO guidelines,2009).

5
50904989 MT5003

Sample size is relatively small limiting chances to detect rare ADR’s (WHO
guidelines,2009).

Often clinical trials are conducted for a short duration which limits the
chances of detecting long term consequences such as Carcinogenesis (WHO
guidelines,2009).

Clinical trials try to mimic real world conditions but cannot predict response
that would be observed. Drug – drug interaction is a situation which is
impossible to predict from clinical trials. Moreover pharmacogenetic
variations in real world could lead to toxicity or reduce the effect of normal
dose which cannot be predicted by the clinical trials (WHO guidelines,2009)

Post marketing Pharmacovigilance is important in overcoming the limitations

of clinical trials and used to detect, evaluate, understand and prevent ADR’s
and any other complication associated with the drug. Understanding of
ADR’s associated with drug will enable practitioners to use drugs with high
degree of safety and efficacy (WHO guidelines,2009).

At MHRA, team of scientist and researchers which includes physicians detect

signals (report which links a reaction to drug which was unrecognized before)
and evaluate the association between drugs and reported reaction. They
analyze and determine the risk leading to ADR on the basis of data collected.
After the risk factor is determined the CHM and Pharmacovigilance expert
advisory group gives suggestions to the MHRA regarding the safety profile of
the drug (WHO guidelines,2009). MHRA can introduce restriction or special
warnings or withdraw the drug depending on the risk associated with the
drug.

In post marketing pharmacovigilance, data is obtained from clinical studies,

epidemiological studies, data from pharmaceutical companies, data from
morbidity and mortality databases, longitudinal patient databases and from
spontaneous reporting (WHO guidelines,2009).

Epidemiological studies include cohort studies, case control and case-cohort

studies.

Cohort studies- Sample population is selected which include 2 subset of

population. One contains exposed patients and other subset includes
unexposed patients, the two subsets are compared from time to time see the
difference in the results (NARANJO, et al.,1982).
E.g. Cohort study was conducted to evaluate the possible association
between MMR vaccine and autism.

6
50904989 MT5003

Case control studies- sample population contains people with disease and
people without disease and they are compared to see the difference in
previous medical exposure (NARANJO, et al.,1982). E.g. Case-control studies
confirmed the association between Reye’s syndrome and Aspirin.

The advantages of post marketing surveillance are- they are good in

detecting reactions, good for evaluation and validation of the ADR’s and
even rare ADR’s could be detected with the help of these studies (Aronson &
Edward, 2000).

The main drawback of this system is that it cannot detect new ADR’s. It
requires huge sample population to conduct these studies and is very
expensive. MHRA and European regulatory authorities are maintaining huge
databases which contain data from spontaneous reporting as well data from
various studies (Aronson & Edward, 2000). These databases require a huge
network of people to feed in information, costing enormous amount of
money. Cohort and case-controlled studies involves complex calculation and
requires highly skilled people to carry out and interpret the results of such
studies (Aronson & Edward, 2000).

●Population statistics

Randomly selected sample from the population are studied and statistical
inferences are drawn from the study. Advantage of such type of study are
large number of people can be studied and are accurate (Aronson & Edward,
2000).

Drawbacks of this study is – it is time consuming, difficult to coordinate such

big studies, information provided could be poor (Aronson & Edward, 2000).

●Use of Pharmacogenetics as a tool to predict ADR’s

Genetic influence on phenotype of individual increases susceptibility to ADR.
With the completion of Human genome project, genotyping is done to reduce
the incidence of adverse and has lead increase in personalized therapy
(Wilkins,2002).

Both the pharmacokinetics of drugs and targets are affected by variations in

genome. Genetic variability not only influences drug metabolizing enzymes
but also affects ion channels, receptors thereby affect the drug response as
a whole (Wilkins,2002).

Single nucleotide polymorphism is single base change in DNA sequence and

it divides patients into ADR group and no ADR group. Genotyping can not

7
50904989 MT5003

only predict the patients susceptibility to ADR’s but can be use to prevent
ADR from occurring (Wilkins,2002).

Knowledge of patient’s genetic profile will enable physicians to reduce the

probability of any adverse reaction in patients and if any issue arises could
be easily resolved by the practitioner (Wilkins,2002).

In spite of this method providing patients with personalized drug therapy and
significantly reducing chances of adverse reaction have lots of drawbacks
like genotyping is expensive and time consuming process, social and ethical
issues are involved, to show the association between genetic variability and
ADR large sample size is required (Wilkins,2002).

●Use of Proteomics as a tool to detect ADR’s

Proteomics is the study of all the protein expressed by the cell in the body.
Fluctuation in the concentration of protein is measured as a diagnostic tool
(Wilkins,2002). With the help of 2-D gels and mass spectrometry number and
mass of proteins in a sample can be estimated and helps in identifying
proteins associated with drug activity (Wilkins,2002).

Proteomics have been used to find protein biomarkers associated with the
drug action and these biomarker proteins are monitored in both therapeutic
and toxic response. Any change in the level of biomarker can be directly
linked to abnormal response to drug (Wilkins,2002).

Proteomics have the ability to recognize the protein biomarkers associated

with drug response which could be used as diagnostic tool (Wilkins,2002).
But proteomics can be used to enhance normal clinical investigation
techniques and prevents occurring of ADR’s.

Disadvantages of this system is inability to process low concentrations of

proteins and non-availability of rapid protein detecting assay (Wilkins,2002).

●Bayesian neural network method for adverse drug reaction signal

generation

Over the past 30 years attempts have been made to enhance the recognition
of adverse effects by “data dredging” or “data mining.” The contribution of
such techniques to detecting adverse reactions has been modest. When
substantial numbers of reports are
available, however, comparing the proportion of reports of an adverse effect
with similar drugs may provide strong hypothesis generating signals, such as
for rhabdomyolysis associated with cerivastatin. Such comparison may even
facilitate some sort of

8
50904989 MT5003

hypothesis testing relative risk assessment. The validity of such techniques,

however, is probably much lower than that of formal designs of
epidemiological studies.

The originally embraced technique of prescription event monitoring has

proved more difficult than expected. Use of research databases such as the
General Practice Research Database for the detection of adverse effects by
generating an hypothesis has had
only limited success. It seems better to restrict the use of such databases to
hypothesis-testing epidemiological studies. (Psaty &Stricker,2004)

●Use of Drosophila melanogaster as a tool to study developmental

and reproductive toxicities in mammals.

When a prospective drug molecule is produced, clinical trials are conducted

to detect any ADR associated with the drug and to further develop the drug
to minimize ADR, if any present, to get FDA approval to market the drug
(Avanesian, et. al., 2009).

Drosophila melanogaster (fruit fly) is used to screen developmental and

reproductive toxicities for drugs in the early developmental stages.
Compared to other non-mammalian models like rodents, worms, zebrafish,
D. melanogaster’s reproductive system show similarity with mammalian
reproductive system making it ideal model system(Avanesian, et. al., 2009).

Although use of D. melanogaster is uncommon but it has been used to

detect drug induced ADR’s in reproductive system. D. melanogaster is also
used for high throughput screening of disease oriented molecule (Avanesian,
et. al., 2009).

Advantage of using D. melanogaster as a model to study ADR’s are- since life

span of fruit fly is normally 20-30 days they are time saving, does not require
much veterinary care, requires small housing space, no ethical issues are
involved and are cheap models to study effects of drug on reproductive
system (Avanesian, et. al., 2009).

The only disadvantage of using fruit fly as model system could be short life
span which prevents long term studies to be carried which assesses
consequences like carcinogenesis (Avanesian, et. al., 2009).

References:

– Annon.(2006)Reporting adverse drug reactions- A guide for healthcare

professionals. British medical association board of science. p 1-42.

9
50904989 MT5003

– Aronson, J.K, Edward, R.I (2000). Adverse drug reactions: definitions,

diagnosis, and management. The Lancet. Vol 356. p 1255-1259

– Aronson, J.K., Ferner, R.I. (2003). BMJ. Vol 327. p 1222-1225.

– Avanesian, A., Semnani, S., Jafari, M. (2009).Can Drosophila

melanogaster represent a model system for the detection of
reproductive adverse drug reactions?. Drug Discovery Today. Vol 14.
No 15.

– Bate, A. (1998). A Bayesian neural network method for adverse drug

reaction signal generation. Eur J Clin Pharmacol.vol 54. p 315- 321.

– British National Formulary [online]. Available from:

http://bnf.org/bnf/bnf/current/29432.htm?
q=”adverse””reactions””to””drugs”#hit [accessed on :27 Nov
2009]

– Drug Safety Research Unit [online]. Available from:

http://www.dsru.org/learningmodule/lm_intro/two.html.[accessed on :
27 Nov 2009]

– Mazzeo, F., et al. (2005).Hospital-based intensive monitoring of

antibiotic-induced adverse events in a university hospital.
Pharmacological research. Vol 51. p 269-274.
– NARANJO, C. A., USOA BUSTO, U., SELLERS, E.M. (1982). Difficulties in
assessing the adverse drug reactions in clinical trials Prog.
Neuro.Psychopharmacol & Biol. Psycfdat. Vol. 6, pp. 651-65.

– Psaty, B.M., Stricker, B.H. (2004).Detection, verification, and

quantification of adverse drug reactions. BMJ VOLUME 329.
– The guardian [online]. Available from:
http://www.guardian.co.uk/society /2008 apr/03/nhs.drugsandalcohol
[accessed 27 Nov 2009].
– University of Liverpool [online]. Available from:
http://www.liv.ac.uk/news/ press_releases/2009/02/adverse-drug-
reactions.htm [ accessed 27 Nov 2009]

– WHO guidelines on the safety of medicine [online]. Available

from: http://whqlibdoc.who.int/hq/2002/WHO_EDM_QSM_2002.2.pdf
[accessed 10 Dec 2009]

– Wilkins, M.R. (2002).What do we want from proteomics in the detection

and avoidance of adverse drug reactions. Toxicology Letters. Vol 127
.p 245–249.

10
50904989 MT5003

11