Psychological therapies for generalised anxiety disorder
Abstract
This is a protocol for a Cochrane Review (Intervention). The objectives are as follows:
To examine the efficacy and acceptability of psychological therapies in comparison with treatment as usual/standard care/waiting list for patients with generalised anxiety disorder
To examine the effiacy and acceptability of cognitive behavioural therapy in comparison with psychodynamic psychotherapy and supportive therapy, for patients with generalised anxiety disorder
Background
Generalised anxiety disorder (GAD) is a highly prevalent condition, characterised by excessive worry or anxiety about everyday events and problems to the point at which the individual experiences considerable distress and difficulty in performing day to day tasks. To meet Diagnostic and Statistical Manual for Mental Disorders (DSM‐IV) criteria for GAD, anxiety and worry should be accompanied by autonomic hyperactivity (rapid heart rate, shortness of breath, dry mouth and dizziness), increased motor tension (fatigue, restlessness, trembling and muscle tension) and increased vigilance (impaired concentration and feeling tense) (APA 1994). The focus of the anxiety and worry is not confined to features of another Axis I disorder such as having a panic attack (as in panic disorder), being embarrassed in public (as in social anxiety disorder) or being contaminated (as in obsessive‐compulsive disorder). Symptoms should be experienced at least one day in two, for a period of six months or longer.
The onset of GAD symptoms is usually gradual, although it may be precipitated by stressful life events. GAD tends to fluctuate in severity (Schweizer 1997) and is recurrent in presentation, with a low rate of remission and recovery (Tonks 2003, Yonkers 1996). It is frequently difficult to diagnose due to its diffuse clinical presentation, coupled with the common occurrence of comorbid medical or psychiatric conditions, with up to 90% of patients with GAD showing concomitant symptoms of depression, dysthymia, somatisation, bipolar disorder or substance abuse (Kessler 1994). As a disorder, GAD tends to be chronic in nature, as demonstrated in the US National Comorbidity Survey (Kessler 1994), which reported that in the 12 months before diagnostic interview, anxiety disorders as a group occurred at a rate of 17%, in comparison with depressive disorders at 11%.
In the general population, the lifetime prevalence of GAD is 5.1%, with a 12‐month prevalence measured at 3.1% (Kessler 1994). Within the primary care setting, the WHO collaborative study on Psychological Disorders in Primary Health Care study (Sartorius 1993) reported that GAD formed the second largest category of psychological disorders in the primary care setting, with a prevalence of just under 7.9%. A survey of high utilisers of medical health care found a particularly high prevalence rate of GAD at 22% (Katon 1990), and the prevalence of GAD in patients visiting physicians' offices has been shown to be twice that found in the community (Schweizer 1997). Women have a higher prevalence rate for GAD than men (Kessler 1994), with the median age of onset occurring during the early 20s (Rickels 1990).
Individuals with GAD report subjective distress due to constant worry, and have difficulty in controlling the worry, resulting in impaired social functioning and quality of life. From a public health perspective, GAD is associated with increased reliance in public assistance, reuced work productivity, impaired social relationships and low ratings of life satisfaction (Massion 1993). It has been suggested that as an independent disorder, GAD has a disabling capacity comparable to that of major depression, and as such should be considered a major public health problem (Kessler 2000). In the UK, the Mental Health Foundation reports that of 91 million working days lost to mental ill health every year, approximately half of those days are lost due to anxiety and stress conditions (MHF 2003), and patients with GAD are more likely than other patients to make frequent medical appointments and to undergo extensive diagnostic testing, with associated cost implications. Direct and indirect costs of anxiety disorders were estimated to be as high as 40 to 50 billion dollars in 1990 in the United States alone (Greenberg 1999), representing approximately a third of all medical expenses incurred during the same period.
In the 1970s benzodiazepines were used extensively in the treatment of anxiety. However, due to their potential for the development of tolerance and dependence, clinical guidelines now recommend that benzodiazepines are prescribed for no longer than 2‐4 weeks in the treatment of GAD (NICE 2004). Azapirones, a form of 5‐HT1 anxiolytic that includes buspirone, are a preferred and reasonably effective alternative to benzodiazepines in treating GAD (Chessick 2006). Antidepressants have become a further pharmacological replacement for benzodiazepines in treating anxiety disorders, with the efficacy of imipramine, venlafaxine and paroxetine against placebo demonstrated in the clinical management of GAD in adults (Kapczinski 2003).
Surveys and opinion polls conducted over the last ten years have consistently indicated that the lay public and primary care attendees prefer psychological therapies to pharmacological treatments as a treatment modality for mental health disorders (Riedel‐Heller 2005, Churchill 2000, Priest 1996). A diverse range of manualised and non‐manualised psychological therapies are now available in treating common mental disorders (CMD), underpinned by cognitive (Beck 1979, Ellis 1962), behavioural (Skinner 1953), psychodynamic (Freud 1949) and humanistic/non‐directive (Rogers 1951, Perls 1976) principles, together with an increasing number of integrative models (Egan 1986). There is a growing and demonstrable evidence base for the effectiveness of psychological therapies in treating CMD (Roth 2005). Cognitive therapy and behavioural interventions such as self‐control desensitisation, self‐monitoring and progressive muscle relaxation, used as stand‐alone treatments or combined within anxiety management programmes (Suinn 1971), appear to be effective compared with standard care for the treatment of GAD in adults (Fisher 1999, Gould 1997), and in the elderly (Wetherell 2005). Oher behavioural approaches such as exposure methods, commonly used in treating other anxiety disorders, may less applicable in GAD, due to the non‐specificity of external triggers (Deacon 2004). Psychotherapeutic approaches seem to be well tolerated by patients with GAD, and the dropout rates in clinical trials appear to be low (Borkovec 2001).
Clinical guidelines now recommend cognitive behavioural therapy (CBT) as a first‐line treatment for GAD (NICE 2004, Ballenger 2001), and a recent UK Government strategy paper has called for an additional 10,000 psychological therapists to be trained in CBT or other evidence‐based therapies to treat mental health disorders (Layard 2004). Based on publications in 1997 and 1998, the Department of Health Treatment Choice in Psychological Therapies and Counselling Evidence‐based Clinical Practice Guideline (DoH 2001) concluded that while cognitive and behavioural therapies were effective in treating GAD, "other psychotherapeutic approaches have not yet been systematically reviewed/evaluated." Previous reviews on psychological therapies for GAD have been limited to a single pooled outcome of clinically significant change (Fisher 1999), have summarised prevailing evidence on anxiety disorders/mental disorders narratively (Butler 2006, Roth 2005, Deacon 2004, DeRubeis 1998), or were published more than eight years ago (Gould 1997). Thus, an in‐depth and up to date comparative investigation of psychological therapy models in a GAD population using meta‐analytic techniques appears to be lacking. The current review aims to provide a comprehensive, updated summary and meta‐analysis on the effectiveness and comparative effectiveness of all psychological therapies for GAD.
Objectives
To examine the efficacy and acceptability of psychological therapies in comparison with treatment as usual/standard care/waiting list for patients with generalised anxiety disorder
To examine the effiacy and acceptability of cognitive behavioural therapy in comparison with psychodynamic psychotherapy and supportive therapy, for patients with generalised anxiety disorder
Methods
Criteria for considering studies for this review
Types of studies
Randomised controlled trials will be eligible for inclusion in the review. Quasi‐randomised controlled trials, in which treatment assignment is decided through methods such as alternate days of the week, will also be included.
Trials that have used a cross‐over design will be included in the review, using data from the first active treatment stage only.
Types of participants
Patient characteristics and setting
Male and female adults, aged between 18‐75 years, treated in a primary, secondary or community setting will be eligible for inclusion in the review. Studies conducted in an in‐patient setting will be excluded.
Diagnosis
The primary diagnosis will comprise generalised anxiety disorder, to include neurotic anxiety, but excluding social phobia, panic disorder, post‐traumatic stress disorder, simple phobias and obsessive‐compulsive disorder. For inclusion in the review, studies should have used a formal standardised interview such as the Affective Disorders Interview Schedule (ADIS) (Di Nardo 1994) to diagnose GAD, based on ICD 9 and ICD‐10 criteria (WHO 1992) or DSM‐III (APA 1980), DSM‐IIIR (APA 1987) and DSM‐IV criteria (APA 1994), conducted by a qualified or trained psychiatric assessor. Studies using validated instruments to identify general anxiety symptoms will be excluded.
Studies in which a minimum of 80% of participants have a primary diagnosis of GAD will be eligible for inclusion. Studies in which fewer than 80% of participants have a primary diagnosis of GAD will also be included in the review if data limited to GAD participants are provided.
Comorbidity
Since comorbidity is known to be a highly prevalent feature of GAD, studies involving participants with comorbid physical or common mental disorders will be eligible for inclusion, as long as the comorbidity is secondary to the diagnosis of generalised anxiety disorder. However, studies involving patients with a comorbid psychiatric diagnosis of substance‐related disorder, schizophrenia or psychotic disorder will be excluded.
Types of interventions
Categorisation of psychological therapies
A diverse range of psychological therapies are now available in treating anxiety disorders. For the purposes of this review, interventions will be classified into three principal categories, according to the theoretical underpinning described by trial authors, together with the references provided. The three planned categories are as follows:
1. Cognitive behavioural therapy
First developed in the 1960s, and later manualised as cognitive therapy (CT) (Beck 1979), cognitive behavioural therapy (CBT) incorporates elements of both behavioural therapy (BT) and cognitive therapy approaches. CBT facilitates the identification of irrational, anxiety‐provoking thoughts, and challenges these negative automatic thoughts and dysfunctional underlying beliefs through collaborative 'hypothesis‐testing', using behavioural tasks of diary‐keeping and validity‐testing of beliefs between sessions, and skills training within sessions. Rational emotive therapy (RET), also known as rational emotive behaviour therapy (REBT) (Ellis 1962), draws from similar principles, through use of the activating event, beliefs and consequences (ABC) model.
For the purposes of this review, the CBT category will include interventions developed and manualised in the treatment of anxiety disorders, including anxiety management training (Suinn 1977), cognitive restructuring, situational exposure and self‐control desensitization (Borkovec 1988), each of which draws from cognitive and/or behavioural principles. For the purposes of this review, stand‐alone CT and BT interventions will be included in the CBT category. The comparative efficacy of CT and BT will also be investigated separately (see Methods section).
Treatments developed for anxiety disorders often include relaxation techniques such as cue controlling, alternative self‐statements, relaxing imagery and meditational relaxation, which may also be manualised as stand‐alone applied interventions (Ost 1987, Berstein 1973). For this reason, whilst acknowledging that relaxation therapy/training (RT) is sometimes used as an attention‐placebo control comparison in psychological treatment trials, it will be considered an active BT intervention in this review, in line with previous reviews examining treatments for GAD (Borkovec 2001, Gould 1997).
2. Psychodynamic therapy
Grounded in psychoanalytic theory (Freud 1949), psychodynamic therapy (PD) uses the therapeutic relationship to explore and resolve unconscious conflict, with development of insight and circumscribed character change as therapeutic goals, and relief of symptomatology as an indirect outcome. Brief therapy models have been devised by Malan 1963, Balint 1972 and Mann 1973. More recently, psychodynamic therapies have been developed and manualised for the purposes of research evaluation, including short‐term anxiety‐provoking psychotherapy (STAPP), which is a focused, psychoanalytically oriented treatment, aiming at the resolution of oedipal, separation and grief problems (Sifneos 1992).
3. Supportive therapy
Supportive therapy (ST) is frequently used in outcome studies as a form of attention‐placebo control comparison against prescriptive or manualised 'active' psychological interventions. Drawing from humanistic principles, the major supportive therapy interventions for inclusion in this review include:
a) Rogerian person‐centred therapy (Rogers 1951) is considered experiential in approach, and core conditions of empathy, acceptance and genuineness are utilised by the therapist within the therapeutic relationship to facilitate the client towards self‐awareness and self‐determination. In recent years, manualised versions of person‐centred therapy have been developed by researchers for use in trials, and include non‐directive therapy (Svartberg 1998), non‐directive counselling (Blowers 1987) and supportive listening (Borkovec 2001).
b) Gestalt therapy (Perls 1976) aims to heighten an individual's self‐awareness and perception of the moment, especially in terms of their relationships with other people and with their environment.
c) Transactional analysis (Berne 1961) is based on an understanding of the interactions (transactions) between patient and therapist and between patient and others in the environment. It focuses primarily on ego states, principally the Parent, Adult, and Child.
As a psychological treatment that draws predominantly from a range of humanistic and/or integrative approaches (Egan 1986), counselling will be included in the supportive therapy category. However, supportive interventions that comprise patient education and provision of information without incorporation of a psychological therapy component will not be eligible for inclusion in the review.
Modality of therapies
The psychological intervention should be delivered face to face between the patient and therapist. Psychological therapies conducted on either an individual or on a group basis will be eligible for inclusion. However, psychological therapies comprising couples therapy and family therapy are excluded, because these therapies work with patterns and dynamics of relating within and between systems, rather than focusing on the individual. Couples therapy for GAD and family therapy for GAD will be covered in separate reviews.
Control comparison
The control comparison will include treatment as usual (TAU) (also called standard care, usual care or no treatment) and waiting list (WL). In each study, the description of a TAU condition will be scrutinised to ensure that it does not comprise an active supportive therapy treatment. Within the TAU condition, participants may receive any appropriate medical care during the course of the study on a naturalistic basis, including pharmacotherapy and/or psychological therapy, as deemed necessary by the clinician. Additional treatment(s) received by participants in both the control and active comparisons for each included study will be carefully documented.
Combination treatment
Combination treatments in which patients are randomised to receive psychological and pharmacological treatment concurrently will be included in the review if the study of interest compares two psychological models and both groups are prescribed the same concomitant pharmacological/placebo intervention. However, combination treatment which is compared against a pharmacological or psychological treatment alone will be excluded from this review, and will be investigated in a separate review.
Main comparisons
The following treatment comparisons will be conducted to test the review hypotheses:
1. All psychological therapies versus treatment as usual/waiting list, stratified by psychological model:
a) Cognitive behavioural therapy versus treatment as usual/waiting list
b) Psychodynamic therapy versus treatment as usual/waiting list
c) Supportive therapy versus treatment as usual/waiting list
2. Cognitive behavioural therapy versus psychodynamic therapy
3. Cognitive behavioural therapy versus supportive therapy
4. Psychodynamic therapy versus supportive therapy
5. Cognitive therapy versus behavioural therapy
Types of outcome measures
Primary outcome
The primary outcome is reduction in anxiety, to be measured in the following two ways:
1. Treatment response, comprising the proportion of participants showing absence vs presence of symptoms, or clinically significant change (treatment response/endpoint functioning) vs no significant change (Borkovec 1993), according to DSM‐III, DSM‐IV, ICD‐9 or ICD‐10 diagnostic criteria for GAD, or through use of a validated diagnostic measure such as the Hamilton Anxiety Scale (HAM‐A) (Hamilton 1959) or a composite of validated measures, according to trialists' definition.
2. Reduction in generalised anxiety symptoms measured using a validated continuous scale, either assessor‐rated, such as the Hamilton Anxiety Scale [HAM‐A] (Hamilton 1959) or self‐report, including the Trait subscale of the Spielberger State‐Trait Anxiety Inventory (STAI‐T) (Spielberger 1983), the Beck Anxiety Inventory (BAI) (Beck 1988), Hospital Depression and Anxiety Scale (HADS) (anxiety subscale, Zigmond 1983), Leeds Anxiety Scale (Snaith 1976) and Zung Self‐Rating of Anxiety Scale (Zung 1975).
Secondary outcomes
1. Reduction in worry/fear symptoms, using validated scales such as the Penn State Worry Questionnaire (PSWQ) (Meyer 1990) and Fear Questionnaire (Marks 1979).
2. Reduction in depression symptomatology, measured using validated observer‐rated scales such as the Hamilton Rating Scale for Depression (HRSD) (Hamilton 1960) or self‐report scales, including the Beck Depression Inventory (BDI) (Beck 1987) and Zung Depression Scale (ZDS) (Zung 1965)
3. Improvement in social functioning, measured using validated self‐report scales such as the Social Adjustment Scale (SAS) (Weissman 1974)
4. Quality of life, using measures such as the MOS 36‐Item Short Form Health Survey (SF‐36) or Short Form 12 (SF‐12) (Ware 1993)
5. Adherence to trial protocol (attrition), measured by the overall number of people dropping out post‐randomisation and during the course of trials
6. The number of participants reporting at least one adverse effect of psychological therapies, including increase in symptomatology levels or behavioural changes (eg increased use of alcohol), and the number of participants dropping out of trials due to adverse treatment effects
7. Treatment acceptability, using self‐report scales measuring satisfaction with care/treatment
8. Cost‐effectiveness outcomes (days of work absence/ability to return to work, number of appointments with primary care physician, number of referrals to secondary services, use of additional treatments, hospitalisation for mental or physical health problems).
For studies comparing different psychological therapy categories, the process of psychological therapy will also be examined, using validated measures of the therapist/client relationship, such as the Relationship Inventory (Barrett‐Lennard 1986).
Outcomes will be classified as post‐treatment, short term follow‐up (up to 6 months post‐treatment), medium term follow‐up (7‐12 months post‐treatment) and long term (longer than 12 months).
Search methods for identification of studies
See: Cochrane Collaboration Depression, Anxiety and Neurosis (CCDAN) Collaborative Review Group search strategy (http://web1.iop.kcl.ac.uk/IoP/ccdan/searches.htm)
1. Electronic databases
a) The two specialised CCDAN registers, CCDANCTR‐Studies and CCDANCTR‐References, will be searched using the following search strategies:
CCDANCTR‐Studies
Diagnosis = "Generalized Anxiety" or "Anxiety Neuros*" or "Neurotic Anxiety")
and
Intervention = *Therapy or Intervention and not "No Intervention"
and not
Age‐group = Child
CCDANCTR‐References
Free‐text = "Generalized Anxiety"
and
Free‐text =*therapy or treatment
b) The Cochrane Central Register of Controlled Trials (CENTRAL) will be searched using the same terms as above.
c) MEDLINE (1966‐2005), PsycINFO (1966‐2005), EMBASE (1980 ‐2005) and LILACS (1982‐2003) will be searched using the search strategy set out in Additional Table 1. A search of SciSearch will also be conducted.
| Search terms |
| #1 ANXIETY |
d) Ongoing studies
controlledtrials.com will be searched for information on trials in progress and recently completed.
2. Conference abstracts and book chapters
Conference abstracts and book chapters will be scrutinised for relevant references.
3. Personal Communication
in order to ensure that as many as possible RCTs and CCTs will be identified, authors of included studies and experts in the field will be consulted to find out if they know of any published or unpublished RCTs/ CCTs of psychological therapies for GAD, which have not been identified through electronic searches.
4. Reference checking
Reference lists of all studies identified as potentially eligible for the review (both those included and those subsequently excluded following scrutiny of whole articles) will be scrutinised to identify potential additional trials. Reference lists of previously published systematic reviews on the same topic will also be scrutinised.
5. Hand‐searching
The following journals will be handsearched
Journal of Anxiety Disorders (1993 onwards)
British Journal of Clinical Psychology (2000 onwards)
Psychology and Psychotherapy (2000 onwards)
Behavioural and Cognitive Psychotherapy (2000 onwards)
Data collection and analysis
Selection of studies
Two review authors (VH and VAT) will screen the abstracts of all publications obtained through the search strategy. For studies where psychological therapies are compared to a different type of psychological therapy or treatment as usual/waiting list, and are indicated to be an RCT or CCT, the full article will be obtained and inspected to assess whether the review inclusion criteria are fully met. Any disagreement on the eligibility of a study will be discussed with a third review author (RC), the decisions documented and, where necessary, the authors of the studies contacted for further information
Data extraction and management
Data will be extracted independently by two review authors (VH and VAT), and entered into spreadsheets designed for the purposes of the review. Any disagreement will be discussed with a third review author (MSL or RC), the decisions documented and, where necessary, the authors of the studies will be contacted for further information. For each included study, information will be recorded on the study population, interventions, randomisation and blinding procedures, sample size, outcome data, follow‐up and methods of statistical analysis.
Assessment of methodological quality of included studies
In order to ensure that variation is not caused by systematic errors in the design of a study, the methodological quality of the selected trials will be assessed by two review authors (VH and VT) independently. Any disagreement will be discussed with a third review author, the decisions documented and, where necessary, the authors of the studies contacted for further information. Methodological quality will be assessed according to the criteria set out in the Cochrane Reviewers' Handbook (Clarke 2002), based on evidence of a strong relationship between allocation concealment and potential for bias in the results (Schulz 1995). The criteria are defined below:
A. Low risk of bias (adequate allocation concealment)
B. Moderate risk of bias (unclear allocation concealment)
C. High risk of bias (inadequate allocation concealment)
An additional quality assessment will be performed using the Cochrane Collaboration Depression and Anxiety Group Quality Rating Scale (QRS) (Moncrieff 2001). The QRS consists of 23 items, including items on sample size, allocation, use of diagnostic criteria, compliance, attrition and statistical analysis. Total scores range from 0‐46. Quality rating scores will be used for descriptive purposes only, and trial exclusions will not be made based on these criteria.
Data analysis
Review Manager software will be used to organise and synthesise the data.
Measures of treatment effect
Continuous outcomes: where studies use the same outcome measure for a comparison, data will be pooled by calculating the weighted mean difference (WMD). Where different measures are used to assess the same outcome for a comparison, data will be pooled by calculating the standardised mean difference (SMD). 95% confidence intervals will be calculated around the pooled mean difference for each outcome. Continuous outcome data are frequently skewed. The standard deviation (SD), when multiplied by 2, should be less than the mean, otherwise it may be assumed that the data are not normally distributed, and the mean is not an appropriate measure of the centre of the distribution (Altman 1996). Trials with data that do not meet this standard will not be used in a continuous outcome comparison, and use of data will be limited to descriptive reporting.
Dichotomous outcomes: dichotomous outcomes will be analysed by calculating a pooled relative risk (RR) for each comparison, with the uncertainty in each result expressed using 95% confidence intervals (CIs). When overall results are significant, the number needed to treat (NNT) to produce one outcome will be calculated by combining the overall relative risk with an estimate of the prevalence of the event in the control group of the trials.
Unit of analysis issues
Where studies have two or more active treatment arms to be compared against TAU, data will be managed as follows:
Continuous data ‐ means, SDs and number of participants for each active treatment group will be pooled across treatment arms as a function of the number of participants in each arm (Law 2003) to be compared against the control group. As an alternative strategy, the active comparison considered to be of greatest relevance will be selected (eg CBT will be selected in preference to CT or BT arms).
Dichotomous data ‐ active treatment groups will be collapsed into a single arm for comparison against the control group.
Dealing with missing data
Missing dichotomous data will be managed through intention to treat (ITT) analysis, in which it will be assumed that patients who dropped out after randomisation had a negative outcome, although it is acknowledged that categorising drop‐outs as treatment failures may overestimate the number of patients with a poor outcome.
Missing continuous data will be either analysed on an endpoint basis, including only participants with a final assessment, or analysed using last observation carried forward to the final assessment (LOCF) if LOCF data are reported by the trial authors. Where SDs are missing, attempts will be made to obtain these data through contacting trial authors. Where SDs are not available from trial authors, they will be calculated from t‐values, confidence intervals or standard errors, where reported in articles (Deeks 1997). If these additional figures are not available or obtainable, the study will not be included in the comparison of interest.
For studies where the number of participants showing clinical response are not presented in the original articles, but means and standard deviations are reported for continuous symptomatology scales, the number of responders will be calculated and imputed from continuous data using a validated statistical method (Furukawa 2005).
Data synthesis
A fixed effects model will be used in the first instance to combine data. Where there is evidence of statistical heterogeneity, results will be recalculated using a random effects model. The random effects model will tend to give a more conservative estimate, but where there is no heterogeneity, the results from the two models should agree.
Assessment of reporting biases
Where sufficient numbers of trials allow a meaningful presentation, funnel plots will be constructed to establish the potential influence of publication bias. Funnel plots provide a graphical display of sample size plotted against effect size. When many studies are located that estimate the same effect, the distribution of points should resemble a funnel shape, with a widening in the spread of effect sizes as sample size decreases. A gap on one side of the wide part of the 'funnel' indicates that some studies have not been published or located.
Assessment of heterogeneity
Statistical heterogeneity will be formally tested using the natural approximate chi‐square test, which provides evidence of variation in effect estimates beyond that of chance. Since the chi‐squared test has low power to assess heterogeneity where a small number of participants or trials are included, the p‐value will be conservatively set at 0.1. Heterogeneity will also be tested using the I square statistic, which calculates the percentage of variability due to heterogeneity rather than chance, with I square values over 50% indicating strong heterogeneity (Higgins 2003).
Subgroup analyses and investigation of heterogeneity
Where sufficient data are available, clinical characteristics will be examined in subgroup analyses to investigate their influence on the size of the treatment effect. Subgroup analyses will be performed for:
1) type of control condition (treatment as usual and waiting list control)
2) modality of treatment (group therapy and individual therapy)
3) number of psychological therapy sessions (up to and including 8 sessions and more than 8 sessions)
4) age (adult and elderly population)
5) concomitant medication use (less than 25% use in sample and 25% or higher use)
6) severity/chronicity of GAD symptomatology at baseline
7) CMD comorbidity (less than 50% comorbidity in sample and 50% or higher comorbidity)
Where significant heterogeneity is indicated, potential sources of clinical heterogeneity will be examined through the use of subgroup analyses.
Sensitivity analyses
Where sufficient data are available, sensitivity analyses will be conducted to test the robustness of the findings obtained, by removing studies based on the following internal validity criteria :
1) overall quality rating on Quality Rating Scale (QRS) of 25 or lower
2) inadequate allocation concealment
3) dropout rates that are higher than 20%
4) lack of formal testing of fidelity to psychological therapy manual
5) use of less stringent diagnostic inclusion criteria
6) psychological therapy allegiance of triallists
Sensitivity analyses will also be used to examine potential sources of methodological heterogeneity.
| Search terms |
| #1 ANXIETY |
