REVIEW
doi: 10.1111/j.1368-5031.2006.00955.x
The neurobiological basis for partial agonist treatment of nicotine
dependence: varenicline
J. FOULDS
Tobacco Dependence Program, UMDNJ School of Public Health, New Brunswick, NJ, USA
SUMMARY
Smoking cessation has major health benefits for men and
women of all ages. However, most smokers are addicted to
nicotine and fail repeatedly in their attempts to quit.
Stimulation of nicotinic receptors in the brain, particularly
a4b2 receptors, releases dopamine in the meso-limbic area of
the brain and is reinforcing. Nicotine abstinence reduces
dopamine release, and this is associated with withdrawal
symptoms and craving for nicotine. Eight current pharmacotherapies – bupropion, nortriptyline, clonidine and
nicotine patch, gum, inhaler, lozenge and nasal spray – are
moderately effective aids to smoking cessation. Each is
significantly better than placebo, but approximately 80% of
patients using one of these medications return to smoking
within the first year. Varenicline, a specific a4b2 nicotinic
receptor partial agonist, is a new pharmacotherapy that stimulates dopamine and simultaneously blocks nicotine receptors. Phase II and III trials have yielded promising results
suggesting that varenicline could be an important advance in
the treatment of nicotine dependence.
Keywords: Smoking; cessation; tobacco; nicotine; dependence; treatment; varenicline pharmacotherapy
2006 Blackwell Publishing Ltd
INTRODUCTION
Tobacco smoking is the number one cause of premature
death in developed countries. It is responsible for approximately 400,000 premature deaths per year in the United
States alone (1) and roughly 4.9 million deaths per year
worldwide, or 8.8% of all global deaths (2). Approximately,
half of all long-term smokers die prematurely as a result of
smoking (3), and the life span of the continuing smoker will
be reduced by an average of 10 years (4).
Smoking cessation confers major health benefits for men
and women of all ages. For example, people who quit smoking by age 50 have half the risk of dying in the next 15 years
compared with continuing smokers (around 10% vs. 20% at
age 50, varying by sex and amount smoked) (5).
Although it is nicotine and its psychological effects that
engender addiction (6,7), it is tobacco’s other components –
the ‘tar’, volatile oxidant gases and carbon monoxide – that
cause the most of the harms to health (7,8). This article aims
to summarise recent research on the neurobiology of nicotine
dependence and discuss the effectiveness of current pharmacotherapies for smoking cessation. The rationale for a
Correspondence to:
Jonathan Foulds PhD, Associate Professor and Director, Tobacco
Dependence Program, UMDNJ School of Public Health, 317
George Street, Suite 210, New Brunswick, NJ 08901, USA
Tel.: þ 1 732 2358213
Fax: þ 1 732 2358297
Email: [email protected]
promising new approach involving partial agonist therapy
will also be presented.
THE CHARACTERISTICS OF NICOTINE
DEPENDENCE
The criteria for nicotine dependence according to both the World
Health Organization’s ‘International Statistical Classification
of Diseases’, 10th Revision (9) and the American Psychiatric
Association’s ‘Diagnostic and Statistical Manual of Mental
Disorders’, Fourth Edition, (10) include (i) unsuccessful
attempts to stop smoking (ii) difficulty controlling tobacco
use and (iii) previous experience of withdrawal symptoms
during a period of abstinence. Withdrawal symptoms occur
following abrupt cessation or reduction of nicotine use and
include depressed mood, insomnia, irritability, anxiety,
difficulty concentrating, restlessness, increased appetite and
cravings for tobacco/nicotine (10). It is this withdrawal
syndrome – together with nicotine’s subtle but powerful
reinforcing effects, repeated 73,000 puffs per year for a
1-pack-per-day smoker – that makes smoking so addictive
(8). Nicotine has a half-life of approximately 2 h; therefore,
the onset of withdrawal symptoms is within 4–6 h of last
nicotine use. These symptoms peak within the first few days
of abstinence and typically resolve within 1 month. However,
most smokers who make a quit attempt relapse within the
first month. How does nicotine act at a neurobiological level
to produce these behavioural effects, and how can new
ª 2006 The Author
Journal compilation ª 2006 Blackwell Publishing Ltd Int J Clin Pract, May 2006, 60, 5, 571–576
572
VARENICLINE FOR TREATMENT OF NICOTINE DEPENDENCE
pharmacotherapies target these neurobiological mechanisms
more effectively?
THE NEUROBIOLOGY OF NICOTINE
DEPENDENCE
The primary effects of nicotine are mediated by nicotinic
acetylcholine receptors (nAChRs), many subtypes of which
are widely distributed throughout the central nervous system.
Seventeen nicotinic receptor subunit genes have been identified to date, and each receptor is composed of five subunits.
The functional properties of each receptor are determined by
its subunit composition. The subtype of nAChRs, composed
of two a4 and three b2 subunits (Figure 1), is known to form
the high-affinity binding sites in the brain (7). A particularly
high concentration of a4 subunits can be found in the ventral
tegmental area (VTA) of the brain, where a dense supply of
dopamine neurones is linked to the brain’s main ‘reward
centre’, the nucleus accumbens. The loss of nicotine selfadministration behaviour in knockout mice lacking the b2
subunit suggests that it contributes to nAChRs relevant to
nicotine dependence (11). The effects of a4-receptor activation
have been shown to be important in dependence, including
reinforcement, tolerance and sensitisation (12). The a4b2
nAChR also has the highest sensitivity to nicotine – 50% of
its maximal activation is produced at a concentration (EC50) of
0.1–1.0 mM, but it can be desensitised by lower concentrations.
Nicotinic receptors pass through three main states. In the
first, or ‘resting’ state, the receptor is not active (ion channel
closed) but is open to activation by contact with agonist
(typically nicotine or acetylcholine). In the ‘active’ state,
binding with an agonist causes the receptor ion channel to
open and remain open for a brief period, during which an
inward flux of Naþ produces local depolarisation. The third,
‘desensitised’ state typically follows activation, in which the
channel is closed to ions and is refractory to activation by
agonist, although agonist can still bind to the receptor. Low
concentrations of agonist can push the receptor into the
desensitised state without going through the open (active)
state, and high concentrations of agonist can stimulate activation of an otherwise resting or desensitised receptor (7).
α4
β2
α4
β2
β2
Surface of
dopamine
neuron
When a sufficient concentration of nicotine is carried in
the blood to activate a4b2 receptors in the VTA, a burst firing
of dopamine neurones occurs (13). The terminals of these
neurones are in the medial shell and core areas of the nucleus
accumbens. This stimulation of dopamine neurones causes an
increased release of extra-synaptic dopamine in the nucleus
accumbens (13). The anatomic locations of these areas of the
brain are shown in Figure 2.
Considerable evidence suggests that repeated nicotine
exposure results in an increase in functional nicotinic receptors in the brain and, specifically, a sensitisation of the mesolimbic dopamine response to nicotine (13). This dopamine
response (i.e. an increase in extra-synaptic dopamine in the
extracellular space between fibres in the accumbens) appears
to be associated with the reinforcing and addictive properties
not only of nicotine but also of other psychostimulant drugs
of abuse (e.g. amphetamine, cocaine) (14). This response
confers hedonic properties on the behaviours associated with
the dopamine activation. An animal that has experienced
repeated nicotine boosts and accumbens dopamine stimulation by pressing a bar (or inhaling on a cigarette) will quickly
learn that the behaviour itself (bar pressing, cigarette puffing)
is enjoyable and comes to acquire reinforcing properties. Over
time and repeated exposures, the smoking ritual (e.g. opening
the pack, lighting the cigarette, feeling the smoke hit the back
of the throat) becomes capable of stimulating meso-limbic
dopamine and therefore acts as a reinforcer itself, even in the
absence of agonist (nicotine)-stimulated dopamine activation
(13). This may be the reason why smokers often state that
they enjoy the ritual of smoking.
The dysphoric symptoms of nicotine withdrawal start to
occur when the regular smoker is deprived of nicotine for at
least 4–6 h and when more nAChRs become resensitised but
unstimulated by nicotine. Animal studies have shown that
Prefrontal
cortex
Nucleus
accumbens
Ventral
tegmental
area
Hippocampus
Figure 1 Simplified structure of a4b2 nicotinic receptor
located on surface of a dopamine cell body
Figure 2 Simplified diagram of the brain showing the
anatomic locations of the ventral tegmental area and
the nucleus accumbens
ª 2006 The Author
Journal compilation ª 2006 Blackwell Publishing Ltd Int J Clin Pract, May 2006, 60, 5, 571–576
573
VARENICLINE FOR TREATMENT OF NICOTINE DEPENDENCE
VTA dopamine neuronal activity is reduced during the first
day of nicotine withdrawal (15).
CURRENT PHARMACOTHERAPIES FOR
TOBACCO DEPENDENCE
The basic rationale for many of the effective pharmacotherapies for nicotine addiction has been to mimic or replace the
effects of nicotine. The most obvious way to do this is by
providing the exogenous agonist itself (i.e. via nicotine gum,
patch, nasal spray, lozenge or inhaler).
Other effective pharmacotherapies, such as bupropion and
nortriptyline, appear to affect neurobiological mechanisms
similar to those affected by nicotine replacement. Typically
they ameliorate nicotine withdrawal by inhibiting reuptake of
dopamine and noradrenaline (norepinephrine) in the central
nervous system, but without the need for a direct agonist
effect (16). Bupropion has also been shown to antagonise
nAChR function. Its principal mode of action appears to be
via reduction of withdrawal symptoms following smoking
cessation via its ability to mimic nicotine effects on dopamine
and noradrenaline (norepinephrine). Thus bupropion
increases dopamine and noradrenaline concentration in the
extracellular space by inhibiting reuptake. Its ability to antagonise nicotinic receptors may prevent relapse by attenuating
the reinforcing properties of nicotine (17). The active bupropion metabolite subtype (2S,3S)-hydroxybupropion is a
potent antagonist of the a4b2 nicotinic receptor (18). While
the primary mechanism of bupropion’s effects on smoking
cessation remains unclear, it seems that these effects are not
limited to an antidepressant action as its efficacy is independent of baseline-depressive symptoms (19).
Nortriptyline is a tricyclic antidepressant that has noradrenergic properties and some dopaminergic activity. It also has
been effective in smoking cessation (16). Other antidepressants, however, such as selective serotonin reuptake inhibitors,
do not appear to be effective aids to smoking cessation.
The a-noradrenergic agonist clonidine suppresses sympathetic activity and has been used for hypertension and to
reduce symptoms associated with alcohol or opiate withdrawal. Both the oral and the patch formulations of clonidine
increased smoking cessation rates in eight of nine trials, but
side effects include sedation and postural hypotension (16).
Meta-analyses of randomised trials of nicotine replacement
therapy, bupropion, nortriptyline and clonidine have shown
these medications to be significantly more effective than
placebo in achieving tobacco abstinence (19–21). However,
as summarised in Table 1, the long-term (i.e. 6–12 months)
tobacco abstinence rates are typically just under double those
achieved by placebo (18% vs. 10%). Evidence suggests that
abstinence rates can be increased when the medications are
combined with more intensive counselling (22–25), or when
combinations of medications are used (22,23,26,27). That
long-term abstinence rates are typically 25%35% even in
ideal circumstances underscores the need for new and more
effective smoking cessation aids.
Although this article focuses on the role of a4b2 nicotinic
receptors, considerable evidence shows that a7 nicotinic receptors likely play a role in the processes that cause nicotine
addiction. These receptors have much lower affinity for nicotine than a4b2 receptors and therefore are not desensitised
rapidly, but they can also stimulate dopamine release via
presynaptic stimulation of glutamatergic afferents. The combined action of these two receptor subtypes has been postulated to produce long-term potentiation of dopamine
stimulation by nicotine (28). Similarly, noradrenergic stimulation likely has a role in nicotine dependence, as suggested by
the efficacy of nortriptyline (which primarily has noradrenergic effects) for smoking cessation (19,24).
T H E R A T I O N A L E F O R A S E L E C T I V E a4 b2
NICOTINIC RECEPTOR PARTIAL AGONIST FOR
SMOKING CESSATION
Compounds that act as a4b2 nAChR partial agonists and
simultaneously block the action of nicotine (29,30) offer a
particularly promising new approach to helping smokers quit.
Partial agonists aim to provide a low-to-moderate level of
dopamine stimulation to reduce craving and withdrawal
symptoms. The lower level of dopamine release may be less
dependence forming than the intermittent spikes in dopamine release produced by inhaled nicotine. The antagonist
effect blocks the reinforcing effects of nicotine and potentially
reduces the risk that a lapse to smoking would turn into a fullblown relapse.
The plant alkaloid, cytisine, has been used for smoking
cessation in Bulgaria and has weak partial agonist activity but
limited absorption in the brain. However, scientists at Pfizer
Inc. were able to modify the structure of the compound to
create varenicline, a new, highly selective and potent a4b2
nAChR partial agonist (30). Varenicline has recently completed phase III trials and is undergoing expedited review by
the US Food and Drug Administration. In rat studies of the
drug, sustained extracellular dopamine levels were observed in
the nucleus accumbens at about half the level of an acute dose
of nicotine, and the effects of a simultaneous dose of nicotine
were blocked.
Figure 3 presents a greatly simplified model for (i) nicotine
activating nicotinic receptors and stimulating dopamine
release (ii) nicotine withdrawal decreasing dopamine release
and (iii) varenicline blocking nicotinic receptors, with the
partial agonist effect producing moderate levels of dopamine
release and reducing withdrawal and craving.
Early studies (including phase II and III clinical trials
involving over two thousand participants) of varenicline
have been presented at scientific meetings prior to their
ª 2006 The Author
Journal compilation ª 2006 Blackwell Publishing Ltd Int J Clin Pract, May 2006, 60, 5, 571–576
574
VARENICLINE FOR TREATMENT OF NICOTINE DEPENDENCE
Table 1 Pharmacotherapies demonstrating efficacy for smoking cessation in the Cochrane Database of Systematic Reviews
Drug
Cochrane review
update
Number of
comparisons
Number of abstinent
active arm (%)
Number of abstinent
control arm (%)
Odds ratio
(95% C. I.)
Nortriptyline
Bupropion
Clonidine
Nicotine gum
Nicotine patch
Nicotine inhaler
Nicotine nasal spray
Nicotine lozenge/tablet
10/27/04
10/27/04
10/21/04
11/02/04
11/02/04
11/02/04
11/02/04
11/02/04
7
21
6
52
42
4
4
5
102/506 (20.2)
835/4158 (20.1)
98/393 (24.9)
1565/8023 (19.5)
1493/10216 (14.6)
84/490 (17.1)
107/448 (23.9)
224/1363 (16.4)
46/515 (8.9)
323/3013 (10.7)
55/383 (14.4)
1125/9760 (11.5)
555/6475 (8.6)
44/486 (9.1)
52/439 (11.8)
121/1376 (8.8)
2.14
1.99
1.89
1.66
1.81
2.14
2.35
2.05
(19)
(19)
(20)
(21)
(21)
(21)
(21)
(21)
publication in peer-reviewed journals. The results of these
studies suggest that varenicline is an effective smoking cessation therapy.
Oncken and colleagues (31) presented data from two phase
II randomised trials in which varenicline produced short-term
(14 weeks) quit rates on the 2 mg/day dose that were
approximately four times higher than the placebo quit rates.
More recently, Tonstad and colleagues (32) presented data
from three phase III randomised trials wherein long-term
(1 years) abstinence rates were more than twice those of
placebo. In one study, those who were quit at 12 weeks
were more likely to remain abstinent at 24 weeks if they
continued on varenicline (32). The phase III placebocontrolled trials included randomisation to bupropion and
found that varenicline produced significantly higher 1-year
abstinence rates than bupropion, which was in turn significantly
better than placebo. Importantly, varenicline appears to have
a good side effect profile (mild to moderate nausea is the most
frequent symptom), with adverse events rates leading to
A
B
Nicotine
receptors
Nicotine
(1.49, 3.06)
(1.73, 2.3)
(1.3, 2.74)
(1.52, 1.81)
(1.64, 2.02)
(1.44, 3.18)
(1.63, 3.38)
(1.62, 2.59)
discontinuation similar to those of placebo. When taken
orally, it reaches a peak blood concentration in 2–4 h and
has a half-life of 20–30 h in healthy smokers. Eighty percent
or more of the drug is excreted unchanged in the urine (33).
COMMENT
Although this article focuses on pharmacotherapy as an
important factor in helping smokers quit, it is recognised
that smoking is a multifaceted phenomenon. Societal interventions such as increases in taxes on cigarettes, laws requiring
that public places be smoke-free and restrictions on the marketing of tobacco have all been shown to impact societal
tobacco use. It is also clear that tobacco dependence is best
conceptualised as a chronic condition. Like other chronic
conditions (e.g. hypertension, diabetes and asthma), tobacco
dependence is frequently not cured by a single short-term
pharmacological intervention and more commonly requires
repeated, and sometimes longer-term (i.e. >3 months)
C
Varenicline ( )
blocks nicotine
receptors
Cell body of dopamine
neuron in ventral
tegmental area
Rapid/burst firing
Dopamine ( ) release
from dopamine terminal
in the nucleus accumbens
Partial agonist
effects stimulate
moderate
dopamine
release
Figure 3 Highly simplified scheme showing effects of (A) nicotine from cigarettes (B) nicotine withdrawal and (C) varenicline on
nicotinic receptors and dopamine release
ª 2006 The Author
Journal compilation ª 2006 Blackwell Publishing Ltd Int J Clin Pract, May 2006, 60, 5, 571–576
575
VARENICLINE FOR TREATMENT OF NICOTINE DEPENDENCE
interventions. More intensive behavioural interventions and
combination of pharmacotherapies improve smoking cessation outcomes (24,25,27). This may also be true for
varenicline.
CONCLUSION
Varenicline is the first smoking cessation treatment specifically designed to target the neurobiological mechanism of
nicotine dependence. If the results of the early clinical trials
can be replicated in clinical practice, varenicline will represent
an important advance in helping patients to quit smoking.
ACKNOWLEDGEMENTS
Jonathan Foulds is primarily funded by a grant from the New
Jersey Department of Health and Senior Services through
New Jersey’s Comprehensive Tobacco Control Program.
While writing this article, he was also receiving support
from the Robert Wood Johnson Foundation, the Cancer
Institute of New Jersey and the National Institute on Drug
Abuse (USA). He has worked as a consultant and received
honoraria from pharmaceutical companies involved in
production of tobacco dependence treatment medications
(including Pfizer Inc., manufacturer of varenicline), as well
as a variety of agencies involved in promoting health (e.g.
WHO, USNIH, etc.). He has also worked as an expert
witness in litigation, including law suits against tobacco
companies. Thanks to Vanessa Sypko for help with the figures
in this article.
REFERENCES
1 Thun MJ, Apicalla LF, Henley SJ. Smoking vs other risk factors
as the cause smoking-attributable deaths: confounding in the
courtroom. JAMA 2000; 284: 706–12.
2 World Health Organization. Tobacco or Health: A Global Status
Report. Geneva, Switzerland: World Health Organization, 1997.
3 Peto R, Lopez AD, Boreham J et al. Mortality From Smoking in
Developed Countries 1950–2000. Oxford, UK: Oxford University
Press, 1994.
4 Doll R, Peto R, Boreham J et al. Mortality in relation to smoking: 50 years’ observations on male British doctors. BMJ 2004;
328: 1519.
5 US Department of Health and Human Services. The Health
Benefits of Smoking Cessation. Washington, DC, US:
Government Printing Office, 1990.
6 US Department of Health and Human Services. The Health
Consequences of Smoking: Nicotine Addiction. Washington, DC,
US: Government Printing Office, 1988.
7 Royal College of Physicians. Nicotine Addiction in Britain:
A Report of the Tobacco Advisory Group of the Royal College of
Physicians. London: Royal College of Physicians, 2000.
8 Foulds J, Burke M, Steinberg M et al. Advances in pharmacotherapy for tobacco dependence. Expert Opin Emerg Drugs
2004; 9: 39–53.
9 World
Health
Organization.
International
Statistical
Classification of Diseases and Related Health Problems, 10th
Revision. Geneva, Switzerland: World Health Organization,
1992.
10 American Psychiatric Association. Diagnostic and Statistical
Manual of Mental Disorders, 4th edn. Washington, DC:
American Psychiatric Association, 1995.
11 Picciotto MR, Zoli M, Rimondini R et al. Acetylcholine receptors containing the ß2 subunit are involved in the reinforcing
properties of nicotine. Nature 1998; 391: 173–7.
12 Tapper AR, McKinney SL, Nashmi R et al. Nicotine activation
of a4* receptors: sufficient for reward, tolerance, and sensitization. Science 2004; 306: 1029–32.
13 Balfour DJK. The neurobiology of tobacco dependence: a
preclinical perspective on the role of the dopamine projections
to the nucleus accumbens. Nicotine Tob Res 2004; 6: 899–912.
14 Kelley AE. Nicotine receptors: addiction’s smoking gun? Nat
Med 2002; 31: 447–9.
15 Liu ZH, Jin W. Decrease of ventral tegmental area dopamine
neuronal activity in nicotine withdrawal rats. Neuroreport 2004;
15: 1479–81.
16 Roddy E. Bupropion and other non-nicotine pharmacotherapies.
BMJ 2004; 328: 509–11.
17 Warner C, Shoaib M. How does bupropion work as a smoking
cessation aid? Addict Biol 2005; 10: 219–31.
18 Damaj MI. Enantioselective effects of hydroxy metabolites of
bupropion on behavior and on function of monoamine
transporters and nicotinic receptors. Mol Pharmacol 2004; 66:
675–82.
19 Hughes JR, Stead LF, Lancaster T. Antidepressants for smoking
cessation (review). Cochrane Database Syst Rev 2004; 4:
CD00031.
20 Gourlay SG, Stead LF, Benowitz NL. Clonidine for smoking
cessation (review). Cochrane Database Syst Rev 2004; 3:
CD00058.
21 Silagy C, Lancaster T, Stead L et al. Nicotine replacement
therapy for smoking cessation (review). Cochrane Database Syst
Rev 2004; 3: CD000146.
22 West R, McNeill A, Raw M. Smoking cessation guidelines for
health professionals: an update. Thorax 2000; 55: 987–99.
23 Fiore MC, Bailey WC, Cohen SJ et al. Treating Tobacco Use and
Dependence. Clinical Practice Guideline. Rockville, MD: US
Department of Health and Human Services, Public Health
Service, 2000.
24 Hall SM, Humfleet GL, Reus VI et al. Extended nortriptyline
and psychological treatment for cigarette smoking. Am J
Psychiatry 2004; 161: 2100–7.
25 Foulds J, Gandhi KK, Steinberg MB et al. Factors associated
with quitting smoking at a tobacco dependence treatment clinic.
Am J Health Behav 2006; 30: 400-12.
26 Bohadana A, Nilsson F, Rasmussen T et al. Nicotine inhaler and
nicotine patch as a combination therapy for smoking cessation: a
randomized, double-blind, placebo-controlled trial. Arch Intern
Med 2000; 160: 3128–34.
ª 2006 The Author
Journal compilation ª 2006 Blackwell Publishing Ltd Int J Clin Pract, May 2006, 60, 5, 571–576
576
VARENICLINE FOR TREATMENT OF NICOTINE DEPENDENCE
27 Steinberg MB, Foulds J, Richardson DL et al. Pharmacotherapy
and smoking cessation at a tobacco dependence clinic. Prev Med
2006; 42: 114–9.
28 Dani JA, Harris RA. Nicotine addiction and comorbidity
with alcohol abuse and mental illness. Nat Neurosci 2005; 8:
1465–70.
29 Coe JW, Brooks PR, Vetelino MG et al. Varenicline: an a4ß2
nicotinic receptor partial agonist for smoking cessation. J Med
Chem 2005; 48: 3474–7.
30 Cohen C, Bergis OE, Galli F et al. SSR591813, a novel selective
and partial a4ß2 nicotinic receptor agonist with potential as
an aid to smoking cessation. J Pharmacol Exp Ther 2003; 306:
407–20.
31 Oncken C, Watsky E, Reeves K et al., and the Varenicline Study
Group. Smoking cessation with varenicline, a selective nicotinic
receptor partial agonist: results from a phase 2 study. [Abstract].
Presented at the Annual Meeting of the Society for Research on
Nicotine and Tobacco, Prague, 2005.
32 Tonstad S, Hays JT, Jorenby DE et al. Smoking cessation
efficacy and safety of an a4ß2 nicotinic receptor partial agonist–
results from varenicline in cessation therapy: optimizing results.
Presented at the Annual Meeting of the American. Heart Dallas:
Association, 2005.
33 Obach RS, Reed-Hagen AE, Krueger SS et al. Metabolism and
disposition of varenicline, a selective a4ß2 acetylcholine receptor
agonist, in vivo and in vitro. Drug Metab Dispos 2006; 34:
121–30.
Paper received February 2006, accepted February 2006
ª 2006 The Author
Journal compilation ª 2006 Blackwell Publishing Ltd Int J Clin Pract, May 2006, 60, 5, 571–576