A psychopharmacological treatment
algorithm for generalised anxiety
JR Davidson (Professor Emeritus), Dept. of Psychiatry and Behavioral Sciences, Duke University Medical Center, Durham, NC 27710, USA.
W Zhang Dept. of Psychiatry and Behavioral Sciences, Duke University Medical Center, Durham, NC 27710, USA.
KM Connor Clinical Neuroscience and Ophthalmology, Merck Research Laboratories, Merck & Co., Inc., North Wales, PA 19454, USA.
J Ji Dept. of Mental Health, Shanghai Medical School, Fudan University, Dept. of Psychological Medicine, Zhongshan Hospital, Shanghai 200032, China.
K Jobson Department of Psychiatry, University of Tennessee, Knoxville, TN 37996, USA.
Y Lecrubier European College of Neuropsychopharmacology, Hôpital La Salpetriere, Paris, France.
AC McFarlane The University of Adelaide, Centre for Military and Veterans’ Health, Adelaide, SA 5000, Australia.
DJ Newport Women’s Mental Health Program, Emory University School of Medicine, Atlanta, Georgia 30322, USA.
DJ Nutt Psychopharmacology Unit, Dept. of Community-based Medicine, University of Bristol, Bristol BS1 3NY, United Kingdom.
DN Osser Department of Psychiatry, Harvard Medical School, VA Boston Healthcare System, Brockton Campus, 940 Belmont Street, Brockton, MA 02301, USA.
DJ Stein Dept. of Psychiatry and Mental Health, University of Capetown, Cape Town, South Africa.
ZN Stowe Women’s Mental Health Program, Emory University School of Medicine, Atlanta, Georgia 30322, USA.
O Tajima Dept. of Mental Health, Kyorin University, School of Health Sciences, Tokyo, Japan .
M Versiani Institute of Psychiatry, Federal University of Rio de Janeiro, Rio de Janeiro, Brazil.
Generalised anxiety disorder (GAD) is defined as excessive and
uncontrollable worry and anxiety about everyday life situations. It is a
chronic disorder, and is associated with substantial somatisation, high
rates of comorbid depression and other anxiety disorders, and significant
disability. The evidence base for pharmacotherapy and psychotherapy has
continued to grow, and a wide range of drug choices for GAD now exists.
Current guidelines for GAD generally restrict themselves to presentation of
the evidence for various treatments, which, as a result, generally do not
offer detailed discussion or recommendation of strategies beyond the first
level of treatment, or take into account the individual circumstances of
the patient. Thus, there is a lack of algorithm-based treatment guidelines
for GAD. Our aim is, therefore, to present an algorithm for the
psychopharmacologic management of GAD, intended for all clinicians who
treat patients with GAD, where issues of pharmacotherapy are under
consideration. We also hope that these GAD algorithms and other
guidelines can help to identify high-priority areas that need further study.
In this algorithm, we provide a sequenced approach to the
pharmacotherapy of GAD, taking into account salient symptomatology and
comorbidity, levels of evidence and extent of response. Special issues,
including comorbidity, insomnia, suicidality, substance abuse, treatment
adherence, pregnancy and lactation, cross-cultural issues, use of
medication in the elderly, psychosocial treatment and dosing issues are
algorithm; generalised anxiety disorder; international psychopharmacology
algorithm project; pharmacotherapy
Journal of Psychopharmacology
00(00) (2008) 1–24
©2008 British Association
SAGE Publications Ltd,
Los Angeles, London,
New Delhi and Singapore
Corresponding author: Dr Wei Zhang, Department of Psychiatry and Behavioral Sciences, Duke University Medical Center, Box 3812, Durham, NC 27710, USA.
J Psychopharmacol OnlineFirst, published on October 2, 2008 as doi:10.1177/0269881108096505
Generalised anxiety disorder (GAD) is defined as excessive and
uncontrollable worry and anxiety about everyday life situa-
tions. It is a chronic disorder and associated with substantial
somatisation, high rates of comorbid depression and other anx-
iety disorders, and significant disability.
The lifetime prevalence of GAD is estimated to be in the
range of 2.8–6.6% among adults worldwide, and 12-month
rates are in the range of 0.9–3.6% (Alonso and Lepine, 2007;
Blazer, et al., 1991b; Carter, et al., 2001; Kessler, et al., 2005a,c;
Offord, et al., 1996; Wittchen and Jacobi, 2005). The highest
rates are often reported in the 45 to 55-year age group, with
women twice as likely to have GAD as men (Carter, et al.,
2001; Offord, et al., 1996; Wittchen, et al., 1994). In the elderly,
one study found GAD to be the most common anxiety disorder,
with a prevalence of 10.2% in this population (Beekman, et al.,
In primary care practice, GAD is often diagnosed by the
International Classification of Disease (ICD) criteria, and is
the most common anxiety disorder, with an 8% prevalence
rate (Maier, et al., 2000). Patients with GAD are in fact more
likely to consult with gastroenterologists than psychiatrists
(Kennedy and Schwab, 1997), and a high proportion of
difficult-to-treat high medical service users of hospital medical
services have GAD (Lin, et al., 1991). GAD is linked to the
overuse of medical services: emergency department visits, hos-
pitalisations, diagnostic and laboratory tests, pharmacy costs
and so on. Recognition of anxiety and depression in primary
care is poor, with only 23% of pure anxiety cases being recog-
nised compared with 56% of depression cases. The various sta-
keholders (patients, family members, employers and insurers)
in a patient’s outcome may act in such a way as to complicate
treatment of anxiety (Roy-Byrne and Wagner, 2004).
GAD follows a chronic course in many cases and it is not
uncommon to find that patients presenting for treatment have
experienced active symptoms of the disorder for more than
10 years. Although it can remit spontaneously, rates of sponta-
neous remission over 5 years are less than 40% in the case of
DSM-IIIR criteria (Yonkers, et al., 2000), and a waxing and
waning course is more characteristic. Lifetime comorbidity
with another Axis I disorder occurs in 90% of subjects with
GAD, depression being found in over 60% (Wittchen, et al.,
1994). When compared with chronic medical disorders, there
is evidence that GAD is as disabling (Kessler, 2000).
The evidence base for pharmacotherapy and psychotherapy
has continued to grow, and a wide range of drug choices for
GAD now exists. Current guidelines for GAD include those of
1) the British Association of Psychopharmacology (BAP)
National Institute of Clinical Excellence (NICE) (http://www.
nice.org.uk/CG022NICEguideline) and (http://www.nice.org.
uk/CG022quickrefguide); 3) the World Federation of Societies
of Biological Psychiatry (WFSBP) guidelines for the pharma-
cological treatment of anxiety, obsessive-compulsive and post-
et al.,2005); 2)the
traumatic stress disorders (Bandelow, et al., 2002); 4) Consen-
sus Statement on GAD from the International Consensus
Group on Depression and Anxiety (Ballenger, et al., 2001a);
5) the Canadian guidelines (Swinson, et al., 2006); 6) the
South African Primary Care Algorithms (http://www.mental-
healthsa.co.za/disclaimer.html); (7) the World Council of Anx-
iety Recommendations for the Long-Term Treatment of GAD
(Allgulander, et al., 2003) and (8) the Singapore Ministry of
Health guidelines (http://www.moh.gov.sg/mohcorp/publica-
These guidelines generally restrict themselves to presenta-
tion of the evidence for various treatments, but do not offer
detailed discussion or recommendation of strategies beyond
the first level of treatment. Thus, there is a lack of algorithm-
based treatment guidelines for GAD. Our aim is, therefore, to
present an algorithm for the psychopharmacologic manage-
ment of GAD. It is important to recognise that two broad
approaches are established, based on good evidence, for treat-
ing GAD – the pharmacological and the psychosocial methods
of treatment. We are unaware of substantial evidence that the
combination of the two adds any further benefit, but acknowl-
edge that common clinical practice combines the two forms of
treatment simultaneously or in sequence (Kuzma and Black,
For decades, pharmacotherapy for generalised anxiety (pre-
viously called anxiety neurosis) was confined to the benzodia-
zepines (BZDs), following the introduction of chlordiazepoxide
and diazepam in the late 1950s to early 1960s. However, drug
treatments expanded with the development of the serotonin
5HT1apartial agonist buspirone in the 1980s. In the 1990s,
the broad spectrum utility of the antidepressants became
apparent, particularly the serotonergic agents, and these are
now considered by many to be first line pharmacotherapy for
GAD. Recent research is focusing on the development of drugs
with novel mechanisms of action, as well as on various phar-
macologic augmentation strategies with other psychotropic
In this algorithm, we provide a sequenced approach to the
pharmacotherapy of GAD, taking into account salient symp-
tomatology and comorbidity, and extent of response. Our
recommendations are based on levels of evidence (LOE, see
Table 2 and appropriate nodes in the text and flowchart)
where this is available, or on informed clinical opinion where
evidence is absent, as tends to be the case for augmentation or
combination treatments, for example. We also cover special
issues, including comorbidity, insomnia, suicidality, substance
abuse, treatment adherence, pregnancy and lactation, cross-
cultural issues, use of medication in the elderly, psychosocial
treatment (PST) and dosing issues. This approach is similar to
one that we have previously taken with another anxiety disor-
der post-traumatic stress disorder (PTSD) [(Davidson, et al.,
2005) and http://www.ipap.org/ptsd/]. In developing these algo-
rithms, we have attempted to use many of the methods that are
thought essential to guideline development, as described, for
example, by the Appraisal of Guidelines Research and Evalua-
tion (AGREE) instrument (AGREE, 2003) (e.g., defining
2Psychopharmacological treatment algorithm for GAD
scope and purpose, obtaining stakeholder involvement, rigor-
ous development, clarity and presentation, applicability and
The International Psychopharmacology Algorithm Project
(IPAP) is a not-for-profit organisation which, through years
of networking, has developed contacts in Europe, Asia, South
America, Africa and Australia to help recruit faculty; it has
historically chosen experts, one disease at a time, to develop
treatment algorithms. This particular group of consultants
was selected to represent all major regions of the world, with
leading authorities participating from China, Japan, Australia,
South America, North America, Europe and Africa, along
with expertise in anxiety disorders. The group conducted its
business via conference calls and emails, in which a draft doc-
ument was circulated and repeatedly revised until agreement
was reached on the final version. Existing guidelines define
levels of scientific evidence in different ways, and consensus
does not exist as to how this should be done; our group
adopted criteria that were essentially similar to those of other
guidelines, using a range of levels, starting with level 1 (the
most stringent) and ranging down to level 4 (the least rigorous).
For level 1, for example, we required more than one placebo-
controlled study in which a sample size of 30 patients was
enrolled. With regard to previous GAD guidelines, some pro-
vide no evidence categories, whereas others use somewhat dif-
ferent definitions. Some accept results in which statistically
significant treatment differences were found on the primary
outcome, whereas others accept evidence based on significant
differences between drug and placebo on important secondary
measures, with trend levels of significance on the primary scale.
In our evaluation of evidence, we followed the latter conven-
tion, noting those (two) studies where we rely on a finding
based on a secondary analysis of outcome.
Intended use and objectives of the algorithms
The algorithms are intended for any clinician, regardless of
experience, who treats patients with GAD, whether generalist
(e.g., primary care) or specialist (psychiatry). Currently, ver-
sions of these algorithms exist in Chinese, English and Spanish.
How the algorithms are used depends on the nature of the
question. Its modular design on the IPAP website (http://www.
ipap.org/gad/) means that the algorithms can readily be used as
isolated topics to inform a specific question, for example, ‘How
do I manage a patient with GAD who is suicidal?’, or ‘What
recommendations are given about using a benzodiazepine in
my patient?’ Some physicians may desire more information
on the diagnosis of GAD by ICD-10 criteria, or which class
of drug to use first. Others may find useful information, for
example, on the question of sleep pathology in a treatment-
resistant case of GAD, or the management of GAD in the
elderly. Such an approach can also work well with the hard
copy–printed document. The flowchart serves as an easy way
to locate the placement of a question in the text; for example,
the management of GAD with comorbid unipolar depression is
identified at node 14 in the flowchart and text information can
be found at the same place in the print version. In some cases,
users who are interested in the complete picture may want to
study the entire document from start to finish.
The principal purpose of these algorithms is to facilitate
clinical management of patients with GAD, with a focus on
pharmacotherapy. As noted above, we recognise that the evi-
dence base is strongest for first and second line forms of mono-
therapy, that is, for antidepressants, BZDs and other drug
classes. Less evidence exists for third line approaches and
beyond, for augmentation, combination treatment, transcul-
tural considerations and for situations where medical/psychiat-
ric comorbidity is present. Thus for these issues, more reliance
is made on clinical opinion, case reports and uncontrolled
trials. We hope that these GAD algorithms and other guide-
lines can help to identify high-priority areas, which need fur-
ther study. For example, the algorithm might be used for
hypothesis generation and in the design of step-wise evaluation
of treatment effectiveness and outcome for GAD, much like
STAR*D (Sequenced Treatment Alternatives to Relieve
Depression) for depression (Fava, et al., 2003) and CATIE
(Clinical Antipsychotic Trials in Intervention Effectiveness)
for schizophrenia (Lieberman, 2007).
The GAD algorithm
The GAD Algorithm Flowchart and Addenda are essential
accompaniments to this article, and can be downloaded at
http://www.ipap.org/gad/. The flowchart is also shown in
Figure 1 in this review, and contains informational nodes with
details described below.
Node 1: Diagnosis of GAD
No diagnostic category in psychiatry has changed as much
over the past 25 years as GAD. The changes reflect, in part,
the results of psychopharmacological studies designed to
improve the specificity of treatments for this disorder (Rickels,
et al., 1993). These changes make GAD one of the more diffi-
cult disorders for making a psychopharmacology algorithm. It
has been a moving target, and to interpret the studies we need
to know which version of DSM was used for the particular
study. An important change, from the standpoint of psycho-
pharmacology, is the movement to the present criteria where
the core problem is conceived to be chronic, excessive worrying
that is difficult to control and causes impairment. This ‘psychic
component’ of the disorder may respond better to antidepres-
sants than BZDs (Rickels, et al., 1993). GAD used to be (in
DSM-III) predominantly a disorder of autonomic, motor and
other somatic manifestations of anxiety. These symptoms may
respond better, in the short term (2 weeks) to BZDs than to
antidepressants (Rickels, et al., 1993), although by week 8 the
antidepressants (imipramine, trazodone) become as effective or
almost as effective. Some patients previously diagnosed with
Psychopharmacological treatment algorithm for GAD3
treatment (Node 3–7a); Green-, second line treatment (Node 7b–11); Blue, third line treatment (Node 7c, 12–16); Purple, assessment and evaluation.
The flowchart is used by permission of the International Psychopharmacology Algorithm Project, http://www.ipap.org.
International Psychopharmacology Algorithm Project (IPAP) Generalised Anxiety Disorder (GAD) Algorithm Flowchart. Yellow, first line
4 Psychopharmacological treatment algorithm for GAD
GAD would now be classified as having somatoform disorders
in DSM-IIIR or IV.
The concept of GAD was first described in DSM-III, as an
anxiety condition. To make the diagnosis required patients to
experience at least 1 month of symptoms from three of four cat-
egories (motor tension often of the back or neck, tension head-
aches and muscle pains secondary to tension and making people
reactive and jumpy to sudden events), autonomic hyperactivity
(sweating, dry mouth, racing heart rate), apprehensive expecta-
tion (a sense of worry about everything in the future, predicting
negative outcomes for future events and occurrences – when this
occurs primarily at night, it can lead to insomnia) and vigilance/
scanning (always looking for threats and problems). There were
sweeping restrictions as to the allowable coexistence of other
Axis I disorders, and GAD was believed to produce no more
than mild impairment. With subsequent editions of the DSM,
the criteria became simpler, shorter and the disorder is now
acknowledged to result in significant impairment. The current
DSM-IV-Text Revision (TR) criteria for GAD (American Psy-
chiatric Association, 2004) and the ICD-10 criteria for GAD
(International Statistical Classification of Diseases and Related
Health Problems, 10th edition, Geneva: World Health Organi-
sation), which differs considerably from those of DSM-IV-TR
(Text Revision), are given in Table 1.
Although DSM-IV specifies the minimal duration of
6 months and at least three associated symptoms, ICD-10
does not require minimal duration or number of symptoms.
ICD-10 criteria place greater emphasis on the presence of
somatic symptoms. Because nearly all the evidence upon
which our algorithms are based comes from studies which used
DSM-IV (or in some cases, DSM-III or IIIR), the algorithm
recommendations are based on DSM, as the evidence is lacking
for the creation of an algorithm based upon the ICD criteria.
At the same time, we recognise that the diagnostic criteria for
DSM-IV GAD may be overly conservative (Kessler, et al.,
2005b; Ruscio, et al., 2005), and that clinicians may therefore
use their judgment in applying the current algorithm to patients
who meet many, but not all, the relevant diagnostic criteria.
Clinical judgment should also be used when applying the
DSM criteria, as opposed to using them as a checklist.
Node 2: Consider diagnosis and other special issues at
At the initial assessment and at intervals during treatment, the
clinician should consider other issues relevant to the patient
with GAD (as listed under ‘CONSIDER AT EACH STAGE’
in Figure 1).
With the extensive degree of psychiatric and medical comor-
bidity that is associated with GAD, it is important to keep in
mind the possibility that ongoing symptoms could be attribut-
able to either separate psychiatric or medical morbidity or a
disorder which is comorbid with GAD. An initial evaluation
needs to include assessment for all relevant comorbid condi-
tions, with appropriate physical examination and laboratory
testing, with attention to thyroid, parathyroid, glucose func-
tion, as well as assessment of current use of prescription and
over-the-counter medications, intake of caffeine, alcohol and
GAD is often found in association with other medical con-
ditions. For example, Castillo, et al. (1993) noted clinically sig-
nificant GAD-like symptoms in 40% of patients with stroke,
27% of whom met full criteria for GAD. In people with
migraine, there was a 10.2% rate of GAD (Guillem, et al.,
1999). Subjects with GAD report a marked increase in the
rate of peptic ulcer disease (Goodwin and Stein, 2002). As
noted by Stein (2001), physical symptoms such as fatigue, pal-
pitations, chest pain, hyperventilation, tension headache,
insomnia, back pain, muscle tension, as well as hypertension,
diabetes and heart disease all make it more likely that the
patient with GAD will make frequent visits to the doctor.
DSM-IV and ICD-10 criteria for generalised anxiety disorder
Diagnostic system Diagnostic codeDiagnostic criteria
DSM-IV300.02a) Excessive anxiety and worry lasting at least 6 months
b) Difficulty in controlling the worry
c) Presence of 3 of the following 6 associated symptoms: restlessness, fatigability, difficulty concentrating,
irritability, muscle tension and sleep disturbance
d) Focus of the anxiety and worry not confined to another anxiety or somatoform disorder
e) Significant distress or functional impairment because of the symptoms
f) Symptoms not because of a substance (e.g., a drug of abuse, a medication) or a general medical
condition (e.g., hyperthyroidism)
a) Generalised and persistent anxiety not restricted to any particular environment (i.e., it is ‘free-floating’)
b) Dominant symptoms including persistent nervousness, trembling, muscular tension, sweating, light-
headedness, palpitations, dizziness, epigastric discomfort and fears of impending illness or an accident
c) Exclusion of neurasthenia
Used by permission of the International Psychopharmacology Algorithm Project, http://www.ipap.org.
DSM-IV, Diagnostic and Statistical Manual of Mental Disorders, 4th edition; ICD-10, International Statistical Classification of Diseases and Related
Health Problems, 10th edition.
Psychopharmacological treatment algorithm for GAD5
Among the concomitant psychiatric disorders and symp-
toms to be included in the differential diagnosis are depression,
bipolar disorder, alcohol and substance use disorders, other
anxiety disorders, suicidal behaviour and attention deficit and
hyperactivity disorder (ADHD). PTSD is often associated with
extensive somatisation (Andreski, et al., 1998), and hypervigi-
lance is seen in both GAD and PTSD. One benefit of consider-
ing PTSD in the differential diagnosis is that it necessitates
obtaining a trauma history, which may declare an aspect of
the patient’s life which is important for the understanding of
their anxiety, vulnerability or prognosis.
There is some evidence that pharmacotherapy for GAD
may be associated with lower risk of developing secondary
major depression, based on a large population study (Goodwin
and Gorman, 2002). Although the two groups were only retro-
spectively randomised, the findings do suggest that pharmaco-
therapy for GAD can confer a preventive effect against depres-
sion. In the case of GAD and other comorbid disorders such as
ADHD, Kessler, et al. (2006) have similarly raised for discus-
sion the importance of adequately treating both conditions.
It is often necessary to address the following issues as part
of an adequate assessment in the diagnosis and treatment of
GAD. The GAD Algorithm Flowchart and Addenda are
essential accompaniments to this article and can be down-
loaded at http://www.ipap.org/gad/.
disorders can be present in patients who are being assessed
for GAD and may complicate accurate diagnosis and treat-
ment. Initially, the patient should have a full psychiatric and
medical history with appropriate consideration or referral for
laboratory and physical examination. As part of the initial
diagnostic evaluation, and after a failed trial of treatment, the
clinician should look for common coexisting conditions, such
as depression, alcohol problems, bipolar disorder and undiag-
nosed medical illness, for example, endocrine (thyroid), pulmo-
nary or cardiac disease. We would note here that according to
the DSM-IV criterion F, if the GAD symptoms occur exclu-
sively during the course of any mood disorder (unipolar or
bipolar), a separate diagnosis of GAD is not made. However,
Zimmerman found that this hierarchical relationship may not
be supported by the evidence. Depressed patients with GAD
(more severely ill) confined to the depressed periods and
GAD not confined to the depressed periods were comparable
on many parameters and differed from a pure depression con-
trol group (Zimmerman and Chelminski, 2003).
A number of newer antidepressants, for example, some selec-
tive serotonin reuptake inhibitors (SSRIs) and serotonin norepi-
nephrine reuptake inhibitors (SNRIs) have been shown to be
effective in both GAD and major depression, suggesting that
they be used first-line when there is overlap. Although anxiety
decreases when such agents are used to treat major depression,
and depression decreases when such agents are used to treat
GAD, there are no prospectively designed trials published on
the treatment of GAD with comorbid depression. A study using
Concomitant psychiatric or medical
post-hoc analysis from large controlled trials has shown that
venlafaxine-XR (Goodman, et al., 2005; Silverstone and Salinas,
2001) is superior to placebo in subgroups with dual diagnoses of
GAD and a secondary diagnosis of major depression. Other rel-
evant studies support the use of antidepressants, such as mirta-
zapine, in this context (Goodnick, et al., 1999). In one small
placebo-controlled study of chromium picolinate in patients
with atypical depression, 87% of whom had concomitant
GAD, the drug was superior to placebo in the full sample and
in the subsample with GAD (Davidson, et al., 2003).
For continuing depression, antidepressant therapy, if toler-
ated, should be titrated to its maximum dose. There may be
some circumstances where the use of a BZD has intensified
depression or promoted its emergence as a side-effect, most
particularly in panic disorder treated with high-dose clonaze-
pam. If it should occur during treatment of GAD, the clinician
is advised to withdraw the offending agent.
For ongoing, or emergent, hostility during the course of
BZD therapy, assessment is recommended as to possible sub-
stance abuse or even abuse of the prescribed BZD, and appro-
priate action taken, which would normally be to either reduce
the BZD dose or taper the drug towards discontinuation. In
such a situation, SRI therapy is recommended if the symptoms
are due to GAD or depression.
Although GAD with comorbid substance use disorder has
not been well studied (see below), there is indirect, albeit
mixed, evidence that a serotonergic antidepressant may be use-
ful in either depressed or anxious patients with comorbid
alcohol-related problems (Nunes and Levin, 2004; Schade,
et al., 2003). It should be noted that anxiety in the context of
substance abuse may reflect intoxication or withdrawal symp-
toms (see below).
For bipolar disorder, other approaches might be considered.
Before introducing a drug for GAD, it would be necessary to
ensure adequate mood stabilisation because anxiety may reflect
poor control of the mood disorder. No randomised controlled
trials have been conducted in patients with bipolar disorder
and any co-occurring anxiety disorder. Among agents with
antimanic or mood-stabilising effects, evidence of anxiolytic
efficacy from placebo-controlled trials exists for valproate in
the treatment of panic disorder (Primeau, et al., 1990; Wood-
man and Noyes, 1994) (LOE 4); lamotrigine (Hertzberg, et al.,
1999) (LOE 3), olanzapine (Butterfield, et al., 2001; Stein,
et al., 2002) (LOE 3) and risperidone (Hamner, et al., 2003;
Reich, et al., 2004) (LOE 3) in PTSD; olanzapine (Bystritsky,
et al., 2004; Shapira, et al., 2004) (LOE 1), quetiapine (Carey,
et al., 2005; Denys, et al., 2004; Fineberg, et al., 2005) (LOE 1)
and risperidone (Erzegovesi, et al., 2005; Hollander, et al.,
2003; McDougle, et al., 2000) (LOE 1) as adjunctive treatment
in SSRI-refractory obsessive-compulsive disorder and risperi-
done as augmentation to a SSRI in GAD (Brawman-Mintzer,
et al., 2005) (LOE 2). Antidepressants from many classes have
efficacy in the treatment of most anxiety disorders, but present
the challenge of minimising switch risk when used in conjunc-
tion with a mood stabiliser. Among novel antiepileptic agents
without proven thymoleptic properties, valproate with sug-
6 Psychopharmacological treatment algorithm for GAD
gested efficacy in panic disorder might be a candidate for the
treatment of GAD comorbid with bipolar disorder (Keck,
et al., 2006) (LOE 5). In one review, Freeman, et al. felt that
antidepressants are best avoided for treating the anxiety disor-
ders in these patients because of risk for ‘switch’ to hypomania
and mania, and preferred anxiolytic antimanics and atypical
antipsychotics. However, as noted below (see node 11b), there
are regional practice differences in respect of the role of anti-
depressants for patients with bipolar depression. For GAD,
they made favourable comments about using BZDs, in that
these agents are efficacious, relatively safe and well-tolerated.
However, they also caution that long-term use of BZD may be
problematic because of tolerance, dependence and withdrawal
issues (Freeman, et al., 2002).
In summary, a careful evaluation for comorbidity is needed
before initiating treatment, there is a limited evidence base spe-
cifically addressing comorbidity, and the choice of treatment in
GAD patients with comorbidity needs to be formulated with
appropriate clinical judgment.
(NCS) did not find GAD to be significantly associated with
suicidal ideation or attempts (Sareen, et al., 2005b), unless
comorbidity was present, increased rates of attempted suicide
Although the National Comorbidity Survey
or suicidal ideation are seen in GAD among young adults aged
14–24 (Wunderlich, et al., 1998), as well as among an adult
population in the Netherlands (Sareen, et al., 2005a). There is
evidence that subthreshold depression associated with GAD
may increase the rate of attempted suicide (Balazs, et al.,
2000). Suicide is certainly a potential risk when patients with
GAD have comorbid MDD. In cases where suicide risk is con-
sidered to be serious, the treatment of this component takes
priority over treating GAD. Although suicidal patients are
excluded from almost all GAD trials, we consider that a ratio-
nal treatment approach would be to institute an antidepressant
drug and avoid the sole use of a BZD or other drug which is
devoid of antidepressant effect, in this circumstance. General
guidelines for assessment and treatment of suicide (e.g., APA
guideline) should also be observed.
There has been considerable attention to the nature of the
relationship between use of antidepressants and emergence of
suicidal thoughts and behaviours in adults who receive these
drugs, with inconclusive results (Gunnell, et al., 2005; Tiiho-
nen, et al., 2006). There is some evidence that venlafaxine
may be associated with greater risk of suicide among high-
risk subjects (i.e., those who were without psychosis and hospi-
talised because of a suicide attempt) in one study (Rubino,
et al., 2007; Tiihonen, et al., 2006) and among those who
were prescribed with the index antidepressant for the first
time according to a general practice research database
(Rubino, et al., 2007). However, when confounding factors
are taken into account, such as family psychiatric history, pre-
vious suicide attempt, severity of illness, this risk abated con-
siderably (Rubino, et al., 2007). Cipriani, et al. (2007) have
observed that systematic reviews have reported an excess risk
of suicide for children and adolescents with major depression
treated with antidepressants (Hammad, et al., 2006), but not
for adults (Gunnell, et al., 2005). Since 2005, the Food and
Drug Administration (FDA) has added to all antidepressant
labels a warning of increased risk of suicidal thoughts and
behaviour in children and adolescents who take antidepres-
sants, and most recently the warning has also expanded to
include adults aged 18–24. The relevance of all these findings
to a population suffering from GAD remains unclear, and fur-
ther studies are urged in this patient group. The extent to which
early agitation, activation or jitteriness plays an aetiological
part in emergent suicidality, especially in GAD, needs further
exploration (Cipriani, et al., 2005).
GAD and may be the presenting complaint, particularly in
the primary care setting. Sleep difficulty may persist even
after an otherwise good response to SRI drugs. A careful
assessment of whether the insomnia is a symptom of GAD,
or a separate disorder, should be conducted early on. In addi-
tion, the initial assessment and treatment should include atten-
tion to sleep hygiene, lifestyle issues such as exercise (especially
in the morning), diet, abstaining from products containing sti-
mulants (coffee, OTC drugs with ephedra, supplements such as
Insomnia is a common and troubling feature of
Level of evidencea
1 = More than one placebo-controlled trial having total sample sizes over
2 = One placebo-controlled trial (or active versus active drug comparison)
with total sample size of 30 or greater
3 = One or more small (n < 30) placebo-controlled trial
4 = Case reports or open-label trials
5 = Expert clinical consensus without published evidence
aTreatment Guidelines for generalised anxiety disorder include.
1) The British Association of Psychopharmacology (BAP) (http://www.
BAP.org.uk) (Baldwin, et al., 2005).
2) The National Institute of Clinical Excellence (NICE) (NHS National Insti-
tute of Clinical Excellence). Anxiety: Management of anxiety (panic disor-
der, with or without agoraphobia, and generalised anxiety disorder) in
adults in primary, secondary and community care (Clinical Guideline 22.
December 2004. Available at http://www.nice.org.uk/CG022NICEguideline
3) The World Federation of Societies of Biological Psychiatry (WFSBP)
guidelines for the pharmacological treatment of anxiety, obsessive-
compulsive and post-traumatic stress disorders (Bandelow, et al., 2002).
4) Consensus Statement on Generalised Anxiety Disorder from the Interna-
tional Consensus Group on Depression and Anxiety (Ballenger, et al.,
5) The Canadian Guidelines (Swinson, et al., 2006).
(6) The South African Primary Care Algorithms (http://www.mental-
Used by permission of the International Psychopharmacology Algorithm
Psychopharmacological treatment algorithm for GAD7
ginseng or Ma Huang, energy drinks, hoodia, myridia), as well
as an evaluation of whether the sleep disturbance is caused or
exacerbated by prescription medications such as some of the
serotonergic antidepressants or stimulants. Some antidepres-
sants have more immediate sleep promoting effects than others;
these include the sedative tricyclic antidepressants plus trazo-
done and mirtazapine. In the longer term, the non-sedating
agents including SSRIs and SNRIs often improve sleep as
well, probably secondary to their anxiolytic actions. Discussion
of ways in which antidepressants effect sleep appear in the
review by Mayers and Baldwin (2005).
In the event of continued poor response, it is important to
consider the possibility of sleep-related breathing disorder, such
as obstructive sleep apnea (OSA), or other sleep disorders, for
example, periodic limb movement disorder, restless leg syn-
drome (RLS), etc. In such circumstances, and if other symp-
toms are present to suggest these disorders, a polysomnogram
can be obtained if resources are available. If a sleep disorder is
identified or diagnosed with a sleep study, then appropriate
treatment (e.g., OSA- continuous positive airways pressure
(CPAP), RLS- pharmacotherapy) can be instituted. Patients
with GAD commonly endorse tiredness, but sleepiness is
found only rarely: when present, it suggests a comorbid pri-
mary sleep disorder or, possibly, a side-effect of some antide-
pressants and sedating anxiolytics.
Patients with current or recent substance dependence or abuse
Individuals with GAD may misuse alcohol, cannabis or other
substances in an effort to ameliorate their anxiety. In time,
such misuse may result in substance abuse or dependence; ade-
quate treatment of GAD may ultimately produce better control
of the substance abuse in many cases. As a general principle
with these patients, conservative and less complicated regimens
are to be preferred. We recommend that patients with GAD
first undergo withdrawal from his or her substance(s) of abuse
or dependence, and make a commitment to abstain from future
use of these substances. Of course, the patient may not succeed
in achieving this goal, and compliance with the commitment to
abstinence must be followed closely. Our recommendations
below are consistent with those of Brady, et al. (2007) and the
American Psychiatric Association (APA) Guidelines (Ameri-
can Psychiatric Association, 2007). Brady, et al. (2007) noted
that the symptoms of GAD are similar with those of alcohol
withdrawal, and that it is difficult to confirm the diagnosis of
GAD until after abstinence has been attained. The APA
Guidelines advise delay in implementing antidepressant drugs
by 1–4 weeks, to allow the clinician to identify those mood syn-
dromes, which remit without medication intervention. How-
ever, the guidelines endorse early use of antidepressants if
there is a history of previous episodes. However, they offer evi-
dence that drinking outcomes improve less than does mood fol-
lowing antidepressants in patients with comorbid depression
and alcohol dependence.
Although we are unaware of pharmacotherapy studies in
GAD with comorbid alcohol use disorder, the evidence is
mixed from a pilot study of paroxetine in social anxiety disor-
der and comorbid alcohol use disorder (Randall, et al., 2001).
Thus, early use of SSRIs or other antidepressants may benefit
these patients. A possible role exists for buspirone in patients
with GAD and alcohol use disorder (Kranzler, et al., 1994),
but BZDs are generally contraindicated in this population,
except for the initial period of detoxification (Lingford-
Hughes, et al., 2004). Nunes and Levin (2004) have compre-
hensively reviewed antidepressant use in dual diagnosis patients
with depression and substance use disorder.
Before beginning pharmacotherapy for comorbid GAD, it
seems reasonable to recommend that the patient should have
completed withdrawal from his or her drug of abuse or depen-
dence and from any drug used for withdrawal, and be abstinent
for at least one additional week. Symptoms present after absti-
nence of less than 1 week may be due in part to the residual
effects of the substance. If symptoms appear to be diminishing
over the first week of sobriety, and there is no history of these
symptoms before onset of the substance abuse/dependency or
during previous periods of extended sobriety, it is reasonable
to wait at least another week before initiating pharmacother-
apy. However, if this is not possible, the initiation of treatment
with an antidepressant, for example, an SSRI should be consid-
ered. Once the patient has shown a therapeutic response, it may
be easier for them to stop the other misused drugs.
Withdrawal from some substances can be prolonged, and
the residual effects of their presence can affect subsequent med-
ication that may be given. For example, methadone has a half-
life of about 2 days and its effects as an inhibitor of cytochrome
P450 2D6 could therefore persist for more than a week.
A group of patients not specifically addressed in this algo-
rithm are individuals who use drugs of abuse, but who do not
meet DSM-IV criteria for abuse or dependency. Should the
sequence of treatments for these ‘recreational’ users be any dif-
ferent from the standard approach? These frequently encoun-
tered patients have received little research attention and a
prudent treatment approach thus involves reasonable and
appropriate clinical judgment in managing such cases. One
important issue to consider in this group is the possible kin-
dling effects of cocaine and amphetamine derivatives where
the effect may be manifest as GAD symptoms. In this group,
abstinence is important to achieve.
Patients with history of, but no current, substance dependence or
If the patient is not actively abusing or dependent upon sub-
stances presently, but has a history of such abuse or depen-
dence, the treatment approach may require relapse prevention
treatment. The patient may have been recently detoxified, be
under unusual stress or may be experiencing strong cravings
for his or her substance and therefore at high risk of relapse.
This would suggest that management of this problem would be
at least as high a priority for treatment compared with treating
8 Psychopharmacological treatment algorithm for GAD
If the patient’s problem is with alcohol, evidence-supported
pharmacotherapy options could include naltrexone (Srisurapa-
nont and Jarusuraisin, 2005), acamprosate (Verheul, et al.,
2005) or topiramate (Johnson, et al., 2004) (All LOE 1 for the
index disorder; LOE 5 for GAD).
Disulfiram may be of value for prevention of relapse in
cocaine-dependent individuals (Carroll, et al., 2004) (LOE 2
for index disorder. LOE 5 for GAD).
All pharmacotherapy options for substance abuse/depen-
dence seem to work best in the context of ongoing intensive,
structured psychotherapeutic treatment focused on abstinence,
compliance and relapse-prevention (O’Malley, 1995).
If the patient with a history of substance abuse has been
assessed and treated (if appropriate), the clinician may then
return to the GAD algorithm. The clinician’s vigilance towards
detecting comorbid substance abuse needs to be continued
expected effects of medication and how to deal with problems
which arise is essential at the initiation of pharmacotherapy, as
well as with the addition of any new medication. Non-
adherence rates to antidepressants in depression can be as
high as 50% within the first 3 months (Lin, et al., 1995, 2003).
It is likely that similar rates might hold true for patients with
GAD. Reasons for non-adherence are myriad, and include side
effects, lack of efficacy, improvement of symptoms, as well as
ambivalence about treatment and the stigma associated with
taking psychotropic medication. Non-adherence to medication
is higher in patients who view their symptoms from a non-
medical perspective. Lin, et al. (1995) have suggested a number
of strategies to enhance treatment adherence in depression, all
of which, in our view, also apply to GAD. The following edu-
cational messages to patients may prove helpful: (a) take medi-
cation daily; (b) antidepressants (but not BZDs) may take some
weeks to work; (c) continue taking medication even when feel-
ing improved; (d) do not stop taking medication without check-
ing with the prescribing physician; (e) instructions on how to
deal with side effects or to resolve other questions about medi-
cation and (f) schedule pleasant activities. Wingerson, et al.
(1993) have reported that GAD subjects with impulsiveness,
novelty seeking traits and who show dislike of regimentation
are more likely to drop out of medication trials, and may there-
fore constitute a high-risk group for treatment non-adherence.
Somatically focused patients are often reluctant to consider
a psychiatric diagnosis and psychotropic medication. A strong
treatment alliance and credible rationale are called for. The
patient may require a detailed rationale about the relationship
between physical symptoms and mode of action of the medica-
tion. The literature on somatic vigilance and selective attention
is relevant in this regard, and provides one explanatory model
by which the effects of psychotropic medication can be under-
stood (Hoehn-Saric, et al., 2004; James, et al., 1990).
Issues relevant to women of childbearing potential
the prevalence of anxiety disorders in women, there is sparse
information regarding the incidence and course of anxiety dis-
orders during pregnancy and the post-partum period. In a large
prospective longitudinal study of a community sample of 8323
pregnant women in England, 21.9% of the women had clini-
cally significant symptoms of anxiety (Heron, et al., 2004).
Most of the women (64%) who reported elevated levels of anx-
iety during pregnancy also reported elevated levels of anxiety
after delivery. Furthermore, antenatal anxiety predicted post-
partum depression at 8 weeks and 8 months, even after control-
ling for the presence of antenatal depression. The course of
panic disorder during pregnancy has also received attention
(Cohen, et al., 1994, 1996; George, et al., 1987; Northcott and
Stein, 1994) with evidence of a either worsening or new onset in
the post-partum period (Cohen, et al., 1994; Metz, et al., 1988;
Sholomskas, et al., 1993).
There is mounting evidence that maternal anxiety during
pregnancy and the post-partum period potentially poses signif-
icant risk to the child. Maternal anxiety during pregnancy has
been associated with behavioural problems at 4 and 6 years of
age (O’Connor, et al., 2002, 2003). Similarly, in older children
aged 8–9, there is a significant correlation between maternal
antenatal anxiety (12–22 weeks gestation) with ADHD, self-
reported anxiety and increased impulsivity as well as lower
scores on the subtests of the WISC-R in 14 to 15-year-old chil-
dren (Van den Bergh and Marcoen, 2004; Van den Bergh,
et al., 2005). A recent prospective study reported that maternal
depression and anxiety during pregnancy predicted higher rates
of conduct disorder in children (Kim-Cohen, et al., 2005).
As such, although the incidence and course of GAD is not
well studied in women during pregnancy and the post-partum
period, there is clear evidence that untreated anxiety may pose
a significant risk for the mother in the post-partum period and
potentially affect the developmental trajectory of the infant.
The primary issues in women of childbearing potential are
best addressed at initial treatment planning. This includes: 1)
documentation of method of birth control for all women of
reproductive capacity at all visits; 2) over 50% of pregnancies
are unplanned, particularly in the 15 to 25-year age group.
Because it is not possible to ensure that all women are with
reproductive capacity are not pregnant, the physician is
encouraged to approach the clinical decision as if the woman
may be pregnant. Clinicians seldom order routine pregnancy
tests. Urine pregnancy tests neither rule out early pregnancy
(1–3 weeks gestation) nor does a recent or current menses. In
cases, where medications are indicated, the treatment should be
conducted with the medications with the majority of obstetrical
outcome data (e.g., fluoxetine, sertraline, citalopram, paroxe-
tine). The recent FDA pregnancy category change for paroxe-
tine to category D must be included in treatment decisions.
Despite the B category rating, there is a dearth of information
for buspirone during pregnancy and/or lactation. Increased risk
of pulmonary hypertension with SSRI in pregnancy has been
reported and should be considered when advising patients
about risk-benefit issues (Chambers, et al., 2006).
Psychopharmacological treatment algorithm for GAD 9
Retrospective cohort review by Ferreira, et al. (2007) sug-
gests that use of SSRI and venlafaxine during the third trimes-
ter of pregnancy is associated in many cases with a brief
neonatal withdrawal state, characterised by CNS or respiratory
signs, which tend to be more prolonged among premature
The primary and/or adjunctive use of BZDs is generally not
recommended in treating GAD during pregnancy. Earlier
reports with diazepam and chlordiazepoxide suggested a higher
rate of cleft lip. Although cohort studies have failed to show
this increased risk, a meta-analysis (Altshuler, et al., 1996) indi-
cated a small but significant risk of birth defects with BZDs.
Similarly, BZDs are considered ‘contra-indicated’ in breast
feeding according to American Academy of Paediatrics com-
mittee on medications in breast feeding report 2002. Should
BZDs be required during pregnancy and/or lactation – the pre-
ferred agents would include clonazepam and lorazepam. Clo-
nazepam has a better pregnancy rating (category C) relative to
other BZDs. Similarly, lorazepam has the advantages of multi-
ple routes of administration, a history of use in children with
status epilepticus, and a pathway of metabolism that does not
require the foetal/neonatal liver. This information regarding
the approach to women in their reproductive years should
influence the algorithm at all points that include use of BZD.
A review of the potential for pregnancy and/or urine preg-
nancy test should be considered at any juncture in the algo-
rithm that adds adjunctive pharmacotherapy (node 9–17).
GAD in the elderly
elderly than are other anxiety disorders, and has a reported
prevalence rate that ranges from 0.7 to 10.2% (Copeland,
et al., 1987; Flint, 1994; Lindesay, et al., 1989). About one-
half of all elderly subjects with GAD report that their condi-
tion is of recent onset, that is, not simply the continuation of a
long-standing problem (Blazer, et al., 1991a). In elderly
patients who present with new onset anxiety, it is also impor-
tant to consider relevant medical aetiologies, as well as iatro-
genic causes, for example, medication side effects or drug–drug
interactions. Appropriate medical work-up is recommended
when evidence suggests that other disorders may be present.
One study has found buspirone to show greater benefit than
placebo in patients over age 65 with GAD (Bohm, et al., 1990)
(LOE2). Retrospectively derived pooled data (LOE2) from five
studies report efficacy of venlafaxine-extended release in
patients with GAD who are age 65 and older (Katz, et al.,
2002). A study which compared sertraline, cognitive beha-
vioural therapy (CBT) and a waiting list (WL) control in a
mixed anxiety population age 60 and older (LOE2 for aggre-
gated diagnoses of anxiety, including GAD), found that both
active treatments were superior to WL, with the pattern of
results generally suggesting a more robust effect for sertraline
over CBT (Schuurmans, et al., 2006). Of all entered subjects,
35% met criteria for GAD, 45% for panic disorder and 20%
other forms of anxiety. There is also a positive placebo-
controlled study of citalopram in elderly patients with anxiety
GAD is more likely to be seen in the
disorders, mainly GAD (Lenze, et al., 2005) (LOE 2 for the
Side effects of antidepressants are of greater concern in the
elderly, as for example the intolerance which was found in one
study of venlafaxine-extended release in frail elderly subjects,
as well as the greater risk of hyponatremia in response to
SSRI drugs in older patients (Jacob and Spinler, 2006; Oslin,
et al., 2003).
BZD use in the elderly is problematic, given their higher
incidence of falls, hip fracture, withdrawal difficulties and
increased risk of cognitive impairment (Bogunovic and Green-
field, 2004; Krasucki, et al., 1999). In addition, there are phar-
macokinetic and pharmacodynamic considerations to be kept
in mind, in particular the greater likelihood of accumulation
of those drugs, which are metabolised by oxidation and which
have longer half-lives. However, drugs with shorter half-lives
may produce more severe withdrawal if used in the long-term.
Elderly people are more likely to be taking other medications,
often for long-term treatment, and thus are at more risk for
Cognitive therapy for GAD has been adapted to the elderly,
with benefit (LOE 2) (Schuurmans, et al., 2006; Stanley and
ing of GAD symptoms in different cultures (Ballenger, et al.,
2001b; Maier, et al., 2000). For example, an internet survey in
Japan (Tajima, 2004) found that the number of patients with
GAD who sought medical treatment is less than 17% and that
subjects tend to see their tension and excessive worry as being
normal reactions to negative life events. There is a need to
screen for GAD with appropriate questions, and to use stan-
dard diagnostic criteria and good clinical judgment, to ensure
that a valid diagnosis is made. Although it has been argued
that somatisation varies in prevalence across different cultures,
GAD is often accompanied by depression and somatisation in
different populations. In formulating a treatment plan, it is,
however, useful to consider patients’ explanatory model of
their illness; this allows the clinician to understand the meaning
of the symptoms for them, including any pertinent cultural
aspects, and to negotiate an agreed upon treatment accord-
ingly. Theories of illness are influenced by cultural factors
and patients’ beliefs should be discussed and carefully
There may be significant differences in report-
Node 3: Treatment
An initial treatment choice may be made as to whether medi-
cation, PST or both will be given. Because GAD is a chronic,
relapsing condition, with only a low rate of spontaneous remis-
sion, in almost all cases of pharmacotherapy, when used, is
recommended for a minimum of 1 year, assuming some degree
of response. Early discontinuation is associated with higher
relapse rates. However, the optimum duration of treatment
10Psychopharmacological treatment algorithm for GAD
varies from one patient to another, and we are unaware of any
reliable predictors of who is at most at risk for relapse.
Recommended doses of each medication are provided in
Table 3, but some patients with GAD are sensitive to medica-
tion side-effects and may require low initial starting doses, to
be followed by a persistent but carefully managed upward dose
titration, at a rate determined by tolerability.
Node 4: Psychosocial treatment
Cognitive behavioural treatment is efficacious for GAD (Bor-
kovec and Ruscio, 2001) (LOE 1). There is no evidence that
combined use of CBT with drug therapy enhances CBT
alone, but CBT in combination with a sub-therapeutic dose of
diazepam produces a greater effect than the same dose of diaz-
epam alone (Power, et al., 1989). Psychotherapeutic treatments
include CBT and relaxation therapy (although there is little evi-
dence base for the latter), as well as social treatments include
problem solving might also be helpful (Catalan, et al., 1984).
Node 5: Monotherapy with SSRI or SNRI: 4 to 6-week
evaluation with adequate dosing
Please refer to Table 3 for initial prescribing and dose titration
First line treatment antidepressants
diagnosis of GAD, the recommended first line choice will be
an SSRI or SNRI drug. Level 1 evidence supports the follow-
ing SSRI drugs for DSM-IV diagnosed GAD: escitalopram
(Davidson, et al., 2004); paroxetine-immediate release (Pollack,
et al., 2001; Rickels, et al., 2003); sertraline (Allgulander, et al.,
2004). Of the above three medications, sertraline would be the
best option for women of reproductive years, from the perspec-
tive of the extent of safety data in pregnancy and lactation.
Level 1 evidence supports the SNRI drugs, venlafaxine-
extended release and duloxetine, in patients who met DSM-
IV criteria for GAD (Allgulander, et al., 2001; Davidson,
et al., 1999; Hartford, et al., 2007; Koponen, et al., 2007; Rick-
els, et al., 2000; Rynn, et al., 2007), including one long-term 6-
month study of extended-release venlafaxine.
Other antidepressants (imipramine and trazodone) have
been reported to show greater efficacy than placebo in one
study of DSM-III–based GAD (level 2) (Rickels, et al., 1993),
but these are not recommended as first choice treatments
because of poorer tolerability and higher risk of potentially
serious side effects.
Lower levels of evidence (LOE 4) support the use of mirta-
zapine in GAD (Gambi, et al., 2005), as well as in GAD with
concomitant major depressive disorder (Goodnick, et al.,
1999). Some evidence exists for citalopram (LOE 2) (Blank,
et al., 2006; Lenze, et al., 2005; Varia and Rauscher, 2002);
nefazodone (LOE 4) (Hedges, et al., 1996) and fluoxetine in
children and adults (LOE 2 for children and for adults) (Birma-
her, et al., 2003; Pollack, et al., 2006). Although the branded
Following a DSM-IV
form of nefazodone (Serzone) has been discontinued, generic
forms of this drug are still available, although we recommend
that the drug should be generally avoided because of liver tox-
icity. In a head-to-head comparison trial of bupropion and
escitalopram, bupropion was equivalent to escitalopram in
treating GAD (LOE 3) (Bystritsky, et al., 2006).
Role of non-antidepressant drugs in GAD
acknowledged that many practitioners use the following drugs
as first line treatment, we recommend their use only as a sec-
ond line form of monotherapy after intolerance to a series of
antidepressants. There is also a place for these drugs in aug-
mentation (see below Nodes 9–17) or on occasion, early in
treatment for marked agitation or severe sleep disturbances.
Although it is
the short-term efficacy of BZD drugs for GAD, as reviewed
by Mitte, et al. (2005), and these data support BZD for all
the recent DSM iterations of GAD, beginning with DSM-III.
Their rapid onset of efficacy, reasonable side-effect profile and
good tolerability make them appealing drugs for many clini-
cians. However, in other quarters, these medications are looked
upon with disfavour because of their abuse potential and asso-
ciation with dependence. In general, we recommend BZDs as
second line treatments, to be chosen after intolerance has been
established to antidepressants. However, Schweizer and Rickels
(1997) propose that BZD are appropriate first line choices in
two circumstances: 1) short-duration GAD type reactions in
response to stress and 2) where somatic symptoms are more
prominent than psychic symptoms (Rickels, et al., 1982,
1993). In this regard, long-half life BZDs may hold merits or
advantages in anxiety disorders except in the elderly because of
their relatively low risk of inter-dose rebound anxiety and with-
drawal symptoms compared with those with short-half life.
Nevertheless, although acknowledging that antidepressants
typically have a greater benefit on psychic than somatic symp-
toms, we recommend the use of an SSRI or SNRI for treating
the somatic symptoms of GAD, based upon their proven effi-
cacy on this symptom cluster. Although antidepressants may
have a slower onset of action than BZDs, they are eventually
as effective, if not more so, and a satisfactory effect is usually
obtained. BZD are not recommended where GAD is charac-
terised by substantial hostility, impatience, irritability and
impulsivity, which can sometimes be made worse by BZDs
(Rosenbaum, et al., 1984). Rickels and Schweizer (1996) have
noted that serotonergic drugs may be more effective in this sit-
uation (LOE 4).
Many authorities suggest the use of BZDs in the early
phases of treatment with SSRIs or other antidepressants to
achieve some symptomatic relief until the antidepressant has
had time to work (generally about 2–3 weeks) and to protect
against the occasional early worsening of anxiety seen at the
beginning of antidepressant therapy. This is perhaps more
likely if comorbid panic disorder is also present. Fewer author-
ities advise that BZDs should be avoided whenever possible
A solid body of level 1 evidence supports
Psychopharmacological treatment algorithm for GAD11
(Woods, et al., 1992), given that some patients benefit from
and do not abuse them (Osser, et al., 1999).
Antidepressants versus BZDs
tive merits of antidepressants and BZD, we note a limited liter-
ature (LOE 1 for imipramine; LOE 2 for paroxetine), which
finds a superior effect for antidepressants (Hoehn-Saric, et al.,
1988; Rickels, et al., 1993; Rocca, et al., 1997). Hoehn-Saric,
et al. (1988) (LOE2) noted that alprazolam was more effective
In deciding upon the compara-
for somatic symptoms, whereas imipramine was more effective
for dysphoria and anticipatory thinking. BZDs are better for
sleep and can be used as hypnotics. The large SSRI (e.g., esci-
talopram) and SNRI (e.g., venlafaxine-XR) databases suggest
that the psychological components of the Hamilton Anxiety
Scale are more responsive to each drug, but the somatic items
do respond, albeit more slowly and to a lesser degree, at least
in the short term. Because of the chronic nature of GAD, some
patients are used to BZDs, buspirone or hydroxyzine; there-
Medication dosing recommendations for treating generalised anxiety disorder
Drug Starting dose
20 mg every 2 weeks
5–10 mg every 1–2 weeks
10–20 mg every 2 weeks
10–20 mg every 2 weeks
12.5–25 mg every 1–2 weeks
Increase to 50 mg within 1 week, then increase by
25–50 mg every 1–2 weeks
Increase by 30 mg after 1–2 weeks
Increase to 75 mg within 1 week, then increase by
37.5–75 mg every 2 weeks
Imipramine25 (HS)100 300 25 mg every 4 days; when at 100 mg, may then increase
in 50 mg increments
15 mg every 1–2 weeks
50 mg every 3–4 days
100 mg every 4–7 days
1–2 (1 BID)
5–15 (5 TID)
1–2 (1 BID)
0.5 mg every 3–4 days
1–2 mg every week
5 mg after 4–7 days; 10 mg every week
1–2 mg every week
10–15 (5 BID-TID)
15–30 (5–10 TID)
5 mg every 3 days
15 mg every 2–4 weeks
50 (25 BID)
150 (75 BID)
50 mg every week
150 mg every 4–7 days
4 (2 BID)4 162–4 mg every week
40 (20 BID), with food
0.5 mg every 1–2 weeks
50 mg after 3–4 days; after 1 week, increase to 200 mg
20 mg every 2–3 days
5050 10050 mg after 1–3 days
Used by permission of the International Psychopharmacology Algorithm Project, http://www.ipap.org.
aAll maximum doses are according to US package insert, for approved non-psychosis indications (i.e., not exclusively for generalised anxiety disorder).
Doses may be reduced in the elderly or those with significant medical conditions at the discretion of the treating clinicians.
bAvailable in liquid formulation.
cUS Food and Drug Administration (FDA) approved for generalised anxiety disorder or its equivalent diagnoses.
12 Psychopharmacological treatment algorithm for GAD
fore, previous medication experience should be taken into
term efficacy in GAD (LOE 1). There have been 10 studies
referred to by Mitte, et al. (2005), many of which were con-
ducted in samples diagnosed with DSM-III or IIIR criteria,
and of which the effect size relative to placebo was acceptable
in eight. In patients with ICD-10 diagnosed, short-duration,
mild symptoms of GAD and not exposed to BZDs, buspirone
may have some role as an initial pharmacotherapy of GAD.
However, because of its slow onset of action, variable tolerabil-
ity and its overall lack of benefit against other comorbid dis-
orders (except possibly for alcohol use disorder), and the lack
of efficacy in recent BDZ users [although not all findings are
consistent in this respect (Delle Chiaie, et al., 1995) (LOE 2)],
we in general do not recommend this drug as a first line treat-
ment for DSM-IV GAD.
Tandospirone, which was launched in Japan in 1996, is
widely used for mild anxious-depressive symptoms, especially
among primary care physicians. The drug was approved for a
wider indication mainly based on ICD-9 criteria of neurosis,
but recently it was launched in China, and the results of a
double-blind, randomised, non-placebo–controlled, trial of
buspirone and tandospirone showed similar efficacy and safety
for both treatments in GAD (Zhan, et al., 2004).
Buspirone is a partial 5HT1aagonist with short-
effective (LOE 1) in studies that have been conducted for as
long as 12 weeks in DSM-IV GAD (Ferreri, et al., 1995;
Lader and Scotto, 1998; Llorca, et al., 2002). In some coun-
tries, hydroxyzine is a widely used anxiolytic, particularly
among primary care doctors, but we recommend its use as a
second line agent in view of its side-effect profile, and lack of
efficacy for comorbid disorders.
The H1antihistaminic drug, hydroxyzine, is
also effective in the short-term treatment of GAD (LOE 1)
(Feltner, et al., 2003; Montgomery, et al., 2006; Pande, et al.,
2003; Pohl, et al., 2005; Rickels, et al., 2005). It is approved
in some European countries, but not in the United States, for
GAD. This drug is available in some countries for adjunctive
treatment of epilepsy (LOE 1) (Arroyo, et al., 2004; Beydoun,
et al., 2005; Elger, et al., 2005; French, et al., 2003), as well as
treating chronic pain associated with diabetes (LOE 1) (Lesser,
et al., 2004; Richter, et al., 2005; Rosenstock, et al., 2004), her-
pes zoster (LOE 1) (Dworkin, et al., 2003; Freynhagen, et al.,
2005; Sabatowski, et al., 2004) and fibromyalgia (LOE 1)
(Crofford, et al., 2005). Although no data exist, to our knowl-
edge, in patients with GAD comorbid with other disorders, the
drug might be considered as a treatment option (LOE 5) for
GAD with comorbid epilepsy, or chronic pain related to diabe-
tes, fibromyalgia or post-herpetic neuralgia. However, we do
not recommend it as a first-line agent for GAD in view of the
The α2δ calcium channel antagonist, pregabalin, is
relative lack of experience to date, and lack of efficacy for
Tiagabine (LOE 2) did not separate from placebo on the
primary measure in a large GAD study, but on some secondary
analyses there was separation in favour of drug (Pollack, et al.,
Evidence for antipsychotic monotherapy in GAD is very
meager. An open label trial suggested benefit for ziprasidone
(LOE 4) (Snyderman, et al., 2005). In the United Kingdom,
flupenthixol is approved for the use of depression, but is widely
used to treat GAD-like states. Most published evidence is lim-
ited to depression, there was one controlled study that showed
flupenthixol was superior to amitriptyline, clotiazepam, or pla-
cebo among subjects with refractory GAD (Wurthmann, et al.,
1995). The latter drug may be useful in patients who have a
mild paranoid element, for example, a feeling that they are
being observed (LOE5). Sulpiride is also used in similar situa-
tions (Bruscky, et al., 1974; Chen, et al., 1994). Although we
are aware of ongoing interest in the use of atypical antipsycho-
tics for GAD, at this point we remain sufficiently concerned
about their tolerability and safety profile that we would not
recommend them as first-line agents.
Riluzole, a presynaptic glutamate release inhibitor, has
shown promise in a small open-label study (LOE 4) (Mathew,
et al., 2005). This was a proof-of-concept trial, and the drug
has significant tolerability concerns.
Node 6: Assessment for initial response
Response to treatment after a trial period is described as remit-
ted, improved, partial response or non-response after 4–
6 weeks. Although reviews for schizophrenia and depression
show that response often appears before 4 weeks, we do not
know of any such reviews for GAD and would avoid recom-
mending too short a treatment trial, in that many patients need
longer to benefit. These response categories are generally
defined as follows:
Remission: at least 70% better or reduction in symptoms
from baseline (Sheehan, 2001).
Improved: at least 50% better or reduction in symptoms
from baseline (Pollack, et al., 2006).
Partial response: 25–49% better or reduction in symptom
severity from baseline.
Non-response: less than 25% better or reduction in symp-
toms from baseline.
For antidepressants, although it may take over 12 weeks before
remission occurs (Bielski, et al., 2005), if at least partial
response (i.e., at least 25% symptom reduction from baseline)
has occurred after 4–6 weeks of adequate trial, we recommend
that the clinician re-evaluate according to Node 7 (see below).
However, if there is no response or the response is less than
25%, we would recommend switch to a different treatment.
Psychopharmacological treatment algorithm for GAD13
We note that this is a clinical recommendation, and look for-
ward to data that will help support it.
For other drug groups, such as BZD and antihistamine,
there is little published data on time until remission, or even
rate of remission, but Schweizer and Rickels (1997) state that
3–4 weeks treatment at a diazepam equivalent dose of 40 mg/
day constitutes an adequate BZD trial. See node 5 for further
information on definitions of response and remission.
In summary, for 1) adequate trial and good response: go to
node 7a; 2) adequate trial and non-response: go to node 11; 3)
adequate trial and partial response: go to node 8; 4) failure to
give an adequate trial: go back up to node 5 for an adequate
trial of an alternative agent.
A number of scales exist for assessing GAD. The most
widely used and well-validated scale in GAD research is prob-
ably the Hamilton Anxiety Rating Scale (HAMA) (Hamilton,
1959), although HAMA is less-likely to be used in clinical prac-
tice because of the length and time it takes to administer the
scale. Among the simpler and best-established are the
physician-rated Clinical Global Impressions of Improvement
(CGI-I) (Guy, 1976), the self-rated Hospital Anxiety and
Depression Scale (HADS) (Zigmond and Snaith, 1983), and
the self-rated Sheehan Disability Scale (SDS) (Sheehan, 1983)
for assessing functional impairment, although neither CGI nor
SDS assess the symptomatology of GAD per se. The HAMA is
a clinician-rated scale that consists of 14 items (scoring 0–4 per
item) that assesses a variety of psychological and physiological
symptoms of anxiety. The CGI-I rates degree of overall
improvement from the start of treatment, along a 7 point
scale. A score of 4 represents no change, 3, minimal improve-
ment; 2, much improvement and 1, very much improvement.
Scores of 5 (minimally worse), 6 (much worse) and 7 (very
much worse) represent degrees of worsening, but are rarely
encountered. A CGI-I score of 1 or 2 is typically considered
to indicate ‘response’, but not necessarily remission. The
HADS contains 14 items rated on a 4-point Likert-type scales
and was initially designed to screen for mood disorders among
the medically ill patients, but has been since used widely among
general population as well as the psychiatrically and medically
ill populations. The SDS comprises of three self-rated items
(work, social life, family life) on a 10-point visual analogue
scale, designed to measure the extent of impairment in patient’s
Node 7: Remission and relapse prevention
After 8–12 weeks of treatment, many patients will experience
improvement, with at least 50% reduction in symptoms. How-
ever, in GAD, there is evidence from two studies that response
and remission rates continue to increase beyond 2 months, and
even beyond 6 months in the case of remission (Bielski, et al.,
2005; Gelenberg, et al., 2000; Stocchi, et al., 2003).
Continued treatment over 8 months or longer is associated
with a reduction in the risk of relapse (LOE 1) (Allgulander,
et al., 2006; Stocchi, et al., 2003). Once a good response to
SRI therapy has been determined, we recommend continuation
of the therapy for at least a year in GAD, given the chronicity
of symptoms, and randomised controlled trials showing relapse
after short-term maintenance treatment. No data exist to our
knowledge on the efficacy of continued treatment with a BZD
or antihistamine in GAD.
Failure to achieve remission should constitute a signal for
the clinicians to either increase the dose to maximal or supra-
maximal levels, to augment or to switch to another drug class.
Although data are not available, we recommend mainte-
nance treatment at the same dose which it took to achieve
response. An exception to this recommendation can be made
where late emerging side effects occur, such as weight gain, sex-
ual difficulty, sleep disruption, behavioural or other mental
changes, such as hostility or impulsivity. In these cases, the
dose may be lowered, discontinued or antidote medicine used,
according to clinical judgment. In primary care, it is possible
that less severe cases of GAD are found as compared with psy-
chiatric settings, and that the required dose may be lower,
although data do not exist on this topic as far as we know.
Because withdrawal from most anti-anxiety medications can
be distressing, with the possible exception of azapirones and
fluoxetine, we recommend a slow taper when the decision has
been made to stop medication.
Node 8: Assessment for partial response
When there is only partial response to an initial trial after 4–
6 weeks of adequate dosing, we recommend that the clinician
re-evaluate and consider either (a) further increase of the dose,
if the maximal dose had not been used, (b) augmentation where
there has been some response to monotherapy (see below node
9, 10) or (c) switch to a different treatment (see below node 11).
The role of non-pharmacologically specific response may be
considered in partial responders, where improvement can
partly be due to the care, diagnostic process, time spent with
the patient, empathic support, investigator bias, etc. that are
part of the interaction with the patient. If the physician deter-
mines that these factors are significant, then drug augmenta-
tion may be of very limited benefit.
Node 9: For partial response with persistent insomnia
Although pharmacologic intervention may be indicated, it is
also important to evaluate lifestyle issues that can impact on
sleep (i.e., sleep hygiene, diet, excess caffeine usage especially
late in the day, alcohol and other substance use, exercise) (see
above, Node 2). When considering pharmacotherapeutic
options, augmentation with proven non-BZD GABAergic hyp-
notic drugs (LOE 1), but which have unproven anxiolytic
effects (LOE 5), can be used, for example, zolpidem, zaleplon
or eszopiclone. BZD drugs can also be considered for sleep
enhancement because they have beneficial effects in GAD.
Other sedating anxiolytic or antidepressant drugs can be used,
although their evidence as hypnotics is weak and their side
effects can be more problematic. Further studies of the
14 Psychopharmacological treatment algorithm for GAD
dose/response/side-effects relationships with antidepressants
used as hypnotics are needed. Alternatively, a sedating antihis-
tamine such as hydroxyzine can be added (LOE 1 for GAD),
but its hypnotic effects are also unproven. The evidence is
mixed, but the more recent studies of another antihistamine
are not encouraging. Tolerance to the sedation from diphenhy-
dramine 50 mg twice daily was complete by day 3 of a placebo-
controlled study (Richardson, et al., 2002). In elderly patients,
major cognitive impairment was found in 426 patients given
diphenhydramine (Agostini, et al., 2001). It is also important
to be mindful that in some instances treatment of GAD may
also exacerbate sleep disturbance.
Node 10: Augmentation strategies for partial response
with general symptom persistence
Almost all clinical trials of GAD have tested the efficacy of
monotherapy, therefore our recommendations are largely
extrapolations from monotherapy trials of the recommended
product. Notable exception applies, however, to the atypical
neuroleptics risperidone and olanzapine. For partial responders
to an SSRI, who showed persistence of the full symptom com-
plex, despite adequate treatment with a variety of antidepres-
sant or non-antidepressant drugs for GAD, the addition of ris-
peridone up to 1.5 mg/day was associated with further
improvement (LOE 2) (Brawman-Mintzer, et al., 2005). In a
study by Pollack, et al. (2006) (LOE2), augmentation of fluox-
etine with olanzapine yielded better response than did augmen-
tation with placebo only on some secondary outcome
measures: failure to separate on the primary measure was
most likely because of underpowering of the study. We con-
sider that augmentation with an atypical antipsychotic drug
has a good foundation based on short-term efficacy data
(rather than long-term safety data). However, many clinicians
may prefer to use drugs that are not associated with risks of
metabolic changes or abnormal involuntary movements. In
that case, we recommend the addition of a drug from other
classes to the primary drug already being given to the patient.
Thus, a BZD, antihistamine, buspirone or tiagabine (LOE 5
each) could be added to an antidepressant. Tiagabine, whose
effects in GAD are modest at best, should be used with caution
in those with a seizure history or predisposition. An antidepres-
sant could be added to any of the aforementioned drugs, in
situations where the patient has been treated, for example,
firstly with an antihistamine or BZD.
CBT could be added, although there is no current evidence
to support a potentiating effect of CBT to ongoing pharmaco-
therapy in GAD (LOE5) (Power, et al., 1989).
Node 11: Switch strategies for partial response or non-
response with general symptom persistence
An alternative strategy for partial response would be to switch
to another antidepressant, either within the same class or in a
different one (e.g., SSRI to SSRI or SSRI to SNRI). No studies
have examined the effects of this strategy in GAD, to our
knowledge, but there are data in MDE that a second SSRI
can be effective in approximately 50% of cases when there is
failure on the first, and there is evidence that a SNRI can be
effective in cases where patients have failed a number of previ-
ous medication trials. There is little systematic guidance as
which strategy is preferable, and when, but there is arguably
greater rationale for switching than for augmentation when
response to the first medication is zero or below 25%.
We do not generally recommend switching from an SSRI/
SNRI to a BZD on the grounds of non-response. If there has
been intolerance to at least two different antidepressants, then
switching to BZD seems reasonable (LOE 5). In addition, if
there were two failed trials with SSRIs, we recommend switch-
ing to an SNRI. Imipramine (LOE 2) may also be a consider-
ation for non-responders to SSRI/SNRI, although no trials
have been done to address this.
Node 12: Assessment for response
See Node 6.
Node 13: Evaluate for comorbidity
If inadequate response is found to be associated with comorbid
depression, stable bipolar disorder or other anxiety disorder
(panic, social anxiety, PTSD, OCD or specific phobia), then
the following steps are recommended.
Node 14: GAD with persisting unipolar depression
For GAD with persisting unipolar depression, we recommend
use of maximum tolerated dose antidepressant drugs or aug-
mentation of an SSRI or SNRI with buspirone (Robinson,
et al., 1990; Trivedi, et al., 2006) (LOE 2 for augmentation of
depression; LOE1 for GAD and MDD, each as monotherapy),
bupropion (Trivedi, et al., 2006) (LOE 1 for MDD as mono-
therapy and augmentation; LOE 3 for GAD) or an atypical
antipsychotic, such as risperidone or olanzapine (LOE 2 for
GAD). In atypical major depression with comorbid GAD,
chromium picolinate showed greater effect than placebo on
symptoms of both conditions (LOE3) (Davidson, et al., 2003).
For severe depression with GAD, it is possible that ECT
would be indicated for the depressive component (LOE 1 for
depression; LOE 5 for GAD). Monoamine oxidase inhibitor
(MAOI) therapy may also be tried at this stage (LOE 1 for
depression and LOE 5 for GAD). Other augmentation strate-
gies, such as lithium and tri-iodothyronine, while effective in
depression (LOE 1) (Carvalho, et al., 2007), have not been
studied in GAD, to our knowledge. However, if other mea-
sures have failed, they could be considered with the expectation
that they might benefit the depressive component.
Psychopharmacological treatment algorithm for GAD15
Node 15: GAD with bipolar spectrum disorder
If the symptom picture suggests bipolar spectrum disorder that
coexists with GAD, we would want to rule out uncontrolled
mood instability. We recommend addition of a drug with
mood stabilising properties such as an anticonvulsant (dival-
proex sodium, lithium) (LOE 1 for bipolar disorder, and LOE
5 for GAD) or atypical antipsychotic (LOE 1 for bipolar dis-
order; LOE 2 for SSRI augmentation in GAD). Of note, some
mood stabilisers and antipsychotics may need periodic labora-
tory monitoring, for example, blood levels of carbamazepine,
valproic acid and lithium, and fasting lipid profile and fasting
blood sugar for some antipsychotics. There are some regional
differences in respect to the use of antidepressants in patients
with bipolar depression, with US physicians preferring not to
recommend their use. Those in Europe, however, use them
more often in practice, backed by substantial long-term data
of antidepressants being associated with lower mortality over
50 years follow-up when used with a mood stabiliser or anti-
psychotic, as compared with treatment without an antidepres-
sant (Angst, 1985). Nevertheless, not all data is consistent, so in
the event that an antidepressant is used, care should be taken
that it does not exacerbate the bipolar disorder (Leverich,
et al., 2006).
Node 16: GAD with other anxiety disorders
For GAD with other anxiety disorders, we recommend the use
of treatments that are beneficial for both states. If monother-
apy with, for example, an SRI or BZD, has failed to ade-
quately treat the comorbid disorder, a second treatment can
be added. Thus, for all anxiety disorders, SSRIs (LOE 1) are
effective, SNRI (LOE 1 for all anxiety disorders except OCD),
TCA (LOE 1 for imipramine or clomipramine in PD; LOE 2
for imipramine in PTSD; LOE 1 for clomipramine, in OCD)
may be beneficial. For panic disorder, a BZD (alprazolam or
clonazepam) may be used (LOE 1). There is some evidence to
support olanzapine monotherapy for SAD (LOE 3) (Barnett,
et al., 2002), augmentation with risperidone or olanzapine in
OCD (LOE 1) (Bystritsky, et al., 2004; Hollander, et al.,
2003; Li, et al., 2005; McDougle, et al., 2000; Shapira, et al.,
2004) and PTSD (LOE 1 or 2) (Bartzokis, et al., 2005; Butter-
field, et al., 2001; Hamner, et al., 2003; Monnelly, et al., 2003;
Reich, et al., 2004; Stein, et al., 2002). There are data to sup-
port use of pregabalin (LOE 1) (Pande, et al., 1999, 2004) and
levetiracetam (LOE 3) (Zhang, et al., 2005) in SAD.
Node 17: Inadequate response without comorbidity
In the case of inadequate response (non-response or partial
response) to the drug combination chosen at Node 9–11, with-
out significant comorbidity, we recommend switching to
another combination, of which one drug should be a serotoner-
gic antidepressant (SSRI, SNRI), noradrenergic and specific
serotonergic antidepressant (NaSSa) or serotonergic TCA, or
adding a third drug of a different class to the two already in
use, although evidence from studies of triple therapy is lacking.
Preliminary evidence (LOE 3) from a small trial in atypical
depression, which was comorbid with GAD in the majority of
subjects, found benefit for monotherapy with chromium pico-
linate (LOE 3) (Davidson, et al., 2003). It is possible that in
combination with established treatments, augmentation with
this agent may be useful (LOE 5) in refractory GAD. MAOI
monotherapy could also be considered at this stage (LOE5).
PST can also be added at this stage (LOE 5).
Node 18: Assessment for response
See Node 6.
Node 19: Diagnostic re-evaluation
Non-response or partial response at this stage would call for
General principles of GAD treatment
We recommend the following general principles be considered
in the diagnosis and treatment of GAD:
At initial and repeated evaluation:
1) GAD is common and often passes unrecognised.
2) The initial evaluation should adhere to DSM-IV or
ICD-10 criteria. Given that the most research on
GAD, and pharmacotherapy in particular, has used
DSM-IV, these criteria may be especially useful.
3) The initial evaluation should include a thorough psychi-
atric assessment, medical history and, where appropri-
ate, referral for laboratory or medical evaluation.
Attention should also be added to sleep hygiene, life-
style issues, as well as medication side effects.
4) Initially and at subsequent points of non-response, or
when there is loss of previous response, assess for key
symptoms that may change management (e.g., suicidal-
ity, psychotic or bipolar symptoms, substance use)
introduction of other medications and treatment non-
5) Patients with bipolar disorder should already be stabi-
lised before introducing antidepressant treatment for
GAD. Latest data suggest avoiding antidepressants:
only 16% respond and many experience a mood switch
(Leverich, et al., 2006), although there are regional dif-
ferences in the approach to using antidepressants for
bipolar depression. Similarly, other comorbid disorders
may need to be stabilised before attempting to treat
16 Psychopharmacological treatment algorithm for GAD
GAD. On other occasions, treatment for both condi-
tions can be instituted simultaneously.
6) Standardised rating scales for symptom severity, quality
of life, function and resiliency (i.e., stress coping ability)
can be useful.
Choice of treatment: medication, psychosocial or both
1) The initial treatment of GAD can be either with medi-
cation, psychotherapy, or the combination. Patient
preference and therapist skill will be important determi-
nants of the choice. Comorbidity and previous response
to treatment are also important considerations.
2) Both approaches have been found efficacious for GAD.
Medications and adequacy of response
1) Patients with GAD who are going to be treated with
medication should, in most cases, receive an SSRI or
SNRI as first line monotherapy. Only if rapid response
is needed, or if insomnia is a dominant symptom, would
a concomitantBZD be
2) The response time to an antidepressant drug in GAD is
generally 4–12 weeks. One expects at least partial
response by 4–6 weeks with adequate dosing, and it is
assumed that the clinician will progressively titrate the
dose upwards according to tolerability of the drug.
Response to BZDs is usually faster than for antidepres-
sants, and if there has not been adequate response after
4–6 weeks, we do not believe that persistence with drug
will produce greater improvement.
3) In our current state of knowledge, we cannot say
whether it is better to increase dose, augment, switch
or wait longer when there has been partial response.
Clinicians may wish to keep their options open as to
the preferred approach. We do, however, recommend
switching treatment where an adequate trial has failed
to elicit at least 25% improvement.
4) Occasionally patients with GAD experience exacerba-
tion of anxiety or jitteriness with onset of SRI treat-
ment. In such instances, we recommend either lowering
of dose to minimal starting levels and then increase in
small steps, or co-prescribe a BZD for a short period of
time, for example, diazepam at 5 mg three times daily.
5) BZD is relatively contraindicated in those with a his-
tory of substance use problems.
6) The patient who has shown an excellent response to
pharmacotherapy should generally be treated for at
least 1 year. Early withdrawal is associated with greater
7) Many scales are available to measure symptom severity
and function in GAD. Their relative merits and limita-
tions are beyond the scope of this report, but three sim-
ple and well validated scales include the CGI-I, HADS,
and the SDS. In addition, HAMA has been widely used
in GAD and other anxiety disorders research.
Managing side effects
1) Patients with an anxiety disorder are often more sensi-
tive than other patients to medication side-effects and
may need lower starting dose and incremental titration
over a longer time than might be the case when treating
2) When patients respond partially, or fail to respond, it is
important to consider if this represents inadequate med-
ication benefit, treatment non-adherence, the result of
side effects, concomitant use of alcohol, substances,
other anxiogenic prescription, over the counter treat-
ments or the need for diagnostic re-evaluation.
3) Antipsychotic drugs are associated with metabolic and
general cardiovascular side-effects. There may be
greater risk of developing type II diabetes or worsening
of previously controlled diabetes, weight gain, abdomi-
nal obesity, increased triglycerides or total and LDL
cholesterol. Appropriate monitoring of metabolic pro-
file is recommended in accordance with current
4) There is the possibility of untoward drug interactions
brought about by the inhibitory properties of some
antidepressant drugs on the CYP 450 isoenzyme sys-
tem. With increased rates of comorbid medical illness,
there is greater likelihood that a patient with GAD will
be taking other medication over the long term. Clini-
cians are, therefore, encouraged to familiarise them-
selves with the more common types of interactions
that could occur with each psychotropic drug, for
example, as reviewed by Oesterheld J, Osser DN and
Sandson N at http://www.genelex.com.
5) Typical SSRI/SNRI discontinuation (or ‘withdrawal’)
symptoms include worsening of anxiety, irritability,
depressed mood and somatic symptoms such as head-
ache, dizziness, nausea, tremor, paresthesia, vivid
dreams and insomnia (Coupland, et al., 1996). As a
general guide to management, we suggest the following:
i) educational material should be provided regarding
the treatment to patients and their care-takers; ii) a
gradual taper is recommended (with an even slower
taper in the elderly and medically ill population) when
a decision is made to discontinue a medication and iii)
liquid form, if available, may be used for easier admin-
istering in the case of an unusually low dose or diffi-
Sometimes an initial rapid response that fades may be indica-
tive of a ‘placebo’ or ‘non-specific’ response, as has been sug-
gested in the depression literature. We do not know to what
extent this is the case for GAD or how it would best be man-
aged. Some hold that under these circumstances a medication
switch would be preferable to augmentation, although there are
no data to inform on the matter. The role of the placebo
response may be considered in partial responders, where
Psychopharmacological treatment algorithm for GAD17
improvement can partly be due to the care, diagnosis, time
spent with the patient, empathic support, investigator bias,
etc. that are part of the interaction with the patient.
Cost is often an important consideration in drug selection.
However, cost of medication must be viewed more broadly as
part of a cost-benefit equation because ‘cheaper’ drugs may
have more problematic side-effects, which bring additional
cost-burdens. Because of the variability in medication costs
from one country to another, we do not make any specific
recommendations about this issue. Related is the issue of risk-
benefit, which should be a consideration in drug selection.
However, taking into account what is available on the formu-
lary in the country or locale of practice, if at any node in the
algorithm there are two or more options of apparently equal
efficacy, similar toxicity and similar acceptability to the
patient, and there is a big difference in cost, it is prudent for
the clinician to prefer the less expensive product.
6) Cost-benefit considerations
This project is supported by a grant from the International Psycho-
pharmacology Algorithm Project (IPAP). Thanks also to Paul Nord-
strom August, Dean Hartley, David Penniman and Oakley Ray as
consultants for the project.
Jonathan RT Davidson: Speaker for Solvay Pharmaceuticals, Pfizer
Inc., GlaxoSmithKline, Wyeth Pharmaceuticals, Lichtwer Pharma,
Forest, American Psychiatric Association.
Research Support from Pfizer, Solvay, Eli Lilly, GlaxoSmithKline,
Wyeth, Organon Inc., Forest, PureWorld, Allergan, Nutrition 21, Bris-
tol Myers Squibb, Johnson and Johnson, Cephalon, Astra Zeneca,
Parke Davis, Pharmacia, Upjohn, UCB, Merck, Janssen.
Advisory Board for Solvay, Pfizer, GlaxoSmithKline, Forest, Eli
Lilly, Ancile, Roche, MediciNova, Jazz, Novartis, Organon, Boehrin-
ger Ingelheim, MedTap, Research Triangle Institute, Astra Zeneca,
Johnson and Johnson, Wyeth, Bristol Myers Squibb, Boots, UCB,
Sanofi-Synthelabo, Alexza, Janssen.
Drugs supplied for other studies from Eli Lilly, Schwabe, Pure-
World, Pfizer Inc.
Royalties from MultiHealth Systems Inc, Guilford Publications,
American Psychiatric Association, Penguin Putnam Publishers, Cur-
rent Medical Science, Martin Dunitz, Taylor and Francis.
Wei Zhang: Grants and Research Support: AstraZeneca; Eli Lilly
and Co; Forest Laboratories, Inc; Glaxo SmithKline; Pfizer, Inc.;
Consultancies: Cephalon, Inc.
Kathryn Connor: Grants and Research Support: Eli Lilly, Pfizer,
Inc., Pure World Botanicals, Inc., Forest Laboratories, Inc, Glaxo
Speakers Bureau: Ortho McNeil, Inc., Pfizer, Inc., Wyeth Ayerst
Pharmaceutical Company, Cephalon Pharmaceuticals, Inc., Solvay
Pharmaceuticals, Forest Laboratories, Inc.
Consultancies: Ortho McNeil, Inc., Pfizer, Inc., Schwarz Pharma
Inc, King Pharmaceuticals, Cephalon Pharmaceuticals, Inc., Predix
Pharmaceuticals, Jazz Pharmaceuticals.
Other Financial or Material Support: Dr Wilmar Schwabe, Nutri-
tion 21, Cephalon,Inc.,Nordic
Jianlin Ji: None applicable.
Ken Jobson: None applicable.
Yves Lecrubier: Advisory boards, consulting, sponsored trials:
Sanofi, Eli Lilly, Novartis, P. Fabre, Merck, Janssen, Wyeth Pederle,
Stock holding: No.
Alexander Macfarlane: Grants and Research Support: Pfizer.
David Nutt: Consultancies: Pfizer (W-L), GlaxoSmithKline, MSD,
Esteve, Novartis, Asahi, Organon, Cypress, Lilly, Janssen, Takeda,
Phamacia, Therasci, Passion for Life.
Speaking honoraria (in addition to above): Wyeth, Reckitt-
Benkiser, Lundbeck, Cephalon.
Grants and Research Support: MSD, GlaxoSmithKline, Novartis,
Servier, Janssen, Yamanouchi, Lundbeck, Pfizer, Wyeth, Organon.
Stock Holding: GlaxoSmithKline.
David Osser: None applicable.
Dan Stein: Research grants and/or consultancy honoraria: Astraze-
neca, Eli-Lilly, GlaxoSmithKline, Lundbeck, Orion, Pfizer, Pharma-
cia, Roche, Servier, Solvay, Sumitomo, and Wyeth.
Osamu Tajima: Speakers Bureau: Meiji, GlaxoSimithKline, Astel-
las, Pfeizer, Asahi-Kasei, Otsuka, Solvay, Jansen, Sumitomo.
Consultancies: Solvay, Shionogi.
Marcio Versiani: Research grants and/or consultancies: Pfizer, Ser-
vier, Organon, Bristol-Myers, Aché.
Agostini, JV, Leo-Summers, LS, Inouye, SK (2001) Cognitive and
other adverse effects of diphenhydramine use in hospitalized older
patients. Arch Intern Med 161: 2091–2097.
AGREE (2003) Development and validation of an international
appraisal instrument for assessing the quality of clinical practice
guidelines: the AGREE project. Qual Saf Health Care 12: 18–23.
Allgulander, C, Bandelow, B, Hollander, E, Montgomery, SA, Nutt,
DJ, Okasha, A, et al. (2003) WCA recommendations for the long-
term treatment of generalized anxiety disorder. CNS Spectr 8: 53–
Allgulander, C, Dahl, AA, Austin, C, Morris, PL, Sogaard, JA,
Fayyad, R, et al. (2004) Efficacy of sertraline in a 12-week trial
for generalized anxiety disorder. Am J Psychiatry 161: 1642–1649.
Allgulander, C, Florea, I, Huusom, AK (2006) Prevention of relapse in
generalized anxiety disorder by escitalopram treatment. Int J Neu-
ropsychopharmacol 9: 495–505.
Allgulander, C, Hackett, D, Salinas, E (2001) Venlafaxine extended
release (ER) in the treatment of generalised anxiety disorder:
twenty-four-week placebo-controlled dose-ranging study. Br J Psy-
chiatry 179: 15–22.
Alonso, J, Lepine, JP (2007) Overview of key data from the European
Study of the Epidemiology of Mental Disorders (ESEMeD). J Clin
Psychiatry 68(Suppl. 2): S3–S9.
Altshuler, LL, Cohen, L, Szuba, MP, Burt, VK, Gitlin, M, Mintz, J
(1996) Pharmacologic management of psychiatric illness during
pregnancy: dilemmas and guidelines. Am J Psychiatry 153: 592–
American Psychiatric Association (2004) Diagnostic and Statistical
Manual of Mental Disorders, 4th edition, Text Revision. Washing-
ton, DC: American Psychiatric Association; 2004.
18 Psychopharmacological treatment algorithm for GAD
American Psychiatric Association Practice guidelines (2007) Treatment
of patients with substance use disorders, 2nd edition. Am J Psychi-
atry 164: 1–124.
Andreski, P, Chilcoat, H, Breslau, N (1998) Post-traumatic stress dis-
order and somatization symptoms: a prospective study. Psychiatry
Res 79: 131–138.
Angst, J (1985) Switch from depression to mania – a record study over
decades between 1920 and 1982. Psychopathology 18: 140–154.
Arroyo, S, Anhut, H, Kugler, AR, Lee, CM, Knapp, LE, Garofalo,
EA, et al. (2004) Pregabalin add-on treatment: a randomized,
double-blind, placebo-controlled, dose-response study in adults
with partial seizures. Epilepsia 45: 20–27.
Balazs, J, Bitter, I, Lecrubier, Y, Csiszer, N, Ostorharics, G (2000)
Prevalence of subthreshold forms of psychiatric disorders in per-
sons making suicide attempts in Hungary. Eur Psychiatry 15:
Baldwin, DS, Anderson, IM, Nutt, DJ, Bandelow, B, Bond, A,
Davidson, JR, et al. (2005) Evidence-based guidelines for the phar-
macological treatment of anxiety disorders: recommendations from
the British Association for Psychopharmacology. J Psychopharma-
col (Oxf) 19: 567–596.
Ballenger, JC, Davidson, JR, Lecrubier, Y, Nutt, DJ, Borkovec, TD,
Rickels, K, et al. (2001a) Consensus statement on generalized anxi-
ety disorder from the International Consensus Group on Depres-
sion and Anxiety. J Clin Psychiatry 62(Suppl. 11): S53–S58.
Ballenger, JC, Davidson, JR, Lecrubier, Y, Nutt, DJ, Kirmayer, LJ,
Lepine, JP, et al. (2001b) Consensus statement on transcultural
issues in depression and anxiety from the International Consensus
Group on Depression and Anxiety. J Clin Psychiatry 62(Suppl. 13):
Bandelow, B, Zohar, J, Hollander, E, Kasper, S, Moller, HJ (2002)
World Federation of Societies of Biological Psychiatry (WFSBP)
guidelines for the pharmacological treatment of anxiety, obsessive-
compulsive and posttraumatic stress disorders. World J Biol Psy-
chiatry 3: 171–199.
Barnett, SD, Kramer, ML, Casat, CD, Connor, KM, Davidson, JR
(2002) Efficacy of olanzapine in social anxiety disorder: a pilot
study. J Psychopharmacol (Oxf) 16: 365–368.
Bartzokis, G, Lu, PH, Turner, J, Mintz, J, Saunders, CS (2005)
Adjunctive risperidone in the treatment of chronic combat-related
posttraumatic stress disorder. Biol Psychiatry 57: 474–479.
Beekman, AT, Bremmer, MA, Deeg, DJ, van Balkom, AJ, Smit, JH,
de Beurs, E, et al. (1998) Anxiety disorders in later life: a report
from the Longitudinal Aging Study Amsterdam. Int J Geriatr Psy-
chiatry 13: 717–726.
Beydoun, A, Uthman, BM, Kugler, AR, Greiner, MJ, Knapp, LE,
Garofalo, EA, et al. (2005) Safety and efficacy of two pregabalin
regimens for add-on treatment of partial epilepsy. Neurology 64:
Bielski, RJ, Bose, A, Chang, CC (2005) A double-blind comparison of
escitalopram and paroxetine in the long-term treatment of general-
ized anxiety disorder. Ann Clin Psychiatry 17: 65–69.
Birmaher, B, Axelson, DA, Monk, K, Kalas, C, Clark, DB, Ehmann,
M, et al. (2003) Fluoxetine for the treatment of childhood anxiety
disorders. J Am Acad Child Adolesc Psychiatry 42: 415–423.
Blank, S, Lenze, EJ, Mulsant, BH, Amanda Dew, M, Karp, JF, Shear,
MK, et al. (2006) Outcomes of late-life anxiety disorders during 32
weeks of citalopram treatment. J Clin Psychiatry 67: 468–472.
Blazer, DG, George, LK, Hughes, C (1991a) Anxiety in the elderly:
treatment and research. In: Salzman, C, Lebowitz, BD (eds), The
Epidemiology of Anxiety Disorders. An Age Comparison. New
York: Springer Publishing Company, pp. 17–30.
Blazer, DG, Hughes, D, George, LK, Swartz, M, Boyer, R (1991b)
Generalized anxiety disorder. In: Robins, L, Regier, DA (eds), Psy-
chiatric Disorders in America: The Epidemiologic Catchment Area
Study. New York, NY: The Free Press, pp. 180–203.
Bogunovic, OJ, Greenfield, SF (2004) Practical geriatrics: use of ben-
zodiazepines among elderly patients. Psychiatr Serv 55: 233–235.
Bohm, C, Robinson, DS, Gammans, RE, Shrotriya, RC, Alms, DR,
Leroy, A, et al. (1990) Buspirone therapy in anxious elderly
patients: a controlled clinical trial. J Clin Psychopharmacol 10:
Borkovec, TD, Ruscio, AM (2001) Psychotherapy for generalized anx-
iety disorder. J Clin Psychiatry 62: 37–42discussion 43–35.
Brady, KT, Tolliver, BK, Verduin, ML (2007) Alcohol use and anxi-
ety: diagnostic and management issues. Am J Psychiatry 164: 217–
Brawman-Mintzer, O, Knapp, RG, Nietert, PJ (2005) Adjunctive ris-
peridone in generalized anxiety disorder: a double-blind, placebo-
controlled study. J Clin Psychiatry 66: 1321–1325.
Bruscky, SB, Caldeira, MV, Bueno, JR (1974) Clinical trials of sulpir-
ide. Arq Neuropsiquiatr 32: 234–239.
Butterfield, MI, Becker, ME, Connor, KM, Sutherland, S, Churchill,
LE, Davidson, JR (2001) Olanzapine in the treatment of post-
traumatic stress disorder: a pilot study. Int Clin Psychopharmacol
Bystritsky, A, Ackerman, DL, Rosen, RM, Vapnik, T, Gorbis, E,
Maidment, KM, et al. (2004) Augmentation of serotonin reuptake
inhibitors in refractory obsessive-compulsive disorder using adjunc-
tive olanzapine: a placebo-controlled trial. J Clin Psychiatry 65:
Bystritsky, A, Kerwin, L, Vapnik, T (2006) A pilot controlled trial of
bupropion versus escitalopram in generalized anxiety disorder
(GAD). Poster presented at the 161th annual meeting of the Amer-
ican Psychiatric Association. Toronto, Canada.
Carey, PD, Vythilingum, B, Seedat, S, Muller, JE, van Ameringen, M,
Stein, DJ (2005) Quetiapine augmentation of SRIs in treatment
refractory obsessive-compulsive disorder: a double-blind, rando-
mised, placebo-controlled study. [ISRCTN83050762]BMC Psychia-
try 5: 5.
Carroll, KM, Fenton, LR, Ball, SA, Nich, C, Frankforter, TL, Shi, J,
et al. (2004) Efficacy of disulfiram and cognitive behavior therapy
in cocaine-dependent outpatients: a randomized placebo-controlled
trial. Arch Gen Psychiatry 61: 264–272.
Carter, RM, Wittchen, HU, Pfister, H, Kessler, RC (2001) One-year
prevalence of subthreshold and threshold DSM-IV generalized anx-
iety disorder in a nationally representative sample. Depress Anxiety
Carvalho, AF, Cavalcante, JL, Castelo, MS, Lima, MC (2007) Aug-
mentation strategies for treatment-resistant depression: a literature
review. J Clin Pharm Ther 32: 415–428.
Castillo, CS, Starkstein, SE, Fedoroff, JP, Price, TR, Robinson, RG
(1993) Generalized anxiety disorder after stroke. J Nerv Ment Dis
Catalan, J, Gath, D, Edmonds, G, Ennis, J (1984) The effects of non-
prescribing of anxiolytics in general practice. I. Controlled evalua-
tion of psychiatric and social outcome. Br J Psychiatry 144: 593–
Chambers, CD, Hernandez-Diaz, S, Van Marter, LJ, Werler, MM,
Louik, C, Jones, KL, et al. (2006) Selective serotonin-reuptake
Psychopharmacological treatment algorithm for GAD19
inhibitors and risk of persistent pulmonary hypertension of the
newborn. N Engl J Med 354: 579–587.
Chen, A, Zhao, Y, Yu, X (1994) The clinical study of antianxiety and
antidepressive effect of sulpiride. Chin J Neurol Psychiatry 27: 220–
Cipriani, A, Barbui, C, Geddes, JR (2005) Suicide, depression, and
antidepressants. BMJ 330: 373–374.
Cipriani, A, Geddes, JR, Barbui, C (2007) Venlafaxine for major
depression. BMJ 334: 215–216.
Cohen, LS, Sichel, DA, Dimmock, JA, Rosenbaum, JF (1994) Post-
partum course in women with preexisting panic disorder. J Clin
Psychiatry 55: 289–292.
Cohen, LS, Sichel, DA, Faraone, SV, Robertson, LM, Dimmock, JA,
Rosenbaum, JF (1996) Course of panic disorder during pregnancy
and the puerperium: a preliminary study. Biol Psychiatry 39: 950–
Copeland, JR, Dewey, ME, Wood, N, Searle, R, Davidson, IA,
McWilliam, C (1987) Range of mental illness among the elderly
in the community. Prevalence in Liverpool using the GMS-
AGECAT package. Br J Psychiatry 150: 815–823.
Coupland, NJ, Bell, CJ, Potokar, JP (1996) Serotonin reuptake inhibi-
tor withdrawal. J Clin Psychopharmacol 16: 356–362.
Crofford, LJ, Rowbotham, MC, Mease, PJ, Russell, IJ, Dworkin, RH,
Corbin, AE, et al. (2005) Pregabalin for the treatment of fibromy-
algia syndrome: results of a randomized, double-blind, placebo-
controlled trial. Arthritis Rheum 52: 1264–1273.
Davidson, J, Bernik, M, Connor, KM, Friedman, MJ, Jobson, KO,
Kim, Y, et al. (2005) A new treatment algorithm for posttraumatic
stress disorder. Psychiatr Ann 35: 887–900.
Davidson, JR, Abraham, K, Connor, KM, McLeod, MN (2003)
Effectiveness of chromium in atypical depression: a placebo-
controlled trial. Biol Psychiatry 53: 261–264.
Davidson, JR, DuPont, RL, Hedges, D, Haskins, JT (1999) Efficacy,
safety, and tolerability of venlafaxine extended release and buspir-
one in outpatients with generalized anxiety disorder. J Clin Psychi-
atry 60: 528–535[comment].
Davidson, JRT, Bose, A, Korotzer, A, Zheng, H (2004) Escitalopram
in the treatment of generalized anxiety disorder: double-blind, pla-
cebo controlled, flexible-dose study. Depress Anxiety 19: 234–240.
Delle Chiaie, R, Pancheri, P, Casacchia, M, Stratta, P, Kotzalidis,
GD, Zibellini, M (1995) Assessment of the efficacy of buspirone
in patients affected by generalized anxiety disorder, shifting to bus-
pirone from prior treatment with lorazepam: a placebo-controlled,
double-blind study. J Clin Psychopharmacol 15: 12–19.
Denys, D, de Geus, F, van Megen, HJ, Westenberg, HG (2004) A
double-blind, randomized, placebo-controlled trial of quetiapine
addition in patients with obsessive-compulsive disorder refractory
to serotonin reuptake inhibitors. J Clin Psychiatry 65: 1040–1048.
Dworkin, RH, Corbin, AE, Young, JP, Sharma, U, LaMoreaux, L,
Bockbrader, H, et al. (2003) Pregabalin for the treatment of post-
herpetic neuralgia: a randomized, placebo-controlled trial. Neurol-
ogy 60: 1274–1283[see comment].
Fava, M, Rush, AJ, Trivedi, MH, Nierenberg, AA, Thase, ME,
Sackeim, HA, et al. (2003) Background and rationale for the
sequenced treatment alternatives to relieve depression (STAR*D)
study. Psychiatr Clin North Am 26: 457–494.
Elger, CE, Brodie, MJ, Anhut, H, Lee, CM, Barrett, JA, Elger, CE, et
al. (2005) Pregabalin add-on treatment in patients with partial sei-
zures: a novel evaluation of flexible-dose and fixed-dose treatment
in a double-blind, placebo-controlled study. Epilepsia 46: 1926–
Erzegovesi, S, Guglielmo, E, Siliprandi, F, Bellodi, L (2005) Low-dose
risperidone augmentation of fluvoxamine treatment in obsessive-
compulsive disorder: a double-blind, placebo-controlled study. Eur
Neuropsychopharmacol 15: 69–74.
Feltner, DE, Crockatt, JG, Dubovsky, SJ, Cohn, CK, Shrivastava,
RK, Targum, SD, et al. (2003) A randomized, double-blind,
placebo-controlled, fixed-dose, multicenter study of pregabalin in
patients with generalized anxiety disorder. J Clin Psychopharmacol
23: 240–249[see comment].
Ferreira, E, Carceller, AM, Agogue, C, Martin, BZ, St-Andre, M,
Francoeur, D, et al. (2007) Effects of selective serotonin reuptake
inhibitors and venlafaxine during pregnancy in term and preterm
neonates. Pediatrics 119: 52–59.
Ferreri, M, Darcis, D, Burtin, B ; Hydroxyzine atFSGf (1995) A mul-
ticenter double-blind placebo-controlled study investigating the
anxiolytic efficacy of hydroxyzine in patients with generalized anx-
iety disorder. Hum Psychopharmacol 139: 402–406.
Fineberg, NA, Sivakumaran, T, Roberts, A, Gale, T (2005) Adding
quetiapine to SRI in treatment-resistant obsessive-compulsive dis-
order: a randomized controlled treatment study. Int Clin Psycho-
pharmacol 20: 223–226.
Flint, AJ (1994) Epidemiology and comorbidity of anxiety disorders in
the elderly. Am J Psychiatry 151: 640–649.
Freeman, MP, Freeman, SA, McElroy, SL (2002) The comorbidity of
bipolar and anxiety disorders: prevalence, psychobiology, and
treatment issues. J Affect Disord 68: 1–23.
French, JA, Kugler, AR, Robbins, JL, Knapp, LE, Garofalo, EA,
French, JA, et al. (2003) Dose-response trial of pregabalin adjunc-
tive therapy in patients with partial seizures. Neurology 60: 1631–
Freynhagen, R, Strojek, K, Griesing, T, Whalen, E, Balkenohl, M,
Freynhagen, R, et al. (2005) Efficacy of pregabalin in neuropathic
pain evaluated in a 12-week, randomised, double-blind, multicen-
tre, placebo-controlled trial of flexible- and fixed-dose regimens.
Pain 115: 254–263.
Gambi, F, De Berardis, D, Campanella, D, Carano, A, Sepede, G,
Salini, G, et al. (2005) Mirtazapine treatment of generalized anxiety
disorder: a fixed dose, open label study. J Psychopharmacol 19:
Gelenberg, AJ, Lydiard, RB, Rudolph, RL, Aguiar, L, Haskins, JT,
Salinas, E (2000) Efficacy of venlafaxine extended-release capsules
in nondepressed outpatients with generalized anxiety disorder: a 6-
month randomized controlled trial. JAMA 283: 3082–3088.
George, DT, Ladenheim, JA, Nutt, DJ (1987) Effect of pregnancy on
panic attacks. Am J Psychiatry 144: 1078–1079.
Goodman, WK, Bose, A, Wang, Q (2005) Treatment of generalized
anxiety disorder with escitalopram: pooled results from double-
blind, placebo-controlled trials. J Affect Disord 87: 161–167.
Goodnick, PJ, Puig, A, DeVane, CL, Freund, BV (1999) Mirtazapine
in major depression with comorbid generalized anxiety disorder. J
Clin Psychiatry 60: 446–448.
Goodwin, RD, Gorman, JM (2002) Psychopharmacologic treatment
of generalized anxiety disorder and the risk of major depression.
Am J Psychiatry 159: 1935–1937.
Goodwin, RD, Stein, MB (2002) Generalized anxiety disorder and
peptic ulcer disease among adults in the United States. Psychosom
Med 64: 862–866.
Guillem, E, Pelissolo, A, Lepine, JP (1999) [Mental disorders and
migraine: epidemiologic studies]Encephale 25: 436–442.
Gunnell, D, Saperia, J, Ashby, D (2005) Selective serotonin reuptake
inhibitors (SSRIs) and suicide in adults: meta-analysis of drug com-
20Psychopharmacological treatment algorithm for GAD
pany data from placebo controlled, randomised controlled trials
submitted to the MHRA’s safety review. BMJ 330: 385.
Guy, W (1976) ECDEU Assessment Manual for Psychopharmacology,
Vol Publication #(ADM) 76-338. Washington, DC: Department of
Health, Education and Welfare.
Hamilton, M (1959) The assessment of anxiety states by rating. Br J
Med Psychol 32: 50–55.
Hammad, TA, Laughren, T, Racoosin, J (2006) Suicidality in pediatric
patients treated with antidepressant drugs. Arch Gen Psychiatry 63:
Hamner, MB, Faldowski, RA, Ulmer, HG, Frueh, BC, Huber, MG,
Arana, GW (2003) Adjunctive risperidone treatment in post-
traumatic stress disorder: a preliminary controlled trial of effects
on comorbid psychotic symptoms. Int Clin Psychopharmacol 18:
Hartford, J, Kornstein, S, Liebowitz, M, Pigott, T, Russell, J, Detke,
M, et al. (2007) Duloxetine as an SNRI treatment for generalized
anxiety disorder: results from a placebo and active-controlled trial.
Int Clin Psychopharmacol 22: 167–174.
Hedges, DW, Reimherr, FW, Strong, RE, Halls, CH, Rust, C (1996)
An open trial of nefazodone in adult patients with generalized anx-
iety disorder. Psychopharmacol Bull 32: 671–676.
Heron, J, O’Connor, TG, Evans, J, Golding, J, Glover, V (2004) The
course of anxiety and depression through pregnancy and the post-
partum in a community sample. J Affect Disord 80: 65–73.
Hertzberg, MA, Butterfield, MI, Feldman, ME, Beckham, JC,
Sutherland, SM, Connor, KM, et al. (1999) A preliminary study
of lamotrigine for the treatment of posttraumatic stress disorder.
Biol Psychiatry 45: 1226–1229.
Hoehn-Saric, R, McLeod, DR, Funderburk, F, Kowalski, P (2004)
Somatic symptoms and physiologic responses in generalized anxi-
ety disorder and panic disorder: an ambulatory monitor study.
Arch Gen Psychiatry 61: 913–921.
Hoehn-Saric, R, McLeod, DR, Zimmerli, WD (1988) Differential
effects of alprazolam and imipramine in generalized anxiety disor-
der: somatic versus psychic symptoms. J Clin Psychiatry 49: 293–
Hollander, E, Baldini Rossi, N, Sood, E, Pallanti, S (2003) Risperi-
done augmentation in treatment-resistant obsessive-compulsive dis-
order: a double-blind, placebo-controlled study. Int J Neuropsy-
chopharmacol 6: 397–401.
Jacob, S, Spinler, SA (2006) Hyponatremia associated with selective
serotonin-reuptake inhibitors in older adults. Ann Pharmacother
James, L, Gordon, E, Kraiuhin, C, Howson, A, Meares, R (1990)
Augmentation of auditory evoked potentials in somatization disor-
der. J Psychiatr Res 24: 155–163.
Johnson, BA, Ait-Daoud, N, Akhtar, FZ, Ma, JZ (2004) Oral topira-
mate reduces the consequences of drinking and improves the qual-
ity of life of alcohol-dependent individuals: a randomized con-
trolled trial. Arch Gen Psychiatry 61: 905–912.
Katz, IR, Reynolds, CF, Alexopoulos, GS, Hackett, D (2002) Venla-
faxine ER as a treatment for generalized anxiety disorder in older
adults: pooled analysis of five randomized placebo-controlled clini-
cal trials. J Am Geriatr Soc 50: 18–25.
Keck, PE, Strawn, JR, McElroy, SL (2006) Pharmacologic treatment
considerations in co-occurring bipolar and anxiety disorders. J Clin
Psychiatry 67(Suppl. 1): S8–S15.
Kennedy, BL, Schwab, JJ (1997) Utilization of medical specialists by
anxiety disorder patients. Psychosomatics 38: 109–112.
Kessler, RC (2000) The epidemiology of pure and comorbid general-
ized anxiety disorder: a review and evaluation of recent research.
Acta Psychiatr Scand 406(Suppl.): S7–S13.
Kessler, RC, Adler, L, Barkley, R, Biederman, J, Conners, CK,
Demler, O, et al. (2006) The prevalence and correlates of adult
ADHD in the United States: results from the National Comorbid-
ity Survey Replication. Am J Psychiatry 163: 716–723.
Kessler, RC, Berglund, P, Demler, O, Jin, R, Walters, EE (2005a)
Lifetime prevalence and age-of-onset distributions of DSM-IV dis-
orders in the National Comorbidity Survey Replication. Arch Gen
Psychiatry 62: 593–602[see comment].
Kessler, RC, Brandenburg, N, Lane, M, Roy-Byrne, P, Stang, PD,
Stein, DJ, et al. (2005b) Rethinking the duration requirement for
generalized anxiety disorder: evidence from the National Comor-
bidity Survey Replication. Psychol Med 35: 1073–1082.
Kessler, RC, Chiu, WT, Demler, O, Walters, EE (2005c) Prevalence,
severity, and comorbidity of 12-month DSM-IV disorders in the
National Comorbidity Survey Replication. Arch Gen Psychiatry
62: 617–627[see comment].
Kim-Cohen, J, Moffitt, TE, Taylor, A, Pawlby, SJ, Caspi, A (2005)
Maternal depression and children’s antisocial behavior: nature
and nurture effects. Arch Gen Psychiatry 62: 173–181.
Koponen, H, Allgulander, C, Erickson, J, Dunayevich, E, Pritchett, Y,
Detke, MJ, et al. (2007) Efficacy of duloxetine for the treatment of
generalized anxiety disorder: implications for primary care physi-
cians. Prim Care Companion J Clin Psychiatry 9: 100–107.
Kranzler, HR, Burleson, JA, Del Boca, FK, Babor, TF, Korner, P,
Brown, J, et al. (1994) Buspirone treatment of anxious alcoholics.
A placebo-controlled trial. Arch Gen Psychiatry 51: 720–731.
Krasucki, C, Howard, R, Mann, A (1999) Anxiety and its treatment in
the elderly. Int Psychogeriatr 11: 25–45.
Kuzma, JM, Black, DW (2004) Integrating pharmacotherapy and psy-
chotherapy in the management of anxiety disorders. Curr Psychia-
try Rep 6: 268–273.
Lader, M, Scotto, JC (1998) A multicentre double-blind comparison of
hydroxyzine, buspirone and placebo in patients with generalized
anxiety disorder. Psychopharmacology (Berl) 139: 402–406.
Lenze, EJ, Mulsant, BH, Shear, MK, Dew, MA, Miller, MD, Pollock,
BG, et al. (2005) Efficacy and tolerability of citalopram in the
treatment of late-life anxiety disorders: results from an 8-week ran-
domized, placebo-controlled trial. Am J Psychiatry 162: 146–150.
Lesser, H, Sharma, U, LaMoreaux, L, Poole, RM, Lesser, H, Sharma,
U, et al. (2004) Pregabalin relieves symptoms of painful diabetic
neuropathy: a randomized controlled trial. Neurology 63: 2104–
Leverich, GS, Altshuler, LL, Frye, MA, Suppes, T, McElroy, SL,
Keck, PE, et al. (2006) Risk of switch in mood polarity to hypoma-
nia or mania in patients with bipolar depression during acute and
continuation trials of venlafaxine, sertraline, and bupropion as
adjuncts to mood stabilizers. Am J Psychiatry 163: 232–239.
Li, X, May, RS, Tolbert, LC, Jackson, WT, Flournoy, JM, Baxter,
LR (2005) Risperidone and haloperidol augmentation of serotonin
reuptake inhibitors in refractory obsessive-compulsive disorder: a
crossover study. J Clin Psychiatry 66: 736–743.
Lieberman, JA (2007) Effectiveness of antipsychotic drugs in patients
with chronic schizophrenia: efficacy, safety and cost outcomes of
CATIE and other trials. J Clin Psychiatry 68: e04.
Lin, EH, Katon, W, Von Korff, M, Bush, T, Lipscomb, P, Russo, J, et
al. (1991) Frustrating patients: physician and patient perspectives
among distressed high users of medical services. J Gen Intern
Med 6: 241–246[see comment].
Psychopharmacological treatment algorithm for GAD 21
Lin, EH, Von Korff, M, Katon, W, Bush, T, Simon, GE, Walker, E,
et al. (1995) The role of the primary care physician in patients’
adherence to antidepressant therapy. Med Care 33: 67–74.
Lin, EH, Von Korff, M, Ludman, EJ, Rutter, C, Bush, TM, Simon,
GE, et al. (2003) Enhancing adherence to prevent depression
relapse in primary care. Gen Hosp Psychiatry 25: 303–310[see
Lindesay, J, Briggs, K, Murphy, E (1989) The Guy’s/Age Concern sur-
vey. Prevalence rates of cognitive impairment, depression and anx-
iety in an urban elderly community. Br J Psychiatry 155: 317–329.
Lingford-Hughes, AR, Welch, S, Nutt, DJ (2004) Evidence-based
guidelines for the pharmacological management of substance mis-
use, addiction and comorbidity: recommendations from the British
Association for Psychopharmacology. J Psychopharmacol 18: 293–
Llorca, PM, Spadone, C, Sol, O, Danniau, A, Bougerol, T, Corruble,
E, et al. (2002) Efficacy and safety of hydroxyzine in the treatment
of generalized anxiety disorder: a 3-month double-blind study. J
Clin Psychiatry 63: 1020–1027[see comment].
Maier, W, Gansicke, M, Freyberger, HJ, Linz, M, Heun, R,
Lecrubier, Y (2000) Generalized anxiety disorder (ICD-10) in pri-
mary care from a cross-cultural perspective: a valid diagnostic
entity. Acta Psychiatr Scand 101: 29–36.
Mathew, SJ, Amiel, JM, Coplan, JD, Fitterling, HA, Sackeim, HA,
Gorman, JM (2005) Open-label trial of riluzole in generalized anx-
iety disorder. Am J Psychiatry 162: 2379–2381.
Mayers, AG, Baldwin, DS (2005) Antidepressants and their effect on
sleep. Hum Psychopharmacol 20: 533–559.
McDougle, CJ, Epperson, CN, Pelton, GH, Wasylink, S, Price, LH
(2000) A double-blind, placebo-controlled study of risperidone
addition in serotonin reuptake inhibitor-refractory obsessive-
compulsive disorder. Arch Gen Psychiatry 57: 794–801.
Metz, A, Sichel, DA, Goff, DC (1988) Postpartum panic disorder. J
Clin Psychiatry 49: 278–279.
Mitte, K, Noack, P, Steil, R, Hautzinger, M (2005) A meta-analytic
review of the efficacy of drug treatment in generalized anxiety dis-
order. J Clin Psychopharmacol 25: 141–150[see comment].
Monnelly, EP, Ciraulo, DA, Knapp, C, Keane, T (2003) Low-dose ris-
peridone as adjunctive therapy for irritable aggression in posttrau-
matic stress disorder. J Clin Psychopharmacol 23: 193–196.
Montgomery, SA, Tobias, K, Zornberg, GL, Kasper, S, Pande, AC
(2006) Efficacy and safety of pregabalin in the treatment of gener-
alized anxiety disorder: a 6-week, multicenter, randomized, double-
blind, placebo-controlled comparison of pregabalin and venlafax-
ine. J Clin Psychiatry 67: 771–782.
Northcott, CJ, Stein, MB (1994) Panic disorder in pregnancy. J Clin
Psychiatry 55: 539–542.
Nunes, EV, Levin, FR (2004) Treatment of depression in patients with
alcohol or other drug dependence: a meta-analysis. JAMA 291:
O’Connor, TG, Heron, J, Golding, J, Beveridge, M, Glover, V (2002)
Maternal antenatal anxiety and children’s behavioural/emotional
problems at 4 years. Report from the Avon Longitudinal Study of
Parents and Children. Br J Psychiatry 180: 502–508.
O’Connor, TG, Heron, J, Golding, J, Glover, V (2003) Maternal ante-
natal anxiety and behavioural/emotional problems in children: a
test of a programming hypothesis. J Child Psychol Psychiatry 44:
O’Malley, SS (1995) Integration of opioid antagonists and psychoso-
cial therapy in the treatment of narcotic and alcohol dependence.
J Clin Psychiatry 56(Suppl. 7): S30–S38.
Offord, DR, Boyle, MH, Campbell, D, Goering, P, Lin, E, Wong, M,
et al. (1996) One-year prevalence of psychiatric disorder in Ontar-
ians 15 to 64 years of age. Can J Psychiatry 41: 559–563[see
Oslin, DW, Ten Have, TR, Streim, JE, Datto, CJ, Weintraub, D,
DiFilippo, S, et al. (2003) Probing the safety of medications in the
frail elderly: evidence from a randomized clinical trial of sertraline
and venlafaxine in depressed nursing home residents. J Clin Psychi-
atry 64: 875–882.
Osser, DN, Renner, J, Bayog, R (1999) Algorithms for the pharmaco-
therapy of anxiety disorders in patients with chemical abuse and
dependence. Psychiatr Ann 29: 285–301.
Pande, AC, Crockatt, JG, Feltner, DE, Janney, CA, Smith, WT,
Weisler, R, et al. (2003) Pregabalin in generalized anxiety disorder:
a placebo-controlled trial. Am J Psychiatry 160: 533–540.
Pande, AC, Davidson, JR, Jefferson, JW, Janney, CA, Katzelnick, DJ,
Weisler, RH, et al. (1999) Treatment of social phobia with gaba-
pentin: a placebo-controlled study. J Clin Psychopharmacol 19:
Pande, AC, Feltner, DE, Jefferson, JW, Davidson, JR, Pollack, M,
Stein, MB, et al. (2004) Efficacy of the novel anxiolytic pregabalin
in social anxiety disorder: a placebo-controlled, multicenter study. J
Clin Psychopharmacol 24: 141–149.
Pohl, RB, Feltner, DE, Fieve, RR, Pande, AC (2005) Efficacy of preg-
abalin in the treatment of generalized anxiety disorder: double-
blind, placebo-controlled comparison of BID versus TID dosing. J
Clin Psychopharmacol 25: 151–158.
Pollack, MH, Roy-Byrne, PP, Van Ameringen, M, Snyder, H, Brown,
C, Ondrasik, J, et al. (2005) The selective GABA reuptake inhibitor
tiagabine for the treatment of generalized anxiety disorder: results
of a placebo-controlled study. J Clin Psychiatry 66: 1401–1408.
Pollack, MH, Simon, NM, Zalta, AK, Worthington, JJ, Hoge, EA,
Mick, E, et al. (2006) Olanzapine augmentation of fluoxetine for
refractory generalized anxiety disorder: a placebo controlled
study. Biol Psychiatry 59: 211–215.
Pollack, MH, Zaninelli, R, Goddard, A, McCafferty, JP, Bellew, KM,
Burnham, DB, et al. (2001) Paroxetine in the treatment of general-
ized anxiety disorder: results of a placebo-controlled, flexible-
dosage trial. J Clin Psychiatry 62: 350–357[see comment] [erratum
appears in J Clin Psychiatry 2001 Aug; 62(8): 658].
Power, KG, Jerrom, DWA, Simpson, RJ (1989) A controlled compar-
ison of cognitive-behavioral therapy, diazepam and placebo in the
management of generalized anxiety. Behav Psychother 17: 1–14.
Primeau, F, Fontaine, R, Beauclair, L (1990) Valproic acid and panic
disorder. Can J Psychiatry 35: 248–250.
Randall, CL, Johnson, MR, Thevos, AK, Sonne, SC, Thomas, SE,
Willard, SL, et al. (2001) Paroxetine for social anxiety and alcohol
use in dual-diagnosed patients. Depress Anxiety 14: 255–262.
Reich, DB, Winternitz, S, Hennen, J, Watts, T, Stanculescu, C (2004)
A preliminary study of risperidone in the treatment of posttrau-
matic stress disorder related to childhood abuse in women. J Clin
Psychiatry 65: 1601–1606.
Richardson, GS, Roehrs, TA, Rosenthal, L, Koshorek, G, Roth, T
(2002) Tolerance to daytime sedative effects of H1 antihistamines.
J Clin Psychopharmacol 22: 511–515.
Richter, RW, Portenoy, R, Sharma, U, Lamoreaux, L, Bockbrader,
H, Knapp, LE, et al. (2005) Relief of painful diabetic peripheral
neuropathy with pregabalin: a randomized, placebo-controlled
trial. J Pain 6: 253–260.
Rickels, K, Downing, R, Schweizer, E, Hassman, H (1993) Antide-
pressants for the treatment of generalized anxiety disorder. A
22Psychopharmacological treatment algorithm for GAD
placebo-controlled comparison of imipramine, trazodone, and diaz-
epam. Arch Gen Psychiatry 50: 884–895.
Rickels, K, Pollack, MH, Feltner, DE, Lydiard, RB, Zimbroff, DL,
Bielski, RJ, et al. (2005) Pregabalin for treatment of generalized
anxiety disorder: a 4-week, multicenter, double-blind, placebo-
controlled trial of pregabalin and alprazolam. Arch Gen Psychiatry
Rickels, K, Pollack, MH, Sheehan, DV, Haskins, JT (2000) Efficacy of
extended-release venlafaxine in nondepressed outpatients with gen-
eralized anxiety disorder. Am J Psychiatry 157: 968–974.
Rickels, K, Schweizer, E (1996) Pharmacological treatment for gener-
alized anxiety disorder. In: Mavissakalian, M, Prien, R (eds),
Long-Term Treatment of Anxiety Disorders. Washington, DC:
American Psychiatric Press, pp. 201–220.
Rickels, K, Weisman, K, Norstad, N, Singer, M, Stoltz, D, Brown, A,
et al. (1982) Buspirone and diazepam in anxiety: a controlled study.
J Clin Psychiatry 43: 81–86.
Rickels, K, Zaninelli, R, McCafferty, J, Bellew, K, Iyengar, M,
Sheehan, D (2003) Paroxetine treatment of generalized anxiety dis-
order: a double-blind, placebo-controlled study. Am J Psychiatry
Robinson, DS, Rickels, K, Feighner, J, Fabre, LF, Gammans, RE,
Shrotriya, RC, et al. (1990) Clinical effects of the 5-HT1A partial
agonists in depression: a composite analysis of buspirone in the
treatment of depression. J Clin Psychopharmacol 10: 67S–76S.
Rocca, P, Fonzo, V, Scotta, M, Zanalda, E, Ravizza, L (1997) Parox-
etine efficacy in the treatment of generalized anxiety disorder. Acta
Psychiatr Scand 95: 444–450.
Rosenbaum, JF, Woods, SW, Groves, JE, Klerman, GL (1984) Emer-
gence of hostility during alprazolam treatment. Am J Psychiatry
Rosenstock, J, Tuchman, M, LaMoreaux, L, Sharma, U, Rosenstock,
J, Tuchman, M, et al. (2004) Pregabalin for the treatment of pain-
ful diabetic peripheral neuropathy: a double-blind, placebo-
controlled trial. Pain 110: 628–638.
Roy-Byrne, PP, Wagner, A (2004) Primary care perspectives on gener-
alized anxiety disorder. J Clin Psychiatry 65(Suppl. 13): S20–S26.
Rubino, A, Roskell, N, Tennis, P, Mines, D, Weich, S, Andrews, E
(2007) Risk of suicide during treatment with venlafaxine, citalo-
pram, fluoxetine, and dothiepin: retrospective cohort study. BMJ
Ruscio, AM, Lane, M, Roy-Byrne, P, Stang, PE, Stein, DJ, Wittchen,
HU, et al. (2005) Should excessive worry be required for a diagno-
sis of generalized anxiety disorder? Results from the US National
Comorbidity Survey Replication. Psychol Med 35: 1761–1772.
Rynn, M, Russell, J, Erickson, J, Detke, MJ, Ball, S, Dinkel, J, et al.
(2007) Efficacy and safety of duloxetine in the treatment of gener-
alized anxiety disorder:a flexible-dose,
placebo-controlled trial. Depress Anxiety 25: 182–189.
Sabatowski, R, Galvez, R, Cherry, DA, Jacquot, F, Vincent, E,
Maisonobe, P, et al. (2004) Pregabalin reduces pain and improves
sleep and mood disturbances in patients with post-herpetic neural-
gia: results of a randomised, placebo-controlled clinical trial. Pain
Sareen, J, Cox, BJ, Afifi, TO, de Graaf, R, Asmundson, GJ, ten Have,
M, et al. (2005a) Anxiety disorders and risk for suicidal ideation
and suicide attempts: a population-based longitudinal study of
adults. Arch Gen Psychiatry 62: 1249–1257.
Sareen, J, Houlahan, T, Cox, BJ, Asmundson, GJ (2005b) Anxiety dis-
orders associated with suicidal ideation and suicide attempts in the
National Comorbidity Survey. J Nerv Ment Dis 193: 450–454.
Schade, A, Marquenie, LA, Van Balkom, AJ, De Beurs, E, Van Dyck,
R, Van Den Brink, W (2003) Do comorbid anxiety disorders in
alcohol-dependent patients need specific treatment to prevent
relapse. Alcohol Alcohol 38: 255–262.
Schuurmans, J, Comijs, H, Emmelkamp, PM, Gundy, CM, Weijnen,
I, van den Hout, M, et al. (2006) A randomized, controlled trial of
the effectiveness of cognitive-behavioral therapy and sertraline ver-
sus a waitlist control group for anxiety disorders in older adults.
Am J Geriatr Psychiatry 14: 255–263.
Schweizer, E, Rickels, K (1997) Strategies for treatment of generalized
anxiety in the primary care setting. J Clin Psychiatry 58(Suppl. 3):
S27–S31discussion 32–23 [see comment].
Shapira, NA, Ward, HE, Mandoki, M, Murphy, TK, Yang, MC,
Blier, P, et al. (2004) A double-blind, placebo-controlled trial of
olanzapine addition in fluoxetine-refractory obsessive-compulsive
disorder. Biol Psychiatry 55: 553–555.
Sheehan, DV (1983) The Anxiety Disease. New York: Bantam.
Sheehan, DV (2001) Attaining remission in generalized anxiety disor-
der: venlafaxine extended release comparative data. J Clin Psychia-
try 62(Suppl. 19): S26–S31.
Sholomskas, DE, Wickamaratne, PJ, Dogolo, L, O’Brien, DW, Leaf,
PJ, et al. (1993) Postpartum onset of panic disorder: a coincidental
event. J Clin Psychiatry 54: 476–480.
Silverstone, PH, Salinas, E (2001) Efficacy of venlafaxine extended
release in patients with major depressive disorder and comorbid
generalized anxiety disorder. J Clin Psychiatry 62: 523–529.
Snyderman, SH, Rynn, MA, Rickels, K (2005) Open-label pilot study
of ziprasidone for refractory generalized anxiety disorder. J Clin
Psychopharmacol 25: 497–499.
Srisurapanont, M, Jarusuraisin, N (2005) Opioid antagonists for alco-
hol dependence. Cochrane Database of Systematic Reviews:
CD001867[update of Cochrane Database Syst Rev 2002; (2):
CD001867; PMID: 12076425].
Stanley, MA, Novy, DM (2000) Cognitive-behavior therapy for gener-
alized anxiety in late life: an evaluative overview. J Anxiety Disord
Stein, DJ (2001) Comorbidity in generalized anxiety disorder: impact
and implications. J Clin Psychiatry 62: 29–34discussion 35–26.
Stein, MB, Kline, NA, Matloff, JL (2002) Adjunctive olanzapine for
SSRI-resistant combat-related PTSD: a double-blind, placebo-
controlled study. Am J Psychiatry 159: 1777–1779.
Stocchi, F, Nordera, G, Jokinen, RH, Lepola, UM, Hewett, K,
Bryson, H, et al. (2003) Efficacy and tolerability of paroxetine for
the long-term treatment of generalized anxiety disorder. J Clin Psy-
chiatry 64: 250–258.
Swinson, RP, Anthony, MM, Bleau, P, Chokka, P, Craven, M, Fallu,
A, et al. (2006) Clinical practice guidelines. Management of anxiety
disorders. Can J Psychiatry 51: 9S–91S.
Tajima, O (2004) Internet survey on generalized anxiety disorder in
Japan. Jpn Med J 4222: 41–46.
Tiihonen, J, Lonnqvist, J, Wahlbeck, K, Klaukka, T, Tanskanen, A,
Haukka, J (2006) Antidepressants and the risk of suicide,
attempted suicide, and overall mortality in a nationwide cohort.
Arch Gen Psychiatry 63: 1358–1367.
Trivedi, MH, Fava, M, Wisniewski, SR, Thase, ME, Quitkin, F,
Warden, D, et al. (2006) Medication augmentation after the failure
of SSRIs for depression. N Engl J Med 354: 1243–1252.
Van den Bergh, BR, Marcoen, A (2004) High antenatal maternal anx-
iety is related to ADHD symptoms, externalizing problems, and
anxiety in 8- and 9-year-olds. Child Dev 75: 1085–1097.
Psychopharmacological treatment algorithm for GAD 23
Van den Bergh, BR, Mennes, M, Oosterlaan, J, Stevens, V, Stiers, P, Download full-text
Marcoen, A, et al. (2005) High antenatal maternal anxiety is
related to impulsivity during performance on cognitive tasks in
14- and 15-year-olds. Neurosci Biobehav Rev 29: 259–269.
Varia, I, Rauscher, F (2002) Treatment of generalized anxiety disorder
with citalopram. Int Clin Psychopharmacol 17: 103–107.
Verheul, R, Lehert, P, Geerlings, PJ, Koeter, MW, van den Brink, W
(2005) Predictors of acamprosate efficacy: results from a pooled
analysisof seven European
dependent patients. Psychopharmacology (Berl) 178: 167–173.
Wingerson, D, Sullivan, M, Dager, S, Flick, S, Dunner, D, Roy-Byrne,
P (1993) Personality traits and early discontinuation from clinical
trials in anxious patients. J Clin Psychopharmacol 13: 194–197.
Wittchen, HU, Jacobi, F (2005) Size and burden of mental disorders in
Europe–a critical review and appraisal of 27 studies. Eur Neurop-
sychopharmacol 15: 357–376.
Wittchen, HU, Zhao, S, Kessler, RC, Eaton, WW (1994) DSM-III-R
generalized anxiety disorder in the National Comorbidity Survey.
Arch Gen Psychiatry 51: 355–364.
Woodman, CL, Noyes, R (1994) Panic disorder: treatment with
valproate. J Clin Psychiatry 55: 134–136.
Woods, JH, Katz, JL, Winger, G (1992) Benzodiazepines: use, abuse,
and consequences. Pharmacol Rev 44: 151–347.
Wunderlich, U, Bronisch, T, Wittchen, HU (1998) Comorbidity pat-
terns in adolescents and young adults with suicide attempts. Eur
Arch Psychiatry Clin Neurosci 248: 87–95.
Wurthmann, C, Klieser, E, Lehmann, E (1995) [Differential pharma-
cologic therapy of generalized anxiety disorders – results of a study
with 30 individual case experiments]Fortschr Neurol Psychiatr 63:
Yonkers, KA, Dyck, IR, Warshaw, M, Keller, MB (2000) Factors pre-
dicting the clinical course of generalised anxiety disorder. Br J Psy-
chiatry 176: 544–549.
Zhan, H, Shu, L, Li, H, Gu, N, Li, T, Ma, C, et al. (2004) Comparison
study on effectiveness and safety of tandospirone and buspirone in
the treatment of generalized anxiety disroder. Chin J Clin Pharma-
col 20: 21–24.
Zhang, W, Connor, KM, Davidson, JR (2005) Levetiracetam in social
phobia: a placebo controlled pilot study. J Psychopharmacol 19:
Zigmond, AS, Snaith, RP (1983) The hospital anxiety and depression
scale. Acta Psychiatr Scand 67: 361–370.
Zimmerman, M, Chelminski, I (2003) Generalized anxiety disorder in
patients with major depression: is DSM-IV’s hierarchy correct. Am
J Psychiatry 160: 504–512.
24 Psychopharmacological treatment algorithm for GAD