Scolaris Content Display Scolaris Content Display

Cochrane Database of Systematic Reviews Protocol - Intervention

Interventions for treating anxiety after stroke

Authors' declarations of interest

Version published: 08 December 2010 Version history

https://doi.org/10.1002/14651858.CD008860

This is not the most recent version

view the current version 23 May 2017
Collapse all Expand all

Abstract

This is a protocol for a Cochrane Review (Intervention). The objectives are as follows:

  1. The primary aim of this review is to assess the effectiveness of pharmaceutical and psychological interventions in treating anxiety disorders or symptoms in stroke survivors.

  2. The secondary aim is to identify whether any of these anxiety interventions have an effect on quality of life, disability, depression, social participation, care‐giver burden or risk of death.

Background

Description of the condition

Stroke is a major health problem. Globally 15 million people suffer a stroke each year and, while historically considered an unfortunate consequence of ageing, approximately 25% of stroke victims are under 65 years of age. Worldwide it is the third leading cause of death (Mackay 2004) and a leading cause of adult disability (Department of Health 2007).     

There is mounting evidence that anxiety after stroke is a significant problem even with considerable passage of time after the event. Anxiety prevalence has been reported in 26% of stroke survivors within the first month of stroke (Lincoln 1998), 16% to 23% at four to six months (Dennis 2000; Sagen 2010), and 3% to 49% at one to five years post‐stroke (House 1991; Burvill 1995; Wilkinson 1997; Langhorne 2000; McCoy 2006). These rates are high compared with those in the general population. For example, a systematic review of the one‐year prevalence of anxiety disorders reported a rate of 11% in adult populations over the age of 18 years (Somers 2006), while studies that focused on individuals over 65 years of age have reported prevalence rates of 3% to 14% (Forsell 1997; Ritchie 2004).   

Unlike depression, which tends to focus on loss, anxiety is future oriented. People with anxiety fear future events or situations. For example, stroke survivors have been reported to fear having a recurrent stroke (Townend 2006), falling (Watanabe 2005) and returning to work (Gilworth 2009). Differentiating between normal anxiety, which everyone experiences at some point in life, and pathological anxiety disorders can be difficult for several reasons. A study investigating barriers in recognising and diagnosing anxiety (Van Rijswijk 2009) found that general practitioners reported that patient factors such as age (e.g. being older), and having limited verbal ability increased the difficulty of diagnosing an anxiety disorder. General practitioners also expressed deficiency in their knowledge of specific anxiety disorders, reported difficulties accessing specialist mental health services, and reported that patients frequently do not present for treatment.

Anxiety disorders are a general term that covers a broad spectrum of syndromes and, while each disorder has certain distinct features, they all share similar hallmark characteristics of excessive and irrational fear, subjective apprehension, and difficulty and distress in managing daily tasks (Gelder 2006). Manifested symptoms may be physical (e.g. heart palpitations, shortness of breath), cognitive (e.g. feeling of losing control), or behavioural (e.g. avoidance of certain stimuli). The gold standard for diagnosis would be one made using a structured or semi‐structured examination of mental status applied to the diagnostic criteria of the DSM‐IV (APA 1994) or ICD‐10 (ICD 1999). There are six major subtypes of anxiety disorder as follows.

  1. General anxiety disorder (GAD): characterised by excessive worry and feelings of apprehension about everyday events or problems with symptoms of muscle and psychic tension, causing significant distress or functional impairment, lasting six months or more.

  2. Panic disorder (with or without agoraphobia): sudden and recurrent attacks of terror, usually accompanied by heart palpitations, sweatiness, weakness, faintness or dizziness. They can produce a sense of unreality, and fear of impending doom or losing control.

  3. Social phobia: persistent and distressing fear of being observed or evaluated negatively by people in social or performance situations; often associated with physical and psychological anxiety symptoms with potential to result in social isolation and loneliness.

  4. Specific phobia: recurring excessive and unreasonable psychological or autonomic symptoms of anxiety in the presence of a specific feared object or situation leading to avoidance. 

  5. Obsessive‐compulsive disorder (OCD): characterised by obsessive ruminations, images or impulses, and/or physical or mental rituals which are time consuming and distressing.

  6. Post‐traumatic stress disorder (PTSD): severe psychological disturbance following a traumatic event characterised by involuntary re‐experiencing elements of the event with symptoms of hyperarousal, avoidance, and emotional numbing.

All subtypes of anxiety disorders have been observed in stroke patients (House 1991; Max 2002). However, post‐stroke GAD has received a substantial amount of attention. It is more common and severe in women and younger stroke survivors (Schultz 1997) and has been found to have an independent negative impact on quality of life (Ahlsio 1984). Co‐morbidity with depression is also very high. A study of 301 stroke patients found that 46% of patients with GAD also had major depression (Castillo 1993), and depression was more severe and longer‐lasting in those with co‐morbid anxiety (Shimoda 1998). Furthermore, those with co‐morbid anxiety and depression had higher levels of impairment in activities of daily living, more cognitive impairment and fewer social ties than those with depression or anxiety alone (Shimoda 1998).

There is also an emerging body of evidence of stroke survivors experiencing PTSD after stroke (Bruggimann 2006; McCoy 2006), even though this condition is typically only experienced by victims of violent crime, accidents, war or major disasters. Research in non‐stroke populations has found that people with PTSD have poorer self‐rated health, and lower levels of work and social participation (Al‐Saffar 2002).  

The disabling capacity of all anxiety disorders has been compared to that of major depression (Kessler 2000), and a US‐based estimate indicates that anxiety disorders cost the US economy $42 to $47 billion each year (DuPont 1996).

Description of the intervention

We are interested in any intervention targeted towards treating anxiety in stroke patients. Given the diversity of anxiety disorders and available therapies, it is not possible to include a comprehensive description a priori of all interventions that may be used. There are, however, evidence‐based guidelines (NICE 2004) outlining pharmaceutical and psychological interventions used to treat certain anxiety disorders in the general population which are described below. 

Pharmacological therapies

Several classes of drugs can be used to treat various anxiety disorders. They vary based on the neurotransmitters in the brain which they effect.

Selective serotonin reuptake inhibitors (SSRIs) are a class of antidepressant drugs used to treat anxiety. Serotonin is a neurotransmitter involved in regulating mood. SSRIs such as fluoxetine (Prozac), sertraline (Zoloft), escitalopram (Lexapro), paroxetine (Paxil) and citalopram (Celexa) are commonly prescribed for panic disorder, OCD, PTSD and social phobia (NIMH 2009). Pharmacologically, SSRIs inhibit the post‐release reuptake of serotonin by pre‐synaptic nerve terminals, hence increasing the level of available serotonin in the brain (Craig 2003). 

Tricyclic antidepressants (TCAs) (e.g. imipramine) are an older generation of antidepressant drug developed in the 1950s, and have been replaced for the most part by SSRIs. They are, however, still recommended in clinical guidelines for treating GAD and panic disorder (NICE 2004). 

Benzodiazepines (e.g. diazepam and alprazolam) are anxiolytics used to treat GAD and social phobia (Baldwin 2005), and in some instances specific phobia (NICE 2004). These drugs enhance the effect of the gamma‐aminobutyric acid (GABA) neurotransmitter which serves to reduce the somatic symptoms associated with anxiety such as muscle tension and insomnia. 

Zopiclone, zaleplone, and zaleplon (Z‐drugs), are hypnotics that can be prescribed to assist for sleep disturbance seen in GAD and PTSD (NICE 2005). These drugs behave in a similar way to benzodiazepines except they have a shorter half‐life. 

Psychological therapies

Various forms of psychological therapies are available for treating anxiety. They may be particularly welcomed by individuals (especially older people) who prefer not to use antidepressant drugs (Givens 2006). This preference is based on concern about dependence, prior negative experiences and the fact that many individuals do not view their psychological symptoms as a medical illness (Givens 2006).   

Behaviour therapy is based on learning theory, and patients are shown ways to develop adaptive ways of behaving. The aim of behaviour therapy is to treat anxiety through techniques designed to reinforce desired behaviours and eliminate undesired ones. 

Cognitive therapy is based on the cognitive model which hypothesises that a person’s emotions and behaviours are influenced by their perception of events. Hence it is not the situation itself that determines how a person feels but rather the way in which they construe the situation (Beck 1979).

Cognitive behaviour therapy (CBT) incorporates elements from both cognitive and behaviour therapy. It seeks to change a person’s thoughts, beliefs, attitudes, expectations and, as in behavioural therapy, change how people act.  It is 'present‐centred' and directs the participant to identify the current issues that are causing them distress, with the support of a trained psychological practitioner. Individuals talk about the specific problems in a structured manner with their therapist and may be given homework in the form of activities to complete before their next session. CBT is characterised as structured, goal‐oriented and time‐limited (Beck 1997).

How the intervention might work

Pharamacological interventions work by altering the level of certain neurotransmitters in the brain, while psychological interventions aim to alter maladaptive behaviour and cognitions in order to improve emotional functioning. A placebo effect is also possible, whereby participants receiving standard care, or waiting to receive the intervention, experience a reduction in anxiety symptoms that is not directly related to the pharmacological action of the treatment.

Why it is important to do this review

Post‐stroke anxiety has received less attention, both clinically and in research, than depression after stroke. Systematic reviews have already been carried out to assess the effectiveness of interventions used to treat depression and emotionalism when they occur after stroke (Hackett 2008; Hackett 2010). Currently there are no published systematic reviews of interventions used to treat anxiety after stroke, hence highlighting a gap in the literature and knowledge base. Studies in stroke (Shimoda 1998) and non‐stroke populations (Wittchen 2003) have shown that anxiety increases the risk and severity of depression. Hence, early treatment of anxiety disorders could reduce the risk of subsequent depression and its associated adverse consequences. Clinical guidelines have been established for treating anxiety, but their effectiveness in stroke populations remains unknown. We have chosen to evaluate both pharmaceutical and psychological interventions as evidence suggests some individuals prefer the latter treatment approach (Hyde 2005; Riedel‐Heller 2005). However, for various reasons including lack of available staff with specialist psychological skills, pharmaceutical drugs are often prescribed, making it necessary to evaluate their effect in the stroke population. 

Objectives

  1. The primary aim of this review is to assess the effectiveness of pharmaceutical and psychological interventions in treating anxiety disorders or symptoms in stroke survivors.

  2. The secondary aim is to identify whether any of these anxiety interventions have an effect on quality of life, disability, depression, social participation, care‐giver burden or risk of death.

Methods

Criteria for considering studies for this review

Types of studies

Randomised controlled trials (RCT) where the aim of the intervention administered was to treat anxiety in people with a clinical diagnosis of stroke (Hatano 1976) will be eligible for inclusion in this review. There will be no restriction on the basis of language or study location. Trials may compare the effect of an intervention against placebo, a different intervention, or different doses or frequency of interventions so long as the primary aim was to treat anxiety. 

Types of participants

All stroke patients enrolled into a RCT must have a clinical diagnosis of an anxiety disorder according to the DSM‐III (APA 1980), DSM‐III‐R (APA 1987), DSM‐IV (APA 1994), DSM‐IV‐TR (APA 2000), ICD‐9 (ICD 1979 ) or ICD 10 (ICD 1999), or have been deemed to have a significant level of anxiety symptoms as established by a pre‐determined cut‐off score above the researcher(s) defined threshold on an anxiety screening tool on entry into the trial. There will be no restriction based on age or gender distribution. We will include studies with mixed populations of ischaemic or haemorrhagic stroke but we will exclude studies assessing treatment effect in an exclusively subarachnoid haemorrhage patient population as the characteristics, treatment, and management of these patients are substantially different to other stroke patients. We will exclude studies treating stroke patients for other conditions such as depression, cognitive impairment or physical disability unless it can be determined that all patients had co‐morbid anxiety upon enrolment  into the trial and treatment of the anxiety was one of the main objectives of the trial.      

Types of interventions

We will evaluate RCTs of pharmaceutical interventions administered to stroke survivors compared with no intervention or standard care. The purpose of administering the drug must be to treat anxiety. We will exclude trials where the drug was administered for other purposes, such as neuroprotection. We will also include psychological interventions compared with standard care or no intervention, which aim to treat anxiety. We expect that these types of interventions will have a clearly defined psychological component, be structured, delivered and supervised by trained staff, and be time‐limited. We will exclude interventions whose purpose is simply to provide information or educate patients. Additionally, we also expect to exclude trials of interventions such as occupational therapy or stroke support co‐ordinator visitation unless they include a definitive psychological component aimed at treating anxiety.

We will include studies in the review that have compared two or more interventions against each other in the absence of a control or standard care comparison, but we will analyse them separately.

Types of outcome measures

Primary outcomes

The primary outcomes of interest will be:

  1. the proportion of stroke survivors without a clinical diagnosis of an anxiety disorder according to the DSM or ICD (APA 1994; ICD 1999) at the end of scheduled follow‐up;

  2. the proportion of stroke survivors scoring outside the anxiety symptom range (as defined by study author), or the change score from baseline on an anxiety rating scale or via self‐report at the end of scheduled follow‐up.  

Secondary outcomes

  1. Co‐morbid depression, as diagnosed by  DSM or ICD or determined by a depression rating scale such as the Beck Depression Inventory (BDI) (Beck 1961), the Hamilton Depression scale (Hamilton 1960) or the Montgomery‐Asberg depression scale (Montgomery 1979).

  2. Quality of life as measured on scales such as the 36‐item short form questionnaire (SF‐36) (Ware 1993).

  3. Social activities as measured on scales such as the Frenchay Activities Index (Wade 1985).

  4. Activities of daily living as measured on scales such as the Barthel Index (Mahoney 1965).

  5. Principal caregiver burden as measured by scales such as the Zarit Caregiver Burden Interview (Zarit 1980).

  6. Any adverse consequence as a result of treatment for anxiety such as drug tolerance, co‐dependence on counsellor and death. We will also record loss to follow‐up rates in different arms of trials as a possible indicator of treatment acceptability. 

Search methods for identification of studies

See the 'Specialized register' section in the Cochrane Stroke Group module.

Electronic searches

We will search the trials registers of the Cochrane Stroke Group and the Cochrane Depression, Anxiety and Neurosis Group. In addition, we will search the following bibliographic databases:

  1. The Cochrane Central Register of Controlled Trials (CENTRAL) (The Cochrane Library, latest issue);

  2. MEDLINE (1950 to present) (Appendix 1);

  3. EMBASE (1947 to present);

  4. PsycINFO (1806 to present);

  5. Allied and Complementary Medicine database (AMED) (1985 to present);

  6. Cumulative Index to Nursing and Allied Health (CINAHL) (1982 to present);

  7. Proquest Digital Dissertations which houses theses from North American and select European universities (1861 to present);

  8. Psychological Database for Brain Impairment Treatment Efficacy (PsycBITE) (2004 to present).

Searching other resources

In an effort to identify further published, unpublished and ongoing trials we will:

  1. search the following ongoing trials registers: ClinicalTrials.gov (http://www.clinicaltrials.gov/), Stroke Trials Registry (www.strokecenter.org/trials/), Current Controlled Trials (www.controlled‐trials.com);

  2. search conference proceedings from the UK Stroke Forum (2006 to 2010), European Stroke Conference (2001 to 2010), the International Stroke Conference (years available to be determined) and the Stroke Society of Australasia annual conference (years available to be determined) not already searched by the Cochrane Stroke Group Trials Search Co‐ordinator;

  3. search PsycBITE (Psychological Database for Brain Impairment Treatment Efficacy) (http://www.psycbite.com/);

  4. use Science Citation Index Cited Reference search for forward tracking of relevant articles;

  5. scan the bibliographies of identified trials;

  6. contact experts known to our research group and researchers with expertise in psychological disorder research, identified by scanning authors of relevant publications;

  7. contact major pharmaceutical companies to request information about any relevant unpublished trials.

We will search for relevant trials in all languages and arrange translation of trial reports published in languages other than English. We will also contact experts in the field and authors of included trials for information on other potentially relevant trials.

Data collection and analysis

Selection of studies

Two review authors will independently screen retrieved reports and will exclude irrelevant citations immediately based on title or abstract. The review authors will compare their lists of potentially eligible citations and will resolve discrepancies using a three‐step approach as required. Initially the two review authors will discuss the study abstract and seek to reach consensus through discussion. If this is not possible, they will ask a third review author to adjudicate and come to a conclusion. The review authors will address failure to reach consensus by reviewing the full text version of the article and through discussion.

Data extraction and management

Two review authors will independently extract data into a database where key information from studies can be recorded. If information is missing, one review author will contact the study authors, either by telephone or email, to request the missing data. After the two authors have reconciled the data extraction, we will export them into Review Manager 5 (RevMan 2008). We will record the following core data elements for analysis.

  1. Details of the study such as author, year and source of publication.

  2. Study methods including information about sampling process, randomisation, allocation, and blinding.

  3. Participants data including the number (i.e. eligible, participated, those loss to follow‐up), setting, diagnostic criteria used for anxiety, age, sex and type of stroke, and whether this was a first ever or recurrent stroke.

  4. Information about the intervention such as intervention type, duration, dose, timing and mode of delivery.

  5. Anxiety outcome (and secondary outcome) measures such as risk ratios, mean difference. We will also collect information about precision and significance of the outcome estimate (e.g. confidence intervals or p‐values), along with the time outcome measures were taken. In the event that a study has used multiple methods to assess anxiety outcomes (e.g. DSM diagnosis and screening tool score), we will report only one such outcome in the review. We will employ a hierarchical process for determining which outcome is selected. The DSM or ICD diagnosis will have priority over validated anxiety screening tools, which in turn will have priority over un‐validated measures of assessing anxiety.

  6. Source of study funding and potential conflicts of interests.

If we cannot obtain core data elements from a study we will describe the study qualitatively but exclude it from the main review.

Assessment of risk of bias in included studies

We will assess study bias in accordance with The Cochrane Collaboration’s tool for assessing risk of bias (Higgins 2008a). This instrument has six domains whereby different types of potential biases can be evaluated. The domains are sequence generation, allocation concealment, blinding (of participants, personnel and outcome assessors), incomplete outcome data, selective outcome reporting and other unspecified types of bias (e.g. conflict of interest). We will identify the respective biases from each study and will display them in a tabular format (e.g. studies listed in rows compared to different categories of bias listed in columns). We will summarise the risks qualitatively and describe their impact on the statistical or clinical relevance of findings. 

Measures of treatment effect

We anticipate that anxiety will be measured as either a dichotomous outcome (e.g. presence/absence) when assessed clinically or on a continuum when evaluated using a standardised scale (e.g. change in mean score). If study heterogeneity is not above 50% and there is more than one study with similar intervention and comparison group, we will conduct a meta‐analysis to pool results from studies reporting a dichotomous measure of anxiety. The outcome measure reported will either be any (or all) anxiety disorders or they may be reported by anxiety subtype (e.g. GAD, PTSD, or panic disorder) depending on the information provided by study authors. The Cochrane Handbook for Systematic Reviews of Interventions (Higgins 2008b) advises that patients and health professionals find it easier to interpret risk ratios as opposed to odds ratios, so we will transform data to estimate the relative risk of anxiety remission in the intervention and control groups. A systematic review on the frequency of anxiety after stroke is being carried out simultaneously to this review, which will provide knowledge of the typical risk of anxiety disorders or symptoms in the absence of treatment.

In studies where the outcome measure assesses the proportion of participants with an anxiety disorder or symptoms at the end of the study period (i.e. continuous outcomes), we will calculate the standard mean difference for each outcome. 

Unit of analysis issues

We expect that the majority of RCTs will have randomised individual stroke participants. If a cluster randomised design has been used, we will use the intra‐cluster correlation coefficient to estimate the effective sample size. Also, the situation where the outcome being observed repeatedly in participants (e.g. follow‐up at six months and one year) may occur in some trials. Given that it will likely be difficult to obtain individual patient data that would allow us to calculate an overall mean, we will take the outcome measurement at the end of treatment and at the longest time‐point post intervention.

Dealing with missing data

We will contact study authors to obtain information about missing data.  If missing outcome data are unobtainable, we will conduct a 'what if' sensitivity analysis exploring the impact the missing data could have on the final outcome. For dichotomous outcome data, the assumptions will be bi‐directional in that we will look at what would happen if all the missing data went for or against the main hypothesis.

Assessment of heterogeneity

We will evaluate the level of heterogeneity between studies using the I2 statistic, which estimates the amount of variation between studies that is due to differences in the way interventions are delivered or the clinical sample rather than just chance.  If heterogeneity is higher than 50% (a level considered to be a moderate to substantial level), we will use the random‐effects meta‐analysis method to measure treatment effect. The random‐effects method assumes that different studies are estimating different but related intervention effects and so provides a more conservative intervention effect estimate and wider confidence intervals (DerSimonian 1986).

Assessment of reporting biases

If there are at least 10 studies in the meta‐analysis, we will construct a funnel plot to estimate the potential influence of reporting bias. If there are no biases, the funnel plot should have a symmetrical distribution. Funnel plot asymmetry tests the effect of small studies and formally examines whether the association between the estimated intervention effects and a measure of study size (e.g. standard error of the intervention) is greater than might be expected to occur by chance (Sterne 2008). We will examine plot asymmetry for continuous outcomes by carrying out a linear regression of the intervention effect estimates on their standard errors, weighting by 1/(variance of the intervention effect estimate). For dichotomous outcomes we would carry out a linear regression of the log odds ratio on its standard error, weighted by the inverse of the variance of the log odds ratio. Although this is the most common method of testing for plot asymmetry there are statistical issues with this approach, so we will consult with a statistician for guidance on the best approach.

Data synthesis

Two review authors will independently extract data from included studies. One review author (AC) will enter data into RevMan (RevMan 2008) and the other will cross‐check the data entry. The review authors will resolve disagreements through reference to the original study report.

Subgroup analysis and investigation of heterogeneity

Several factors may have an impact on heterogeneity of studies and effect size, hence we will attempt to undertake subgroup analyses on certain clinically relevant factors such as:

  • specific type of anxiety disorder (e.g. GAD, PTSD, social phobia);

  • severity of anxiety;

  • length of time treatment was administered (for both pharmaceutical and psychological interventions);

  • stroke lesion location (the influence of which on mood outcome is contested);  

  • length of time since stroke at entry into the trial;

  • participant age;

  • participant gender;

  • co‐morbid depression.

Sensitivity analysis

We will carry out a methodological sensitivity analysis to test the robustness of findings and examine the degree to which the effect size is influenced by assumptions made.  We will re‐analyse data to look at the impact of including only those studies in which the following indicators have been executed to the highest standard:

  • allocation concealment;

  • blinding of participants and outcome assessor;

  • fidelity to administered intervention.

These sensitivity analyses will examine potential sources of methodological heterogeneity.