Passiflora for anxiety disorder

Abstract

Background

Anxiety is a very common mental health problem in the general population and in the primary care setting. Herbal medicines are popularly used worldwide and could be an option for treating anxiety if shown to be effective and safe. Passiflora (passionflower extract) is one of these compounds.

Objectives

To investigate the effectiveness and safety of passiflora for treating any anxiety disorder.

Search methods

The following sources were used: electronic databases: Cochrane Collaboration Depression, Anxiety and Neurosis Cochrane Controlled Trials Register (CCDANCTR‐Studies), Medline and Lilacs; Cross‐checking references; contact with authors of included studies and manufacturers of passiflora.

Selection criteria

Relevant randomised and quasi‐randomised controlled trials of passiflora using any dose, regime, or method of administration for people with any primary diagnosis of general anxiety disorder, anxiety neurosis, chronic anxiety status or any other mental health disorder in which anxiety is a core symptom (panic disorder, obsessive compulsive disorder, social phobia, agoraphobia, other types of phobia, postraumatic stress disorder). Effectiveness was measured using clinical outcome measures such as Hamilton Anxiety Scale (HAM‐A) and other scales for anxiety symptoms.

Data collection and analysis

Two reviewers independently selected the trials found through the search strategy, extracted data, performed the trial quality analyses and entered data. Where any disagreements occured, the third reviewer was consulted. Methodological quality of the trials included in this review was assessed using the criteria described in the Cochrane Handbook. For dichotomous outcomes, relative risk with 95% confidence intervals (CI) were calculated, and for continuous outcomes, weighted mean difference with 95%CI was used.

Main results

Two studies, with a total of 198 participants, were eligible for inclusion in this review. Based on one study, a lack of difference in the efficacy of benzodiazepines and passiflora was indicated. Dropout rates were similar between the two interventions. Although the findings from one study suggested an improvement in job performance in favour of passiflora (post‐hoc outcome) and one study showed a lower rate of drowsiness as a side effect with passiflora as compared with mexazolam, neither of these findings reached statistical significance.

Authors' conclusions

RCTs examining the effectiveness of passiflora for anxiety are too few in number to permit any conclusions to be drawn. RCTs with larger samples that compare the effectiveness of passiflora with placebo and other types of medication, including antidepressants, are needed.

Plain language summary

Passiflora for the treatment of anxiety disorders in adults

Anxiety is a very common mental health problem in the general population. Passiflora, a herbal medicine, could be an option for treating anxiety if shown to be effective and safe. This review summarised the evidence from currently available studies on passiflora. Only two studies were eligible for inclusion, involving a total of 198 participants. One study showed that passiflora was as effective as benzodiazepines, with similar dropout rates between the two treatments. Given the lack of studies, it is not possible to draw any conclusions on the effectiveness or safety of passiflora in the treatment of anxiety disorders.

Background

Anxiety is a very common mental health disturbance in the general population and in the primary care setting. In the US National Comorbidity Survey Kessler 1994 found a one year prevalence for anxiety disorders of 17% and a lifetime prevalence of almost 25%. Using the Composite International Diagnostic Interview (CIDI) in 1996‐99, Bijl 1998 included 7076 people in 90 municipalities in the Netherlands, and detected a prevalence rate for anxiety of 19.3% in the general population. In Brazil the prevalence of anxiety was reported at 12.1% in Brasilia, 6.9% in São Paulo and 5.4% in Porto Alegre (Almeida‐Filho 1997). One study found a marked reduction in quality of life and psychosocial functioning in people with anxiety disorders (Mendlowicz 2000).

Although anxiety is a treatable disease, it is often not diagnosed and treated properly. The majority of patients suffering from anxiety consult their general practitioner, and often their complaint presents as a physical symptom such as headache, palpitations, breathing difficulties or chest pain (Walley 1994). Anxiety may also be associated with physical disorders, such as diabetes, arthritis and cancer. It may also be the main symptom of a specific psychiatric disorder such as generalized anxiety disorder, postraumatic stress disorder, panic or obsessive compulsive disorder.

Benzodiazepines are effective in short term treatment but their overuse may cause dependence (Priest 1988). Psychotherapy is effective and commonly used to treat anxiety (Taylor 1978, Taylor 1988, Borkovec 1987, Borkovec 1993). A systematic review on psychotherapy for generalized anxiety disorder has been conducted and is expected to be published soon in the Cochrane Library (Hunot 2006). Another systematic review found that cognitive behavioural therapies and pharmacological treatments including buspirone, trazodone, imipramine and ritanserin significantly improved anxiety (Gould 1997). However psychotherapy may be an unrealistic option in public health settings in many countries as it is costly and time consuming, and although pharmacological treatments are effective, they may be limited by their side effects and cost. Herbal medicines are popularly used worldwide and could be an option if shown to be effective and safe. A systematic review on the effectiveness of kava kava for anxiety showed its superiority as compared with placebo (Pittler 2002). There is no systematic review on passionflower for anxiety.

Passiflora (or Passionflower) is a folk remedy used for anxiety. In Brazil it is called 'maracujá' and its juice is used as a popular beverage. The plant is of the Passifloraceae family with about 400 species is native to North, Central and South America. The beautiful flower that blossoms out of the unassuming bud was compared in the 15th century to the Passion of Christ, and named passionflower by the Spanish explorers who came to Peru (Olin 1989).

Most toxicity studies concern Passiflora incarnata. The plant extract contains flavonoids and alkaloids (harman or passiflorin) (Olin 1989). The herbal extract may be found in tablet forms (500 mg), dried herb for oral use or by infusion, liquid extract or tincture (Fisher 2000; Newall 1996). It has been used for many years in America, and has not been associated with acute or chronic toxicity. However, the pharmacological profile of the extracts suggests that large doses may result in central nervous system depression and bradycardia, prolonged QTc and ventricular tachycardia (Fisher 2000). Although it has been used for a long time in treating anxiety disorders, its efficacy and tolerability are not yet established.

Objectives

To investigate the effectiveness and safety of passiflora for treating any anxiety disorders.

Methods

Criteria for considering studies for this review

Types of studies

Relevant randomised and quasi‐randomised controlled trials.

Types of participants

People with any primary diagnosis of anxiety disorders or anxiety neurosis or chronic anxiety status, or any other disease in which anxiety was a core symptom (panic disorder, obsessive compulsive disorder, social phobia, agoraphobia, other types of phobia, postraumatic stress disorder), irrespective of gender, and ethnic background, aged 16 and over.

Studies in which anxiety was a minor symptom of a different disorder (for example depression or any other psychiatric diagnoses) were excluded.

Types of interventions

Passiflora extract of any dose, regime, or method of administration. Trials where passiflora was used in association with another drug were not included.

Comparison: placebo, other drugs, psychotherapy or no intervention.

Types of outcome measures

Primary outcome 
 Effectiveness, measured using clinical outcome measures such as Hamilton Anxiety Scale (HAM‐A) (Hamilton 1959) and other scales for anxiety symptoms: 
 1) improvement in anxiety (continuous outcome) 
 2) absence of treatment response (dichotomous outcome)

Secondary outcome measures 
 1) Acceptability of treatment : 
 ‐ number of participants reporting side effects 
 ‐ number of participants dropping out due to side effects. 
 2) Side effects 
 3) Dropout rates 
 4) Suicide attempts 
 5) Use/misuse of substances 
 6) Use of health services 
 7) Death 
 8) Quality of life

Search methods for identification of studies

Electronic databases 
 The Cochrane Collaboration Depression, Anxiety and Neurosis Cochrane Controlled Trials Register (CCDANCTR‐Studies), Lilacs (2005) and Medline (2005) were searched using standard phrases for 'controlled trials' and for 'anxiety disorders' associated to 'passiflora OR passionflower OR passion flower'. There was no restriction related to language or date of publication of the studies.

Reference lists 
 The references of all identified studies were inspected for additional studies.

Personal communication 
 The first author of each included study was contacted for information regarding unpublished trials and additional data when necessary. The manufacturers of passiflora products and experts in the field were also contacted for information regarding unpublished trials.

Data collection and analysis

Selection of trials 
 Two review authors (LSM and BGOS) independently selected the trials found through the search strategy, extracted the data, assessed trial quality and analysed the results. Where any disagreements occurred, the third review author (ANA) was consulted, and if consensus was not yet reached, data were not included in the review until the author of the trial provided further clarification.

Data extraction 
 Review authors screened the abstracts of all publications obtained by the search strategy, and considered eligible in fulfilling the inclusion criteria. Data concerning participant characteristics, intervention details and outcome measures from the included studies was extracted independently using a standard extraction form with the following items: 
 1) General information: published or not, title, authors, contact address, country, resource, publication idiom, publication year, duplication of publishing, sponsor. 
 2) Characteristics of the study: design, length, whether randomised and method, allocation procedure, blinding (patients, administrator care and outcome appraiser), allocation check. 
 3) Intervention: placebo inclusion, intervention and controls (dose, route of administration and duration). 
 4) Patients: inclusion and exclusion criteria, total number of randomised and analysed subjects, sex, age, basic characteristics, diagnostic criteria, length of disease and similarity of groups on the basic characteristics (including any co‐morbidity), assessment on complications, sub‐groups. 
 5) Outcomes: those specified in the section 'Types of outcomes', any other outcome assessed, other events, extension of attendance, and quality of related outcomes. 
 6) Results: outcomes and evaluation time (including the evaluative measure), where necessary converted to the measurement of the effects specified below; intention to treat analysis.

Quality assessment 
 Methodological quality of the trials included in this review was assessed using the criteria described in the Cochrane Handbook (Clarke 2000). The Cochrane Handbook criteria are based on the evidence of a strong relationship between allocation concealment and potential for bias in the results (Schulz 1995). The categories for these criteria are as follows: 
 A ‐ Low risk of bias (adequate allocation concealment) 
 B ‐ Moderate risk of bias (unclear allocation concealment) 
 C ‐ High risk of bias (inadequate allocation concealment)

For the purpose of the analysis in this review, trials were included if they met the criteria A or B. Additionally, a cut‐off of two points in the Jadad scale was used to check the assessment made by the Cochrane Handbook criteria (Jadad 1996). However, the Jadad scale was not used to exclude trials in this review.

Measures of treatment effect

Dichotomous outcomes 
 Dichotomous outcomes were analysed by calculating relative risks (RR) for each trial with the uncertainty in each result being expressed using 95% confidence intervals (CI) and by combining individual trials in a meta‐analysis. The estimates of RR were based on the random effects model, as it took into account any between study differences (even if there was no statistically significant heterogeneity) and gave the same result as the fixed effects model when there was no between study variance. When overall results were significant, it was planned to calculate the number needed to treat (NNT) or number needed to harm (NNH) to produce one outcome.

Continuous outcomes 
 Where data become available in future updates, continuous outcomes will be analysed according to their difference in mean treatment effects and its standard deviation. When multiple outcome measures are described in a single study, for the purposes of pooling results, a single 'best available' outcome measure will be chosen for each study, according to the authors' specification as the principle outcome or what is reported first. Continuous outcomes of individual studies will be combined by calculating the standardised mean difference.

Dealing with missing data 
 Information on missing data was clarified where possible through contact with the authors.

Assessment of heterogeneity 
 Heterogeneity in the results of the trials was assessed both by inspection of graphical presentations and by calculating a chi‐square test of heterogeneity. Heterogeneity was assumed to be present when the significance level was lower than 0.10 (p < 0.10).

Subgroup analysis and investigation of heterogeneity 
 Where additional trials with data for meta‐analysis are included in future updates and significant heterogeneity is indicated to be present, an attempt will be made to explain the differences based on the clinical characteristics of the included studies. Data from these trials will be included in the meta‐analysis. Potential sources of heterogeneity may include differences in sample characteristics, diagnostic criteria and differences in medication dosage.

Assessment of reporting biases 
 A funnel plot was produced to look for the possibility of publication bias.

Sensitivity analysis 
 Where further studies are included in future updates of the review, sensitivity analyses will be performed based on the exclusion of quasi‐randomised studies.

Results

Description of studies

Results of the search 
 The electronic search identified two relevant references in MEDLINE (1966‐2005), and three in CCCDANCTR‐Studies. No references for relevant studies were found in LILACS. From the five references, three possible relevant papers were selected for further evaluation, because they evaluated the use of passiflora for anxiety. Two trials (Akhondzadeh 2001; Mori 1993) were included, and one was excluded (Bourin 1997).

Manufacturers of passiflora (Marjan, Brasmedica, Infabra, Millex Roux, Biolab Sanus, Hertz, DM Ind Farmaceutica) were contacted to ask for information on unpublished trials, but no further trials were found.

Included studies 
 Mori 1993, a multicentre double blind RCT conducted in Japan, compared mexazolam with passiflora over four weeks. Akhondzadeh 2001, a double blind RCT conducted in Iran, compared oxazolam with passiflora, also over a four‐week period.

Participants: 
 In Akhondzadeh 2001, 36 subjects with generalised anxiety disorder (DSM‐IV), were included, and in Mori 1993, 162 subjects with neurosis/anxiety (no specific diagnostic criteria) were included. The majority of participants in the two trials were female (59%).

Interventions: 
 Passiflora was prescribed in distinct conditions in the two included studies: 45 drops/day in Akhondzadeh 2001 (+ placebo tablet) and a tablet of 90 mg/day in Mori 1993. The distinct formulations of doses used in each study make it difficult to combine data in comparisons of passiflora against control groups.

Control groups: 
 The control groups used were oxazolam 30 mg/day (+ placebo drops) in Akhondzadeh 2001 and mexazolam 1.5 mg/day in Mori 1993.

Excluded studies 
 Bourin 1997 compared a combination of six plant extracts with placebo, with no group receiving passiflora only.

Studies awaiting assessment 
 There are no studies awaiting assessment.

Risk of bias in included studies

Allocation 
 Mori 1993 reported that randomisation was carried out by a 'central committee of randomisation', a method considered as A according to the Cochrane Handbook classification of allocation concealment. Akhondzadeh 2001 did not explain how randomisation was done, and was classified as B.

Blinding 
 Both trials employed a double blind design. Akhondzadeh 2001 used a double‐dummy technique, since passiflora was used as drops and oxazolam as tablets.

Dropout rates 
 The trial by Mori 1993 reported 16 dropouts in the passiflora group (completers=63) and 12 in the mexazolam group (completers=71). The trial by Akhondzadeh 2001 had two dropouts in each group.

Jadad scores 
 The total Jadad score for the study conducted by Mori 1993 was 5, and the total Jadad score for the study by Akhondzadeh 2001 was 3.

Effects of interventions

Two randomised double blind controlled trials included a total of 198 patients assigned to treatment.

01 PASSIFLORA vs BENZODIAZEPINES

Primary outcome

Effectiveness 
 In the trial by Mori 1993, 37% of the passiflora group and 44% of the mexazolam group showed a moderate to marked improvement (RR 1.13, 95%CI 0.85 to 1.49).

Akhondzadeh 2001 presented some outcome data that could not be re‐analysed in this review, showing a more rapid onset of action with oxazolam as compared with passiflora, but no difference detected at the end of the trial. The reduction in HAM‐A scores was similar for both arms.

Secondary outcomes

Acceptability of treatment 
 In the trial by Mori 1993, 81% of the passiflora group and 79% of the mexazolam group reported no side effects (RR 0.90, 95%CI 0.46 to 1.78).

Side effects 
 For types of side effects, a meta‐analysis pooling data from both studies was performed for drowsiness, dizziness and allergic reactions only. Those taking passiflora were less likely to be sedated (5%) than those receiving benzodiazepines (21%), but this finding did not reach statistical significance (RR 0.56, 95%CI 0.20 to 1.59, n=166). Allergic reactions (RR 0.94, 95% CI 0.22 to 4.07, n=166) and dizziness (RR 0.89, 95%CI 0.44 to 1.82, n=166) were equally present between the two interventions. Other adverse events described in the trial by Mori 1993 did not differ significantly between passiflora and mexazolam.

Dropout 
 Combining data from both included studies, no significant difference in dropout rates was found between passiflora and benzodiazepines (RR 1.43, 95%CI 0.77 to 2.69).

Job performance (post‐hoc outcome) 
 Akhondzadeh 2001 reported that impairment of job performance was greater for the oxazolam group (61%) when compared with the passiflora group (44%), but this finding did not reach statistical significance (RR 0.67, 95%CI 0.38 to 1.17).

Other secondary outcomes 
 No data were available for other secondary outcomes, including suicide attempts, use/misuse of substances, use of health services, death and quality of life.

Discussion

Passiflora is available on the market in a range of different preparations, mainly in tablet form (500mg) of the dried herb for oral use or by infusion, as liquid extract or as tincture (Fisher 2000). In addition to variation in preparation, several different manufacturers produce formulations of passiflora, making it even more difficult to compare the efficacy of the distinct preparations.

In this review, two studies, evaluating two distinct formulations, were not able to differentiate passiflora from benzodiazepines for any of the outcome measures. This lack of statistical difference may be interpreted in two ways: firstly, the medications were equally effective or secondly, the absence of difference may explained due to type 2 error (sample sizes not large enough or insufficient number of studies for inclusion).

One important limitation of the present review is the lack of direct comparison between passiflora and antidepressants, which are now the most commonly used medication for anxiety disorders. One systematic review found that trazodone, imipramine or ritanserin were significantly more effective in the treatment of generalized anxiety disorder than placebo (Gould 1997). However, another review that compared antidepressants to benzodiazepines found no significant differences between the two classes (Rickels 1993), suggesting a lack of clarity on this topic.

Authors' conclusions

Implications for practice.

The number of randomised controlled trials examining the effectiveness of passiflora for anxiety disorders are too few to permit clear conclusions to be drawn. Evidence from two small studies show a similar pattern of response for passiflora and benzodiazepines under evaluation in short‐term treatment of anxiety disorders. There are no comparisons of passiflora with the most commonly prescribed medications for these disorders, antidepressants, which limits the ability to draw any practical conclusions for clinical practice.

Implications for research.

RCTs with larger samples, comparing passiflora to other interventions used in anxiety disorders, such as antidepressants, are needed. Furthermore it is important to conduct a proper comparison of passiflora to placebo to establish the primary efficacy and safety of this intervention.

What's new

Date	Event	Description
3 November 2008	Amended	Converted to new review format.

History

Protocol first published: Issue 4, 2003
 Review first published: Issue 1, 2007

Date	Event	Description
13 November 2006	New citation required and conclusions have changed	Substantive amendment

Data and analyses

Comparison 1. Passiflora vs Benzodiazepines.

Outcome or subgroup title	No. of studies	No. of participants	Statistical method	Effect size
1 Lack of moderate/marked improvement (FGIR)	1		Risk Ratio (M‐H, Random, 95% CI)	Subtotals only
1.1 Mexazolam	1	134	Risk Ratio (M‐H, Random, 95% CI)	1.13 [0.85, 1.49]
2 Number of subjects reporting side‐effects	1	134	Risk Ratio (M‐H, Random, 95% CI)	0.90 [0.46, 1.78]
3 Side‐effects	2		Risk Ratio (M‐H, Random, 95% CI)	Subtotals only
3.1 agression	1	32	Risk Ratio (M‐H, Random, 95% CI)	0.0 [0.0, 0.0]
3.2 allergic reaction	2	166	Risk Ratio (M‐H, Random, 95% CI)	0.94 [0.22, 4.07]
3.3 ataxia	1	32	Risk Ratio (M‐H, Random, 95% CI)	0.50 [0.11, 2.35]
3.4 confusion	1	32	Risk Ratio (M‐H, Random, 95% CI)	1.0 [0.30, 3.32]
3.5 disinhibition	1	32	Risk Ratio (M‐H, Random, 95% CI)	0.0 [0.0, 0.0]
3.6 dizziness	2	166	Risk Ratio (M‐H, Random, 95% CI)	0.89 [0.44, 1.82]
3.7 drowsiness	2	166	Risk Ratio (M‐H, Random, 95% CI)	0.56 [0.20, 1.59]
3.8 dyspnea	1	32	Risk Ratio (M‐H, Random, 95% CI)	0.2 [0.01, 3.86]
3.9 hyporeflexia	1	32	Risk Ratio (M‐H, Random, 95% CI)	0.33 [0.01, 7.62]
3.10 respiratory depression	1	32	Risk Ratio (M‐H, Random, 95% CI)	0.14 [0.01, 2.56]
3.11 slurred speech	1	32	Risk Ratio (M‐H, Random, 95% CI)	0.0 [0.0, 0.0]
3.12 abdominal fullness	1	134	Risk Ratio (M‐H, Random, 95% CI)	0.38 [0.02, 9.04]
3.13 abnormality of salivation	1	134	Risk Ratio (M‐H, Random, 95% CI)	3.38 [0.14, 81.39]
3.14 anorexia	1	134	Risk Ratio (M‐H, Random, 95% CI)	5.63 [0.28, 114.99]
3.15 blurred vision	1	134	Risk Ratio (M‐H, Random, 95% CI)	3.38 [0.14, 81.39]
3.16 chest discomfort	1	134	Risk Ratio (M‐H, Random, 95% CI)	3.38 [0.14, 81.39]
3.17 constipation	1	134	Risk Ratio (M‐H, Random, 95% CI)	0.38 [0.02, 9.04]
3.18 diarrhea	1	134	Risk Ratio (M‐H, Random, 95% CI)	5.63 [0.28, 114.99]
3.19 discouragement	1	134	Risk Ratio (M‐H, Random, 95% CI)	0.38 [0.02, 9.04]
3.20 fatigue	1	134	Risk Ratio (M‐H, Random, 95% CI)	5.63 [0.28, 114.99]
3.21 headache	1	134	Risk Ratio (M‐H, Random, 95% CI)	0.38 [0.02, 9.04]
3.22 heart pressure	1	134	Risk Ratio (M‐H, Random, 95% CI)	3.38 [0.14, 81.39]
3.23 inarticulateness	1	134	Risk Ratio (M‐H, Random, 95% CI)	0.38 [0.02, 9.04]
3.24 light head feeling	1	134	Risk Ratio (M‐H, Random, 95% CI)	0.23 [0.01, 4.60]
3.25 nausea	1	134	Risk Ratio (M‐H, Random, 95% CI)	0.38 [0.02, 9.04]
3.26 numbness	1	134	Risk Ratio (M‐H, Random, 95% CI)	1.13 [0.07, 17.65]
3.27 stomach ache	1	134	Risk Ratio (M‐H, Random, 95% CI)	1.13 [0.07, 17.65]
3.28 thirst	1	134	Risk Ratio (M‐H, Random, 95% CI)	0.38 [0.02, 9.04]
3.29 weakness	1	134	Risk Ratio (M‐H, Random, 95% CI)	3.38 [0.14, 81.39]
4 Dropout	2	170	Risk Ratio (M‐H, Random, 95% CI)	1.43 [0.77, 2.69]
4.1 Oxazepam	1	36	Risk Ratio (M‐H, Random, 95% CI)	1.0 [0.16, 6.35]
4.2 Mexazolam	1	134	Risk Ratio (M‐H, Random, 95% CI)	1.50 [0.77, 2.93]
5 Impairment of job performance	1	32	Risk Ratio (M‐H, Random, 95% CI)	0.67 [0.38, 1.17]

1.1. Analysis.

Comparison 1 Passiflora vs Benzodiazepines, Outcome 1 Lack of moderate/marked improvement (FGIR).

1.2. Analysis.

Comparison 1 Passiflora vs Benzodiazepines, Outcome 2 Number of subjects reporting side‐effects.

1.3. Analysis.

Comparison 1 Passiflora vs Benzodiazepines, Outcome 3 Side‐effects.

1.4. Analysis.

Comparison 1 Passiflora vs Benzodiazepines, Outcome 4 Dropout.

1.5. Analysis.

Comparison 1 Passiflora vs Benzodiazepines, Outcome 5 Impairment of job performance.

Characteristics of studies

Characteristics of included studies [ordered by study ID]

Akhondzadeh 2001.

Methods	RCT double blind 4 weeks non‐ITT Iran
Participants	GAD (DSM‐IV) N=36 20 female age range=19‐47 outpatient
Interventions	1. Passiflora 45 drops/day + placebo tablet n=18 2. Oxazolam tablet 30 mg + placebo drops n=18
Outcomes	HAM‐A Side effects Impairment of job performance Dropout
Notes	Data from HAM‐A presented on graphic
Risk of bias
Bias	Authors' judgement	Support for judgement
Allocation concealment?	Unclear risk	B ‐ Unclear

Mori 1993.

Methods	RCT (randomisation by a central commitee) double blind 4 weeks multicentre, Japan
Participants	Neurosis/ Anxiety N=162 79 female age range=15‐70
Interventions	1. Passiflora 90 mg n=63 2. Mexazolam 1.5 mg n=71
Outcomes	FGIR OSR Dropout
Notes
Risk of bias
Bias	Authors' judgement	Support for judgement
Allocation concealment?	Low risk	A ‐ Adequate

BZD=Benzodiazepines 
 FGIR=Final Global Improvement Rate 
 GAD=Generalised Anxiety Disorder 
 GUR=Global utility Rate 
 HAM‐A=Hamilton Anxiety 
 OSR=Overall Safety Rate

Characteristics of excluded studies [ordered by study ID]

Study	Reason for exclusion
Bourin 1997	combination of plant extract multicenter double blind placebo controlled adjustment disorder with anxious mood 28 days HAM‐A

Contributions of authors

LSM ‐ protocol writing, searching, trial selection, data extraction, completion of report

ANA ‐ protocol writing, expertise advice, completion of report

BGOS ‐ protocol writing, searching, trial selection, data extraction, completion of report

Sources of support

Internal sources

• Brazilian Cochrane Centre, Brazil.

External sources

• No sources of support supplied

Declarations of interest

None known.

Edited (no change to conclusions)