Pharmacological interventions for adjustment disorders in adults

  • Protocol
  • Intervention



This is the protocol for a review and there is no abstract. The objectives are as follows:

To assess the efficacy and tolerability of pharmacological interventions for the treatment of AD in adults.


Description of the condition

Adjustment disorder (AD) is defined as an abnormal reaction that occurs in response to a stressful event, such as work-related problems, marital difficulties, familial problems and a range of other stressors, either acute or long term (Despland 1995). AD is distinguished from normal adaptive reactions to stressful events by the severity of symptoms or the impairment in functioning, or both. In psychiatry it is recognised that life events often have a role in triggering a range of illnesses such as major depressive disorder (MDD) but their presence is not essential, unlike in AD, where the diagnosis is contingent upon the presence of an identifiable and recent stressor. Psychiatric disorders are defined using two international systems of classification. The International Classification of Diseases 10th edition (ICD-10) (WHO 1992) is the World Health Organization compendium and the Diagnostic and Statistical Manual 4th edition (DSM-IV) is the American Psychiatric Association equivalent (APA 1994). Both recognise AD although there are some minor differences between the terminology and criteria used in each.

The symptoms of AD are similar to those of MDD and generalised anxiety disorder (GAD); therefore, distinguishing AD from other syndromes is difficult (Casey 2006). As well as symptoms, there may also be disturbance in functioning and sometimes this may be the dominant feature. ICD-10 requires the presence of both symptoms and impaired functioning before a diagnosis can be made, while DSM-IV specifies that one or the other may be present.

Several subtypes of AD are described in both DSM-IV and ICD-10, based on the dominant symptom patterns or the behaviour exhibited. These consist broadly of AD with depression, AD with anxiety, AD with disturbance of conduct and AD with mixed states (WHO 1992; APA 1994). AD may be acute (less than one month in ICD-10 or less than six months in DSM-IV) or chronic (more than one month but less than two years in ICD-10 or more than six months in DSM-IV). It is a self limiting condition and the prognosis is excellent with complete symptomatic and functional resolution being common unless there are persistent stressors. Even among those with AD who require psychiatric admission there are significantly fewer readmissions than among those with GAD, MDD or dysthymia/sub-threshold depression (Jones 2002), and less frequent use of outpatient and psychotherapeutic services (Bronisch 1991).

In certain groups of people AD is common. Among those who deliberately self harm (intentional self injury or self poisoning irrespective of the underlying NICE definition motivation) (NICE 2004), AD is the most common clinical diagnosis (Taggart 2006). It is also prominent in consultation-liaison psychiatry, being diagnosed in about 12% of psychiatric referrals in general hospitals (Strain 1998; Huyse 2001). Among acutely ill medical inpatients (Silverstone 1996) and in obstetric/gynaecology consultation-liaison (Rigatelli 2002) it is more common than other mood disorders including MDD. In a palliative care setting a recent meta-analysis found that the prevalence of AD was 15.4% and only marginally less than that for major depression (16.5%), while in oncological settings AD was more common than major depression (19.4% versus 14.9%) (Mitchell 2011). In primary care there are few studies but a prevalence range from 1% to 18% of consulters (Casey 1984; Blacker 1988) has been described, while a recent study identified AD in 2.4% of primary care consulters with emotional problems (Fernandez 2012). Within the psychiatric services, AD was identified in 11% to 36% of new outpatient referrals, depending on the assessment method (Shear 2000), and in around 9% of consecutive admissions to a public sector psychiatric unit (Koran 2002). The public health implications of AD are unclear since the data suggest that it is an uncommon disorder in the general population, being identified in only around 1% of the population (Ayuso-Mateos 2001). Internationally, the focus of service provision has been based on the very high prevalence of MDD as a major cause of disease burden (Ustün 2004). This has led to the establishment of guidelines for the treatment of depression (NICE 2009) in primary and secondary care including both pharmacological and psychological treatments. However, if it is further shown (as some studies have done) that AD might be the more appropriate diagnosis in many, due to conflation of the two conditions (Casey 2006), this could have major implications for the type and duration of treatment that is offered, including whether any specific intervention other than general support is required for these self limiting conditions. Major cost implications would flow from this, with significant savings on the drugs budget.

There has been general neglect of AD in research (Casey 2001), in particular, the psychobiology of the condition has received little attention. The focus in the few studies that have examined this has been on the hypothalamic-pituitary-adrenal (HPA) axis. Cortisol levels after dexamethasone suppression have shown different patterns in those with AD (with depressive features) in comparison to MDD (Lindqvist 2008). There have been no studies on the psychobiology of the other subcategories of AD. Thus, the treatment of AD is not at present underpinned by any biological parameters.

For a condition that is as common as AD, especially in general hospital psychiatry as well as in emergency settings involving self harming patients, it is crucial that treatment recommendations are based on firm evidence and that the risks and benefits of recommended interventions are carefully weighed.

Description of the intervention

The importance of AD is that it requires only limited treatment due to its tendency to be short lived and to resolve spontaneously (Strain 2008). Pharmacological agents such as benzodiazepines and hypnotics are recommended for symptomatic relief of anxiety and insomnia in AD (Uhlenhuth 1995; Shaner 2000). Others recommend the use of alternative agents such as valerian and kava-kava since these are sometimes used in the treatment of GAD in preference to anxiolytics because of their non-addictive properties (Volz 1997). However, herbal remedies will not be included in the present review as per Cochrane policy. The role of antidepressive agents is much less clear in AD but one investigator (Stewart 1992) recommends their use in the treatment of minor depression, a term that is often used interchangeably, albeit inaccurately, with AD. A retrospective case note study (Hameed 2005) found that AD with depression showed a better response to antidepressants than did MDD. No particular group of antidepressants has been shown to be more effective than any other. Recommendations for the use of pharmacological agents in AD are not accompanied by any guidance on dosage or duration of treatment. Neither has there been any discussion on which antidepressive agents might be helpful in the treatment of AD with depression, an important consideration in light of its spontaneous and, usually, rapid resolution and the delay in onset of effect of antidepressants. On the other hand there have been recommendations that the focus should be on psychological therapies (Strain 2008) such as "mirror therapy", a form of holistic intervention used in those with AD post myocardial infarct (Gonzáles-Jaimes 2003), cognitive therapy for those who have AD associated with occupational dysfunction (van der Klink 2003), "ego-enhancing" therapy for older adults (Frankel 2001) and general support (De Leo 1989). The recommendations for pharmacotherapy are based on the opinions of individuals rather than on any examination of the evidence base for these interventions.

How the intervention might work

The psychobiology of AD has received scant attention. Therefore, the biological rationale for using pharmacological agents is unclear, apart from the pragmatic approach to prescribing for symptomatic relief irrespective of the underlying psychobiology of the illness. This assumes that the pathophysiology of subsyndromal conditions such as AD and full syndromes such as MDD and GAD are the same and that the response to treatment will therefore be the same. This view was reinforced by a recent systematic review that found antidepressants to be effective in depression (MDD, AD and dysthymia) with physical illness (Rayner 2010).

In the treatment of MDD, antidepressants are believed to act by enhancing the activity of monoamines (serotonin, adrenaline and dopamine) in the central nervous system and this might be one possible mechanism by which this occurs in AD. A possible impact on the HPA axis, thought to be involved in stress reactions and in MDD (Pariante 2008), might also be a possibility although there is little firm evidence in the literature to support this in the case of AD.

Another view on the psychobiology of AD is that it is a stress reaction similar to acute stress reaction and post-traumatic stress disorder (PTSD) (Maercker 2008). Following from this, it is arguable that broadly similar conditions such as AD might benefit from similar treatments that include the selective serotonin reuptake inhibitors (SSRIs), as these have been shown to be efficacious in this condition in a heterogeneous group of traumas of varying duration and severity (Stein 2006). Their efficacy is thought to result from their impact in controlling the dysregulation of the neurotransmitter systems and neuroendocrine systems (HPA axis), some of which may also be abnormal in AD.

With regard to benzodiazepines in AD with anxiety disorders, it is likely that they will work in a manner similar to that in GAD, by enhancing the action of gamma-amino-butyric acid (GABA) although the role of GABA has not been studied in the AD with anxiety subtype.

Why it is important to do this review

Recommendations for pharmacological treatments for AD have been developed by expert opinion rather than as a result of randomised controlled trials. Since it is a self limiting disorder, the treatment recommendations are for brief interventions with an emphasis on psychological therapies (Strain 2008) although there may be a role for the symptomatic treatment of insomnia and anxiety symptoms with hypnotics or tranquillizers (Uhlenhuth 1995; Shaner 2000).

Despite the limited evidence of the benefits of pharmacological treatments, in particular antidepressants, there are indications that antidepressants are increasingly being used in the treatment of AD (Diefenbacher 2002), due to what some authors describe as the "culture of prescribing" (Strain 2008). Recent data indicate that the condition which has shown the greatest increase in antidepressant usage is AD, with the rate of prescription changing from 22.26 per 100 cases in 1996 to 39.37 per 100 cases in 2005 (Olfson 2009). Therefore, it is important to consider the evidence, if any, for the use of pharmacological agents in general, and antidepressants in particular, in the treatment of AD. Since suicidal ideation and behaviour is common in those with AD (Kryzhanovskaya 2001), it is also of clinical relevance to identify whether pharmacological agents assist in reducing these.


To assess the efficacy and tolerability of pharmacological interventions for the treatment of AD in adults.


Criteria for considering studies for this review

Types of studies

Published and unpublished studies will be included if they are double blind (participants and personnel), randomised controlled trials. Those that have a cross-over design and cluster randomised trials will also be included.

There will be no language restrictions.

Types of participants

Participant characteristics

Participants will be adults, both male and female, over the age of 18. There will be no ethnic restriction.


Participants must have a diagnosis of AD according to the following criteria:

a) a clinical diagnosis based on ICD-10 (WHO 1992) or DSM-IV (APA 1994) or earlier versions of ICD and DSM criteria; or

b) made by a valid diagnostic instrument such as the SCAN (Wing 1990), SCID (First 1995) or earlier versions of SCAN or SCID; or

c) using some other validated diagnostic instrument.

Participants must meet the criteria for acute AD (< 1 month in ICD-10 or < 6 months in DSM-IV) or chronic AD (> 1 month but < 2 years in ICD-10 or > 6 months in DSM-IV).

Since AD was introduced in 1984 in DSM-III (APA 1980) and introduced in ICD-10 in 1992 (WHO 1992), trials that predate these criteria will be included if the participants are diagnosed with the older term 'situational disturbance' or the interchangeable terms 'reactive depression', provided the trials were undertaken prior to the introduction of the current diagnostic terms.


There will be no restriction on the setting of these studies, e.g. whether in general practice, inpatient psychiatric units, outpatient psychiatric clinics or consultation-liaison psychiatry settings.


We will include medically ill participants.

We will exclude studies where participants have another axis 1 psychiatric disorder although this is unlikely to be a problem since a diagnosis of AD cannot be made in the presence of another axis 1 diagnosis.

Types of interventions

Experimental interventions
  1. Antidepressants (tricyclics, SSRIs, others) given in any dosage and for any period of time. The effect of different dosages will be examined.

  2. Benzodiazepines (anxiolytics and hypnotics).

  3. Non-benzodiazepine hypnotics.

  4. Other pharmacological interventions (e.g. mood stabilisers, antipsychotic agents).

These may be used as monotherapy or in combination.

Comparator interventions
  1. Waiting list control.

  2. Placebo.

  3. Watchful waiting.

  4. Treatment as usual.

  5. Psychological therapies (behavioural, cognitive behavioural, third wave cognitive behavioural therapy, psychodynamic, humanistic/supportive and integrative).

  6. Any other pharmacological intervention.

  7. Combinations of the above.

Types of outcome measures

Primary outcomes
  1. An evaluation of a change in total symptom scores. Symptoms that will be considered include:

    1. depression, measured using the Hamilton Rating Scale for Depression (Hamilton 1960), the Montgomery-Asberg Depression Rating Scale (Montgomery 1979) or the Beck Depression Inventory (Beck 1960) measured either as a continuous variable (total score) or dichotomous (above or below the threshold for possible 'caseness');

    2. anxiety, measured using the Hospital Anxiety Depression Rating Scale (Zigmond 1983) measured either as a continuous variable (total score) or dichotomous (above or below the threshold for possible 'caseness').

  2. Adverse effects of treatment as measured by:

    1. the total number of dropouts from each arm due to side effects as a proportion of the total randomised;

    2. any increase in deliberate self harm in the treatment group in comparison to the control group.

The list of scales is not exhaustive and others will also be examined provided they have been validated as symptom measures of anxiety or depression. There are currently no validated scales to specifically measure AD.

Secondary outcomes
  1. Changes to social function score using a validated measure such as the Global Assessment of Function (Luborsky 1962) (continuous measure).

  2. Changes to quality of life score using measures such as the SF-36 (Ware 1993), or the HoNOS (Wing 1994) (continuous measure).

  3. Changes to suicidal ideation scores using item 9 of the BDI (Beck 1960) or some specific suicidal ideation scale such as the Beck Scale for Suicidal Ideation (Beck 1979) (continuous measure).

  4. Changes to 'caseness' status if diagnostic instruments such as SCAN (Wing 1990) or SCID (First 1995) are used (dichotomous measure).

  5. The number of participants from each arm experiencing at least one adverse event as a proportion of the total randomised (dichotomous measure).

  6. The number of dropouts from each arm due to inefficacy as a proportion of the total randomised (dichotomous measure).

The list of scales mentioned above is not exhaustive and others will also be examined provided they have been validated.

Search methods for identification of studies

The Cochrane Depression, Anxiety and Neurosis Review Group's Specialised Register (CCDANCTR)

The Cochrane Depression, Anxiety and Neurosis Group (CCDAN) maintain two clinical trials registers at their editorial base in Bristol, UK: a references register and a studies based register. The CCDANCTR-References Register contains over 31,500 reports of trials in depression, anxiety and neurosis. Approximately 65% of these references have been tagged to individual, coded trials. The coded trials are held in the CCDANCTR-Studies Register and records are linked between the two registers through the use of unique Study ID tags. Coding of trials is based on the EU-Psi coding manual. Please contact the CCDAN Trials Search Coordinator for further details. Reports of trials for inclusion in the Group's registers are collated from routine (weekly), generic searches of MEDLINE (1950-), EMBASE (1974-) and PsycINFO (1967-); quarterly searches of the Cochrane Central Register of Controlled Trials (CENTRAL) and review specific searches of additional databases. Reports of trials are also sourced from international trials registers c/o the World Health Organization's trials portal (ICTRP), drug companies, the handsearching of key journals, conference proceedings and other (non-Cochrane) systematic reviews and meta-analyses.

Details of CCDAN's generic search strategies (used to identify RCTs) can be found on the Group's website.

Electronic searches

The CCDANCTR-Studies Register will be searched using the following terms (condition only):
Diagnosis = (“adjustment disorder*” or ((reactive or mild or minor or situational or subclinical or sub-clinical or subthreshold or sub-threshold) and depress*) or “reactive disorder*” or “situational disturbance*” or “anniversary reaction*” or “maladaptive behav*” or “complicated grief”).

The CCDANCTR-References Register will be searched using free-text terms to find additional reports of RCTs not yet tagged to individual studies:
("adjustment disorder*" or "reactive depressi*"  or "situational depress*" or "mild depress*" or "minor depress* or "subclinical depress*" or "sub-clinical depress*" or "subthreshold depress*" or sub-threshold depress*" or "subsyndromal depress*" or "sub-syndromal depress*" or "reactive disorder*" or "situational disturbance*" or "anniversary reaction*" or "maladaptive behav*" or "maladaptive coping" or "maladaptive reaction*" or "maladaptive trait*" or grief or bereavement or stressor* or "emotional symptom*").

An additional search of PsycINFO will be conducted, using a more sensitive set of terms for 'adjustment disorders', exceeding those routinely used by CCDAN, to check that no studies have been missed (Appendix 1). Complementary searches will also be carried out on LILACS and Dissertation Abstracts.

Details of unpublished or ongoing trials, or both, will be sourced from international trials registers c/o the WHO trials portal (ICTRP) and

The following grey literature databases will also be searched:

OpenGrey (

No date or language restrictions will be applied to the search.

Searching other resources

Handsearches of relevant conference proceedings (last three years) will be carried out, for example:

  • World Psychiatric Association (WPA) (;

  • American Psychiatric Association (;

  • European College of Neuropsychopharmacology (ECNP) Congress ( (ECNP abstracts published in European Neuropsychopharmacology).

The reference lists of all included and excluded studies will be screened to help identify any additional studies and a cited reference search will be conducted on the Web of Science. Chapters in relevant books will be scrutinised and experts in the field contacted to identify any unpublished studies.

Data collection and analysis

Selection of studies

The initial screening, to decide on potential eligibility, will be carried out by PC and LW by reading the abstract. Those studies that do not deal with adjustment disorder or the terms listed above (situational disturbance/depression or reactive depression) will be rejected. Doubtful studies will be included at this stage.

Thereafter, the selection of studies will be carried out by two of the authors independently (PC and LW or DP) by reading the abstract and if it seems to meet the inclusion criteria then the paper will be read in full to evaluate its suitability for inclusion in this review. Where the abstract is unclear the paper will be included and read in full. Where there is disagreement an attempt will be made to resolve it by discussion. If this is not successful the paper will be sent to a third reviewer (AM) who will make the final decision on inclusion.

Data extraction and management

Two authors (BK and PC) will carry out data extraction, collecting the data independently of each other onto an extraction sheet developed and piloted for this purpose, as recommended by Higgins 2008. One author will carry out the initial extraction (PC) and this will be cross-checked by the other author (BK) to ensure reliability. We will contact trial authors when data are missing or when discrepancies arise during the extraction process that cannot be resolved by discussion.

The following information will be collated from each trial.

  • Description of the trials, including the primary researcher, the year of publication, the country in which the study was carried out and the source of funding.

  • Characteristics of trial methodology including the diagnostic criteria (e.g. DSM-IV or ICD-10), the exclusion criteria employed, whether the diagnosis was based on clinical or structured interview, the use of a placebo run-in.

  • The diagnostic subtype (if specified).

  • Characteristics of the subjects and controls, including gender distribution and age distribution and their baseline measures of anxiety, depressive, suicidal symptoms, quality of life and social functioning.

  • Characteristics of the interventions, including the number of participants randomised to the treatment and control groups, the number of total dropouts per group as well as the number that dropped out due to poor tolerability or due to inefficacy.

  • Outcome measures for subjects and controls including measures of anxiety, depressive, suicidal symptoms, quality of life and social functioning.

  • Changes to 'caseness' status and information on attrition due to inefficacy, adverse events, or loss to follow-up will also be extracted, as will the percentage with at least one adverse event.

  • Quality assessment - data from the risk of bias assessment will also be extracted (see Assessment of risk of bias in included studies).

  • Additional information will be included, such as whether results were based on the intention-to-treat (ITT) analysis with last observation carried forward (LOCF) for completers only, and the minimal period required for inclusion of participants in the LOCF analyses.

Comparisons will be made between the outcomes of interest in the following groups:
  1. Antidepressants vs controls (placebo, waiting list control, watchful waiting combined).

  2. Antidepressants vs anxiolytics.

  3. Antidepressants vs other pharmacological treatments (excluding anxiolytics).

  4. Antidepressants vs psychological therapies.

  5. Anxiolytics vs controls (placebo, waiting list control, watchful waiting).

  6. Anxiolytics vs psychological therapies.

  7. Anxiolytics vs other pharmacological treatments (excluding antidepressants).

  8. Pharmacological interventions versus psychological therapies.

  9. Where combinations of interventions are used comparisons will made using the structure detailed above.

Where studies have allowed medications for physical illnesses to continue, the principal investigator will be contacted in order to establish what these medications were and if they have any effect on psychological symptoms. The information obtained will be analysed by means of a sensitivity analysis or a subgroup analysis, depending on the responses received (or not received).

If different interventions are used in an individual subtype the analysis will be stratified by class of intervention, so that the results of multiple analyses can be presented in a single graph with a subtotal summary score for each instead of an overall summary effect.

Timepoint management - outcomes will be based on measures at three months and six months.

Assessment of risk of bias in included studies

Risk of bias will be assessed independently by two authors (PC and AM) in the following domains:

  • random sequence generation;

  • concealment of allocation;

  • blinding of participants and personnel and blinding of outcome;

  • incomplete outcome data (including attrition);

  • selective outcome reporting;

  • other sources of bias (e.g. study design, baseline imbalance, sponsorship or some other aspect of the study not included above).

The risk of bias in each domain and overall will be rated as high, low or unclear, according to the criteria in the Cochrane Handbook for Systematic Reviews of Interventions (Higgins 2008). Each assessment will be accompanied by a support for judgement comment.

If the information provided in the paper is inadequate, we will contact the authors for further details.

Where there is disagreement this will be resolved by discussion. If consensus is still not reached, a third author (BK) will adjudicate.

Measures of treatment effect

For continuous data, we will measure treatment effect using mean differences (MDs) or standardised mean differences (SMDs) and their 95% confidence intervals (CI). The MD will be used when the outcome of interest is measured using the same scale in all studies while the SMD will be implemented when the scales used differ across studies.

For dichotomous data, we will use risk ratios (RRs) in response to treatment, including 95% confidence intervals, instead of odds ratios (ORs) since the latter are more difficult to interpret. If only ORs are provided, these will be converted to RR using the formula in Higgins 2008. Numbers needed to treat (NNT) will be calculated from risk differences (RD) according to the formula provided in Higgins 2008.

Unit of analysis issues

Multiple treatment groups

Where there are multiple treatment groups (e.g. different doses, different medications) only one intervention will be evaluated for each meta-analysis, thus satisfying the assumption of independent observations and avoiding bias through multiple comparisons with the same control group. Where the same treatment is given to different groups (e.g. men and women) data from the groups will be averaged to produce a single mean and standard deviation. For other combinations we will take care to avoid loss of information as recommended by Higgins 2008.

Cross-over trials

These studies will be included in the analysis when it is possible to extract data from the active treatment and control arms for the first treatment period only.

Cluster randomised trials

These will be included in the analysis and the data extracted will take account of the cluster design as recommended by Higgins 2008. To avoid unit of analysis errors a summary score will be calculated for each cluster so that the sample size will be the number of clusters and the analysis will proceed as if the clusters represent individuals. If the information required for this analysis is not presented in the study, we will contact the authors to seek the required data, including:

  • the number of clusters randomised to each arm or the mean of each cluster;

  • the outcome data for the individuals in each cluster;

  • an estimation of the intracluster correlation.

Dealing with missing data

Should more than 40% of participants be lost by completion of the study, the data will be included in the initial analysis but the effect of this inclusion evaluated in a sensitivity analysis.

We will contact the authors to obtain missing information where possible and we will discuss the likely impact on the results.

The reasons for missing data, their number, characteristics and whether they differed significantly from completers will be considered.

If data from intention-to-treat analyses are unavailable, the study will be excluded from the intention-to-treat meta-analysis. If available, best/worst case scenario calculations will be carried out so that the boundaries of the treatment effect can be described. If data are available on completers only these will still be used.

For continuous data either end point analysis or last observation carried forward (LOCF) analysis will be carried out if such data are provided in the papers or by the authors when contacted.

If some statistics are missing (e.g. standard deviations) and are not available from trial authors we will attempt to calculate them from P values.

The likely impact of missing data will be considered in the discussion.

Assessment of heterogeneity

An assessment of clinical heterogeneity will be made by examining differences in study populations and interventions.

We will examine statistical heterogeneity by studying the degree of overlap of the confidence intervals for individual studies in a forest plot.

We will also carry out more formal assessments using a Chi2 test with the P value set at 0.1. As this statistic has low power to detect diversity when the number of studies is low or sample size is small, we may also need to calculate I2. As I2 only provides a rough estimate of the variability due to heterogeneity, the following overlapping bands will guide the interpretation of the I2 statistic, as suggested in the Cochrane Handbook for Systematic Reviews of Interventions (Higgins 2008).

  • 0% to 40% might not be important;

  • 30% to 60% may represent moderate heterogeneity;

  • 50% to 90% may represent substantial heterogeneity;

  • 75% to 100% represents considerable heterogeneity.

However, as there is an a priori expectation that clinical and statistical heterogeneity will be present, it is anticipated that a random-effects model will be used.

Assessment of reporting biases

A detailed exploration for possible unpublished trials will be carried out in several ways.

1. Trials registers (detailed in Search methods for identification of studies) will be examined for trials registered but not identified in the search. Information on these will be sought from the investigators.

2. Authors who have previously carried out trials will be contacted for information on other possible unpublished material.

3. We will assess reporting bias by creating a funnel plot (sample size plotted against effect size) based on the published studies, using Review Manager software. This will be visually inspected for asymmetry. Formal tests for asymmetry will only be carried out if there are sufficient studies (i.e. greater than 10) to distinguish true from chance asymmetry and provided the studies are not of similar size (Higgins 2008). However, there are causes of funnel plot asymmetry other than reporting bias (e.g. small sample size) that will also be considered (Higgins 2008).

Data synthesis

A meta-analysis will be carried out if there are sufficient studies. It is likely that a random-effects model will be used but this will depend on the level of heterogeneity. The outcomes will be expressed in terms of an average effect size for each outcome measure and their 95% confidence intervals.

If there are insufficient studies the review will be written in narrative form.

Subgroup analysis and investigation of heterogeneity

We plan to carry out a subgroup analysis for the various subtypes of adjustment disorder categorised in DSM-IV and ICD-10 (1 to 4 listed below) since treatments for these are likely to differ (Higgins 2008):

  1. adjustment disorder with depression;

  2. adjustment disorder with anxiety;

  3. adjustment disorder with disturbance of conduct;

  4. adjustment disorder mixed type;

We also plan to perform subgroup analysis by control condition where appropriate (i.e. placebo, waiting list control, watchful waiting).

Each of these subgroups will be examined for the comparisons described in Data Extraction and Management above and for the same time points.

Sensitivity analysis

We plan to carry out a sensitivity analysis to determine the impact of decisions made during the review process on the robustness of the conclusions. We will examine the impact of:

  1. excluding studies where randomisation or the level of blinding is low or unclear;

  2. excluding studies that potentially use older terms for adjustment disorders (such as reactive depression, acute situational disturbance or situational depression) in order to be confident that the conclusions will apply to those with definite AD;

  3. including studies in which more than 40% of participants had dropped out.


The authors wish to thank members of the editorial team of the Cochrane Review Group who provided guidance during the development of the protocol. Search strategies were devised in collaboration with CCDAN's Trials Search Co-ordinator (TSC).

CRG funding acknowledgement

The National Institute for Health Research (NIHR) is the largest single funder of the Cochrane Depression, Anxiety and Neurosis Group. 


The views and opinions expressed therein are those of the authors and do not necessarily reflect those of the NIHR, NHS or the Department of Health.


Appendix 1. OVID PsycINFO search

1. Adjustment Disorders/
2. Adjustment/
3. exp Emotional Adjustment/
4. Occupational Adjustment/
5. School Adjustment/
6. Social Adjustment/
7. Reactive Depression/
8. ((adjustment or reacti*) adj disorder*).tw.
9. ((reacti* or mild or minor or situational or subclinical or sub-clinical or subthreshold or sub-threshold) adj1 depress*).tw.
10. situational disturbance*.tw.
11. anniversary reaction*.tw.
12. Anniversary Events/
13. (maladapt* adj (behav* or coping or reaction* or trait*)).tw.
14. complicated
15. stressor*.tw.
16. (emotional adj2 (symptom* or trauma*)).tw.
17. or/1-16
18. (pharma* or psychopharm* or or psychotropic* or drug* or antidepress* or anti depress* or MAOI or monoamine oxidase inhibit* or ((serotonin or norepinephrine or noradrenaline or neurotransmitter* or dopamine*) and (uptake or reuptake or re uptake)) or SSRI* or SNRI* or TCA* or tricyclic* or tetracyclic* or heterocyclic* or benzo* or anxiolytic* or hypnotic*).af.
19. clinical
20. or/18-19
21. treatment effectiveness
22. clinical
23. mental health program
25. placebo$.ti,ab.
26. randomly.ab.
27. randomi#ed.ti,ab.
28. trial.ti,ab.
29. ((singl$ or doubl$ or trebl$ or tripl$) adj3 (blind$ or mask$ or dummy)).mp.
30. (control$ adj3 (trial$ or study or studies or group$)).ti,ab.
31. factorial$.ti,ab.
38. allocat$.ti,ab.
32. assign$.ti,ab.
33. volunteer$.ti,ab.
34. (crossover$ or cross over$).ti,ab.
35. (quasi adj (experimental or random$)).mp.
36. "2000".md.
37. or/21-36
38. (17 and 20 and 37)

Contributions of authors

PC will be the guarantor of the review.

PC has conceived and designed the review. She is also the co-ordinator of the review. She has written the draft protocol except for the search strategy.

AR has assisted in developing the search strategy. AM made suggestions for this section also.

BK, AM, DP and LW had input into drafting this protocol.

Declarations of interest

None from any author.

Sources of support

Internal sources

  • DP was an employee of University College Dublin and she utilised her research sessions to participate in writing this review, Ireland.

    The support was the availability of sessional time in which to carry out this work

  • LW was an employee of the Mater Misericordiae University Hospital, Ireland.

    The support was the availability of sessional time in which to carry out this work

External sources

  • No sources of support supplied