SEARCH

SEARCH BY CITATION

Keywords:

  • Adolescent;
  • Anxiety Disorders [*drug therapy];
  • Randomized Controlled Trials as Topic;
  • Serotonin Uptake Inhibitors [*therapeutic use];
  • Child;
  • Humans

Abstract

  1. Top of page
  2. Abstract
  3. Background
  4. Objectives
  5. Methods
  6. Results
  7. Discussion
  8. Authors' conclusions
  9. Characteristics of studies
  10. Data and analyses
  11. Appendices
  12. What's new
  13. History
  14. Contributions of authors
  15. Declarations of interest
  16. Sources of support
  17. Differences between protocol and review
  18. Acknowledgements
  19. REFERENCES

Background

Anxiety disorders are a potentially disabling group of disorders which are prevalent in childhood and adolescence. The recognition of the early onset of anxiety disorders, and their successful treatment with medication in adults, has led to the growing interest in using medication for paediatric anxiety disorders.

Objectives

To assess the efficacy and tolerability of medication for treating paediatric anxiety disorders.

Search strategy

We searched the Cochrane Depression, Anxiety & Neurosis Group specialised register (CCDANCTR-Studies), MEDLINE (via PubMed 1966 to August 2008), EMBASE (1966 to August 2008), and PsycINFO (1972 to August 2008). Various electronic registers were searched for unpublished studies. Reference lists of retrieved articles were searched for additional studies.

Selection criteria

All randomised controlled trials (RCTs) of pharmacotherapy in childhood/adolescent anxiety disorders.

Data collection and analysis

Two raters independently assessed RCTs for inclusion in the review, collated trial data, and assessed trial quality. Investigators were contacted to obtain missing data. Summary statistics were stratified by medication class, and by medication agent for the selective serotonin reuptake inhibitors (SSRIs). Dichotomous and continuous measures were calculated using a random effects model, heterogeneity was assessed, and subgroup/sensitivity analyses were undertaken.

Main results

22 short-term (<= 16 weeks) RCTs were included in the analysis (2519 participants). The majority of the trials assessed the efficacy of the SSRIs (N = 15).

Medication and placebo response occurred in 58.1% and 31.5% of patients, respectively (Number of studies (N) = 14, Number needed to treat (NNT) = 4). Medication was more effective than placebo in reducing overall symptom severity in OCD in a post-hoc comparison (N = 7, Weighted Mean Difference (WMD) = -4.45, 95%CI = -5.94, -2.97, n = 765). Medication was less well tolerated than placebo overall, though the absolute proportion of participants who withdrew due to drug-related adverse events was low (4.9%).

Authors' conclusions

Medication treatments can be effective in paediatric anxiety disorders, acting to reduce core symptoms, and should be considered as part of the treatment of these disorders. The greatest number of trials showing efficacy to date have assessed the SSRIs in treating paediatric OCD.

There is no clear evidence to show that any particular class of medication is more effective or better tolerated than any other. As quantitative data was only available for the SSRIs and venlafaxine the routine use of benzodiazepines cannot be recommended, especially given concerns of dependency and treatment -related emergent adverse events associated with this class of drugs.

Future RCTs could help identify potential clinical moderators of treatment efficacy. Studies of the long-term efficacy of medication treatment, optimal dosage, as well as direct comparisons of pharmacotherapy and psychotherapy are also warranted.

Plain Language Summary

A systematic review and meta-analysis of randomised controlled trials of medication in treating anxiety disorders in children and adolescents

Anxiety disorders are a potentially disabling group of disorders which frequently occur in childhood and adolescence. Increasing recognition of the early onset of anxiety disorders and of the effectiveness of medication in treating adult anxiety disorders has contributed to a growing interest in the use of medication in treating paediatric patients. This systematic review of randomised controlled trials (RCTs) of pharmacotherapy of anxiety disorders in children and adolescents identified 22 short-term (<= 16 weeks) randomised controlled trials which were eligible for inclusion (2519 participants). Treatment response was significantly greater after treatment with medication (58.1%) than with placebo (31.5%) in 14 trials. Medication was more effective than placebo in reducing overall symptom severity across all of the anxiety disorders (number of studies (N) = 9). The greatest number of trials were for obsessive compulsive disorder (OCD), for which treatment efficacy in reducing symptom severity was also observed . The greatest number of trials showing efficacy to date have used the selective serotonin reuptake inhibitors (SSRIs). No controlled evidence could be found for the effectiveness of benzodiazepines, despite their continued prescription for paediatric anxiety disorders. Medication was less well tolerated than placebo, as indicated by the significant proportion of children and adolescents who dropped out due to adverse effects during the short term trials. Furthermore, while few incidences of suicidal behaviour/ideation in the included trials were attributed to study medication, it is important to be aware of the need for careful monitoring after initiation of SSRIs in treating this population. In conclusion, medication should be considered as part of the treatment of paediatric anxiety disorders over the short-term. Additional research into the optimal dose and duration of medication treatment, as well as the effects of age on the efficacy and tolerability of medication is warranted.


Background

  1. Top of page
  2. Abstract
  3. Background
  4. Objectives
  5. Methods
  6. Results
  7. Discussion
  8. Authors' conclusions
  9. Characteristics of studies
  10. Data and analyses
  11. Appendices
  12. What's new
  13. History
  14. Contributions of authors
  15. Declarations of interest
  16. Sources of support
  17. Differences between protocol and review
  18. Acknowledgements
  19. REFERENCES

Description of the condition

Childhood/adolescent anxiety disorders include both those traditionally associated with childhood (separation anxiety disorder and the DSM-III (APA 1987) diagnoses of overanxious disorder and avoidant disorder of childhood) and those traditionally seen as adult disorders (panic disorder with or without agoraphobia, generalised anxiety disorder (GAD), social phobia, specific phobia (SP), obsessive-compulsive disorder (OCD), and post-traumatic and acute stress disorders). These disorders have increasingly been recognised as frequently having their onset during childhood or adolescence (Kessler 2007). Prevalence in the paediatric community has been reported as lying between 4% and 7% when assessed using DSM-IV criteria (adjusted for impairment), with the cumulative incidence of anxiety disorders as high as 10% by 16 years of age (Canino 2004; Ford 2003).

Anxiety disorders in childhood and adolescence tend to be chronic if left untreated, and are potentially disabling (Keller 1992, Woodward 2001). Research has shown that the long term outcome of young people with anxiety disorders is poorer in terms of physical and mental health, quality of life and achievement than that of children without such disorders (Woodward 2001). Early onset of these disorders may impair social and personal developmental processes, thereby contributing towards the finding that adults with anxiety disorders have poorer subjective views of their physical and mental health, social relationships, occupation, and home and family life (Olatunji 2007).

Description of the intervention

Effective management of early onset anxiety disorders may improve outcome, thus it is important that guidelines for treatment be available.

Traditional approaches to anxiety disorders in childhood were predominantly psychological. Few controlled studies assessing the efficacy of such treatment modalities are available (Target 1994), but studies of cognitive-behavioural strategies in adolescent anxiety disorders are encouraging (Kendall 1997; Flannery 2000). The results of a Cochrane review of cognitive behavioural therapy (CBT) for paediatric anxiety disorders (James 2005), as well as behavioural therapy and CBT for OCD in children and adolescents (O'Kearney 2006) also emphasise a role for this treatment modality. Nevertheless, the increasing elucidation of the biological substrate of anxiety disorders and the discovery of effective pharmacological treatments for adults argues for investigating the efficacy and tolerability of pharmacotherapy for paediatric anxiety disorders.

Reports of the use of pharmacotherapy in childhood anxiety disorders began to appear in the early 1960s (D'Amato 1962). With the growing tendency to use psychotropic medications in younger patients (Olfson 2002), it seems imperative that the efficacy of such strategies be evaluated. For instance, a review of medical claims in the US discovered a 2 to 3- fold increase in the use of psychotropic agents amongst children and adolescents between 1987 and 1996, with antidepressants ranking as the second most utilized medication class (Zito 2003). Neverthess, for many reasons, including failure to recognise the extent of mental health problems in children and adolescents (Sayal 2006), and ethical problems surrounding the use of newly developed pharmacotherapeutic agents in this population group, child psychiatry has lagged behind in the development of biologically based treatments.

The perception of paediatric anxiety disorders as the initial manifestation of the same disorders found in adults (APA 1994; Hawkridge 2000; Kessler 2007) is consistent with expert consensus that selective serotonin reuptake inhibitors (SSRIs) should be regarded as first line agents in treating anxiety disorders in both children/adolescents and adults. Sertraline, fluoxetine and fluvoxamine have all been approved by the FDA in treating paediatric obsessive compulsive disorder (OCD) (Flament 2007). The SSRIs have also been compared favourably to other medication agents in treating paediatric anxiety disorders (Biederman 2004). They typically give rise to a smaller number of side effects, and a longer treatment effect than the tricyclic antidepressants (TCAs), and do not possess the risk of dependency associated with the benzodiazepines. Nevertheless, the necessity of a systematic comparison of the risk/benefit profiles of the different medication classes in treating paediatric anxiety disorders is highlighted by concerns that the benefits of certain SSRIs in treating depression do not outweigh the risks associated with their use amongst patients under 18 years of age (MHRA 2003).

Why it is important to do this review

Particular problems have been associated in comparing the responses to different medications in the available studies in this area. From the time of the early studies of Gittelman-Klein et al (Gittelman-Klein 1971) to the large randomised controlled trials (RCTs) of venlafaxine extended release in treating paediatric generalized anxiety disorder (March 2007; Rynn 2007), nosology in child psychiatry has changed significantly, making comparisons of early and late trials difficult. This has been compounded by clinical and methodological variation between trials. Reviews have generally concluded that while there is evidence that some pharmacotherapeutic agents are effective in some paediatric onset anxiety disorders, there is a need for further large, rigorous controlled trials in many areas (e.g.,Greenhill 1998, Hawkridge 2000; Geller 2003). A systematic review would provide guidelines for practical management of these disorders and help identify areas and disorders in which research should be focused.

Objectives

  1. Top of page
  2. Abstract
  3. Background
  4. Objectives
  5. Methods
  6. Results
  7. Discussion
  8. Authors' conclusions
  9. Characteristics of studies
  10. Data and analyses
  11. Appendices
  12. What's new
  13. History
  14. Contributions of authors
  15. Declarations of interest
  16. Sources of support
  17. Differences between protocol and review
  18. Acknowledgements
  19. REFERENCES

1) To compare the efficacy of different medication classes in treating different childhood/adolescent anxiety disorders, relative to a placebo control.

2) To assess whether treatment effects are influenced by methodological characteristics of the trials or the clinical characteristics of the participants included in the trials.

3) To document and describe the drug-related adverse events associated with each agent.

4) To provide guidance for practice by describing RCT trials which are eligible for inclusion in the review, but could not be quantitatively analysed. This includes RCT trials which are excluded from the meta-analyses due to missing data and/or the use of unreliable measurement scales.

Methods

  1. Top of page
  2. Abstract
  3. Background
  4. Objectives
  5. Methods
  6. Results
  7. Discussion
  8. Authors' conclusions
  9. Characteristics of studies
  10. Data and analyses
  11. Appendices
  12. What's new
  13. History
  14. Contributions of authors
  15. Declarations of interest
  16. Sources of support
  17. Differences between protocol and review
  18. Acknowledgements
  19. REFERENCES

Criteria for considering studies for this review

Types of studies

All randomised controlled trials, published and unpublished, were considered for inclusion.Studies employing cross-over as well as parallel - group designs were potentially eligible, as were trials employing cluster randomised designs. Differences between trials (for example, sample size, trial duration, trial quality) were not used to exclude studies. The availability of outcome data was not an inclusion criteria for trials in this review.

Types of participants

All studies of children and/or adolescents diagnosed with anxiety disorders according to DSM-III (APA 1980), DSM-III-R (APA 1987) and DSM-IV (APA 1994) diagnostic criteria were included.

These included studies of:

  • anxiety disorders associated with childhood/adolescence, such as separation anxiety disorder and the DSM-III (APA 1987) diagnoses of overanxious disorder and avoidant disorder of childhood. Treatment studies of selective mutism, in which children refuse to speak in certain specific social contexts, has frequently been conceptualised as a form of SP (Chavira 2007), and were therefore also included.

  • anxiety disorders traditionally regarded as adult disorders -- panic disorder with or without agoraphobia, GAD, social phobia, SP, OCD, and post-traumatic and acute stress disorders.

Studies concerning anxiety symptoms that are part of the clinical presentation of another primary diagnosis (e.g., attention deficit/hyperactivity disorder (ADHD), disruptive behaviour disorders, obsessive-compulsive behaviour in autistic disorder) were excluded. Comorbid psychiatric diagnoses were not grounds for exclusion, on the condition that these disorders were considered by the authors to be secondary to the anxiety disorder being treated. Concurrent treatment of these secondary disorders with psychotherapy or medication would constitute grounds for exclusion, however.

Age of participants was restricted to 18 years and under, as this is the most frequently used age limit in controlled trials. Where studies included both the target population and older participants, attempts were made to extract the relevant data, on the condition that randomised age blocks were used.

Types of interventions

The following interventions will be compared:

  • medication versus placebo control

Separate categories were created for trials which included agents falling within one of the three major medication classes (benzodiazepines, SSRIs and TCAs), while other agents were placed under the generic category of "Other medications". This included medications such as buspirone, venlafaxine, nefazodone, clonidine, and other beta-blocker and antipsychotic medications. Where available, placebo-controlled dose comparison trials were also included.

Types of outcome measures
Primary outcomes

Treatment response (number of responders) was determined from:

1) The improvement item of the Clinical Global Impressions scale (CGI-I) (Guy 1976). The CGI-I is a valid clinician rated seven-point categorical scale that has demonstrated validity in predicting improvement on "gold standard" symptom severity scales for anxiety disorders and MDD (Bandelow 2006). Treatment response was defined in this review as a score of 1 (very much improved) or 2 (much improved) on the CGI-I, in line with common practice.

2) In trials in which the CGI-I was not used, summary statistics from closely related measures (such as the Clinical Global Impressions scale for Social Phobia (CGI-SP)) were used. In these cases, the reviewers adopted the criteria for treatment response employed by the trial investigators.

Anxiety symptom and symptom cluster response were determined from established clinician-rated DSM-based anxiety scales such as the:

  • Paediatric Anxiety Rating Scale (PARS) (RUPPASG 2001)

  • the Child Yale-Brown Obsessive-Compulsive Scale (CY-BOCS) (Scahill 1997).

Secondary outcomes

1) The response of comorbid symptoms to medication was measured by scales such as the Childhood Depression Rating Scale (Poznanski 1984).

2) Quality of life measures, as well as measures of functional disability (e.g., social functioning, school attendance or dropout records) were also included when provided.

3.1) Medication acceptability was determined by the proportion of participants who withdrew from the RCTs due to treatment emergent adverse events. This surrogate measure was included in the absence of other more direct indicators of acceptability.

3.2) Significant differences between the medication and control groups in the prevalence of the most common drug-related adverse events (defined as occurring in at least 10% of the participants given medication) was tabulated and described as part of a narrative review. The emergence of adverse events during long-term treatment will also be described.

4.1) Treatment response over the longer-term on the primary outcome measures was described as part of a narrative review.

4.2) A comparison of the number of participants who relapsed (as defined by the trial investigators )and a descriptive review of adverse events following long-term maintenance and treatment discontinuation were also conducted.

Search methods for identification of studies

Electronic searches

1) The CCDANCTR-Studies and CCDANCTR-References databases were searched in August 2008 using the following search terms:

Age-Group = Child or Adolescent

and

Diagnosis = Anxiety or Panic or Phobic or Obsessive or Compulsive or "Post-Traumatic"

and

Intervention = Placebos

2) Literature searches were conducted using the electronic databases MEDLINE (via PubMed, 1966 - August 2008), EMBASE (1966-August 2008) and PsycINFO (1972-August 2008). The MEDLINE search query was derived from the search strategy for RCTs developed by Robinson and Dickersin (Robinson 2002) (see Appendix 1 for the search strategies for all three databases).

3) Ongoing trials were located using the metaRegister of Controlled Trials database (mRCT)( http://www.controlled-trials.com) as well as the National Institute of Health's Computer Retrieval of Information on Scientific Projects (CRISP) service (1972-2008). The search terms "(child* OR adolesc*) AND anxiety" and "(anxiety AND children)" were entered into the search interfaces for these databases, respectively.

4) Unpublished studies were located through browsing through trials listed in the online clinicalstudyresults.org database (http://www.clinicalstudyresults.org). All of the trials funded by the pharmaceutical companies identified as sponsors of RCTs in this review were scanned. Additional unpublished trials were retrieved through searching the Dissertational Abstracts International database - a database of unpublished dissertations included as part of PsycINFO.

Searches were not limited to any particular language. An initial broad strategy was undertaken to find not only RCTs, but also open-label trials as well as journal and chapter reviews of the pharmacotherapy of anxiety disorders in children and adolescents.

Searching other resources
Reference lists

The reference lists of the retrieved articles were checked for additional trials.

Correspondence

1) Attempts were made to obtain published and unpublished trials from key researchers, as identified by the frequency with which they are cited in the bibliographies of RCTs and open-label studies.

2) Pharmaceutical companies were also contacted. They were identified through the source of funding cited in published RCTs, as well as the companies with which the authors were affiliated.

Data collection and analysis

Selection of studies

Studies for which additional information is required in order to determine their suitability for inclusion in the review have been listed in the "studies awaiting assessment" table in the Review Manager (RevMan) software, pending the availability of this information.

Data extraction and management

RCTs which have been identified were independently assessed for inclusion by two raters, based on information included in the abstract and/or main body of the trial report. The raters also independently collated data from RCTs which they both regarded as satisfying the inclusion criteria specified for this review. Any disagreements in the independent trial assessment and data collation procedures were resolved by discussion with a third reviewer.

Spreadsheet forms were designed for the purpose of recording descriptive information, as well as the summary statistics of the outcome measures, the quality scale ratings, and associated commentary. These data were subsequently exported to RevMan.

The following information was collated from each trial:

(a) Description of the trials, including whether it was industry funded, primary researcher and year of publication

(b) Characteristics of the interventions, including the number of participants randomised to the medication and control groups, the number of total drop-outs per group, as well as those who dropped out due to adverse effects, the dose of medication and the period over which it was administered, the name of the medication, and the class (SSRIs, TCAs, benzodiazepines, and "Other medication") to which it belongs.

(c) Characteristics of trial methodology, including the diagnostic (eg. DSM-IV (APA 1994)) criteria employed, the screening instrument used (eg. the Structured Clinical Interview for DSM-IV (SCID) (Spitzer 1996)) for both the primary diagnosis and comorbidity, the use of a placebo run-in and a minimal severity criterion, the number of centres involved, and the trial's methodological quality on the CCDAN Quality of Research Scale.

(d) Characteristics of participants, including their sex distribution, mean age, age range, the total number of children and adolescents in the sample (as defined in the trial), the mean length of time that the participants have been diagnosed with the disorder, whether they have previously been exposed to medication belonging to the same class as the current agent (treatment naivety), the number of participants in the sample with MDD, and the baseline severity of the anxiety disorder, as assessed by the trial's primary outcome measure or another commonly employed scale.

(e) Outcome measures employed, and summary continuous (baseline and endpoint means and standard deviations, or change scores) and dichotomous (number of responders) data. Additional information was included, such as whether these data reflected the intent-to-treat (ITT) with last observation carried forward (LOCF) or completer/observed cases (OC) sample. Other methods of estimating the outcome for participants who dropped out of the study, such as the mixed effects (ME) model, were also recorded.

(f) Quality assessment, including the number of randomised participants who were not included in the analysis (lost to follow-up (LTF)), whether blinding occurred for the assessor/s, patients, or those who administered medication, as well as whether the allocation of medication was randomised and the allocation sequence concealed (the methods used in implementing these respective bias reduction measures were also documented).

Assessment of risk of bias in included studies
Assessment of methodological quality of included studies

The quality of the trials included in this review was independently assessed by two raters, using the CCDAN Quality of Research Scale (CCDAN-QRS) (Moncrieff 2001). This 23 item scale assesses a range of features such as sample size, the duration of the intervention, inclusion and exclusion criteria, and whether or not the power of the trial to detect a treatment effect was calculated. Disagreements between raters which could not be resolved were referred to a third reviewer for discussion (DS). In addition, data for other trial characteristics which have been recognised as potential sources of bias, such as the method used in generating the allocation sequence, how allocation was concealed, whether outcome assessment was blinded, and the number of participants who where lost to follow up, were independently determined by two raters, as part of the data collation process. This is regarded as necessary given doubts concerning the usefulness of an overall quality score from a scale composed of multiple items (Alderson 2003).

Risk of bias assessment

There is overlap between the quality items assessed using the procedure employed in this review and the domains assessed as part of the Cochrane Risk of Bias tool (Higgins 2008). The Risk of Bias tool covers a total of six domains (sequence generation, allocation concealment, blinding of participants, personnel and assessors, incomplete outcome data, selective outcome reporting and other sources of bias), and is now recommended for assessing risk of bias in studies included in Cochrane reviews. In future updates of this review, studies will be assessed for risk of bias using all domains of this tool.

Measures of treatment effect
Categorical data

Relative risk (RR) of response to treatment was used as the summary statistic for the dichotomous outcome of interest (CGI-I). Relative risk ratios was used instead of odds ratios, as odd ratios are less easily interpreted. Odds ratios also tend to overestimate the size of the treatment effect when confused with relative risks. This is especially the case when the occurrence of the outcome of interest is common (as anticipated in this review, with an expected response greater than 20%) (Deeks 2008a).

The number needed to treat (NNT) was also calculated. The NNT is interpreted as the number of patients who require treatment before a single patient responds to medication, relative to a control. It was calculated using the risk ratio as per recommendations in the Cochrane Handbook (Schünemann 2008).

Pearson's chi-square tests were used to detect differences between the comparison groups in side effect prevalence (using the R statistical software (R 2005)), regardless of whether the results of significance tests were reported. A probability statistic less than 0.10 was interpreted as evidence of a difference in the frequency of a particular drug-related adverse event.

Continuous data

Weighted mean difference (WMD) scores were calculated for comparisons of continuous data from the same scale, while standardised mean differences (SMD) were determined for those comparisons which employed a range of scales, such as the comorbidity comparisons (Deeks 2008a). This method of analysis standardises the differences between the means of the treatment and control groups in terms of the variability observed in the trial.

Unit of analysis issues
Cross-over trials

Cross-over trials were only included in the calculation of summary statistics when it was (a) possible to extract medication and placebo/comparator data from the first treatment period, or (b) when the inclusion of these data from both treatment periods was justified through a wash-out period of sufficient duration as to minimise the risk of carry-over effects. The minimum wash-out period required was assessed on the basis of the plasma half-life of the particular agent, as determined by consulting the Lundbeck Psychotropics website (Lundbeck 2003). This period was adjusted for individual agents according to evidence from the literature for faster metabolism of psychotropic agents in children (Soldin 2000).

Dealing with missing data

All analyses of dichotomous data were intention-to-treat (ITT). The total number of participants randomised to the different comparison groups were used as the denominator in comparisons of treatment response. Only data from trials which provide information on the original group size (prior to drop-outs) were included in the analyses of treatment response. In trial reports in which multiple forms of data imputation were conducted, preference was given to the inclusion of summary statistics for continuous outcome measures derived from mixed effects models (ME), followed by last observation carried forward (LOCF) and observed cases (OC) summary statistics (in that order). This is in line with evidence that ME methods are more robust to bias than LOCF analyses (Verbeke 2000).

The reviewers contacted investigators by email in an attempt to obtain missing information

Assessment of heterogeneity

Heterogeneity of treatment response, that is whether the differences among the results of trials were greater than would be expected by chance alone, was assessed by means of the chi-squared test of heterogeneity. A significance level of less than 0.10 was interpreted as evidence of heterogeneity, given the low power of the chi-square statistic when the number of trials is small (Higgins 2008). In addition, the I-squared heterogeneity statistic reported by RevMan was used to quantify the inconsistency of the trial results within each comparison (Higgins 2003).

Differences in treatment response between subgroups were assessed by means of Deeks' stratified test of heterogeneity (2001). This method subtracts the sum of the chi-squared statistics available for each of the subgroups in the study from the chi-statistic available for all the trials, to provide a measure (Qb) of heterogeneity between subgroups. Differences in treatment response on the CGI-I was determined by whether the confidence intervals for the effect sizes of the subgroups overlap. This method was chosen in preference to the stratified test, due to inaccuracies in the calculation in RevMan of the chi-square statistic for dichotomous measures (Deeks 2008).

Assessment of reporting biases

Publication is not necessarily related to study quality and indeed publication may imply certain biases (Dickersin 1992; Song 2000 ). The possibility that the overall treatment effect obtained in this review was biased by selective publication of small studies with positive results was assessed by visual inspection of a funnel plot of treatment response.

Data synthesis

Summary statistics for categorical and continuous measures were obtained from a random effects model (Higgins 2008). The summary statistics were expressed in terms of the size and precision of the average treatment effect. Data from different scales were transformed prior to synthesis so that higher scores always represented improved outcomes for the treatment group.

Subgroup analysis and investigation of heterogeneity

A priori subgroup analyses (Thompson 1994) were undertaken in order to assess the degree to which clinical and methodological differences between trials systematically influence differences observed in the primary treatment outcomes.

The trials were grouped according to the following potential methodological sources of heterogeneity:

  • the involvement of participants from a single centre or multiple centres. Single centre trials are more likely to be associated with lower sample size but less variability in clinician ratings.

  • whether or not trials were industry funded. In general, published trials which are sponsored by pharmaceutical companies appear more likely to report positive findings than trials which are not supported by for-profit companies (Als-Nielsen 2003; Baker 2003).

The following clinical sources of heterogeneity between trials were also assessed:

  • the presence or absence of major depression. This will assist in determining the extent to which the efficacy of a medication agent in treating paediatric anxiety disorders is independent of its ability to reduce symptoms of depression, an important consideration given the classification of many of these medications as antidepressants.

  • the age classification of participants. There is evidence that age-related differences in metabolism, and consequent differences in the rate at which medications are broken down in the body, may affect the efficacy and tolerability of pharmacotherapy (Simeon 1989). In RCTs which contain both children and adolescents (as defined in the trial reports) the outcomes for these 2 age bands will be calculated separately, to determine whether response to medication is affected by the onset of puberty.

In recognition of the possibility of differential effects for different types of medication, all of the comparisons were stratified by medication class. Medications which could not be classified as either benzodiazepines, SSRIs, or TCAs were placed in a separate category, labelled "other medication". Post-hoc comparisons of treatment response and symptom severity were conducted for trials of OCD, as well as for a group of anxiety disorders comprising GAD, Separation Anxiety Disorder (SAD), and SP. The decision to conduct these analyses was justified by the large number of controlled trials of pharmacotherapy for paediatric OCD, while GAD, SAD and SP are highly comorbid in children and adolescents and are often treated together in the intervention literature (Birmaher 2003; RUPPASG 2001) .

Sensitivity analysis

Sensitivity analyses determine the robustness of the reviewer's conclusion to methodological assumptions made in conducting the meta-analysis. The planned comparison of the use of treatment response versus non-response as a summary statistic was not performed. Although this comparison is regarded as necessary given concerns that treatment response may result in less consistent summary statistics than non-response (Deeks 2002), empirical evidence that this is a danger only extends to trials in which the control group event rate is high. As this was not the case for the majority of the trials included in this review, the reviewers decided that such an analysis was not necessary.

A "worst case/best case" analysis was conducted to determine the influence of the exclusion of participants who were lost to follow up (LTF) on the findings of treatment efficacy (Higgins 2008). In the worst case, all the missing data for the treatment group were recorded as non-responders, whereas in the best case, all the missing data in the control group were treated as non-responders. Should the conclusions regarding treatment efficacy not differ between these two comparisons, it can be assumed that missing data in trial reports did not have a significant influence on outcome.

Results

  1. Top of page
  2. Abstract
  3. Background
  4. Objectives
  5. Methods
  6. Results
  7. Discussion
  8. Authors' conclusions
  9. Characteristics of studies
  10. Data and analyses
  11. Appendices
  12. What's new
  13. History
  14. Contributions of authors
  15. Declarations of interest
  16. Sources of support
  17. Differences between protocol and review
  18. Acknowledgements
  19. REFERENCES

Description of studies

See: Characteristics of included studies; Characteristics of excluded studies; Characteristics of studies awaiting classification; Characteristics of ongoing studies.

Results of the search

The search of the CCDAN Controlled Trials Registry yielded 16 results. The PubMed, PsycINFO and EMBASE searches retrieved a total of 472, 161 and 746 references, respectively. A total of 210 abstracts were screened (PubMed: 60, PsycINFO:29, EMBASE: 121). Two additional eligible RCTS (Graae 1994; Geller 2001b) were identified through searching reference lists, and one trial was identified by an anonymous reviewer (Walkup 2008). The reasons for excluding trials from the review are provided in the "Characterisitcs of excluded studies" table, and summarised below. In searching for ongoing, unpublished trials, the mRCT database provided 655 studies, and the CRISP database 193.

Included studies

A total of 22 short term (<= 16 weeks) randomised controlled trials (2519 participants) were included in this review (see the "Characteristics of included studies" table for additional information). All of the publications were reported in English.

Design

Parallel-group designs were employed in the majority of the studies, with a crossover design employed in four of the trials (Flament 1985; Graae 1994; Riddle 1992; Rapoport 1980).

Sample sizes

Samples ranged in size between 15 (Graae 1994) and 322 (Wagner 2004) (mean = 110; standard deviation = 103). Seven of the 8 trials taking place at single centres employed samples of 30 or fewer participants, while an average of 165 participants were included in the 14 multi-centre trials (range: 2 to 48 centres).

Setting

The majority of the trials recruited participants exclusively through centres based in the US (17) or Canada (1), with multicultural studies including centres based in Canada (3), Belgium (1) or South Africa (1).

Participants

The mean age of patients in the included studies was 12 years. Male and female participants were almost equally represented across these trials (52.1% and 47.9%, respectively).

Interventions

Of the ten trials which contained add-on components, one included a maintenance phase (Liebowitz 2002), one a discontinuation phase (POTS 2004), four open-label extensions (DeVeaugh-Geiss 1992; March 1998; Rynn 2007; Walkup 2008), and four follow-up assessments (Beidel 2007; Geller 2004; Simeon 1992; Wagner 2004). Outcome statistics from these trials include data from a randomised controlled discontinuation extension of an open-label trial of paroxetine for obsessive compulsive disorder (Geller 2001b). Half of the trials (11) consisted of OCD interventions. The remaining RCTs evaluated the efficacy of pharmacotherapy in treating GAD (N = 2), SP (N= 3), overanxious and avoidant disorders (N = 1) and selective mutism (N = 1). Three trials included patients diagnosed with either SP, GAD or SAD (or their combination), while a single RCT included a variety of anxiety disorders. The medication agents most commonly tested in placebo controlled trials were the SSRIs (N = 15: 6 fluoxetine, 2 fluvoxamine, 3 paroxetine, 4 sertraline), followed by the serotonin-norepinephrine reuptake inhibitors (SNRIs: N = 5, 3 clomipramine, 2 venlafaxine), the benzodiazepines (N = 2: 1 alprazolam, 1 clonazepam), and the tricyclic antidepressants (TCAs: N = 1, 1 desipramine).

Outcomes

In those trials which identified an outcome measure as primary, the CGI-I was most commonly employed (N = 13), followed by the CY-BOCS (N = 9).

Duration

Patients were treated for an average of 11 weeks (range: 4 to 16 weeks).

Trials excluded from quantitative analyses

It was decided not to include summary statistics from the first period of treatment in Riddle 1992, due to lack of information, as well as concerns regarding the possibly biased nature of these trial data (the dropout rate in this trial was abnormally high, relative to other trials of fluoxetine). Insufficient data was provided in the report for the crossover trial of clonazepam (Graae 1994) and for one of the clomipramine crossover trials (Rapoport 1980) for inclusion in the meta-analyses. Data from the other cross-over trial of clomipramine (Flament 1985) were not included, as the inclusion of data from both periods may have resulted in a unit of analysis error (Deeks 2008).

Ongoing studies

Trials currently underway and which potentially satisfy the inclusion criteria specified for this review include the following: a multisite, randomised controlled trial comparing CBT and fluvoxamine for children and adolescents with anxiety disorders (March 2006), a fMRI imaging study comparing 8 weeks of (a) fluoxetine treatment with (b) either cognitive-behavioral therapy or interpersonal therapy or (c) random assignment to either placebo or fluoxetine (NCT00018057) for youth and adults diagnosed with SP, SAD, or GAD, and an 18 week CBT-augmented comparison of sertraline and placebo treatment for OCD (NCT00382291).

Excluded studies

The most common reason for excluding trials from the review was the treatment of conditions which are not recognised as DSM-based anxiety disorders, such as anxiety (Fatovich 1995; Feld 1989; Kain 1999; Lal 2001; Liacouras 1998; Lindsay 1985; Ljungman 2000; Roelofse 1993; Shane 1994; Sherwin 2000; Theroux 1993; Walker 1996) and school phobia (Berney 1981; Bernstein 2000; Gittelman-Klein 1971). Other reasons for excluding studies included the presence of comorbid psychiatric disorders (Abikoff 2005; Diamond 1999; Gadow 2002; Kurlan 1993; Tannock 1995; Weir 2000), publication prior to the development of the DSM-III diagnostic taxonomy (Cremonesi 1974; Dahlstrom 1973; Lader 1970), the concurrent treatment of patients with behavioural psychotherapy (Klein 1992;Cohen 2007) or medications (Procter 2001), the failure to report separate outcome data for adolescents in trials in which the majority of the participants were adults (Hollander 2003; Romano 2001), lack of a placebo control (Asbahr 2005; Leonard 1989; de Haan 1998; Neziroglu 1999; Steiner 2007), the non-random allocation of patients to treatment groups (Bernstein 1990), the analysis of a subset of a sample from another study (March 1990), and the implementation of a medication substitution design (Leonard 1991). An RCT of imipramine in treating paediatric acute stress disorders (Robert 1999) will be assessed for inclusion in a Cochrane review of pharmacotherapy for the prevention of PTSD (Ipser 2006).

Risk of bias in included studies

Allocation
Generation of allocation sequence

Ten trials provided descriptions of the randomisation procedure employed. A computer-generated randomisation list was employed in 8 of these trials (Beidel 2007; Geller 2004; March 1998; March 2007; POTS 2004; Wagner 2004). Five RCTs implemented a randomised blocking procedure, with blocks of 4 participants employed in two trials (POTS 2004; Rynn 2001), and gender and age used as blocking variables in the remaining trials (March 2007; Rynn 2007; Walkup 2008). One of the RCTs employed a biased coin toss design method (Birmaher 2003), while another described participants as being randomly assigned by the research pharmacist (Riddle 1992).

Allocation Concealment

Only 3 of the 22 trials provided sufficient information to determine the adequacy of allocation concealment (Birmaher 2003; Riddle 1992; Walkup 2008). Post-assignment analyses conducted in POTS 2004 provided no evidence of selection bias in allocation procedures.

Blinding

The clinicians who assessed treatment outcomes were blinded to treatment assignment in at least 9 of the 22 trials (Black 1994; Beidel 2007; Liebowitz 2002; POTS 2004; Riddle 1992; RUPPASG 2001; Rynn 2001; Wagner 2004; Walkup 2008). An independent evaluator assessed outcome in 4 of these trials (Beidel 2007; Liebowitz 2002; POTS 2004; Walkup 2008).

Incomplete outcome data
Loss to follow up

On average, 27.4% of participants withdrew from the RCTs prior to study completion. Equivalent proportions of participants in the medication (25.4%) and placebo groups (28.1%) dropped out.

Other potential sources of bias
Quality score on CCDAN-QRS

The average quality score on the CCDAN-QRS for the short term trials was 27 (range: 9-40) out of a maximum of 46 points. On this scale, 15 of the 22 published trials failed to assess compliance with experimental treatments, 15 provided inadequate details of the side effects experienced by each group, 7 trials failed to either provide a record of the exclusion criteria used or report the number of people excluded by these criteria, and 4 RCTs did not provide information about funding.

Effects of interventions

Primary outcome measures
Treatment response

Treatment response was significantly greater in the medication than the placebo groups (number of trials (N) = 14, relative risk of response (RR) = 1.9, 95% CI = 1.6, 2.26 , number of participants (n) = 2102). Almost twice as many patients responded to medication (58.1%) than to placebo (31.5%) (Analysis 1.1).

Using an assumed control risk of 0.364 (estimated from a recent meta-analysis of paediatric antidepressant treatment Bridge 2007) the number needed to treat is 4, indicating that four patients would have to be treated with medication for one additional treatment responder, relative to the control condition. We also calculated NNT assuming a control risk of 0.1 and 0.5. The NNT was 12 and 3 respectively.

Significant heterogeneity in treatment response on the CGI-I was observed across all trials (chi-squared = 27.33, p = 0.01, I-squared = 52%). This did not appear to be explained by differences between the individual SSRI agents or differences in efficacy between the SSRIs and the SNRI trials, as indicated by overlap in confidence intervals .

In post-hoc analyses, a significantly larger response to treatment was observed following medication than placebo in trials of both OCD (N = 5, RR = 1.65, 95%CI = 1.32, 2.06, n = 654) (Analysis 1.2) and non-OCD anxiety disorders (N = 9, RR = 2.01, 1.59, 2.55, n = 1448) (Analysis 1.3). However, differences in treatment response across trials was observed for the non-OCD anxiety disorders only (chi-squared = 23.5, p < 0.01, I-squared = 66%). Treatment efficacy was observed for the SSRIs (N = 12, RR = 2.07, 95%CI = 1.73, 2.47, n = 1496) and the SNRI venlafaxine (N = 2, RR = 1.46, 95% CI = 1.25, 1.71, n = 606).

Reduction in symptom severity

Medication was more effective than placebo in reducing overall symptom severity across all of the anxiety disorders (N = 9, SMD = -0.69, 95%CI = -0.94, -0.44, n = 810). In post-hoc comparisons, medication treatment resulted in a decrease of approximately 4.5 points on the CY-BOCS in OCD patients (N = 7, WMD = -4.45, 95%CI = -5.94, -2.97, n = 765). Statistically significant reductions in symptom severity after short-term treatment with SSRIs were also observed on a variety of scales for other anxiety disorders ( n = 428, SMD = -0.82, 95%CI = -1.3, -0.33, N=4).

A substantial degree of heterogeneity in the effect sizes reported for this outcome measure was observed across the different RCTs (chi-squared = 21.19, p < 0.01, I-squared = 62%). Although post-hoc analyses detected a moderate difference across trials of OCD (chi-squared = 8.81, p = 0.18, I-squared = 32%), heterogeneity was more prononunced for the non-OCD disorders (chi-squared = 13.8, p < 0.01, I-squared = 78%). This difference remained when comparing symptom severity between trials of GAD, SP and SAD (chi-squared = 7.12, p = 0.03, I-squared = 71.9%). No differences in the reduction of symptom severity were detected between the individual SSRI agents.

Secondary outcome measures
1) Levels of Co-morbid Depression

Levels of comorbid depression amongst the child and adolescent anxiety disorder samples were not noticeably affected by treatment with SSRIs (N = 4, SMD = -0.29, 95%CI = -0.88, 0.3, n = 250), the only trials that provided data for this comparison. The significant difference between trials in the effect of medication on depression symptoms (chi-squared = 13.42, p < 0.01, I-squared = 78%) can be attributed to the strong positive effect observed in the single trial of sertraline (Rynn 2001).

2) Quality of Life

Improvement was observed in quality of life following short-term treatment with fluoxetine or sertraline , as assessed by the Children's Global Assessment Scale (CGAS) (Beidel 2007; Birmaher 2003; Walkup 2008) and the Child Obsessive Compulsive Impact Scale (COIS-P) (Liebowitz 2002) (N = 4, RR = 0.55, 95%CI = 0.34, 0.76, n = 390).

3.1) Acceptability of medication

Drug-related adverse events were significantly more frequent following medication treatment (N = 12, RR = 1.91, 95%CI = 1.2, 3.05, n = 1997), and accounted for approximately twice as many withdrawals (65 versus 29) from the medication than the placebo groups. An average of 7.3% of participants who received SSRIs withdrew for this reason. The dropout rates were relatively consistent across trials (chi-squared = 11.82, p =0.38, I-squared = 7%), and no medication agent contributed more to the dropout rate than any other.

3.2) Adverse events associated with particular medication agents (narrative review)

The table in Appendix 2 contains a breakdown of the treatment-related adverse events that occured significantly more frequently in the medication than placebo groups. This data is stratified by anxiety disorder and agent, with information provided on the doses employed.

Regarding suicidality, one OCD patient had to discontinue treatment with fluoxetine following the emergence of suicidal thoughts (Riddle 1992). Treatment with paroxetine over 10 to 16 weeks resulted in the emergence of suicidal ideation for some of the participants (Geller 2004: n = 1, Wagner 2004: n = 4). Suicidal ideation was observed in four patients in an 8 week trial of venlafaxine (Rynn 2007), and in three patients in a 16 week trial of the same agent (March 2007). The ideation was severe enough in Rynn (2007) to result in one patient withdrawing for this reason. No cases of completed suicide were reported for any of the RCTs.

Drug-related changes in behaviour associated with an "activation syndrome" have been hypothesised as increasing the likelihood of engaging in suicidal behaviours in paediatric populations (Goodman 2007). Symptoms hypothesised as falling with the activation syndrome include irritability, agitation, panic attacks, restlessness, hostility, aggressivity, insomnia, disinhibition, emotional lability, impulsivity, restlessness, social withdrawal and odd behaviour. These symptoms were observed in some of the RCTs included in this review. Although the benzodiazepine clonazepam did not result in higher scores on the side effects rating scale in the one trial which tested this medication (Graae 1994), drowsiness, irritability and oppositional behaviour were described as the most frequent side effects observed in treating a variety of anxiety disorders. Two children dropped out of this trial due to serious disinhibition with marked irritability, tantrums and aggressivity. Approximately 17% of 133 GAD, SP and SAD patients treated with sertraline for 12 weeks exhibited activation symptoms. Of the 11 participants who dropped out as a result of an adverse response to medication or placebo in this trial, four experienced behavioural adverse effects (three for agitation or disinhibition and one for hyperactivity) (Walkup 2008). Five youths (7%) had to be discontinued from a 12 week fixed-dose trial of fluoxetine due to excitement, giddiness or disinhibition (Birmaher 2003). Behaviours regarded as changed/abnormal were reported for 2% of youth (6 of 293) receiving venlafaxine ER, with one patient manifesting time-limited hypomania (March 2007).

4.1) Long-term treatment (Analysis 4.2 and narrative review)

Continued reduction of symptom severity on the CY-BOCS and improvement of treatment response on the CGI-I was observed in 8 weeks' maintenance treatment of responders to a short-term (8 week) placebo-controlled RCT of fluoxetine (Liebowitz 2002). Similarly, a one year open-label extension of an 8 week RCT of clomipramine for OCD reduced the total score on the CY-BOCS by more than half (23 to 9.8) amongst completers (DeVeaugh-Geiss 1992). Nevertheless, the relapse-prevention phases of two trials of fluoxetine for the treatment of paediatric OCD (Geller 2001b; Liebowitz 2002) failed to demonstrate overall superiority to placebo in the long-term efficacy of this medication in reducing relapse (RR = 0.78, 95%CI = 0.55, 1.11, n = 211).

4.2) Adverse events during long-term treatment and its discontinuation (narrative review)

With regards to the side-effect profile of medication taken over the long term, a relatively high withdrawal rate due to adverse events (16/137) was observed in a 52 week open-label maintenance intervention for OCD patients who completed 12 weeks of controlled treatment with sertraline (March 1998). The most common of these were hyperkinesia and aggressive reactions in four and two children, respectively. Grand mal convulsions were observed in two adolescents in this study, with convulsions in one of the patients considered drug-related. A suicide attempt and a suicidal gesture in one patient was attributed to concurrent depression, while suicidal ideation in another was considered to be related to diagnosis of comorbid post-traumatic stress disorder. Approximately three quarters (76%) of patients who participated in the extension phase of March 1998 experienced newly emergent treatment-related adverse events.

Discontinuation of acute treatment with paroxetine resulted in a greater incidence of nausea and vomiting in Geller 2004, and more frequent instances of nausea, dizziness and abdominal pain in Wagner 2004. A significantly greater proportion of participants in the medication group experienced withdrawal-related side effects in the latter study. A withdrawal syndrome characterised by confusion and agitation was observed in a 10-year-old girl 24 hours after the last taper dose was administered in an RCT of venlafaxine (Rynn 2007).

Subgroup analyses

There was no evidence of a difference in reduction of symptom severity or response on the CGI-I for trials conducted in multiple versus single centres (though only one single-centre study was included in the former comparison). Greater treatment efficacy was observed for industry-funded trials on measures of symptom severity (N= 8, Qb = 2.94, p = 0.09) but not treatment response. Finally, the inclusion of participants with major depression did not appreciably moderate the efficacy of medication on treatment response (as indicated by confidence interval overlap) or symptom severity reduction.

The only trials to report outcomes separately for adolescents and children were Wagner 2004 and Rynn 2007. Wagner 2004 found little difference in treatment response between these age groups following treatment of SP with paroxetine, while Rynn 2007 detected significant reductions in the severity of GAD symptoms for both adolescents and children receiving venlafaxine XR. These findings were confirmed by an RCTof sertraline for GAD which controlled for age when conducting efficacy analyses (Rynn 2001). Geller 2004 and Riddle 2001, on the other hand, reported a higher response rate in children than adolescents following short-term treatment of OCD with fluvoxamine and paroxetine, respectively.

Sensitivity analyses

The short-term paroxetine trials (Geller 2004; Wagner 2004) were the only parallel RCTs which failed to include all of the participants randomised to the treatment groups in the responder analysis. Nevertheless, the comparison of separate efficacy analyses in which the omitted participants were regarded as non-responders in either the medication or placebo groups supported the robustness of the evidence of this agent's efficacy in treating anxiety disorders (RR = 1.82, 95%CI = 1.04, 3.17 and RR = 1.91, 95%CI = 1.12, 3.28, respectively).

Publication bias

The funnel plots for the primary outcomes provide little evidence of publication bias with respect to treatment response or reduction in symptom severity (refer to Figure 1 and Figure 2 under “Additional Figures”).

thumbnail image

Figure 1. Funnel plot of comparison: 1 Medication versus Placebo: Primary outcomes, outcome: 1.1 Clinical Global Impressions scale - Improvement item (CGI-I).

Download figure to PowerPoint

thumbnail image

Figure 2. Funnel plot of comparison: 1 Medication versus Placebo: Primary outcomes, outcome: 1.4 Symptom severity reduction on DSM-based measures.

Download figure to PowerPoint

Discussion

  1. Top of page
  2. Abstract
  3. Background
  4. Objectives
  5. Methods
  6. Results
  7. Discussion
  8. Authors' conclusions
  9. Characteristics of studies
  10. Data and analyses
  11. Appendices
  12. What's new
  13. History
  14. Contributions of authors
  15. Declarations of interest
  16. Sources of support
  17. Differences between protocol and review
  18. Acknowledgements
  19. REFERENCES

Summary of main results

This review provides evidence of the efficacy of medication in the short-term treatment of childhood and adolescent anxiety disorders. Almost twice as many patients responded to treatment with medication (59%) as placebo (31%). This translates to four paediatric patients requiring treatment with medication before one responds to the agent. An overall effect of medication in reducing the severity of anxiety disorder symptoms was also observed. Post-hoc analyses revealed that the effects of medication in reducing symptom severity and increasing treatment response held for both paediatric OCD, as well as certain other anxiety disorders.

The overwhelming majority of evidence of efficacy was for the SSRIs, with the most evidence in paediatric OCD. This finding is consistent with expert consensus that the SSRIs be regarded as first line treatments for paediatric OCD (Bandelow 2002). This review found little evidence for differences between SSRI agents in efficacy. Preliminary evidence of an improvement in functioning following short-term treatment with the SSRIs fluoxetine and paroxetine was observed. This is a potentially significant result given the chronicity of anxiety disorders if left untreated, and their potential to disrupt social and psychological development in paediatric populations.

Inconsistent findings have been reported for trials assessing the long-term efficacy of treatment with medication. Although the observation of continued response to maintenance treatment with fluoxetine supports the consensus that long-term maintenance therapy is indicated for paediatric anxiety disorders, the synthesis in this review of relapse rates for two fluoxetine trials (Geller 2001b; Liebowitz 2002) failed to detect evidence that this agent was effective in reducing relapse. While open-label treatment of OCD with sertraline over a year led to significant improvements in treatment response beyond that detected after acute treatment, this finding should be weighed against the high rate of newly emergent drug-related adverse events observed (March 1998).

Drug-related adverse events

A significant proportion of children and adolescents included in this review withdrew due to drug-related adverse events. No agent was more likely to result in treatment discontinuation for this reason. Nevertheless, the finding of particularly high drop-out rates amongst children and adolescents receiving higher doses of sertraline for OCD (March 1998), and no evidence in controlled trials of a relationship between dosage and efficacy amongst the SSRIs (RUPPASG 2001; Schweizer 2001), has led to the suggestion that SSRI agents be initiated at a relatively low dose. There was no difference in the tolerability of the SNRI venlafaxine and the SSRIs, which, when combined with the evidence of this agent's efficacy in treating paediatric GAD, suggests that venlafaxine should be considered as a treatment option for this disorder. The finding of a significant difference in the height of patients after medication treatment in an 8 week placebo-controlled trial of venlafaxine extended-release (Rynn 2007) indicates that potential delays in physical development should be monitored during long-term treatment with this medication.

Considerable attention has been devoted recently to the possibility that treatment with antidepressants (with a focus on SSRIs) may induce suicidal behaviour (Möller 2008). A recent meta-analysis commissioned by the FDA concluded that antidepressants as a group increase the risk of suicidal behaviour and ideation (Hammad 2006). This finding was based on an analysis of both MDD and anxiety disorder clinical trials. A review of RCTs of sertraline in which suicidality was separated by disorder concluded, on the other hand, concluded that the benefit-risk ratio for treating OCD in children and adolescents with SSRIs was more favourable than the use of the same agents for MDD (March 2006). In the antidepressant trials included in this study, only 13 patients experienced suicidal behaviour or ideation, with 5 and 7 of these occuring in RCTs of paroxetine and venlafaxine, respectively. The relatively high proportion of such suicidality in paroxetine trials is consistent with the recommendation by the FDA against the use of this agent for paediatric depression (FDA Talk 2003), and the reports of increased rates of suicides in some cohort studies of venlafaxine in adults (Rubino 2007;Tiihonen 2006). It should be noted, however, that the absolute rate of suicidal ideation for both these agents was low (approx. 1%), and that no completed suicides were observed in any of the trials included in this review. The perceived risks of pharmacotherapy for paediatric anxiety disorders also needs to be weighed against increased suicidal ideation and behaviour  associated with failing to treat these disorders, as illustrated by reports of increases in suicide attempts after FDA warnings led to fewer prescriptions for antidepressants (Gibbons 2006). Nevertheless, it is important to be aware of the need for careful monitoring after initiation of SSRIs in treating this population (CSM 2004).

Methodological and clinical moderators of treatment response

No evidence was detected in this review of a difference in treatment effect between studies conducted in multi-centre versus single-centre trials. Trials funded by the pharmaceutical industry demonstrated greater reductions in the severity of anxiety disorder symptoms than those that were funded by other sources. Source of funding has been associated with biased reporting of results in other health-care fields (Als-Nielsen 2003; Baker 2003). This review failed to detect an effect of excluding patients with MDD on the primary outcomes of this review . Although the power of this comparison to detect an effect was low, it is consistent with the lack of evidence in this review that pharmacotherapy reduces scores on depression scales. This suggests that reductions in the severity of anxiety disorder symptoms as a result of treatment are probably not mediated by the effect of medication on symptoms of depression.

An insufficient number of RCTs in this review stratified outcome data by age group to determine whether efficacy and tolerability differed for children and adolescents. A greater incidence of drug-related adverse effects were observed after treatment of OCD with paroxetine amongst children than adolescents (Geller 2004). Hyperkinesia was the most common symptom leading to treatment withdrawal in this study, with dropouts for this reason occuring in children only (Geller 2004). Activation has generally been reported as more commonly observed in children than adolescents following administration of SSRIs, and as one of the major causes of withdrawal (Hammad 2006; Safer 2006). Questions remain about the adjustment of dosage for the different age bands, though it would be prudent to start treatment in children with the minimum therapeutic dose. It has been recommended that children receive at most two-thirds (200 mg/d) of the maximum dose of fluvoxamine administered to adolescents (300 mg/d), on the basis of the finding that weight-adjusted steady-state plasma concentrations were 2 to 3 times higher in the younger group (Cheer 2001).

Agreements and disagreements with other studies or reviews

A number of meta-analyses of RCTs of pharmacotherapy for paediatric anxiety disorders have recently been published (Bridge 2007; Watson 2008). The relatively greater reduction in the severity of symptoms of non-OCD disorders compared to OCD observed in this review (SMD = -0.82 versus -0.52, respectively) is consistent with the Bridge 2007 finding that the effects of antidepressants are strongest for non-OCD anxiety disorders, followed by OCD and then MDD. Likewise, the finding of significant variation in the size of treatment effects for non-OCD RCTs was confirmed in Bridge 2007, who also reported a trend for a greater reduction in symptom severity in adolescents for non-OCD anxiety disorders. This review adds to the Bridge 2007 findings by including RCTs published after 2006, and also demonstrating larger and less consistent effects for non-OCD than OCD trials on a dichotomous measure of treatment response (CGI-I). The observation in both Watson 2008 and an earlier meta-analysis of SSRIs for paediatric OCD (Geller 2003) that clomipramine has the largest effect of all medication agents was replicated in this review. Nevertheless, this finding must be qualified by the observation that it is based on data from a single small trial of clomipramine (DeVeaugh-Geiss 1992). Clomipramine is generally regarded as a second-line agent in paediatric OCD, given its more severe side-effect profile and the need to monitor for blood and cardiac irregularities.

Watson 2008 concluded that both CBT and medication are effective in reducing the severity of symptoms of OCD in children and adolescents. Although the effect sizes they reported were larger for CBT than medication RCTs (ES = 1.45 versus 0.48, respectively) a Cochrane review assessing the overall effect of CBT on all paediatric anxiety disorders reported an effect size on the Revised Children's Manifest Anxiety Scale (SMD = -0.58) and the Fear Survey for Children Revised (SMD = -0.55) of similar magnitude to that observed in this review (James 2005). Caution is advised in interpreting the results of such indirect comparisons of the efficacy of different treatment modalities, however, given the non-equivalence of control groups typically employed in RCTs of CBT and pharmacotherapy. Recently a number of RCTs have compared the efficacy of CBT and medication directly (Beidel 2007; POTS 2004; Walkup 2008). A 12 week program of Social Effectiveness Therapy for Children (SET-C) was superior to fluoxetine in reducing social distress and behavioural avoidance and in increasing general functioning and social skills (Beidel 2007). Although there was no difference between sertraline and CBT in reducing symptom severity over 12 weeks, CBT proved superior in an analysis of remission and clinical significance (POTS 2004). Notably, the largest effects in all comparisons were obtained when combining CBT with sertraline. This is consistent with the patterns of results reported in a study employing a similar design (Walkup 2008), in which combination of sertraline and CBT demonstrated superior efficacy to either modality on its own. The observation in two of these studies of a levelling off of the effects of medication but not CBT on symptom severity after 8 weeks suggests that differences in efficacy between these modalities may emerge with long term treatment (Beidel 2007; Walkup 2008).

Overall completeness, applicability and quality of evidence

There are a number of factors limiting the strength of the conclusions that can be supported by this review. The majority of the trials excluded patients with comorbid psychiatric diagnoses, thereby excluding individuals who are arguably more resistant to treatment, more prone to relapse following discontinuation of pharmacotherapy, and more typical of general population samples (Geller 2001b). Most of the trials assessed short-term treatment with medication, and therefore do not speak to the long-term efficacy of pharmacotherapy. Insufficient data was available to compare treatment response for children and adolescents, despite requests to trial investigators for this information. Finally, the findings of this review were based on a small number of RCTs. This is largely due to the paucity of research in this area, despite inducements such as the Best Pharmaceuticals for Children Act (BPCA 2002).

No controlled evidence could be found for the efficacy of benzodiazepines (Graae 1994; Simeon 1992) to warrant their increasing prescription for paediatric patients (Zito 2003). The lack of evidence of efficacy, combined with their potential for abuse and dependence, has led to the suggestion that trials making use of benzodiazepines be kept short and confined to acute cases of anxiety (Masi 2001). Further support for this approach comes from the finding of a high dropout rate due to disinhibition and the occurrence of treatment-emergent adverse events following even low-dose (0.25 - 2 mg/d ) acute treatment with clonazepam (Graae 1994). This is despite suggestions that the faster breakdown of these agents in children renders them relatively non-toxic (Simeon 1989).

Anxiety disorders in children and adolescents were regarded for the purposes of this review as forming a continuum with adult anxiety disorders, despite evidence that certain forms of childhood anxiety disorders, such as the Pediatric Autoimmune Neuropsychiatric Disorder (PANDAS) variant of OCD, are qualitatively distinct from the adult form. Moreover, the observation of significant variability for both treatment response and symptom severity in post-hoc analyses of outcome data from non-OCD populations suggests that the decision to treat children and adolescents diagnosed with SP, SAD or GAD as a unitary group may not be warranted.

Authors' conclusions

  1. Top of page
  2. Abstract
  3. Background
  4. Objectives
  5. Methods
  6. Results
  7. Discussion
  8. Authors' conclusions
  9. Characteristics of studies
  10. Data and analyses
  11. Appendices
  12. What's new
  13. History
  14. Contributions of authors
  15. Declarations of interest
  16. Sources of support
  17. Differences between protocol and review
  18. Acknowledgements
  19. REFERENCES

Implications for practice

Medication treatments can be effective in childhood and adolescent anxiety disorders, acting to reduce core symptoms, and should be considered as part of the treatment of these disorders. Although there is also no clear evidence to show that any particular class of medication is more effective or better tolerated than any other, the majority of trials contributing evidence of efficacy to date have been with the SSRIs. Evidence of withdrawal from treatment due to drug-related adverse events, absence of evidence that clearly identifies clinical predictors of treatment response, and a lack of available evidence of a relationship between medication dosage and efficacy indicates that it would be prudent to initiate treatment at the lower end of the recommended dose range. Conversely, care should be taken to ensure that adequate doses are administered, given the finding that medications are generally well-tolerated over the short-term. The routine use of benzodiazepines can not be recommended, as there is insufficient efficacy data from controlled trials to offset potential adverse effects of this agent. The evidence supporting the value of long-term medication interventions is limited and inconsistent, and needs to be weighed against the possible emergence of adverse events with continued treatment.

Implications for research

Controlled studies of the efficacy and tolerability of pharmacotherapy for paediatric anxiety disorders are remarkably sparse, given recognition of the early age of onset for anxiety disorders, and the poor prognosis for many young people who do not receive treatment. Future RCTs could help identify potential moderators of treatment efficacy through assessing the impact of clinical heterogeneity on treatment outcome. Examples include conducting studies designed to tease apart the influence of psychiatric comorbidity on treatment response, and facilitating the synthesis of data across trials by improving the quality of reporting (e.g. stratifying outcomes by age group). Reporting outcomes separately for different diagnostic groups might also help explain some of the variation in treatment response. Finally, future direct comparisons of pharmacotherapy and psychotherapy, dose-comparison studies, and controlled assessments of the long-term efficacy and tolerability of medication treatment are warranted.

Characteristics of studies

  1. Top of page
  2. Abstract
  3. Background
  4. Objectives
  5. Methods
  6. Results
  7. Discussion
  8. Authors' conclusions
  9. Characteristics of studies
  10. Data and analyses
  11. Appendices
  12. What's new
  13. History
  14. Contributions of authors
  15. Declarations of interest
  16. Sources of support
  17. Differences between protocol and review
  18. Acknowledgements
  19. REFERENCES

Characteristics of included studies [ordered by study ID]

Beidel 2007
MethodsDESIGN Description: randomised, double-blind fixed dose placebo-controlled, parallel trial, one-year follow-up BLINDING Participants: Yes Assessors: Yes Administrators: Unclear ALLOCATION CONCEALMENT Method: Unclear RANDOMISATION Method: Computer-generated randomisation list
ParticipantsSAMPLE Description: 80 DSM-IV SP, 48.9% female, no MDD, average age: 11.6 years SCREENING Primary diagnosis: ADIS -child and parent version, => 4 on Clinical Severity Rating Scale Comorbidity: ADIS -child and parent version
InterventionsDescription: fluoxetine 10 mg/d (wk 1 & 2) , 20 mg/d (wk 3 & 4), 30 mg/d (wk 5 & 6) - 40 mg/d (wk 7+) versus Social Effectiveness Therapy for Children (SET-C) versus placebo x 12 weeks
OutcomesPrimary outcomes: CGI-I, absence of SAD diagnosis, high-end state functioning Secondary outcomes: clinician-rated (CGI-S, CGAS, BARS), self-rated (SPAI-C, MASC, CDI, loneliness scale, EPQ-junior), parent-rated (CBCL); behavioural assessment Data estimation: LOCF
NotesINDUSTRY SUPPORT Industry funded: No (medication provided by industry) Any of the authors work for industry: No ADDITIONAL INFORMATION Drop-out rates: 15 on fluoxetine and 8 on placebo (including pretreatment dropouts) Quality rating score: 33
Birmaher 2003
MethodsDESIGN Description: randomised, double-blind fixed-flexible dose placebo-controlled, parallel trial BLINDING Participants: Yes Assessors: Yes Administrators: Unclear ALLOCATION CONCEALMENT Method: Unclear RANDOMISATION Method: Unclear
ParticipantsSAMPLE Description: 74 DSM-IV SAD (47%), GAD (63%), SP (54%), 54.1% female, no MDD, average age: 11.8 years, 54% female, average duration of illness: 5.2 years, baseline severity on PARS: fluoxetine (15.6), placebo (14.9) SCREENING Primary diagnosis: CGAS <= 60, SADS-SCPL Comorbidity: FPH
InterventionsDescription: fluoxetine 10mg/d - 20mg/d versus placebo x 12 weeks
OutcomesCGI-I, CGI-S, PARS, SCARED-C, SCARED-P, CDI (<= 12yrs), BDI (>12yrs), CGAS, SEFCA (no distinction made between primary and secondary outcomes) Data estimation: LCOF (claimed)
NotesINDUSTRY SUPPORT Industry funded: No (medication provided by industry) Any of the authors work for industry: No ADDITIONAL INFORMATION Drop-out rates: 9 on fluoxetine and 6 on placebo Quality rating score: 32
Black 1994
MethodsDESIGN Description: random-assignment, placebo-controlled, parallel, flexible dose, double-blind, 2 week single-blind placebo run-in BLINDING Participants: Unclear Assessors: Yes Administrators: Yes ALLOCATION CONCEALMENT Method: Unclear RANDOMISATION Method: Unclear
ParticipantsSAMPLE Description: 16 DSM-III-R elective mutism, 60% (9/15) female, average age of OC sample: 8.5 years, average duration of illness in OC sample: 5.8 years, no MDD, all diagnosed with SP or avoidant disorder, baseline severity not provided SCREENING Primary diagnosis: CPQ (modified), TRS, PARIS Comorbidity: Unclear
InterventionsDescription: fluoxetine 0.2 mg/kg - 0.6mg/kg (average maximum dose: 21.4mg/d) versus placebo x 12 weeks
OutcomesParents: mutism CGI-C, anxiety CGI-C, shyness CGI-C, global CGI-C; Psychiatrist: mutism CGI-C, anxiety CGI-C, global CGI-C; Teachers: mutism CGI-C, anxiety/nervousness CGI-C, shyness CGI-C, TRS (abbreviated) (no distinction made between primary and secondary outcomes) Data estimation: OC
NotesDrop out rate = 0 Quality rating scale: 22 INDUSTRY SUPPORT Industry funded: Unclear Any of the authors work for industry: No ADDITIONAL INFORMATION Drop-out rates: 0 on fluoxetine, 1 on placebo Quality rating score: 22
DeVeaugh-Geiss 1992
MethodsDESIGN Description: random-assignment, placebo-controlled, parallel, flexible dose, double-blind, 2 week single blind placebo run-in BLINDING Participants: Unclear Assessors: Unclear Administrators: Unclear ALLOCATION CONCEALMENT Method: Unclear RANDOMISATION Method: Unclear
ParticipantsSAMPLE Description: 60 DSM-III OCD, 65% male, no MDD, average age: 14.3 years, age range: 10-17 years, baseline severity on CY-BOCS: clomipramine (27.1), placebo (28.4) SCREENING Primary diagnosis: YBOCS >= 16, NIMH-GOCS >= 7 Comorbidity: HAM-D <= 16 (item 1 <=2)
InterventionsDescription: clomapramine 25mg/d - 200mg/d (or 3mg/kg) versus placebo x 8 weeks
OutcomesPrimary outcomes: CY-BOCS, Secondary outcomes: NIMH-GOCS, Patient Self-Rating Scale, Physicians global evaluation of therapeutic change Data estimation: LOCF
NotesINDUSTRY SUPPORT Industry funded: Yes Any of the authors work for industry: Yes ADDITIONAL INFORMATION Drop-out rates: 4 on clomipramine and 2 on placebo Quality rating score: 24 Concomittant use of supportive psychotherapy. One participant on CMI discontinued due to elevated hepatic enzyme levels, though not regarded as dropping out due to adverse effects.
Flament 1985
MethodsDESIGN Description: random-assignment, placebo-controlled, crossover, flexible dose, double-blind, 1 week placebo run-in BLINDING Participants: Unclear Assessors: Unclear Administrators: Unclear ALLOCATION CONCEALMENT Method: Unclear RANDOMISATION Method: Unclear
ParticipantsSAMPLE Description: 23 DSM-III OCD, 66.7% (18/27) male, mean age of OC sample: 14.5 years, age range of OC sample: 10-18 years, average duration of illness for OC sample: 4.3 years, baseline severity on OCRS for OC sample: 13.5 SCREENING Primary diagnosis: Unclear Comorbidity: Unclear
InterventionsDescription: clomipramine 100 - 200 mg/d (average dose: 141 mg/d) versus placebo x 10 weeks
OutcomesLOI-CV, OCR, CPRS, NIMH-GS, BPRS, NIMH-SS (no distinction made between primary and secondary outcomes) Data estimation: OC
NotesINDUSTRY SUPPORT Industry funded: No Any of the authors work for industry: No ADDITIONAL INFORMATION Drop-out rates: 2 on clomipramine, 0 on placebo Quality rating score: 20 Children received individual supportive psychotherapy during trial as needed
Geller 2001a
MethodsDESIGN Description: random-assignment, placebo-controlled, parallel, fixed dose, double-blind, 1 week pre-randomisation evaluation period BLINDING Participants: Unclear Assessors: Unclear Administrators: Unclear ALLOCATION CONCEALMENT Method: Unclear RANDOMISATION Method: Unclear
ParticipantsSAMPLE Description: 103 DSM-IV OCD, mean age (LOCF): 11.4 years, 75 between 6-12yrs, and 28 between 13 and 17yrs, 5 MDD (secondary), 52% LOCF sample female, average age of LOCF sample: 11.4, age range: 7-17 years, baseline severity on CY-BOCS: fluoxetine(24.5), placebo (26.3) SCREENING Primary diagnosis: CGI-S >=4, CY-BOCS >= 16, NIMH-GOCS >= 7 Comorbidity: CDRS-R <= 40
InterventionsDescription: fluoxetine 20 mg/d - 60 mg/d + 2 tablets placebo daily (mean daily dosage: 24.6 mg/d) versus placebo 3 tablets daily x 13 weeks
OutcomesPrimary outcomes: CY-BOCS Secondary outcomes: CGI-S, CGI-I, PGI, NIMH-GOCS, OCD-IS, CDRS-R, MASC Data estimation: LCOF
NotesDrop out rate = 34 (33%) Quality rating scale: 27 INDUSTRY SUPPORT Industry funded: Yes Any of the authors work for industry: Yes ADDITIONAL INFORMATION Drop-out rates: 22 (31%) on fluoxetine and 12 (37.5%) on placebo Quality rating score: 28
Geller 2001b
MethodsDESIGN Description: Double blind, placebo-controlled, randomised relapse prevention extension of an open-label acute trial BLINDING Participants: Unclear Assessors: Unclear Administrators: Unclear ALLOCATION CONCEALMENT Method: Unclear RANDOMISATION Method: Unclear
ParticipantsSAMPLE Description: 193 DSM-IV OCD, average age: 11.7 years, baseline severity on CY-BOCS at beginning of relapse-prevention phase: 9.9 (paroxetine), 9.6 (placebo) SCREENING Primary diagnosis: K-SADS, CY-BOCS >= 16 Comorbidity: K-SADS
InterventionsDescription: paroxetine 10 mg/d - 60 mg/d (mean dose: 32 mg/d) for open-label phase versus placebo x 16 weeks, followed by 16 weeks relapse prevention
OutcomesPrimary outcomes: CY-BOCS, CGI-I Secondary outcomes: None declared Data estimation: LOCF
NotesINDUSTRY SUPPORT Industry funded: Yes Any of the authors work for industry: Yes ADDITIONAL INFORMATION Drop-out rates: 55.8% (53/95) on paroxetine, 66.3% (65/98) on placebo Quality rating score: 21
Geller 2004
MethodsDESIGN Description: random-assignment, placebo-controlled, parallel, flexible dose, double-blind, 1 week evaluation prior to randomisation BLINDING Participants:Unclear Assessors:Unclear Administrators:Unclear ALLOCATION CONCEALMENT Method:Unclear RANDOMISATION Method:Computer-generated randomisation list
ParticipantsSAMPLE Description: 207 DSM-IV OCD, LOCF population - most common comorbidity: ADHD: 19(9.4%), 58% male (117/203), average age : 11.4 years, average duration of illness in LOCF sample: 4.2 years, baseline severity on CY-BOCS: paroxetine (24.4), placebo (25.3) SCREENING Primary diagnosis: K-SADS-PL, CY-BOCS >= 16 Comorbidity: Unclear
InterventionsDescription: paroxetine 10 mg/d - 50 mg/d (max mean: 30.1mg/d; children: 25.4mg/d; adolescents: 36.5mg/d) versus placebo x 10 weeks, tapered discontinuation, followed by 2 weeks follow-up if no open-label continuation
OutcomesPrimary outcomes: CY-BOCS Secondary outcomes: CGI-I (NIMH), CGIS (NIMH), GAF Data estimation: LCOF
NotesDrop out rate = 58 (28%) Quality rating scale: 30 INDUSTRY SUPPORT Industry funded: Yes Any of the authors work for industry: Yes ADDITIONAL INFORMATION Drop-out rates: 35 on paroxetine and 27 on placebo Quality rating score: 28 A greater proportion of children dropped out due to adverse effects (13.8%) than adolescents (5%). Greater improvement was seen on the CY-BOCS in children than adolescents, irrespective of treatment
Graae 1994
MethodsDESIGN Description: random-assignment, placebo-controlled, crossover, flexible dose, double-blind, no washout (tapering) BLINDING Participants: Unclear Assessors: Unclear Administrators: Unclear ALLOCATION CONCEALMENT Method: Unclear RANDOMISATION Method: Unclear
ParticipantsSAMPLE Description: 15 DSM-III-R separation anxiety disorder (14), overanxious disorder (6), avoidant disorder (2), SP (5), simple phobia (5), oppositional disorder (3), conduct disorder (1), ADHD (3), 53% male, average age:9.8 years, average duration of SAD: 3.8 years, 53.3% males, baseline severity not provided SCREENING Primary diagnosis: DISC 2.1 Comorbidity: Unclear
InterventionsDescription: clonazepam 0.25mg/d - 2mg/d versus placebo x 8 weeks
OutcomesDISC 2.1 (parent & child), BPRS, CGI-I (1-8), SERS, CMAS (no distinction made between primary and secondary outcomes) Data estimation: OC
NotesINDUSTRY SUPPORT Industry funded: Yes Any of the authors work for industry: Yes ADDITIONAL INFORMATION Drop-out rates: 3 during clonazepam phase, 0 during placebo Quality rating score: 21 Child psychiatrist provided concurrent supportive therapy
Liebowitz 2002
MethodsDESIGN Description: random-assignment, placebo-controlled, parallel arm, flexible dose, double-blind, acute and maintenance component BLINDING Participants: Yes Assessors: Yes Administrators: Yes ALLOCATION CONCEALMENT Method: Unclear RANDOMISATION Method: Unclear
ParticipantsSAMPLE Description: 43 DSM-III-R or DSM-IV OCD, 58% male, 17 children (8-12 yrs) and 26 adolescents (13-17 yrs), average age: 12.7 years, most common comorbid disorder: GAD - 16%, baseline severity on CY-BOCS: fluoxetine (22.5), placebo (23.82) SCREENING Primary diagnosis: SADS-SAC, DISC, NIMH-OC >= 7, CY-BOCS >=16 Comorbidity: Unclear, presumably diagnosed using SADS-SAC, DISC
InterventionsDescription: fluoxetine 20 mg/d (week 1 - 2) , 40 mg/d (week 3-4 ), 60 mg/d (week 5-6), up to 80 mg/d afterwards (average: 64.8 mg/d) versus placebo (average dose: 59 mg/d) x 8 weeks, followed by 8 weeks maintenance component for responders
OutcomesPrimary outcomes: CY-BOCS, CGI-I Secondary outcomes: NIMH-OC, CGI-S, HAM-D, COIS-P Data estimation: LOCF for participants with baseline data
NotesINDUSTRY SUPPORT Industry funded: Yes Any of the authors work for industry: No ADDITIONAL INFORMATION Drop-out rates: 1 in fluoxetine and 4 in placebo group Quality rating score: 34 Participants were recruited over 7 year period (1991-1998), medication group significantly more depressed at baseline than placebo
March 1998
MethodsDESIGN Description: random-assignment, placebo-controlled, parallel, flexible dose, double-blind, 1 week single-blind placebo run-in BLINDING Participants: Unclear Assessors: Unclear Administrators: Unclear ALLOCATION CONCEALMENT Method: Unclear RANDOMISATION Method: computer generated randomisation algorithm
ParticipantsSAMPLE Description: 189 DSM-III-R OCD, mean age: 12.6 years, 107 children (6-12 yrs) and 80 adolescents (13-17 yrs), 4 (2%) depression, 53% (99/187) male, most common form of comorbidity is ADHD: 9(4.7%), baseline severity on CY-BOCS: sertraline (23.4), placebo (22.2) SCREENING Primary diagnosis: NIMH-GOCS >= 7 and HAMD-D total <=17 (and 0 or 1 on item 1) Comorbidity: Unclear
InterventionsDescription: sertraline for children: 25 mg/d - 200 mg/d, sertaline for adolescents: 50 mg/d - 200 mg/d (total maximum average: 167 mg/d) versus placebo (total maximum average: 180 mg/d) x 12 weeks
OutcomesPrimary outcomes: CY-BOCS, NIMH-GOCS, NIMH CGI-S & CGI-I scales Secondary outcomes: None mentioned Data estimation: LOCF
NotesINDUSTRY SUPPORT Industry funded: Yes Any of the authors work for industry: No ADDITIONAL INFORMATION Drop-out rates: 20 on sertraline and 13 on placebo Quality rating score: 26
March 2007
MethodsDESIGN Description: random-assignment, placebo-controlled, parallel, flexible dose, double-blind, BLINDING Participants: Yes Assessors: Yes Administrators: Yes ALLOCATION CONCEALMENT Method: Centralised computerised randomisation RANDOMISATION Method:Computerised randomisation into 1:1 ratio with site as stratification variable and gender as blocking variable
ParticipantsSAMPLE Description: 293 DSM-IV SP, mean age: 13.6 years, 43.5% (124/285) male, baseline severity on SAS-CA: venlafaxine (64.8), placebo (66.2) SCREENING Primary diagnosis: ADIS - child and parents version Comorbidity: ADIS - child and parents version
InterventionsDescription: venlafaxine 37.5 mg/d - 112.5 mg/d (25-33 kg), 37.5 - 150 mg/d (34-49 kg), 37.5 mg/d - 225 mg/d (>50kg); mean dose for 0 to 112 days of exposure (141.5 mg/d), mean dose for > 112 days (155 mg/d) versus placebo x 16 weeks
OutcomesPrimary outcomes: SAS-CA (SASC-C (ages 8-11), SAS-A (ages 12-17)) Secondary outcomes: CGI-I, CGI-S, LSAS-CA Data estimation: LOCF
NotesINDUSTRY SUPPORT Industry funded: Yes Any of the authors work for industry: Yes ADDITIONAL INFORMATION Drop-out rates: 38 on venlafaxine and 53 on placebo Quality rating score: 32
POTS 2004
MethodsDESIGN Description: random-assignment, placebo-controlled, parallel, flexible dose, double-blind BLINDING Participants: Yes Assessors: Yes (IE) Administrators: Yes ALLOCATION CONCEALMENT Method: Unclear RANDOMISATION Method: computer generated randomised permuted blocking procedure with block size of 4
ParticipantsSAMPLE Description: 112 DSM-IV OCD. medication and placebo groups only: average age: 12 years, 55% male, 63% of total sample had an anxiety and/or affective disorder, baseline severity on CY-BOCS: sertraline (23.5), placebo (25.2) SCREENING Primary diagnosis: CY-BOCS, ADIS-C, CY-BOCS > 16, NIMH-GS > 7 Comorbidity: Unclear
InterventionsDescription: sertraline 25mg/d - 200mg/d (maximum average dose: 170mg/d) versus placebo (maximum average dose: 176mg/d) x 12 weeks, followed by 16 week discontinuation trial for responders
OutcomesPrimary outcomes: CY-BOCS Secondary outcomes: None mentioned Data estimation: Mixed Methods, except for treatment response on CY-BOCS (LOCF)
NotesINDUSTRY SUPPORT Industry funded: Yes Any of the authors work for industry: No ADDITIONAL INFORMATION Drop-out rates: 2 on sertraline, 7 on placebo Quality rating score: 38 10% of sample taking concurrent psychostimulant for ADHD
Rapoport 1980
MethodsDESIGN Description: random-assignment, placebo-controlled, parallel, fixed dose, double-blind, cross-over trial BLINDING Participants: Unclear Assessors: Unclear Administrators: Unclear ALLOCATION CONCEALMENT Method: No RANDOMISATION Method: Unclear
ParticipantsSAMPLE Description: 9 OCD (assumed to be diagnosed on the DSM-III), average age: 15.2 years, 78% male SCREENING Primary diagnosis: Unclear Comorbidity: Unclear
InterventionsDescription: chlorimipramine (clomipramine) 50 mg/d - 150 mg/d (in doses of 50 mg) versus desmethylimipramine (desipramine) 50 mg/d - 150 mg/d (in doses of 50 mg) versus placebo 16 weeks, followed by 6 month to 2 year follow-up
OutcomesLOI, BPRS, OI, CPRS (3 patients) (no distinction between primary and secondary outcomes) Data estimation: OC
Notes 
Riddle 1992
MethodsDESIGN Description: random-assignment, placebo-controlled, crossover, fixed dose, double-blind, no washout BLINDING Participants: Unclear Assessors: Yes (inferred) Administrators: Unclear ALLOCATION CONCEALMENT Method: Unclear RANDOMISATION Method: randomly assigned by research pharmacist
ParticipantsSAMPLE Description: 14 DSM-II-R OCD, average age: 11.8 years, 57% female, most common comorbidity: separation anxiety disorder: 3, no MDD: 2, baseline severity on CY-BOCS: fluoxetine (24.3), risperidone (20.2) SCREENING Primary diagnosis: Comorbidity:
InterventionsDescription: fluoxetine 20 mg/d versus placebo 20mg/d x 8 weeks first phase, 12 weeks second phase
OutcomesPrimary outcomes: CY-BOCS, CGI-OCD Secondary outcomes: LOI-CV, CMAS-R, CDI, CGAS (modified), SATEE Data estimation: LOCF
NotesINDUSTRY SUPPORT Industry funded: No (medication provided by industry) Any of the authors work for industry: No ADDITIONAL INFORMATION Drop-out rates: 2 on fluoxetine and 1 on placebo after first phase (8 weeks) Quality rating score: 30 Used data from first leg of crossover trial, as drop-outs for entire trial too high. Seven participants continued with individual supportive or psychodynamic psychotherapy
Riddle 2001
MethodsDESIGN Description: random-assignment, placebo-controlled, parallel, flexible dose, double-blind, 7 - 14 day single-blind placebo run-in BLINDING Participants: Unclear Assessors: Unclear Administrators: Unclear ALLOCATION CONCEALMENT Method: Unclear RANDOMISATION Method: Unclear
ParticipantsSAMPLE Description: 120 DSM-III-R OCD, average age: 13 years, 59 children (8-12yrs) and 61 adolescents (13-17yrs), 53% male, average duration of illness: 3.6 years, baseline severity on CY-BOCS: fluvoxamine (24.2), placebo (24.2) SCREENING Primary diagnosis: CY-BOCs > 15, NIMH-GOCS > 7 & > than NIMH-G scales for depression, mania, psychosis & anxiety Comorbidity: CDRS-R < 40
InterventionsDescription: fluvoxamine 50 mg/d - 200 mg/d (average maximum dose for children (8-12): 157 mg/d, for adolescents (13-17): 170 mg/d) versus placebo 2 - 8 capsules (average maximum number of capsules: 7) x 10 weeks.
OutcomesPrimary outcomes: CY-BOCS Secondary outcomes: NIMH-GOCS, CGI (clinican), CGI (parent), CGI (patient) Data estimation: LOCF
NotesINDUSTRY SUPPORT Industry funded: Yes Any of the authors work for industry: Yes ADDITIONAL INFORMATION Drop-out rates: 19 (33.3%) on fluvoxamine, 27 (42.9%) on placebo Quality rating score: 31 Sample relatively free of psychiatric comorbidity Non-specific supportive and/or behavioural therapy was permitted during study
RUPPASG 2001
MethodsDESIGN Description: random-assignment, placebo-controlled, parallel, flexible dose, double-blind, 3 week initial psychotherapy BLINDING Participants: Unclear Assessors: Yes (although blinding possibly compromised through clinicians rating efficacy and side-effects) Administrators: Yes ALLOCATION CONCEALMENT Method: Unclear RANDOMISATION Method: Unclear
ParticipantsSAMPLE Description: 128 DSM-IV SP (65.6%) or SAD (59.4%) or GAD (57%), average age: 10.3 years, 51% male, 95 children (6-12yrs) and 33 adolescents (13-17yrs), baseline severity on PARS: fluvoxamine (18.7), placebo (19) SCREENING Primary diagnosis: PARS, CGAS < 60, K-SADS Comorbidity: Unclear
InterventionsDescription: 50mg/d - 300mg/d fluvoxamine for adolescents (13-17), 50 mg/d - 250 mg/d children (6-12) (average final dose for both groups: 4 mg/kg/d) versus placebo (average dose: 3.8 mg/kg/d) x 8 weeks
OutcomesPrimary outcomes: PARS, CGI-I Secondary outcomes: MASC, SCARED Data estimation: MM
NotesINDUSTRY SUPPORT Industry funded: Yes Any of the authors work for industry: No ADDITIONAL INFORMATION Drop-out rates: 10 (15.9%) on fluvoxamine, 14 (21.5%) on placebo Quality rating score: 28 Participants received 3 weeks of open-label supportive psychotherapy prior to treatment, as well as during the treatment phase
Rynn 2001
MethodsDESIGN Description: random-assignment, placebo-controlled, parallel, fixed dose, double-blind, 2 - 3 week evaluation period, 1 week taper BLINDING Participants: Unclear Assessors: Yes (at least for one of the assessors) Administrators: Unclear ALLOCATION CONCEALMENT Method: Unclear RANDOMISATION Method: blocked randomisation in groups of 4
ParticipantsSAMPLE Description: 22 DSM-IV GAD, average age: 11.7, age range: 5-17, 77% male, baseline severity on HAM-A: sertraline (20.6), placebo (23.3) SCREENING Primary diagnosis: ADISC-R Comorbidity: HAM-A >= 16
InterventionsDescription: sertraline 50mg/d versus placebo x 9 weeks
OutcomesPrimary outcomes: HAM-A, CGI-S, CGI-I Secondary outcomes: MASC, CAMS-R, HAM-D Data estimation: MM and LOCF
NotesINDUSTRY SUPPORT Industry funded: No Any of the authors work for industry: No ADDITIONAL INFORMATION Drop-out rates: 1 (9%) on sertraline and 2 (18%) on placebo Quality rating score: 28
Rynn 2007
MethodsDESIGN Description: random-assignment, placebo-controlled, parallel, fixed dose, double-blind, 2 week optional taper, 4 to 10 day single-blind placebo run-in BLINDING Participants: Unclear Assessors: Unclear Administrators: Unclear ALLOCATION CONCEALMENT Method: Unclear RANDOMISATION Method: Unclear
ParticipantsSAMPLE Description: 313 DSM-IV GAD, average age: 11.3, age range: 6-17, 57.5% male, baseline severity on 9 items of C-KIDDIE-SADS GAD: venlafaxine (74.4), placebo (74.8) Results reported for pooled analysis of Rynn 2007 (1) and Rynn 2007 (2) SCREENING Primary diagnosis: C-KIDDIE-SADS GAD; total severity score on 5 severity questions >= 20; total impairment score on 4 impairment questions >= 7; score of >= 4 on following items of scale: (1) severity of anxiety and worry, (2) difficulty controlling the worry, (3) severity of associated symptoms; Score of at least 4 on following severity items: (1) frequency of anxiety and worry during average week, (2) frequency of associated symptoms during average week; Score of at least 4 on following impairment item: (1) global impairment in functioning. CDRS-R score < 45. CGI-S >= 4. Anxiety symptoms for at least 6 months before entry, not actually suicidal Comorbidity: No information provided
InterventionsDescription: venlafaxine extended-release (37.5 mg/d - 112.5 mg/d (25-39 kg), 37.5 - 150 mg/d (40-49 kg), 37.5 mg/d - 225 mg/d (>50kg)) versus placebo x 8 weeks
OutcomesPrimary outcomes: C-KIDDIE-SADS GAD (9 items) Secondary outcomes: C-KIDDIE-SAD GAD severity and impairment components; PARS; HAM-A; SCARED Parent and Patient scores; CGI-S; CGI-I Data estimation: LOCF
NotesINDUSTRY SUPPORT Industry funded: Yes Any of the authors work for industry: Yes ADDITIONAL INFORMATION Drop-out rates: 37 (23.6%) from medication group, 41 (25.2%) from placebo group Quality rating score: 30
Simeon 1992
MethodsDESIGN Description: random-assignment, placebo-controlled, parallel, flexible dose, double-blind BLINDING Participants: Unclear Assessors: Unclear Administrators: Unclear ALLOCATION CONCEALMENT Method: Unclear RANDOMISATION Method: Unclear
ParticipantsSAMPLE Description: 30 (DSM-III-R?) overanxious (21) and avoidant disorder (9), mean age: 12.6 years, 77% male, 14 children (8-12 yrs) and 16 adolescents (13-16.9 yrs) SCREENING Primary diagnosis: Unclear Comorbidity: Unclear
InterventionsDescription: alprazolam 0.25mg/d (< 40kg) or 0.50mg/d (> 40 kg) - 0.04mg/kg/d (maximum average daily dose: 1.57 mg) vs. placebo x 4 weeks, followed by 1 week placebo substitution, another week of placebo, and 4 weeks follow-up
OutcomesBPRS, CGI, ARC, CRS, SPQ, SAIC, CMAS, SRQ (no distinction made between primary and secondary outcomes) Data estimation: Unclear
NotesINDUSTRY SUPPORT Industry funded: No Any of the authors work for industry: No ADDITIONAL INFORMATION Drop-out rates: No data on dropouts provided Quality rating score: 9
Wagner 2004
MethodsDESIGN Description: Double-blind, randomised, placebo-controlled parallel trial, with taper and +-14 day follow-up BLINDING Participants: Yes Assessors: Yes Administrators: Yes ALLOCATION CONCEALMENT Method: Unclear RANDOMISATION Method: Computer-generated randomization list
ParticipantsSAMPLE Description: 322 DSM-IV SP, 91 of ITT sample (319) children (8-11 years) and 228 adolescences (12-17 years), average age: 13.2 years, 160 (50.2%) male, most common comorbid condition: specific phobia (24.8%) and GAD (23.5%), baseline severity on LSAS-CA: paroxetine (77.6), placebo (77.7) SCREENING Primary diagnosis: ADIS Comorbidity: Unclear
InterventionsDescription: paroxetine 10 mg/d - 50 mg/d (average endpoint dosage: 32.6 mg/d) versus placebo x 16 weeks
OutcomesPrimary outcomes: CGI-I Secondary outcomes: CGI-S, LSAS-CA, SPAI-C, SPAI-A, K-GSADS-A, GAF Data estimation: LOCF
NotesINDUSTRY SUPPORT Industry funded: Yes Any of the authors work for industry: Yes ADDITIONAL INFORMATION Drop-out rates: 41 (24.8%) on paroxetine and 53 (33.8%) on placebo Quality rating score: 30
Walkup 2008
  1. Acronyms for scales: ADIS-C: Anxiety Disorders Interview Schedule for Children; ADISC-R: Anxiety Disorders Interviews Schedule for Children - Revised; ARC: Anxiety Rating for Children scale; BPRS: Brief Psychiatric Rating Scales; C-KIDDIE-SADS GAD: Columbia-Kiddie Schedule for Affective Disorders and Schizophrenia GAD subsection; CDI: Children's Depression Inventory; CDRS-R: Children's Depression Rating Scale-Revised; CGAS: Children's Global Assessment Scale; CGI-S: Clinical Global Impression of Severity; CGI-I Clinical Global Impression of Improvement; COIS-P: Child Obsessive Compulsive Impact Scale; CMAS: Children's Manifest Anxiety Scale; CPQ: Connors' Parents Questionnaire; CPRS: Comprehensive Psychopathological Rating Scale; CRS: Connors Rating Scales; CTQ: Connors' Teachers Questionniare; CY-BOCS: Children's Yale-Brown Obsessive Compulsive Scale; DICA: Diagnostic Interview for Children and Adolesecents; DISC: Diagnostic Interview Schedule for Children; FPH: Family Psychiatric History;GAF: Global Assessment of Functioning; HAM-D Hamilton Depressional scale; STESS: Subjective Treatment Emergent Symptoms Scale; K-GSADS-A: Kutcher Generalized Social Anxiety Disorder Scale for Adolescents; K-SADS: Schedule for Affective Disorders and Schizophrenia; K-SADS-SCPL: Schedule for Affective Disorders and Schizophrenia for School-Age Children-Present and Lifetime Version; LOI: Leyton Obsessional Inventory; LOI-CV: Leyton Obsessional Inventory-Child version; LSAS-CA: Liebowitz Social Anxiety Scale for Children and Adolescents; MASC: Multidimensional Anxiety Scale for Children; NIMH-GS: National Institute of Mental Health - Global Scales; NIMH-GOCS: National Insititute of Mental Health Obsessive Compulsive Rating Scale;NIMH-SS: National Institute of Mental Health Self-rating Scale; OCD-IS: OCD Impact Scale; OCR: Obsessive Compulsive Rating; PARIS: Parent as Respondent Informant Schedule; PARS: Pediatric Anxiety Rating Scale; PGI: Patients Global Impressions; SADS: Schedule for affective disorders and Schitzophrenia; SAIC: State-Trait Anxiety Inventory for Children; SATEE: Systematic Assessment for Treatment Emergent Events; SCARED: Screen for Child Anxiety Related Emotional Disorders (SCARED-P: parents version; SCARED-C: child version); SEFCA: Side Effects form for Children and Adolescents; SERS: Side Effects Rating Scale; SPAI-C: Social Phobia and Anxiety Inventory for Children; SPQ: Sleep Pattern Questionnaire, SRQ: Self-Report Questionnaire

MethodsDESIGN Description: Double-blind, randomised, placebo-controlled parallel trial, with a 6 month open-label continuation phase for responders BLINDING Participants: Yes Assessors: Yes Administrators: Yes ALLOCATION CONCEALMENT Method: Allocation sequence maintained at central pharmacy RANDOMISATION Method: Computer-generated randomisation list
ParticipantsSAMPLE Description (for full sample): 488 DSM-IV-TR SP, SAD or GAD, 362 of ITT sample (488) children (7-12 years) and 126 adolescences (13-17 years), average age: 10.7 years, 246 (50.4%) male, most common comorbid condition: other internalising disorders (including anxiety disorders and dysthymia) specific phobia (43.6%), baseline severity on PARS: sertraline (18.8), placebo (19.6) SCREENING Primary diagnosis: ADIS Comorbidity: Unclear
InterventionsDescription: sertraline 25 - 200 mg/d (mean dose at final visit: 133.7 mg/d) versus placebo 50-200 mg/d (mean dose at final visit: 175 mg/d) versus CBT (14 60-minute sessions based on Coping Cat protocol) versus combination of sertraline and CBT x 12 weeks
OutcomesPrimary outcomes: CGI-I, PARS Secondary outcomes: CGI-S, CGAS Data estimation: MM, LOCF
NotesINDUSTRY SUPPORT Industry funded: No (medication provided by industry) Any of the authors work for industry: No ADDITIONAL INFORMATION Drop-out rates: 23 (17.3%) on sertraline and 15 (19.7%) on placebo Quality rating score: 40

Characteristics of excluded studies [ordered by study ID]

Abikoff 2005Participants diagnosed with ADHD and comorbid anxiety disorders. Course of methylphenidate blindly augmented with fluvoxamine.
Alibeu 1990Patients treated for post-operative agitation with Aconite, a homeopathic treatment
Asbahr 2005Did not include a placebo control group
Berney 1981School refusal not a DSM diagnosis
Bernstein 1990Participants were not randomly allocated to treatment groups, and include patients suffering from depression alone
Bernstein 2000Efficacy of imipramine determine concurrently with the administration of CBT. School refusal not a DSM diagnosis
Cohen 2007Placebo-controlled medication augmentation of trauma-focused CBT
Cremonesi 1974Trial published prior to publication of DSM-III
Dahlstrom 1973Trial published prior to publication of DSM-III
de Haan 1998Comparison of behaviour therapy versus open-label pharmacotherapy. Trial did not include a placebo control.
Diamond 1999Patients were diagnosed with ADHD and comorbid anxiety
Fatovich 1995Patients treated with oral midazolam and buffered lidocaine for anxiety associated with suturing lacerations
Feld 1989Oral transmucosal fentanyl citrate (OTFC) was tested in this RCT as a pre-anaesthetic medication. Anxiety and separation scores were included in the outcome measures.
Gadow 2002Patients with comorbid ADHD and tic disorders
Gittelman-Klein 1971School refusal not a DSM diagnosis
Hennes 1990Patients not diagnosed with DSM anxiety disorder
Hollander 2003Separate outcome data not provided for adolescent (16-18 years) proportion of sample
Kain 1999Patients not diagnosed with DSM anxiety disorder
Klein 1992Imipramine administered to SAD patients who were resistant to an open-label trial of behavioural therapy. Medication was administered concurrently with psychotherapy.
Kurlan 1993Children diagnosed with Tourettes.
Lader 1970Trial conducted prior to publication of DSM-III
Lal 2001Patients not diagnosed with DSM anxiety disorder
Layne 2003School refusal not DSM diagnosis. Concurrent treatment with CBT
Leonard 1989Lack of placebo control group
Leonard 1991Medication substitution study
Liacouras 1998Patients not diagnosed with DSM anxiety disorder
Lindsay 1985Patients not diagnosed with DSM anxiety disorder
Ljungman 2000Patients not diagnosed with DSM anxiety disorder
Lustig 2002Report on establishing a psychopharmacology trial on an inpatient paediatric psychiatry service. Trial itself never implemented.
March 1990Data reported from subset of sample used in DeVeaugh-Geiss 1992
Neziroglu 1999Comparison of combination medication/behaviour therapy treatment with medication alone. No placebo control group included in trial
Procter 2001One of the 3 participants was taking a concurrent course of risperidone
Robert 1999To be included in a Cochrane review of pharmacotherapy for the prevention of PTSD
Roelofse 1993Patients not diagnosed with DSM anxiety disorder
Romano 2001Separate outcome data not provided for adolescent (14-18 years) proportion of sample
Shane 1994Patients not diagnosed with DSM anxiety disorder
Sherwin 2000Patients not diagnosed with DSM anxiety disorder
Steiner 2007Trial does not include a placebo control
Tannock 1995Participants diagnosed with ADHD and comorbid anxiety disorders
Theroux 1993Patients not diagnosed with DSM anxiety disorder
Walker 1996Patients not diagnosed with DSM anxiety disorder
Weir 2000Participants diagnosed with comorbid OCD and tic disorders

Characteristics of studies awaiting assessment [ordered by study ID]

Bouvard 1996
MethodsDouble-blind study comparing buspirone to placebo over six weeks.
Participants42 adolescents (aged 15-18 years) meeting DSM-IIIR criteria for GAD
InterventionsBuspirone (30mg) per day vs placebo
Outcomes'Improvement', adverse effects, tolerability
Notes 
Carlson 1999
MethodsDouble-blind, placebo-controlled trial of sertraline within a replicated multiple baseline/across participants research design, 4 & 20 week follow-up
Participants5 children diagnosed with selective mutism
Interventions16 sertraline versus placebo
Outcomesmeasures assessing mutism, anxiousness, and shyness
Notes 

Characteristics of ongoing studies [ordered by study ID]

March 2006
Trial name or titleChild and Adolescent Anxiety Multisite Study (CAMS)
MethodsRandomised, parallel, short-term trial, assessed blindly by independent evaluators, 6 maintenance for responders
Participants318 DSM-IV separation anxiety, social phobia, and/or generalized anxiety disorder
Interventions12 weeks CBT versus fluvoxamine versus combination CBT/fluvoxamine
Outcomescomprehensive parent, child, clinician, and teacher reports
Starting date21 September, 2002
Contact informationJohn March jsmarch@acpub.duke.edu
Notes 
NCT00018057
Trial name or titleClinical Trial of Fluoxetine in Anxiety and Depression in Children, and Associated Brain Changes
MethodsfMRI imaging study including a randomised placebo-controlled pharmacotherapy component
ParticipantsTotal sample of 630 children, adolescents, and adults will include 150 juveniles with only a current anxiety disorder and 150 juveniles with no psychiatric disorder. Children will be diagnosed on the K-SADS with social phobia, separation anxiety or generalized anxiety disorder, will remain symptomatic after two weekly sessions of talk therapy and will require a score greater than 9 on PARS and a score less than 60 on the CGAS.
Interventions8 week fluoxetine versus CBT versus interpersonal therapy (based on patient choice) or random assignment to placebo or fluoxetine
OutcomesVerbal and written symptom ratings
Starting dateJune 2001
Contact informationAllison M. Detloff tel: (301) 451-6817 detloffa@mail.nih.gov
NotesSubjects may elect to receive attention-retraining as part of their treatment. Patients who have not improved by the end of the study will be offered other treatment for 1 to 3 months, and the clinicians will help with finding subsequent aftercare. Adolescent subjects will be re-tested after eight-weeks using only the attention, memory, and conditioning paradigms.
NCT00382291
Trial name or titleEffectiveness of Sertraline and Cognitive Behavioral Therapy in Treating Pediatric Obsessive-Compulsive Disorder
MethodsRandomized, double blind, placebo-controlled, parallel trial
ParticipantsPrincipal diagnosis of OCD for minimum of 6 months. OCD diagnosed according to clinical interview
Interventions18 weeks sertraline versus placebo
OutcomesPrimary: TE-ASAP and existing behavioral measures of irritability, impulsivity/aggression, restlessness, and mania
Starting dateFebruary 2008
Contact informationPam Allen, RN tel: 352-392-8373 pallen@psychiatry.ufl.edu
NotesParticipants will receive supportive psychotherapy during their first 3 visits. At Visit 4, participants will begin receiving 60-minute CBT sessions, which will continue until the final visit.

Data and analyses

  1. Top of page
  2. Abstract
  3. Background
  4. Objectives
  5. Methods
  6. Results
  7. Discussion
  8. Authors' conclusions
  9. Characteristics of studies
  10. Data and analyses
  11. Appendices
  12. What's new
  13. History
  14. Contributions of authors
  15. Declarations of interest
  16. Sources of support
  17. Differences between protocol and review
  18. Acknowledgements
  19. REFERENCES
Table Comparison 1.. Medication versus Placebo: Primary outcomes
Outcome or subgroup titleNo. of studiesNo. of participantsStatistical methodEffect size
1 Clinical Global Impressions scale - Improvement item (CGI-I)142102Risk Ratio (M-H, Random, 95% CI)1.90 [1.60, 2.26]
1.1 fluoxetine5314Risk Ratio (M-H, Random, 95% CI)2.06 [1.49, 2.85]
1.2 fluvoxamine2248Risk Ratio (M-H, Random, 95% CI)2.40 [1.69, 3.40]
1.3 paroxetine2516Risk Ratio (M-H, Random, 95% CI)1.88 [1.10, 3.22]
1.4 sertraline3418Risk Ratio (M-H, Random, 95% CI)2.16 [1.28, 3.65]
1.5 venlafaxine2606Risk Ratio (M-H, Random, 95% CI)1.46 [1.25, 1.71]
2 Treatment response for OCD on CGI-I5654Risk Ratio (M-H, Random, 95% CI)1.65 [1.32, 2.06]
2.1 fluoxetine2146Risk Ratio (M-H, Random, 95% CI)2.27 [1.35, 3.80]
2.2 fluvoxamine1120Risk Ratio (M-H, Random, 95% CI)1.88 [0.94, 3.76]
2.3 paroxetine1201Risk Ratio (M-H, Random, 95% CI)1.41 [1.00, 1.98]
2.4 sertraline1187Risk Ratio (M-H, Random, 95% CI)1.61 [1.07, 2.43]
3 Treatment response for GAD/SAD/SP on CGI-I91448Risk Ratio (M-H, Random, 95% CI)2.01 [1.59, 2.55]
3.1 fluoxetine3168Risk Ratio (M-H, Random, 95% CI)1.98 [1.25, 3.15]
3.2 fluvoxamine1128Risk Ratio (M-H, Random, 95% CI)2.61 [1.74, 3.90]
3.3 paroxetine1315Risk Ratio (M-H, Random, 95% CI)2.44 [1.91, 3.12]
3.4 sertraline2231Risk Ratio (M-H, Random, 95% CI)3.61 [0.95, 13.74]
3.5 venlafaxine2606Risk Ratio (M-H, Random, 95% CI)1.46 [1.25, 1.71]
4 Symptom severity reduction on DSM-based measures9810Std. Mean Difference (IV, Random, 95% CI)-0.69 [-0.94, -0.44]
4.1 fluoxetine3219Std. Mean Difference (IV, Random, 95% CI)-0.50 [-0.78, -0.23]
4.2 fluvoxamine2244Std. Mean Difference (IV, Random, 95% CI)-0.71 [-1.49, 0.07]
4.3 sertraline3287Std. Mean Difference (IV, Random, 95% CI)-0.80 [-1.39, -0.21]
4.4 clomipramine160Std. Mean Difference (IV, Random, 95% CI)-1.15 [-1.70, -0.60]
5 Symptom severity reduction for OCD on CY-BOCS7765Mean Difference (IV, Random, 95% CI)-4.45 [-5.94, -2.97]
5.1 fluoxetine2146Mean Difference (IV, Random, 95% CI)-5.49 [-8.63, -2.36]
5.2 fluvoxamine1120Mean Difference (IV, Random, 95% CI)-2.70 [-5.76, 0.36]
5.3 paroxetine1196Mean Difference (IV, Random, 95% CI)-3.44 [-5.65, -1.23]
5.4 sertraline2243Mean Difference (IV, Random, 95% CI)-3.82 [-5.83, -1.81]
5.5 clomipramine160Mean Difference (IV, Random, 95% CI)-8.90 [-12.73, -5.07]
6 Symptom severity reduction for SAD, SP and/or GAD on DSM-based measures4428Std. Mean Difference (IV, Random, 95% CI)-0.82 [-1.30, -0.33]
6.1 fluoxetine173Std. Mean Difference (IV, Random, 95% CI)-0.41 [-0.87, 0.06]
6.2 fluvoxamine1124Std. Mean Difference (IV, Random, 95% CI)-1.11 [-1.49, -0.73]
6.3 sertraline2231Std. Mean Difference (IV, Random, 95% CI)-1.06 [-2.43, 0.31]
thumbnail image

Figure Analysis 1.1. Comparison 1 Medication versus Placebo: Primary outcomes, Outcome 1 Clinical Global Impressions scale - Improvement item (CGI-I).

Download figure to PowerPoint

thumbnail image

Figure Analysis 1.2. Comparison 1 Medication versus Placebo: Primary outcomes, Outcome 2 Treatment response for OCD on CGI-I.

Download figure to PowerPoint

thumbnail image

Figure Analysis 1.3. Comparison 1 Medication versus Placebo: Primary outcomes, Outcome 3 Treatment response for GAD/SAD/SP on CGI-I.

Download figure to PowerPoint

thumbnail image

Figure Analysis 1.4. Comparison 1 Medication versus Placebo: Primary outcomes, Outcome 4 Symptom severity reduction on DSM-based measures.

Download figure to PowerPoint

thumbnail image

Figure Analysis 1.5. Comparison 1 Medication versus Placebo: Primary outcomes, Outcome 5 Symptom severity reduction for OCD on CY-BOCS.

Download figure to PowerPoint

thumbnail image

Figure Analysis 1.6. Comparison 1 Medication versus Placebo: Primary outcomes, Outcome 6 Symptom severity reduction for SAD, SP and/or GAD on DSM-based measures.

Download figure to PowerPoint

Table Comparison 2.. Medication versus Placebo: Secondary outcomes
Outcome or subgroup titleNo. of studiesNo. of participantsStatistical methodEffect size
1 Comorbid symptoms of depression4250Std. Mean Difference (IV, Random, 95% CI)-0.29 [-0.88, 0.30]
1.1 fluoxetine2108Std. Mean Difference (IV, Random, 95% CI)0.00 [-0.49, 0.50]
1.2 fluvoxamine1120Std. Mean Difference (IV, Random, 95% CI)0.04 [-0.32, 0.40]
1.3 sertraline122Std. Mean Difference (IV, Random, 95% CI)-1.84 [-2.87, -0.81]
2 Quality of life4390Std. Mean Difference (IV, Random, 95% CI)0.55 [0.34, 0.76]
2.1 fluoxetine3181Std. Mean Difference (IV, Random, 95% CI)0.66 [0.36, 0.96]
2.2 sertraline1209Std. Mean Difference (IV, Random, 95% CI)0.45 [0.17, 0.74]
3 Drop-outs due to adverse effects121997Risk Ratio (M-H, Random, 95% CI)1.91 [1.20, 3.05]
3.1 fluoxetine3305Risk Ratio (M-H, Random, 95% CI)3.42 [1.14, 10.25]
3.2 fluvoxamine2248Risk Ratio (M-H, Random, 95% CI)5.27 [0.90, 30.76]
3.3 paroxetine1322Risk Ratio (M-H, Random, 95% CI)4.28 [0.94, 19.51]
3.4 sertraline3452Risk Ratio (M-H, Random, 95% CI)2.60 [1.10, 6.15]
3.5 clomipramine160Risk Ratio (M-H, Random, 95% CI)2.81 [0.12, 66.40]
3.6 venlafaxine2610Risk Ratio (M-H, Random, 95% CI)0.82 [0.41, 1.63]
4 Relapse2211Risk Ratio (M-H, Random, 95% CI)0.78 [0.55, 1.11]
thumbnail image

Figure Analysis 2.1. Comparison 2 Medication versus Placebo: Secondary outcomes, Outcome 1 Comorbid symptoms of depression.

Download figure to PowerPoint

thumbnail image

Figure Analysis 2.2. Comparison 2 Medication versus Placebo: Secondary outcomes, Outcome 2 Quality of life.

Download figure to PowerPoint

thumbnail image

Figure Analysis 2.3. Comparison 2 Medication versus Placebo: Secondary outcomes, Outcome 3 Drop-outs due to adverse effects.

Download figure to PowerPoint

thumbnail image

Figure Analysis 2.4. Comparison 2 Medication versus Placebo: Secondary outcomes, Outcome 4 Relapse.

Download figure to PowerPoint

Table Comparison 3.. Subgroup analyses - Methodological criteria
Outcome or subgroup titleNo. of studiesNo. of participantsStatistical methodEffect size
1 Single versus multi-centre trials: Treatment response142102Risk Ratio (M-H, Random, 95% CI)1.90 [1.60, 2.26]
1.1 Single centre trials3110Risk Ratio (M-H, Random, 95% CI)2.17 [1.10, 4.28]
1.2 Multi-centre trials111992Risk Ratio (M-H, Random, 95% CI)1.89 [1.57, 2.27]
2 Single versus multi-centre trials on DSM-based measures8788Std. Mean Difference (IV, Fixed, 95% CI)-0.61 [-0.75, -0.46]
2.1 Single centre trials173Std. Mean Difference (IV, Fixed, 95% CI)-0.41 [-0.87, 0.06]
2.2 Multi-centre trials7715Std. Mean Difference (IV, Fixed, 95% CI)-0.63 [-0.78, -0.47]
3 Industry versus non-industry funded trials132087Risk Ratio (M-H, Random, 95% CI)1.90 [1.59, 2.28]
3.1 Industry funded trials91703Risk Ratio (M-H, Random, 95% CI)1.81 [1.49, 2.20]
3.2 Non-industry funded trials4384Risk Ratio (M-H, Random, 95% CI)2.47 [1.50, 4.06]
4 Industry versus non-industry funded trials on DSM-based measure8788Std. Mean Difference (IV, Random, 95% CI)-0.63 [-0.86, -0.40]
4.1 Industry funded trials6506Std. Mean Difference (IV, Random, 95% CI)-0.71 [-1.01, -0.41]
4.2 Non-industry funded trials2282Std. Mean Difference (IV, Random, 95% CI)-0.44 [-0.68, -0.19]
thumbnail image

Figure Analysis 3.1. Comparison 3 Subgroup analyses - Methodological criteria, Outcome 1 Single versus multi-centre trials: Treatment response.

Download figure to PowerPoint

thumbnail image

Figure Analysis 3.2. Comparison 3 Subgroup analyses - Methodological criteria, Outcome 2 Single versus multi-centre trials on DSM-based measures.

Download figure to PowerPoint

thumbnail image

Figure Analysis 3.3. Comparison 3 Subgroup analyses - Methodological criteria, Outcome 3 Industry versus non-industry funded trials.

Download figure to PowerPoint

thumbnail image

Figure Analysis 3.4. Comparison 3 Subgroup analyses - Methodological criteria, Outcome 4 Industry versus non-industry funded trials on DSM-based measure.

Download figure to PowerPoint

Table Comparison 4.. Subgroup analyses - Clinical criteria
Outcome or subgroup titleNo. of studiesNo. of participantsStatistical methodEffect size
1 Inclusion of major depression vs. non-inclusion: CGI-I111626Risk Ratio (M-H, Random, 95% CI)1.89 [1.56, 2.29]
1.1 Trials including patients with major depressive disorder2230Risk Ratio (M-H, Random, 95% CI)1.66 [1.16, 2.37]
1.2 Trials excluding patients with major depressive disorder91396Risk Ratio (M-H, Random, 95% CI)1.95 [1.55, 2.45]
2 Inclusion of major depression vs. non-inclusion: Symptom severity on DSM-based measure6591Std. Mean Difference (IV, Fixed, 95% CI)-0.57 [-0.74, -0.41]
2.1 Trials including patients with major depressive disorder143Std. Mean Difference (IV, Fixed, 95% CI)-0.37 [-0.97, 0.23]
2.2 Trials excluding patients with major depressive disorder5548Std. Mean Difference (IV, Fixed, 95% CI)-0.59 [-0.77, -0.42]
thumbnail image

Figure Analysis 4.1. Comparison 4 Subgroup analyses - Clinical criteria, Outcome 1 Inclusion of major depression vs. non-inclusion: CGI-I.

Download figure to PowerPoint

thumbnail image

Figure Analysis 4.2. Comparison 4 Subgroup analyses - Clinical criteria, Outcome 2 Inclusion of major depression vs. non-inclusion: Symptom severity on DSM-based measure.

Download figure to PowerPoint

Table Comparison 5.. Sensitivity analysis
Outcome or subgroup titleNo. of studiesNo. of participantsStatistical methodEffect size
1 "Worst case" lost-to-follow-up analysis2524Risk Ratio (M-H, Random, 95% CI)1.82 [1.04, 3.17]
2 "Best case" lost-to-follow-up analysis2521Risk Ratio (M-H, Random, 95% CI)1.91 [1.12, 3.28]
thumbnail image

Figure Analysis 5.1. Comparison 5 Sensitivity analysis, Outcome 1 "Worst case" lost-to-follow-up analysis.

Download figure to PowerPoint

thumbnail image

Figure Analysis 5.2. Comparison 5 Sensitivity analysis, Outcome 2 "Best case" lost-to-follow-up analysis.

Download figure to PowerPoint

Appendices

  1. Top of page
  2. Abstract
  3. Background
  4. Objectives
  5. Methods
  6. Results
  7. Discussion
  8. Authors' conclusions
  9. Characteristics of studies
  10. Data and analyses
  11. Appendices
  12. What's new
  13. History
  14. Contributions of authors
  15. Declarations of interest
  16. Sources of support
  17. Differences between protocol and review
  18. Acknowledgements
  19. REFERENCES

Appendix 1. Search strategy for electronic databases

PubMed search strategy:

(randomized controlled trial [pt] OR controlled clinical trial [pt] OR randomized controlled trials [mh] OR random allocation [mh] OR double-blind method [mh] OR single-blind method [mh] OR clinical trial[pt] OR clinical trials [mh] OR ("clinical trial" [tw]) OR ((singl* [tw] OR doubl* [tw] OR trebl* [tw] OR tripl* [tw]) AND (mask* [tw] OR blind* [tw])) OR ("latin square" [tw]) OR random* [tw] OR research design [mh:noexp] OR comparative study [pt] OR evaluation studies [pt] OR follow-up studies [mh] OR prospective studies [mh] OR cross-over studies [mh] OR control* [tw] OR prospectiv* [tw] OR volunteer* [tw]) NOT (animal [mh] NOT human [mh]) AND ((placebos [mh] OR placebo* [tw]) AND (anxiety disorder [mh:noexp] OR "anxiety disorder" [tw] OR phobic disorders [mh:noexp] OR obsessive-compulsive disorder [mh] OR "obsessive-compulsive" [tw] OR stress disorders, post-traumatic [mh:noexp] OR "post-traumatic" [tw]) AND ("paediatric" [tw] OR "pediatric" [tw] OR child*[tw] OR child [mh] OR adolesc* [tw] OR adolescent [mh]))

EMBASE search strategy:

(random OR "placebo":de. OR "double blind") AND (adolescent:de OR adolesc* OR child:de OR child* OR "paediatric" OR "pediatric") AND ("anxiety disorder":de OR "phobia" OR "obsessive compulsive" OR "OCD" OR "post-traumatic" OR PTSD)

PsycINFO search strategy:

("randomisation" OR "randomization") OR "controlled" AND ("drug" OR "medication") AND ("child*" OR "adolesc*" ) AND "anxiety".

Appendix 2. Frequent treatment-emergent adverse events in medication group  

MedicationDisorderStudySide effects*Final dosage
Alprazolamoveranxious & avoidant disordersSimeon 1992No info0.25mg/d (< 40kg) or 0.50mg/d (> 40 kg) - 0.04mg/kg/d (maximum average daily dose: 1.57 mg)
ClonazepamVariousGraae 1994No info0.25mg/d - 2mg/d
ClomipramineOCDDeVeaugh 1992dry mouth, somnolence, dizziness, fatigue, tremor, constipation, anorexia75 - 200 mg/d (or 3 mg/kg). Max. dose: 75 mg/d (25-30 kg), 100 mg/d (31-40 kg), 150 mg/d (46-60kg), 200 mg/d (>60kg)
 OCDFlament 1985tremor, dry mouth, dizziness, constipation and sweating100 - 200 mg/d (mean dose: 141 mg/d)
 OCDRapoport 1980No info50 mg/d - 150 mg/d
FluoxetineGAD/SAD/SPBirmaher 2003abdominal pain and nausea10 - 20 mg/d
 OCDGeller 2001aNone20 - 60 mg/d (mean daily dose: 24.6 mg/d)
 OCDLiebowitz 2002apalpitations, weight-loss, decreased appetite20 - 80 mg/d (mean: 64.8mg/d)
 SPBeidel 2007nausea**40 mg/d
 OCDRiddle 1992None20 mg/d
FluvoxamineOCDRiddle 2001insomnia, hyperkinesia and somnolence50  - 200 mg/d (mean for children (8-12): 157 mg/d, for adolescents (13-17): 170 mg/d)
 OCDRUPPASG 2001abdominal discomfort, increased motor activity, vomiting, tiredness/fatigue, muscle/joint pain, decreased appetite 300 mg/d max. dose for adolescents, 250 mg/d for children ( mean: 4 mg/kg/d)
ParoxetineOCDGeller 2004trauma, decreased appetite, hostility, diarrhoea, asthenia10 50 mg/d paroxetine (mean: 30.1 mg/d; children: 25.4 mg/d; adolescents: 36.5 mg/d)
 SPWagner 2004asthenia, insomnia, decreased appetite, vomiting and conjunctivitis10 50 mg/d (mean: 26.5 mg/d - children, 35 mg/d - adolescents)
 GADRynn 2001None50mg/d
SertralineGAD, SAD, SPWalkup 2008None25 - 200 mg/d (mean: 133.7 mg/d)
 OCDMarch 1998insomnia, nausea, agitation and tremor**25 - 200 mg/d (mean: 167mg/d)
 OCDPOTS 2004decreased appetite, diarrhea, nausea, stomache-ache25 - 200 mg/d  (mean: 170 mg/d)
 GADRynn 2007anorexia37.5 - 112.5 mg/d (25-39 kg), 37.5 - 150 mg/d (40-49 kg), 37.5 - 225 mg/d (>50kg)
VenlafaxineSPMarch 2007asthenia, anorexia, nausea, weight loss, abnormal/changed behaviour, pharyngitis, mydriasis**37.5 - 112.5 mg/d (25-33 kg), 37.5 - 150 mg/d (34-49 kg), 37.5 mg/d - 225 mg/d (>50kg); mean dose for <= 112 days (141.5 mg/d), mean dose for > 112 days (155 mg/d)

* p < 0.1

** reported as statistically significant by authors

What's new

  1. Top of page
  2. Abstract
  3. Background
  4. Objectives
  5. Methods
  6. Results
  7. Discussion
  8. Authors' conclusions
  9. Characteristics of studies
  10. Data and analyses
  11. Appendices
  12. What's new
  13. History
  14. Contributions of authors
  15. Declarations of interest
  16. Sources of support
  17. Differences between protocol and review
  18. Acknowledgements
  19. REFERENCES

Last assessed as up-to-date: 7 August 2008.

11 May 2010AmendedMinor corrections have been made to the review for Issue 6, 2010, as follows: hyperlinks to the primary outcome have been added / corrected; data have been added to the table for the Bouvard 1996 'Ongoing Study'; the choice of effects model for the forest plot for severity of symptoms for non-OCD studies has been changed to match the (already correct) text.

History

  1. Top of page
  2. Abstract
  3. Background
  4. Objectives
  5. Methods
  6. Results
  7. Discussion
  8. Authors' conclusions
  9. Characteristics of studies
  10. Data and analyses
  11. Appendices
  12. What's new
  13. History
  14. Contributions of authors
  15. Declarations of interest
  16. Sources of support
  17. Differences between protocol and review
  18. Acknowledgements
  19. REFERENCES

Protocol first published: Issue 1, 2005

Review first published: Issue 3, 2009

28 April 2009New citation required and major changesSubstantive amendment
23 October 2008AmendedConverted to new review format.

Contributions of authors

  1. Top of page
  2. Abstract
  3. Background
  4. Objectives
  5. Methods
  6. Results
  7. Discussion
  8. Authors' conclusions
  9. Characteristics of studies
  10. Data and analyses
  11. Appendices
  12. What's new
  13. History
  14. Contributions of authors
  15. Declarations of interest
  16. Sources of support
  17. Differences between protocol and review
  18. Acknowledgements
  19. REFERENCES

Sue Hawkridge compiled the original protocol, and provided feedback on the draft version of the review. Dan Stein coordinated the work on the protocol and review, and provided feedback for both. Jonathan Ipser revised the protocol, provided methodological support for the review, wrote the draft version of the review and responded to editorial feedback. Lara Hoppe provided assistance with the writing of the review.

Declarations of interest

  1. Top of page
  2. Abstract
  3. Background
  4. Objectives
  5. Methods
  6. Results
  7. Discussion
  8. Authors' conclusions
  9. Characteristics of studies
  10. Data and analyses
  11. Appendices
  12. What's new
  13. History
  14. Contributions of authors
  15. Declarations of interest
  16. Sources of support
  17. Differences between protocol and review
  18. Acknowledgements
  19. REFERENCES

The MRC Anxiety and Stress Disorders Research Unit has received funding from almost all pharmaceutical companies involved with psychiatry in South Africa.

Potential conflicts of interest for individual reviewers

Jonathan Ipser has no known conflicts of interest outside of his employment by the MRC Unit on Anxiety Disorders.

Dan Stein has received research grants and/or consultancy honoraria from Astrazeneca, Eli-Lilly, GlaxoSmithKline, Lundbeck, Orion, Pfizer, Pharmacia, Roche, Servier, Solvay, Sumitomo, and Wyeth. He has participated in a number of ongoing trials, and has presented data from some of these trials on behalf of the sponsoring companies.

Sue Hawkridge has previously received travel/conference grants from Eli Lilly, Lundbeck and Roche, speaker's honoraria from Lundbeck and has participated in clinical trials for Eli Lilly, GlaxoSmithKline and Janssens.

Lara Hoppe has no known conflicts of interest.

Differences between protocol and review

  1. Top of page
  2. Abstract
  3. Background
  4. Objectives
  5. Methods
  6. Results
  7. Discussion
  8. Authors' conclusions
  9. Characteristics of studies
  10. Data and analyses
  11. Appendices
  12. What's new
  13. History
  14. Contributions of authors
  15. Declarations of interest
  16. Sources of support
  17. Differences between protocol and review
  18. Acknowledgements
  19. REFERENCES

The stratification of the primary outcomes by whether they were reported for RCTs of OCD was added to the review, in recognition (a) that this is consistent with how these disorders have been grouped by other researchers (Birmaher 2003; RUPPASG 2001), and (b) that this distinction has the potential to describe a substantial amount of heterogeneity observed in the primary outcomes.

A comparison of the number of patients who relapsed (as defined by the trial investigators), as well as a descriptive review of adverse events following long-term maintenance and treatment discontinuation have been added as secondary outcomes.

The section on studies with multiple treatment groups in the Methods section of the protocol has been omitted from the review. This section described the procedure involved in the analysis of data from RCTs comparing multiple medications as well as different dosages of medication. The only study comparing different medications (Rapoport 1980) did not provide data for inclusion in the comparisons, while there were no dosage-comparison studies that were eligible for inclusion. In the event of studies being identified in future which are configured in this way and have extractable data, advice may be sought from CCDAN statisticians as to the advisability of multiple treatments meta-analysis (MTM, also known as network meta-analysis).

Additional information provided in the protocol on the procedure followed for the analysis of outcome data from cross-over trials which either provided data from the first period of treatment or included a washout period of sufficient duration was omitted from the review, as none of the cross-over trials included in this review satisfied these criteria. The protocol included the following paragraph:

"For cross-over trials in which the washout period was regarded as adequate, data from both periods will be included when it is possible to determine the standard error of the mean difference in response between groups (Elbourne 2002). The summary statistics required to derive the standard error of interest will be obtained from the trial report, or for trials for which this information was missing, and will be imputed through averaging the relevant statistic from other included crossover trials with comparable control conditions." In the event of studies being identified in future with relevant data, these methods or others recommended by CCDAN statisticians will be employed.

Acknowledgements

  1. Top of page
  2. Abstract
  3. Background
  4. Objectives
  5. Methods
  6. Results
  7. Discussion
  8. Authors' conclusions
  9. Characteristics of studies
  10. Data and analyses
  11. Appendices
  12. What's new
  13. History
  14. Contributions of authors
  15. Declarations of interest
  16. Sources of support
  17. Differences between protocol and review
  18. Acknowledgements
  19. REFERENCES

The authors are supported by the MRC Research Unit on Anxiety and Stress Disorders (Cape Town, South Africa). We would like to thank Drs Boris Birmaher, Daniel Geller, Arifulla Khan, John March, Daniel Pine and Benedetto Vitiello for responding to queries for additional trial data. We would also like to acknowledge Wyeth and Pfizer for providing data for RCTs of venlafaxine XR for GAD and sertraline for OCD, respectively. Finally, we must thank CCDAN for the generous support they provided us in conducting this review.

REFERENCES

  1. Top of page
  2. Abstract
  3. Background
  4. Objectives
  5. Methods
  6. Results
  7. Discussion
  8. Authors' conclusions
  9. Characteristics of studies
  10. Data and analyses
  11. Appendices
  12. What's new
  13. History
  14. Contributions of authors
  15. Declarations of interest
  16. Sources of support
  17. Differences between protocol and review
  18. Acknowledgements
  19. REFERENCES
  • References to studies included in this review
  • Beidel 2007 {published data only}
  • * Beidel DC, Turner SM, Sallee FR, Ammerman RT, Crosby LA, Pathak S. SET-C versus fluoxetine in the treatment of childhood social phobia. Journal of the American Academy of Child and Adolescent Psychiatry  2007;46(12):1622-32.
  • Birmaher 2003 {published data only}
  • * Birmaher B, Axelson DA, Monk K, Kalas C, Clark DB, Ehmann M, Bridge J, Heo J, Brent DA. Fluoxetine for the treatment of childhood anxiety disorders. Journal of the American Academy of Child and Adolescent Psychiatry 2003;42(4):415-24.
  • Black 1994 {published data only}
  • * Black B, Uhde TW. Treatment of elective mutism with fluoxetine: A double-blind, placebo-controlled study. Journal of the American Academy of Child and Adolescent Psychiatry 1994;33(7):1000-6.
  • DeVeaugh-Geiss 1992 {published data only}
  • * DeVeaugh-Geiss J, Moroz G, Biederman J, Cantwell D, Fontaine R, Greist JH, Reichler R, Katz R, Landau P. Clomipramine hydrochloride in childhood and adolescent obsessive-compulsive disorder - a multicenter trial. Journal of the American Academy of Child and Adolescent Psychiatry 1992;31(1):45-9.
  • Flament 1985 {published data only}
  • * Flament MF, Rapoport JL, Berg CJ, Sceery W, Kilts C, Mellstrom B, Linnoila M. Clomipramine treatment of childhood obsessive-compulsive disorder. Archives of General Psychiatry 1985;42:977-83.
  • Geller 2001a {published data only}
  • * Geller DA, Hoog SL, Heiligenstein JH, Ricardi Rk, Tamura R, Kluszynski S, Jacobson JG, FPOCDST. Fluoxetine treatment for obsessive-compulsive disorder in children and adolescents: A placebo-controlled clinical trial. Journal of the American Academy of Child and Adolescent Psychiatry 2001;40(7):773-9.
  • Geller 2001b {published data only}
  • * Geller DA, Biederman J, Stewart SE, Mullin B, Farrell C, Wagner KD, Emslie G, Carpenter D. Impact of comorbidity on treatment response to paroxetine in pediatric obsessive-compulsive disorder: Is the use of exclusion criteria empirically supported in randomized clinical trials?. Journal of Child and Adolescent Psychopharmacology 2003;13(Suppl 1):S19-29.
  • Geller 2004 {published data only}
  • * Geller DA, Wagner KD, Emslie G, Murphy T, Carpenter DJ, Wetherhold E, Perera P, Machin A, Gardiner C. Paroxetine treatment in children and adolescents with obsessive-compulsive disorder: A randomized, multicenter, double-blind, placebo-controlled trial. Journal of the American Academy of Child and Adolescent Psychiatry 2004;43(11):1387-96.
  • Graae 1994 {published data only}
  • * Graae F, Milner J, Rizzotto L, Klein RG. Clonazepam in childhood anxiety disorders. Journal of the American Academy of Child and Adolescent Psychiatry 1994;33(3):372-6.
  • Liebowitz 2002 {published data only}
  • * Liebowitz MR, Turner SM, Piacentini J, Beidel DC, Clarvit SR, Davies SO, Graae F, Jaffer M, Lin S, Sallee FR, Schmidt AB, Simpson HB. Fluoxetine in children and adolescents with OCD: A placebo-controlled trial. Journal of the American Academy of Child and Adolescent Psychiatry 2002;41(12):1431-8.
  • March 1998 {published data only}
  • Cook EH, Wagner KD, March JS, Biederman J, Landau P, Wolkow R, Messig M. Long-term sertraline treatment of children and adolescents with obsessive-compulsive disorder. J Am Acad Child Adolesc Psychiatry 2001;40(10):1175-81.
  • * March JS, Biederman J, Wolkow R, Safferman A, Mardekian J, Cook EH, Cutler NR, Dominguez R, Ferguson J, Muller B, Riesenberg R, Rosenthal M, Sallee FR, Steiner H, Wagner KD. Sertraline in children and adolescents with obsessive-compulsive disorder: A multicenter randomized controlled trial. Journal of the American Medical Association 1998;280(20):1752-6.
  • Wagner KD, Cook EH, Chung H, Messig M. Remission status after long-term sertraline treatment of pediatric obsessive-compulsive disorder. Journal of Child and Adolescent Psychopharmacology 2003;13(Suppl. 1):S53-60.
  • March 2007 {published data only}
  • * March JS, Entusah RA, Rynn M, Albano AM, Tourian KA. A randomized controlled trial of venlafaxine ER versus placebo in pediatric social anxiety disorder. Biological Psychiatry 2007;62(10):1149-54.
  • POTS 2004 {published data only}
  • * The Pediatric OCD Treatment Study (POTS) team. Cognitive-behavior therapy, sertraline, and their combination for children and adolescents with obsessive-compulsive disorder. Journal of the American Medical Association 2004;292(16):1969-76.
  • Rapoport 1980 {published data only}
  • * Rapoport J, Elkins R, Mikkelsen E. Clinical controlled trial of chlorimipramine in adolescents with obsessive-compulsive disorder. Psychopharmacology Bulletin 1980;16(3):61-3.
  • Riddle 1992 {published data only}
  • * Riddle MA, Scahill L, King RA, Hardin MT, Anderson GM, Ort SI, Smith JC, Leckman JF, Cohen DJ. Double-blind, crossover trial of fluoxetine and placebo in children and adolescents with obsessive-compulsive disorder. Journal of the American Academy of Child and Adolescent Psychiatry 1992;31(6):1062-9.
  • Riddle 2001 {published data only}
  • * Riddle MA, Reeve EA, Yaryura-Tobias JA, Yang HM, Claghorn JL, Gaffney G, Greist JH, Holland D, McConville BJ, Pigott T, Walkup JT. Fluvoxamine for children and adolescents with obsessive-compulsive disorder: A randomized, controlled, multicenter trial. Journal of the American Academy of Child and Adolescent Psychiatry 2001;40(2):222-9.
  • RUPPASG 2001 {published data only}
  • Ginsburg GS, Riddle MA, Davies M. Somatic symptoms in children and adolescents with anxiety disorders. Journal of the American Academy of Child and Adolescent Psychiatry 2006;45(10):1179-87.
  • * The Research Unit on Pediatric Psychopharmacology Anxiety Study Group. Fluvoxamine for the treatment of anxiety disorders in children and adolescents. New England Journal of Medicine 2001;344(17):1279-85.
  • Walkup JT, Labellarte MJ, Riddle MA, Pine D, Greenhill L, Klein R, Davies M, Sweeney M, Fu C, Abikoff H, Hack S, Klee B, McCracken J, Bergman L, Piacentini J, March J, Compton S, Robinson J, O'Hara T, Baker S, Vitiello B, Ritz L, Roper M, Research Units on Pediatric Psychopharmacology Anxiety Study Group. Searching for moderators and mediators of pharmacological treatment effects in children and adolescents with anxiety disorders. Journal of the American Academy of Child and Adolescent Psychiatry 2003;42(1):13-21.
  • Rynn 2001 {published data only}
  • * Rynn MA, Siqueland L, Rickels K. Placebo-controlled trial of sertraline in the treatment of children with generalized anxiety disorder. American Journal of Psychiatry 2001;158(12):2008-14.
  • Rynn 2007 {published data only}
  • Kunz NR, Khan A,  Lamm LW,  Nicolacopoulos E,  Jenkins L. Efficacy and safety of venlafaxine extended release in children and adolescents with generalised anxiety disorder. European Neuropsychopharmacology 2002;12(3):358.
  • * Rynn MA, Riddle MA, Yeung PP, Kunz NR. Efficacy and safety of extended-release venlafaxine in the treatment of generalized anxiety disorder in children and adolescents: Two placebo-controlled trials. The American Journal of Psychiatry 2007;164(2):290-300.
  • Simeon 1992 {published data only}
  • * Simeon JG, Ferguson B, Knott V, Roberts N, Gauthier B, Dubois C, Wiggins D. Clinical, cognitive, and neurophysiological effects of alprazolam in children and adolescents with overanxious and avoidant disorders. Journal of the American Academy of Child and Adolescent Psychiatry 1992;31(1):29-33.
  • Wagner 2004 {published data only}
  • * Wagner KD, Berard R, Stein MB, Wetherhold E, Carpenter DJ, Perera P, Gee M, Davy K, Machin A. A multicenter, randomized, double-blind, placebo-controlled trial of paroxetine in children and adolescents with social anxiety disorder. Archives of General Psychiatry 2004;61:1153-62.
  • Walkup 2008 {published data only}
  • * Walkup JT, Albano AM, Piacentini J, Birmaher B, Compton SN, Sherrill JT, Ginsburg GS, Rynn MA, McCracken J, Waslick B, Iyengar S, March JS, Kendall PC. Cognitive behavioral therapy, sertraline, or a combination in childhood anxiety. New England Journal of Medicine 2008;359(26):2753-66.
  • References to studies excluded from this review
  • Abikoff 2005 {published data only}
  • * Abikoff H, McGough J, Vitiello B, McCracken J, Davies M, Walkup J, Riddle M, Oatis M, Greenhill L, Skrobala A, March J, Gammon P, Robinson J, Lazell R, McMahon DJ, Ritz L. Sequential pharmacotherapy for children with comorbid attention-deficit/hyperactivity and anxiety disorders. Journal of the American Academy of Child and Adolescent Psychiatry 2005;44(5):418-27.
  • Alibeu 1990 {published data only}
  • * Alibeu JP, Jobert J. Aconite in homeopathic relief of post-operative pain and agitation in children. Pediatrie 1990;45(7-8):465-6.
  • Asbahr 2005 {published data only}
  • * Asbahr FR, Castillo AR, Ito LM, Latorre MR, Moreira MN, Lotufo-Neto F. Group cognitive-behavioral therapy versus sertraline for the treatment of children and adolescents with obsessive-compulsive disorder. Journal of the American Academy of Child and Adolescent Psychiatry 2005;44(11):1128-36.
  • Berney 1981 {published data only}
  • * Berney T, Kolvin I, Bhate SR, Garside RF, Jeans J, Kay B, Scarth L. School phobia: A therapeutic trial with clomipramine and short term outcome. British Journal of Psychiatry 1981;138:110-18.
  • Bernstein 1990 {published data only}
  • * Bernstein G, Garfinkel BD, Borchardt CM. Comparative studies of pharmacotherapy for school refusal. Journal of the American Academy of Child and Adolescent Psychiatry 1990;29(5):773-81.
  • Bernstein 2000 {published data only}
  • * Bernstein GA, Borchardt CM, Perwien AR, Crosby RD, Kushner MG, Thuras PD, Last CG. Imipramine plus cognitive-behavioural therapy in the treatment of school refusal. Journal of the American Academy of Child and Adolescent Psychiatry 2000;39(3):276-83.
  • Cohen 2007 {published data only}
  • * Cohen JA, Mannarino AP, Perel JM, Staron V. A pilot randomized controlled trial of combined trauma-focused CBT and sertraline for childhood PTSD symptoms. Journal of the American Academy of Child and Adolescent Psychiatry 2007;46(7):811-9.
  • Cremonesi 1974 {published data only}
  • * Cremonesi E, Silva LF, da, Curras JS, Atunes AH, Moraes R, Sartoretto JN. Evaluation of lorazepam as an oral preanaesthetic medication: A comparative double-blind study with diazepam and placebo. Current Medical Research and Opinion 1974;2(4):244-8.
  • Dahlstrom 1973 {published data only}
  • * Dahlstrom H, Fedor Freybergh P, Kareland H, Vahlne L. Benzoctamine and medazepam in treatment of anxiety of children and adolescents: A comparative study. Acta Psychiatrica Scandinavica 1973;49(6):735-43.
  • de Haan 1998 {published data only}
  • * de Haan E, Hoogduin KA, Buitelaar JK, Keijsers GP. Behavior therapy versus clomipramine for the treatment of obsessive-compulsive disorder in children and adolescents. Journal of the American Academy of Child and Adolescent Psychiatry 1998;37(10):1022-9.
  • Diamond 1999 {published data only}
  • * Diamond IR, Tannock R, Schachar RJ. Response to methylphenidate in children with ADHD and comorbid anxiety. Journal of the American Academy of Child and Adolescent Psychiatry 1999;38(4):402-9.
  • Fatovich 1995 {published data only}
  • * Fatovich DM, Jacobs IG. A randomized, controlled trial of oral midazolam and buffered lidocaine for suturing lacerations in children (the SLIC Trial). Annals of Emergency Medicine 1995;25:209-14.
  • Feld 1989 {published data only}
  • * Feld LH, Champeau MW, Van Steennis CA, Scott JC. Preanesthetic medication in children: A comparison of oral transmucosal fentanyl citrate versus placebo. Anesthesiology 1989;71:374-7.
  • Gadow 2002 {published data only}
  • * Gadow KD, Nolan EE, Sverd J, Sprafkin J, Schwartz J. Anxiety and depression symptoms and response to methylphenidate in children with attention-deficit hyperactivity disorder and tic disorder. Journal of Clinical Psychopharmacology 2002;22(3):267-74.
  • Gittelman-Klein 1971 {published data only}
  • * Gittelman-Klein R, Klein DF, Oaks G. Controlled imipramine treatment of school phobia. Archives of General Psychiatry 1971;25:204-7.
  • Hennes 1990 {published data only}
  • * Hennes HM, Wagner V, Bonadio WA, Glaeser PW, Losek JD, Walsh-Kelly CM, Smith DS. The effect of oral midazolam on anxiety of preschool children during laceration repair. Annals of Emergency Medicine 1990;19(9):1006-9.
  • Hollander 2003 {published data only}
  • * Hollander E, Allen A, Steiner M, Wheadon DE, Oakes R, Burnham DB. Acute and long-term treatment and prevention of relapse of obsessive-compulsive disorder with paroxetine. Journal of Clinical Psychiatry 2003;64(9):113-21.
  • Kain 1999 {published data only}
  • * Kain ZN, Mayes LC, Wang SM, Hofstadter MB. Postoperative behavioral outcomes in children: Effects of sedative premedication. Anesthesiology 1999;90(3):758-65.
  • Klein 1992 {published data only}
  • * Klein RG, Koplewicz HS, Kanner A. Imipramine treatment of children with separation anxiety disorder. Journal of the American Academy of Child and Adolescent Psychiatry 1992;31(1):21-8.
  • Kurlan 1993 {published data only}
  • * Kurlan R, Como PG, Deeley C, McDermott M, McDermott MP. A pilot controlled study of fluoxetine for obsessive-compulsive symptoms in children with Tourette's syndrome. Clinical Neuropharmacology 1993;16(2):167-72.
  • Lader 1970 {published data only}
  • * Lader MH, Mathews AM. Comparison of methods of relaxation using physiological measures. Behaviour Research and Therapy 1970;8(4):331-7.
  • Lal 2001 {published data only}
  • * Lal MK, McClelland J, Phillips J, Taub NA, Beattie RM. Comparison of EMLA cream versus placebo in children receiving distraction therapy for venepuncture. Acta Paediatrica 2001;90(2):154-9.
  • Layne 2003 {published data only}
  • * Layne AE, Bernstein GA, Egan EA, Kushner MG. Predictors of treatment response in anxious-depressed adolescents with school refusal. Journal of the American Academy of Child and Adolescent Psychiatry 2003;42(3):319-26.
  • Leonard 1989 {published data only}
  • * Leonard HL, Swedo SE, Rapoport JL, Koby EV, Lenane MC, Cheslow DL, Hamburger SD. Treatment of obsessive-compulsive disorder with clomipramine and desipramine in children and adolescents. Archives of General Psychiatry 1989;46:1088-92.
  • Leonard 1991 {published data only}
  • * Leonard HL, Swedo SE, Lenane MC, Rettew DC, Cheslow DL, Hamburger SD, Rapoport JL. A double-blind desipramine substitution during long-term clomipramine treatment in children and adolescents with obsessive-compulsive disorder. Archives of General Psychiatry 1991;48:922-7.
  • Liacouras 1998 {published data only}
  • * Liacouras CA, Mascarenhas M, Poon C, Wenner WJ. Placebo-controlled trial assessing the use of oral midazolam as a premedication to conscious sedation for pediatric endoscopy. Gastrointestinal Endoscopy 1998;47(6):455-60.
  • Lindsay 1985 {published data only}
  • * Lindsay SJ, Yates JA. The effectiveness of oral diazepam in anxious child dental patients. British Dental Journal 1985;159(5):149-53.
  • Ljungman 2000 {published data only}
  • * Ljungman G, Kreuger A, Andreasson S, Gordh T, Sorensen S. Midazolam nasal spray reduces procedural anxiety in children. Pediatrics 2000;105(1):73-8.
  • Lustig 2002 {published data only}
  • * Lustig SL, Botelho C, Lynch L, Nelson SV, Eichelberger WJ, Vaughan BL. Implementing a randomized clinical trial on a pediatric psychiatric inpatient unit at a children's hospital: The case of clonidine for post-traumatic stress. General Hospital Psychiatry 2002;24(6):422-9.
  • March 1990 {published data only}
  • * March JS, Johnston H, Jefferson JW, Kobak KA, Greist JH. Do subtle neurological impairments predict treatment resistance to clomipramine in children and adolescents with obsessive compulsive disorder?. Journal of Child and Adolescent Psychopharmacology 1990;1(2):133-40.
  • Neziroglu 1999 {published data only}
  • * Neziroglu F, Yaryura-Tobias JA, Walz J, McKay D. The effect of fluvoxamine and behavior therapy on children and adolescents with obsessive-compulsive disorder. Journal of Child and Adolescent Psychopharmacology 2000;10(4):295-306.
  • Procter 2001 {published data only}
  • * Procter E, McNicholas F, Baird G. Low-dose sertraline in children with obsessive-compulsive disorder. Clinical Child Psychology and Psychiatry 2001;6(4):545-50.
  • Robert 1999 {published data only}
  • * Robert R, Blakeney PE, Villarreal C, Rosenberg L, Meyer WJ 3rd. Imipramine treatment in pediatric burn patients with symptoms of acute stress disorder: A pilot study. Journal of the American Academy of Child and Adolescent Psychiatry 1999;38(7):873-82.
  • Roelofse 1993 {published data only}
  • * Roelofse JA, Van der Bijl P. Comparison of rectal midazolam and diazepam for premedication in pediatric dental patients. Journal of Oral and Maxillofacial Surgery 1993;51(5):525-9.
  • Romano 2001 {published data only}
  • * Romano S, Goodman W, Tamura R, Gonzales J, Deltito J, DuPont R, Hertzman M, Burnie G, Jenike M, Landbloom R, Pigott T, Shear K, Stahl S, Stewart R, Zajecka J. Long-term treatment of obsessive-compulsive disorder after an acute response: A comparison of fluoxetine versus placebo. Journal of Clinical Psychopharmacology 2001;21(1):46-52.
  • Shane 1994 {published data only}
  • * Shane SA, Fuchs SM, Khine H. Efficacy of rectal midazolam for the sedation of preschool children undergoing laceration repair. Annals of Emergency Medicine 1994;24:1065-73.
  • Sherwin 2000 {published data only}
  • * Sherwin TS, Green SM, Khan A, Chapman DS, Dannenberg B. Does adjunctive midazolam reduce recovery agitation after ketamine sedation for pediatric procedures? A randomized, double-blind, placebo-controlled trial. Annals of Emergency Medicine 2000;35(3):229-38.
  • Steiner 2007 {published data only}
  • * Steiner H, Saxena KS, Carrion V, Khanzode LA, Silverman M, Chang K. Divalproex sodium for the treatment of PTSD and conduct disordered youth: A pilot randomized controlled clinical trial. Child Psychiatry and Human Development 2007;38(3):183-93.
  • Tannock 1995 {published data only}
  • * Tannock R, Ickowicz A, Schachar R. Differential effects of methylphenidate on working memory in ADHD children with and without comorbid anxiety. Journal of the American Academy of Child and Adolescent Psychiatry 1995;34(7):886-96.
  • Theroux 1993 {published data only}
  • * Theroux MC, West DW, Corddry DH, Hyde PM, Bachrach SJ, Cronan KM, Kettrick RG. Efficacy of intranasal midazolam in facilitating suturing of lacerations in preschool children in the emergency department. Pediatrics 1993;91(3):624-7.
  • Walker 1996 {published data only}
  • * Walker J. The use of oral midazolam in accident and emergency to reduce anxiety in children. Accident and Emergency Nursing 1996;4(3):110-3.
  • Weir 2000 {published data only}
  • * Weir E. Treating obsessive-compulsive and tic disorders. Canadian Medical Association Journal 2000;163(1):82.
  • References to studies awaiting assessment
  • Bouvard 1996 {unpublished data only}
  • Bouvard MP, Braconnier AJ, Dissoubray C. Buspirone in adolescents with anxiety disorders. APA 149th Annual Meeting. 1996.
  • Carlson 1999 {published data only}
  • * Carlson JS, Kratochwill TR, Johnston HF. Sertraline treatment of 5 children diagnosed with selective mutism: A single-case research trial. Journal of Child and Adolescent Psychopharmacology 1999;9(4):293-306.
  • References to ongoing studies
  • March 2006 {unpublished data only}
  • * March JS. Child and Adolescent Anxiety Multisite Study (CAMS). CRISP (http://crisp.cit.nih.gov/crisp/CRISP_LIB.getdoc?textkey=6893719&p_grant_num=5U01MH064107-04&p_query=(anxiety+%26+children)&ticket=16509556&p_audit_session_id=77879437&p_audit_score=21&p_audit_numfound=30&p_keywords=anxiety+children). [: Grant Number: 5U01MH064107-04]
  • NCT00018057 {published data only}
  • * Detloff AM, Pine DS. Clinical Trial of Fluoxetine in Anxiety and Depression in Children, and Associated Brain Changes. www.controlled-trials.com (http://clinicaltrials.gov/show/NCT00018057) 2001. [: ClinicalTrials.gov identifier: NCT00018057]
  • NCT00382291 {published data only}
  • * Murphy TK. Effectiveness of Sertraline and Cognitive Behavioral Therapy in Treating Pediatric Obsessive-Compulsive Disorder. www.controlled-trials.com 2006. [: ClinicalTrials.gov identifier: NCT00382291]
  • Additional references
  • Alderson 2003
  • Alderson P, Green S, Higgins JP. Assessment of study quality. In: AldersonP, GreenS, HigginsJPT editor(s). Cochrane Reviewers' Handbook 4.2.1 [updated December] Section 8;. Chichester, UK: John Wiley & Sons, Ltd., 2004.
  • Als-Nielsen 2003
  • Als-Nielsen B, Chen W, Gluud C, Kjaergard LL. Association of funding and conclusions in randomized drug trials: A reflection of treatment effect or adverse events?. Journal of the American Medical Association 2003;290:921-8.
  • APA 1980
  • American Psychiatric Association. Diagnostic and Statistical Manual of Mental Disorders (DSM-III). 3rd Edition. Washington, DC: American Psychiatric Press, 1980.
  • APA 1987
  • American Psychiatric Association. Diagnostic and statistical manual of mental disorders (DSM-III-R). 3rd Edition. Washington DC: American Psychiatric Association, 1987.
  • APA 1994
  • American Psychiatric Association. Diagnostic and statistical manual of mental disorders (DSM-IV). 4th Edition. Washington DC: American Psychiatric Association, 1994.
  • Baker 2003
  • Baker CB, Johnsrud MT, Crismon ML, Rosenheck RA, Woods SW. Quantitative analysis of sponsorship bias in economic studies of antidepressants. British Journal of Psychiatry 2003;183:498-506.
  • Bandelow 2002
  • Bandelow B, Zohar J, Hollander E, Kasper S, Moller HJ. World Federation of Societies of Biological Psychiatry (WFSBP) guidelines for the pharmacological treatment of anxiety, obsessive-compulsive and posttraumatic stress disorders. The World Journal of Biological Psychiatry: The Official Journal of the World Federation of Societies of Biological Psychiatry 2002;3(4):171-99.
  • Bandelow 2006
  • Bandelow B, Baldwin DS, Dolberg OT, Andersen HF, Stein DJ. What is the threshold for symptomatic response and remission for major depressive disorder, panic disorder, social anxiety disorder, and generalized anxiety disorder?. Journal of Clinical Psychiatry 2006;67(9):1428-34.
  • Biederman 2004
  • Biederman J, Spencer T, Wilens T. Psychopharmacology. In: WienerJM, DulcanMK editor(s). Textbook of Child and Adolescent Psychiatry. 3rd Edition. Washington DC: American Psychiatric Publishing, Inc., 2004:931-74.
  • BPCA 2002
  • FDA. Best Pharmaceuticals for Children Act. http://www.fda.gov/opacom/laws/pharmkids/contents.html.
  • Bridge 2007
  • Bridge JA, Iyengar S, Salary CB, Barbe RP, Birmaher B, Pincus HA, Ren L, Brent DA. Clinical response and risk for reported suicidal ideation and suicide attempts in pediatric antidepressant treatment: a meta-analysis of randomized controlled trials. Journal of the Americal Medical Association 2007;297(15):1683-96.
  • Canino 2004
  • Canino G, Shrout PE, Rubio-Stipec M, Bird HR, Bravo M, Ramirez R, Chavez L, Alegria M, Bauermeister JJ, Hohmann A, Ribera J, Garcia P, Martinez-Taboas A. The DSM-IV rates of child and adolescent disorders in Puerto Rico: prevalence, correlates, service use, and the effects of impairment. Archives of General Psychiatry 2004;61(1):85-93.
  • Chavira 2007
  • Chavira DA, Shipon-Blum E, Hitchcock C, Cohan S, Stein MB. Selective mutism and social anxiety disorder: all in the family?. Journal of the American Academy of Child and Adolescent Psychiatry 2007;46(11):1464-72.
  • Cheer 2001
  • Cheer SM, Figgitt DP. Fluvoxamine: A review of its therapeutic potential in the management of anxiety disorders in children and adolescents. Paediatric Drugs 2001;3(10):763-81.
  • CSM 2004
  • Committee on Safety of Medicines (CSM) Expert Working Group (EWG) on SSRIs. Report of the CSM Expert Working Group on the safety of selective serotonin reuptake inhibitor antidepressants. www.mhra.gov.uk/home/groups/pl-p/documents/drugsafetymessage/con019472.pdf (date last accessed, May 2009). Medicines and Healthcare products Regulatory Agency, 2004.
  • D'Amato 1962
  • D'Amato G. Chlordiazepoxide in the management of school phobia. Diseases of the Nervous System 1962;23:292-5.
  • Deeks 2002
  • Deeks JJ. Issues in the selection of a summary statistic for meta-analysis of clinical trials with binary outcomes. Statistics in medicine 2002;21(11):1575-600.
  • Deeks 2008
  • Deeks J, Higgins JPT, Altman D (editors). Analysing and presenting results. In: HigginsJPT, GreenS editor(s). Cochrane Reviewers' Handbook 5.0.0 [updated February] Section 9;. Chichester, UK: John Wiley & Sons, Ltd., 2008.
  • Dickersin 1992
  • Dickersin K, Min YI, Meinert CL. Factors influencing publication of research results: Follow-up of applications submitted to two institutional review boards. The Journal of the American Medical Association 1992;267(3):374-8.
  • Elbourne 2002
  • Elbourne ER, Altman DG, Higgins JPT, Curtin F, Worthington HV, Vail A. Meta-analyses involving cross-over trials: Methodological issues. International Journal of Epidemiology 2002;31:140-9.
  • FDA Talk 2003
  • US Food, Drug Administration. FDA Statement Regarding the Anti-Depressant Paxil for Pediatric Population. FDA Talk Paper 2003.
  • Flament 2007
  • Flament MF, Geller D, Irak M, Blier P. Specificities of treatment in pediatric obsessive-compulsive disorder. CNS Spectrums 2007;12(2 Suppl 3):43-58.
  • Flannery 2000
  • Flannery-Schroeder EC, Kendall PC. Group and individual cognitive-behavioral treatments for youth with anxiety disorders: A randomized clinical trial. Cognitive Therapy and Research 2000;24:251-78.
  • Ford 2003
  • Ford T, Goodman R, Meltzer H. The British Child and Adolescent Mental Health Survey 1999: the prevalence of DSM-IV disorders. Journal of the American Academy of Child and Adolescent Psychiatry 2003;42(10):1203-11.
  • Geller 2003
  • Geller DA, Biederman J, Stewart SE, Mullin B, Martin A, Spencer T, Faraone SV. Which SSRI? A meta-analysis of pharmacotherapy trials in pediatric obsessive-compulsive disorder. American Journal of Psychiatry 2003;160(11):1919-28.
  • Gibbons 2006
  • Gibbons RD, Hur K, Bhaumik DK, Mann JJ. The relationship between antidepressant prescription rates and rate of early adolescent suicide. American Journal of Psychiatry 2006;163(11):1898-904.
  • Goodman 2007
  • Goodman WK, Murphy TK, Storch EA. Risk of adverse behavioral effects with pediatric use of antidepressants. Psychopharmacology 2007;191(1):87-96.
  • Greenhill 1998
  • Greenhill LL, Pine D, March J, Birmaher B, Riddle M. Assessment issues in treatment research of pediatric anxiety disorders: What is working, what is not working, what is missing and what needs improvement. Psychopharmacology Bulletin 1998;34(2):155-64.
  • Guy 1976
  • Guy W. ECDEU assessment manual for psychopharmacology. Washington, DC: U.S. National Institute of Health, 1976.
  • Hammad 2006
  • Hammad TA, Laughren T, Racoosin J. Suicidality in pediatric patients treated with antidepressant drugs. Archives of General Psychiatry 2006;63(3):332-9.
  • Hawkridge 2000
  • Hawkridge SM, Stein DJ. A risk-benefit assessment of pharmacotherapy for anxiety disorders in children and adolescents. In: PalmerKJ editor(s). Topics in paediatric psychiatry. Auckland: Adis International, 2000:11-28.
  • Higgins 2003
  • Higgins JTP, Thompson SG, Deeks JJ, Altman DG. Measuring inconsistency in meta-analyses. British Medical Journal (Clinical Research ed.) 2003;327(6):557-60.
  • Higgins 2008
  • Deeks J, Higgins J, Altman D. Cochrane Reviewers' Handbook 5.0.0 [updated February 2008]. In: HigginsJ, GreenS editor(s). The Cochrane Library. Chichester, UK: John Wiley & Sons, Ltd, 2008.
  • Ipser 2006
  • Ipser JC, Seedat S, Stein DJ. Pharmacotherapy for prevention of post-traumatic stress disorder. Cochrane Database of Systematic Reviews 2006, Issue 4. [DOI: 10.1002/14651858.CD006239 : CD006239]
  • James 2005
  • James A, Soler A, Weatherall R. Cognitive behavioural therapy for anxiety disorders in children and adolescents (Review). Cochrane Database of Systematic Reviews 2005, Issue 4. [DOI: 10.1002/14651858.CD004690.pub2; : CD004690.]
  • Keller 1992
  • Keller MB, Lavori PW, Wunder J, Beardslee WR, Schwartz CE, Roth J. Chronic course of anxiety disorders in children and adolescents. Journal of the Americal Academy of Child and Adolescent Psychiatry 1992;31(4):595-9.
  • Kendall 1997
  • Kendall PC. Therapy for youths with anxiety disorders: A second randomized clinical trial. Journal of Consulting and Clincal Psychology 1997;65(3):366-80.
  • Kessler 2007
  • Kessler RC, Angermeyer M, Anthony JC, DE Graaf R, Demyttenaere K, Gasquet I, DE Girolamo G, Gluzman S, Gureje O, Haro JM, Kawakami N, Karam A, Levinson D, Medina Mora ME, Oakley Browne MA, Posada-Villa J, Stein DJ, Adley Tsang CH, Aguilar-Gaxiola S, Alonso J, Lee S, Heeringa S, Pennell BE, Berglund P, Gruber MJ, Petukhova M, Chatterji S, Ustün TB. Lifetime prevalence and age-of-onset distributions of mental disorders in the World Health Organization's World Mental Health Survey Initiative. World Psychiatry 2007;6(3):168-76.
  • Lundbeck 2003
  • Lundbeck. Psychotropics. http://www.psychotropics.dk/ 2003.
  • Masi 2001
  • Masi G, Mucci M, Millepiedi S. Separation anxiety disorder in children and adolescents: Epidemiology, diagnosis and management. CNS Drugs 2001;15(2):93-104.
  • MHRA 2003
  • MHRA. Safety review of antidepressants used by children completed. http://www.mhra.gov.uk/NewsCentre/Pressreleases/CON002045 (date of last access May 2009) 2003.
  • Moncrieff 2001
  • Moncrieff J, Churchill R, Drummond DC, McGuire H. Development of a quality assessment instrument for trials of treatments for depression and neurosis. International Journal of Methods in Psychiatric Research 2001;10(3):126-33.
  • Möller 2008
  • Möller HJ, Baldwin DS, Goodwin G, Kasper S, Okasha A, Stein DJ, Tandon R, Versiani M, WPA Section on Pharmacopsychiatry. Do SSRIs or antidepressants in general increase suicidality? WPA Section on Pharmacopsychiatry: consensus statement. Eur Arch Psychiatry Clin Neurosci 2008;258(Suppl 3):3-23.
  • O'Kearney 2006
  • O'Kearney RT, Anstey KJ, Von Sanden C. Behavioural and cognitive behavioural therapy for obsessive compulsive disorder in children and adolescents. Cochrane Database of Systematic Reviews 2006, Issue 4. [DOI: 10.1002/14651858.CD004856.pub2]
  • Olatunji 2007
  • Olatunji BO, Cisler JM, Tolin DF. Quality of life in the anxiety disorders: a meta-analytic review. Clinical Psychology Review 2007;27(5):572-81.
  • Olfson 2002
  • Olfson M, Gameroff MJ, Marcus SC, Jensen PS. National Trends in the Use of Psychotropic Medications by Children. The American Journal of Psychiatry 2003;160(6):1071-7.
  • Poznanski 1984
  • Poznanski EO, Grossman JA, Buchsbaum Y, Banegas M, Freeman L, Gibbons R. Preliminary studies of the reliability and validity of the children's depression rating scale. Journal of the American Academy of Child Psychiatry 1984;23(2):191-7.
  • R 2005
  • R Development Core Team. R: A Language and Environment for Statistical Computing. Vienna: R Foundation for Statistical Computing, 2005. [: ISBN 3-900051-07-0]
  • Robinson 2002
  • Robinson KA, Dickersin K. Development of a highly sensitive search strategy for the retrieval of reports of controlled trials using PubMed. International Journal of Epidemiology 2002;31:150-3.
  • Rubino 2007
  • Rubino A, Roskell N, Tennis P, Mines D, Weich S, Andrews E. Risk of suicide during treatment with venlafaxine, citalopram, fluoxetine, and dothiepin: retrospective cohort study. British Medical Journal 2007;334(7587):242-7.
  • Safer 2006
  • Safer DJ, Zito JM. Treatment-emergent adverse events from selective serotonin reuptake inhibitors by age group: Children versus adolescents. Journal of Child and Adolescent Psychopharmacology 2006;16(1-2):159-69.
  • Sayal 2006
  • Sayal K. Annotation: Pathways to care for children with mental health problems. Journal of Child Psychology and Psychiatry, and Allied Disciplines 2006;47(7):649-59.
  • Scahill 1997
  • Scahill L, Riddle MA, McSwiggin-Hardin M, Ort SI, King RA, Goodman WK, Cicchetti D, Leckman JF. Children's Yale-Brown Obsessive Compulsive Scale: Reliability and validity. Journal of the American Academy of Child and Adolescent Psychiatry 1997;36(6):844-52.
  • Schweizer 2001
  • Schweizer E, Rynn M, Mandos LA, Demartinis N, García-España F, Rickels K. The antidepressant effect of sertraline is not enhanced by dose titration: Results from an outpatient clinical trial. International Clinical Psychopharmacology 2001;16(3):137-43.
  • Schünemann 2008
  • Schünemann H, Oxman AD, Vist GE, Higgins JPT, Deeks JJ, Glasziou P, Guyatt GH, on behalf of the Cochrane Applicability and Recommendations Methods Group. Interpreting results and drawing conclusions. In: HigginsJPT, GreenS editor(s). Cochrane Handbook for Systematic Reviews of Interventions Version 5.0.1 [updated September 2008]. Available from www.cochrane-handbook.org. The Cochrane Collaboration, 2008.
  • Simeon 1989
  • Simeon JG. Pediatric psychopharmacology. Canadian Journal of Psychiatry 1989;34(2):115-22.
  • Soldin 2000
  • Soldin SJ, Steele BW. Mini-Review: Therapeutic Drug Monitoring in Pediatrics. Clinical Biochemistry 2000;33(5):333-5.
  • Song 2000
  • Song F, Eastwood AJ, Gilbody S, Duley L, Sutton AJ. Publication and related biases. Health Technology Assessment 2000;4:1-115.
  • Spitzer 1996
  • Spitzer R, Williams J, Gibbons M. Instructional manual for the structured clinical interview for DSM-IV. New York. NY: New York State Psychiatric Institute., 1996.
  • Target 1994
  • Target M, Fonagy P. Efficacy of psychoanalysis for children with emotional disorders. Journal of the American Acadamy of Child and Adolescent Psychiatry 1994;33(3):361-71.
  • Thompson 1994
  • Thompson SG. Why sources of heterogeneity in meta-analysis should be investigated. British Medical Journal (Clinical Research ed.) 1994;309(6965):1351-5.
  • Tiihonen 2006
  • Tiihonen J, Lönnqvist J, Wahlbeck K, Klaukka T, Tanskanen A, Haukka J. Antidepressants and the risk of suicide, attempted suicide, and overall mortality in a nationwide cohort. Archives of General Psychiatry 2006;63(12):1358-67.
  • Verbeke 2000
  • Verbeke M, Molenberghs G. Linear Mixed Models for Longitudinal Data. New York: Springer, 2000.
  • Watson 2008
  • Watson HJ, Rees CS. Meta-analysis of randomized, controlled treatment trials for pediatric obsessive-compulsive disorder. Journal of Child Psychology and Psychiatry 2008;49(5):489-98.
  • Woodward 2001
  • Woodward LJ, Fergusson DM. Life course outcomes of young people with anxiety disorders in adolescence. Journal of the American Academy of Child and Adolescent Psychiatry 2001;40(9):1086-93.
  • Zito 2003
  • Zito JM, Safer DJ, DosReis S, Gardner JF, Magder L, Soeken K, Boles M, Lynch F, Riddle MA. Psychotropic practice patterns for youth: A 10-year perspective. Archives of Pediatrics and Adolescent Medicine 2003;157(1):17-25.