Eye movement desensitisation and reprocessing for survivors of life‐threatening medical events

Objectives

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

To assess the effects of eye movement desensitisation and reprocessing (EMDR), a psychological intervention programme, on symptoms related to traumatic stress in survivors of life‐threatening medical events.

Secondary objectives: to evaluate whether the effects of EMDR differ according to the nature of the medical event (associated diagnosis or setting), measured outcome (post‐traumatic stress disorder (PTSD), anxiety, depression, or quality of life), or intervention (online, face‐to‐face, group or individual sessions).

Background

Description of the condition

Post‐traumatic stress disorder (PTSD) is a psychiatric condition characterised by specific symptom clusters that cause significant distress and/or functional impairment following exposure to a life‐threatening event. While experiencing psychological distress is a common and normal response to such trauma, the nature, severity, and persistence of symptoms are highly variable. Fear and acute stress frequently diminish over time; however, a substantial portion of the exposed population experience serious and enduring psychological problems. These include disorders of attachment, social engagement, and adjustment and are categorised in the Diagnostic and Statistical Manual of Mental Disorders 5th Edition, Text Revision (DSM‐5‐TR) as trauma‐ and stressor‐related disorders (APA 2022).

PTSD is a severe, disabling, and widely studied trauma‐related disorder. A diagnosis requires each of the following criteria to be present.

A. Exposure to death, life‐threat, or violence.

B. Symptoms of intrusion, e.g. recurrent, unwanted memories, distressing dreams, flashbacks, psycho‐ or physiological distress following internal or external reminders of the traumatic event.

C. Persistent avoidance of distressing thoughts, memories, or external reminders related to the index trauma.

D. Negative alterations in cognition and/or mood, e.g. the inability to remember elements of the trauma, persistent negative self‐belief, emotional state, distorted cognition, diminished interest, detachment, or inability to express positive emotions.

E. Marked alterations in arousal, e.g. irritability, reckless behaviour, hypervigilance to perceived threat, exaggerated startle response, concentration problems, sleep disturbance.

Symptoms must be:

• present for more than one month;

• causing clinically significant distress or functional impairment, or both;

• not attributable to the physiological effects of a substance or medical condition.

The World Health Organization (WHO) World Mental Health Survey reports total population lifetime prevalence of 3.9%, rising to 5.6% among those exposed to trauma (Koenen 2017). In 1994, the DSM‐IV acknowledged that being diagnosed with a life‐threatening illness could be a potential index trauma resulting in PTSD (APA 1994). DSM‐5‐TR asserted that a life‐threatening illness or debilitating medical condition is not necessarily a traumatic event. The interpretation of criterion A, exposure to death, threatened death, or serious injury, following 'sudden, catastrophic medical events, such as waking during surgery or anaphylaxis', is left to the discretion of the diagnosing clinician. This uncertainty has raised conceptual and methodological concerns about the use of medical events as index events for PTSD diagnosis.(Cordova 2017; Greimel 2013). Despite fulfilling established diagnostic criteria, individuals with PTSD stemming from medical events may have experienced an additional biological component, necessitating a nuanced analysis of this specific subpopulation. Regardless, a large body of literature has documented patients experiencing traumatic stress symptoms following life‐threatening medical events. According to meta‐analyses, the incidence of PTSD following a cancer diagnosis is 11% (Abbey 2015), cardiac events 12% (Edmondson 2012), stroke 23% (Edmondson 2013), and 20% following intensive care admission (Righy 2019). PTSD is associated with adverse physical health outcomes, including increased all‐cause mortality (Boscarino 2006), and increased risk of cardiovascular disease (Edmondson 2013(2); Edmondson 2017), cancer (Roberts 2019), chronic inflammation, oxidative stress (Miller 2018), and immunological dysregulation (Sumner 2020).

Furthermore, a person suffering from PTSD is more likely to have a lower quality of life (Balayan 2014), increased healthcare cost, (von der Warth 2020), harmful alcohol and drug use (Hassan 2017), poor medication adherence (Kronish 2012), and postponed return to work (McPeake 2019).

Description of the intervention

Eye movement desensitisation and reprocessing (EMDR) is a trauma‐focused psychological therapy believed to reduce symptoms by helping to process and integrate distressing memories into existing, positive emotional and cognitive neural networks (Novo Navarro 2018). Individuals are guided to recount traumatic memories while receiving bilateral stimulation, most frequently by performing saccadic eye movements by following the therapist’s finger movements (Hase 2021). American psychologist Francine Shapiro observed that side‐to‐side eye movements, while walking in a park, reduced the intensity of disturbing thoughts associated with a traumatic event. She developed, tested, and refined the first eye‐movement desensitisation protocol, by bringing the saccadic eye movements under voluntary control, while recounting distressing memories (Shapiro 2018). By 2001, the eight‐phase EMDR protocol had been established, incorporating a memory reprocessing component. The eight phases are as follows.

1. History taking: obtain background information, determine suitability for treatment, and identify target memories for processing.

2. Preparation: the client and therapist establish a therapeutic alliance while explaining the EMDR process, addressing concerns and questions, and preparing clients with stress‐reduction techniques to support emotional equilibrium between processing phases.

3. Assessment: access a target stressful memory and rate subjective units of distress (SUD), from 0 (no distress) to 10 (worst distress imaginable) and Validity of Cognition (positive and negative), from 1 (completely false) to 7 (completely true).

4. Desensitisation: the client relates the target memory, associated emotions and beliefs, while performing side‐to‐side eye movements, guided by the therapist. Continues until diminution of distress.

5. Installation: the target memory is associated with positive cognition identified in phase 2.

6. Body‐scan: the patient scans their body for residual somatic discomfort associated with the target memory. Repeat phases 4 and 5 if distress remains. If SUD is rated at ≤ 1, return to phase 3 to identify another target memory.

7. Closure: the therapist and patient use relaxation and guided imagery to ensure stability following the session and to address distress that may arise between sessions.

8. Reassessment: evaluate treatment effects, and patient stability.

This protocolised approach allows for fidelity‐rating and testing in controlled studies (Korn 2018). Several meta‐analyses report a positive treatment effect following a range of traumatic experiences (Bisson 2013; Cuijpers 2020; Mavranezouli 2020; Watts 2013) and comparable or better effect than alternative therapies, such as cognitive behavioural therapy (Chen 2015) and pharmacological approaches (Watts 2013). This evidence suggests that EMDR is an effective therapeutic approach for helping individuals to process and overcome traumatic events. Consequently, the WHO (van Ommeren 2013), American Psychiatric Association (APA) (Ursano 2010), International Society for Traumatic Stress Studies (ISTSS) (Bisson 2019), and the National Institute for Health and Care Excellence (NICE) (NICE 2018) recommend EMDR as a candidate therapy for treating PTSD. However, while EMDR has been effective in treating populations who have experienced armed combat, natural disasters, and assault, less is known about its efficacy in treating psychological trauma caused by medical events.

How the intervention might work

The clinical efficacy of EMDR has generated a variety of conjectural theories regarding its underlying mechanism. These theories are largely in line with the Adaptive Information Processing (AIP) model (Hase 2021). Memories, thoughts, and emotions associated with new experiences are believed to be processed into an 'adaptive resolution' within our existing neural networks. This process strengthens our positive emotional and cognitive database, enabling us to draw upon our experiences and use them to our advantage. Conversely, an excitatory neurobiological response to psychological trauma, mediated by neurotransmitters such as cortisol and adrenaline, can disrupt information processing, leading to maladaptive storage of memories in a disturbed state. This can cause memories to be triggered by both internal and external stimuli, resulting in the intrusive and arousal symptoms associated with PTSD. The hypothesis appears to be consistent with observed physiological changes associated with PTSD, such as alterations in neuroendocrine concentrations (Sbisa 2023), a hyperactive limbic system (Landin‐Romero 2018), and reduced thalamic activity (Lanius 2003).

The two core features of EMDR ‐ bringing the traumatic memory into working memory while focusing on bilateral stimulation ‐ are hypothesised to reduce the intensity of the physiological response to the associated memory, creating optimal neurobiological conditions for AIP (Amano 2016). Furthermore, the saccadic eye movements may replicate brain conditions observed during rapid eye‐movement (REM) and slow‐wave sleep, when memories are thought to be consolidated into existing general semantic networks (Pagani 2017; Stickgold 2002). Other potential explanations include enhanced memory processing through increased inter‐ and intra‐hemispheric connectivity (Keller 2014), and attention to dual‐stimuli taxing the working memory system, thereby diminishing associated vividness and emotionality (van den Hout 2014). Despite many studies, a conclusive mechanistic explanation for EMDR remains elusive. However, the suggested theories are not necessarily contradictory, and growing interest in integrative models plausibly suggests that multiple interacting components may be generating the observed positive clinical effect (Coubard 2016).

Why it is important to do this review

Experiencing a life‐threatening medical event can result in persistent and disabling post‐traumatic stress symptoms. This review is important to determine if EMDR may be considered a candidate therapy for patients following life‐threatening medical events, and to identify knowledge gaps that may inform future research.

NICE guidance 2018 recommends EMDR as a treatment for PTSD; however, the evidence base for this guidance was generated using homogeneous populations, primarily survivors of armed combat, assault, and natural and man‐made disasters. Epidemiological research suggests that PTSD following a life‐threatening medical event affects a substantial proportion of survivors, with high associated healthcare costs, but it is unclear if EMDR can be considered a reasonable treatment strategy.

PTSD following external trauma such as armed combat or assault may differ fundamentally from PTSD following internal trauma such as medical events. Researchers have reported fear of illness recurrence and progression (Fait 2018), and elevated internal threat perception (Meli 2017), as the prominent symptom features following medical‐event PTSD. It is therefore important to evaluate therapeutic approaches in this distinct population.

Secondly, a systematic review of randomised controlled trials provides additional information regarding response to EMDR within subgroups of medical populations, for example whether EMDR is more effective for individuals who have suffered a life‐threatening diagnosis or emergency life‐saving medical treatment. This systematic review may also provide additional information regarding the effectiveness of EMDR among particular diagnostic or treatment subgroups.

PTSD commonly co‐exists with anxiety and depressive symptoms. This review may provide information regarding the impact of comorbid mental health issues on clinical response to EMDR.

Finally, previous systematic reviews of psychotherapeutic interventions following life‐threatening medical events did not find any EMDR studies (Birk 2019); seven studies where EMDR was used following cancer diagnosis only (Portigliatti Pomeri 2022); and three studies where EMDR was more effective than alternative psychological therapies (Haerizadeh 2020). However, the risk of bias was unclear, and the overall strength of evidence was weak.

Given these inconclusive reports, a systematic review that adheres to Cochrane guidelines for the identification of trials, heterogeneity assessment, assessment of methodological quality, and measurement of clinical effect is warranted.

Objectives

To assess the effects of eye movement desensitisation and reprocessing (EMDR), a psychological intervention programme, on symptoms related to traumatic stress in survivors of life‐threatening medical events.

Secondary objectives: to evaluate whether the effects of EMDR differ according to the nature of the medical event (associated diagnosis or setting), measured outcome (post‐traumatic stress disorder (PTSD), anxiety, depression, or quality of life), or intervention (online, face‐to‐face, group or individual sessions).

Methods

Criteria for considering studies for this review

Types of studies

We will include randomised controlled trials (RCTs), cluster‐RCTs, and cross‐over RCTs. Peer‐reviewed abstracts, theses, and preliminary reports will be eligible for inclusion. We will not exclude trials based on sample size or language of publication.

Types of participants

Participant characteristics: we will include studies with adult participants (≥ 16 years of age) who have experienced medical event‐related traumatic exposure consistent with the DSM‐5‐TR qualifying criteria ‘life‐threatening medical illness, injury or treatment’. In cases where a study encompasses both eligible and ineligible participants, we will aim to extract data solely from the subset of eligible participants. If this is not possible, we will contact the corresponding author to request the necessary data. If the data are unobtainable, we will exclude the study.

Comorbidities: we will not apply restrictions based on comorbidities.

Setting: we will not apply restrictions on setting of EMDR delivery. We will consider studies where EMDR was delivered remotely, via the internet.

Types of interventions

Experimental intervention

We will include studies where an established EMDR protocol is delivered, either in groups or individually, face‐to‐face or online. Mandatory components of EMDR include focus on the traumatic memory while performing guided saccadic eye movements or other forms of bilateral stimulation (Shapiro 2018).

Comparators

Comparators will include, but not be limited to:

• usual care;

• medication;

• placebo;

• other psychological therapies that are not EMDR;

• wait‐list;

• watch‐and‐wait repeated assessment;

• psychoeducation.

Due to the wide range of comparisons to be examined, we will consider all comparisons for analysis and list all treatment arms in the 'Characteristics of included studies' table. We anticipate grouping the included studies according to the specific comparator used (e.g. EMDR versus placebo, EMDR versus other therapies, EMDR versus wait‐list).

We will analyse all EMDR studies, including those where EMDR is combined with another intervention, as this is common practice, but will undertake a supplementary analysis in which EMDR is delivered as monotherapy.

Types of outcome measures

We will include studies irrespective of whether they evaluate the outcome measures listed below. We will report all outcomes according to time point, defining short term as less than six months; medium term as seven to 12 months; and long term as over 12 months.

Primary outcomes

EMDR is a trauma‐focused psychological intervention programme recommended for the treatment of symptoms associated with post‐traumatic stress. The primary outcome of this review will be changes to continuous measures of post‐traumatic stress symptom severity, using a validated scale. These will include, but not be limited to, clinician assessments such as Clinician‐Administered PTSD Symptom Scale (CAPS‐5) (Blake 1995), and patient‐reported outcome measures such as PTSD Checklist (PCL‐5) (Blevins 2015) or the Impact of Events Scale ‐Revised (IES‐R) (Weiss 1996). We will report data on treatment fidelity where available.

Secondary outcomes

• Severity of depressive symptoms using a validated scale, such as Hospital Anxiety and Depression Scale ‐ Depression (HADS‐D) (Zigmond 1983) or Beck Depression Inventory (BDI) (Beck 1996).

• Severity of anxiety symptoms using a validated scale such as Generalized Anxiety Disorder‐7 (GAD‐7) (Spitzer 2006).

• Adverse events associated with EMDR intervention, as a surrogate marker for acceptability.

• Quality of life using a validated scale such as EQ‐5D‐5L (Herdman 2011).

Search methods for identification of studies

Electronic searches

We will conduct searches of the following electronic databases with the assistance of an academic librarian at the University of Southampton. EMDR was developed and first reported in 1989, therefore search coverage dates will be from 1989 to date of search.

• Cochrane Common Mental Disorders Controlled Trials Register (CCMDCTR)

• Cochrane Central Register of Controlled Trials (CENTRAL) (latest issue)

• MEDLINE Ovid (1989 to date of search)

• MEDLINE PubMed (1989 to date of search)

• Embase (1989 to date of search)

• American Psychological Association PsycINFO (1989 to date of search)

• CINAHL (Cumulative Index to Nursing and Allied Health Literature) (1989 to date of search)

• Web of Science (1989 to date of search)

We will search the following trial registers for ongoing or unpublished trials.

• ClinicalTrials.gov (www.clinicaltrials.gov)

• World Health Organization International Clinical Trials Registry Platform (WHO ICTRP) (apps.who.int/trialsearch/)

• EU Clinical Trials Register (www.clinicaltrialsregister.eu)

The MEDLINE search strategy (Appendix 1) will be translated into appropriate strategies (keywords, subject headings, search syntax) for the remaining databases (Lefebvre 2023).

Searching other resources

We will also search the following resources.

• Reference lists from relevant systematic reviews revealed by the search

• Academic theses (about.proquest.com/en/dissertations/)

• Personal contact with corresponding authors for missing data and ongoing studies

• Treatment guidelines (such as the ISTSS and WHO) (Bisson 2019; van Ommeren 2013)

Data collection and analysis

We will use RevMan software to perform all analyses reported in this review (RevMan 2024).

Selection of studies

We will upload candidate studies identified by the search to Covidence for screening (Covidence). Two review authors (AB and RC) will independently screen the titles and abstracts of records identified by the search against the eligibility criteria. If AB or RC was involved in the conduct, analysis, or publication of a study potentially eligible for inclusion in the review, eligibility of that study will be determined by two other review authors (NH and SC). We will obtain the full‐text reports of those studies deemed potentially eligible and assess these for inclusion in the review. Any disagreements will be resolved through consultation with a third, senior review author. We will record reasons for exclusion and present study selection data using a PRISMA flow diagram and 'Characteristics of excluded studies' table (Liberati 2009; Moher 2009). We will group multiple reports published from a single study under a single study ID.

Data extraction and management

Two review authors (AB and RC) will independently extract study data. Where AB and RC are investigators of an included study, NH and SC will extract data from this study. These data will be spot‐checked against the study reports by a third, senior review author, with any discrepancies resolved by group review. Where data are missing or uncertain, we will contact the manuscript corresponding author or senior listed author, or both.

We will extract the following study characteristics and and enter the data into RevMan software.

• Study details: authors, source (journal, year, volume, pages), language.

• Methods: study design, study duration, setting, recruitment, number of centres.

• Participants: sample size, mean age, age range, gender, characteristics of index medical trauma (primary diagnosis, hospital length of stay, symptom severity), inclusion and exclusion criteria, comorbidity (physical and psychological), withdrawal rates, and exposure to previous/childhood trauma. We will report the specific diagnostic criteria used by studies, and symptom domains, where reported.

• Intervention: we will provide details of the EMDR protocol applied, which components of the protocol were used, number of sessions, adherence, EMDR fidelity assessment, and deviations, where reported.

• Outcomes: means and standard deviation (SD) at baseline and follow‐up to determine treatment effect. Means and SD at baseline and follow‐up for relevant secondary outcomes.

• Additional notes: funding sources and potential conflicts of interest.

We will use a data extraction template developed in accordance with Section 7.5 of the Cochrane Handbook for Systematic Reviews of Interventions (Higgins 2011a).

Assessment of risk of bias in included studies

Two review authors (AB and RC) will independently assess risk of bias in the included studies using the Cochrane RoB 2 tool. We will consider the effect of assignment at baseline, regardless of whether the intervention was received as intended, in an 'intention‐to‐treat' analysis, to maintain the benefit of randomisation (Menerit 2012). Two review authors (NH and SC) will independently assess the risk of bias in studies for which AB and RC are investigators. Where possible, we will undertake a risk of bias assessment for each of the main outcomes for each study, as listed above. Any discrepancies will be resolved through discussion, with a third, senior review author acting as 'judge' where agreement cannot be reached. We will assess risk of bias in cluster‐randomised and cross‐over trials using the Cochrane RoB 2 variant specific to these study designs (Sterne 2019). We will use the RoB 2 Excel tool to manage risk of bias assessment (Sterne 2019). We will assess risk of bias for our primary outcome (PTSD) and secondary outcomes (anxiety, depression, adverse events, and quality of life) at the predetermined time points: short term (less than six months), medium term (seven to 12 months), and long term (over 12 months).

We will assess the following risk of bias domains, as described in the Cochrane Handbook for Systematic Reviews of Interventions (Higgins 2022).

• Bias arising from the randomisation process

• Bias due to deviations from intended interventions

• Bias due to missing outcome data

• Bias in measurement of the outcome

• Bias in selection of reported result

We will use a series of signalling questions, as follows.

• Was the allocation sequence random?

• Was the allocation sequence concealed until participants were enrolled and assigned to interventions?

• Did baseline differences between intervention groups suggest a problem with the randomisation process?

We will map the responses to these questions onto an algorithm to generate a risk of bias judgement and rate each domain as having low risk of bias, some concerns, or high risk of bias. We will include free text boxes to justify responses to the risk of bias judgement.

The risk of bias assessment will inform the GRADE assessment and summary of findings table.

Measures of treatment effect

We will perform data analysis following the guidance in Chapter 10 of the Cochrane Handbook for Systematic Reviews of Interventions (Deeks 2022). We will present data with a consistent direction of effect.

Continuous data (PTSD, depressive and anxiety symptom severity, and quality of life): where studies use the same measures to assess a given outcome, we will report mean difference (MD) and 95% confidence interval (CI). Where studies use different measures to assess the same outcome, we will convert continuous outcome data to standardised mean difference (SMD) and 95% CI.

Dichotomous data (adverse events): we will analyse the risk ratio (RR) with 95% CI.

Unit of analysis issues

We will take into account the level at which randomisation occurred.

For cluster‐randomised trials, if we are unable to extract a direct estimate of the required effect size from an analysis that properly accounts for the cluster design, we will request individual patient data to obtain an estimate for the intracluster correlation coefficient, or will attempt to estimate this in order to re‐analyse the trial data.

We will include relevant cross‐over designs in the review and will attempt to approximate a paired analysis. If carry‐over is thought to be a problem, we will consider including only data collected in the first period of the study (Li 2015).

Dealing with missing data

If there are missing individuals from reported results, summary data for outcomes, or study‐level characteristics, we will contact the corresponding authors and senior authors to collect key missing data. If appropriate, we will consider imputation of missing data, and conduct a sensitivity analysis to determine the effect of this on the results.

Assessment of heterogeneity

We anticipate the presence of clinical, methodological, and statistical diversity across studies. We will measure the degree of statistical heterogeneity using the I² statistic, which will determine suitability for meta‐analysis, and comment on the strength of certainty of the treatment effect following the guidance in the Cochrane Handbook for Systematic Reviews of Interventions (Higgins 2022). We will interpret I² as follows.

• 0% to 40%: might not be important

• 30% to 60%: may represent moderate heterogeneity

• 50% to 90%: may represent substantial heterogeneity

• 75% to 100%: considerable heterogeneity

We will also investigate clinical heterogeneity by examining the variability in participants, interventions, and outcomes. We will inspect variability in study design, outcome measurement tools, and risk of bias to describe methodological heterogeneity.

Assessment of reporting biases

We will assess reporting bias using domain five (selection of reported results) of the RoB 2 tool, and compare the published protocol papers and clinical trial registries and methods sections of papers against the results sections of the manuscript. If we include 10 or more studies, we will use a funnel plot and formal Egger test for statistical asymmetry (Egger 1997), which may indicate publication bias. If there are fewer than 10 included studies, we will report a narrative assessment of the risk of publication bias.

Data synthesis

Meta‐analysis of numerical data

We will undertake a random‐effects meta‐analysis of intervention effect, where we can identify two or more studies that are:

• clinically homogenous studies (participants, intervention outcomes);

• investigating EMDR following a life‐threatening medical event, with comparable follow‐up timing (up to or greater than three months from intervention delivery);

• investigating EMDR versus a homogenous comparator (psychotherapy, placebo, usual care, pharmacology).

We will consider group and individual interventions separately due to uncertainty pertaining to therapeutic 'dose' within group therapies.

Synthesis using other methods

If we are unable to pool outcome data, we will apply the guidance in Chapter 12 of the Cochrane Handbook for Systematic Reviews of Interventions (McKenzie 2022).

Subgroup analysis and investigation of heterogeneity

We do not know whether the underlying aetiology of traumatic medical events differs between receipt of a life‐threatening diagnosis or the nature or setting of life‐saving medical treatment. We will conduct a formal test for subgroup differences using RevMan software to explore whether treatment effect of EMDR differs between:

• index medical event‐related trauma according to a life‐threatening diagnosis (including, but not limited to, cancer diagnosis, cardiac event, stroke);

• index medical event‐related trauma according to treatment modality and/or setting (including, but not limited to, emergency department or intensive care unit admission).

Sensitivity analysis

We will conduct the following sensitivity analyses.

• Excluding studies with an overall high risk of bias, thereby removing potentially biased effect from the meta‐analysis.

• Excluding cluster‐randomised and cross‐over trials.

• If imputation is used to account for missing data, we will run the analysis on the full imputed data, and then use available data only.

• Only including studies where PTSD was the primary outcome.

• Only including studies where PTSD symptom severity was clinician‐assessed or patient‐reported.

• Using just post‐EMDR symptom severity.

Summary of findings and assessment of the certainty of the evidence

We will report findings following the guidance in Chapter 3 of the Cochrane Handbook for Systematic Reviews of Interventions (Cumpston 2022). We will include summary of findings tables, created using GRADEpro GDT software (GRADEpro GDT), to report the following main outcomes of the review.

• Change in PTSD symptom severity

• Depressive symptoms

• Anxiety

• Quality of life

• Incidence of adverse events

We will report outcomes according to predefined time points: short term (less than six months), medium term (seven to 12 months), and long term (over 12 months).

Where meta‐analysis is possible, we will present the certainty of the evidence using the five GRADE considerations (overall risk of bias, consistency of effect, imprecision, indirectness, and publication bias) as recommended in the Cochrane Handbook for Systematic Reviews of Interventions (Schünemann 2022). Two review authors (AB and RC) will independently perform the GRADE assessment, with any discrepancies resolved through discussion or consultation with a third, senior review author (MG) if necessary. We will classify the certainty of evidence as high, moderate, low, or very low. We will provide justification for downgrading the certainty of the evidence in footnotes or a comments column, or both. We will also report additional outcome data not reported in the meta‐analysis, and comment on whether these data support or contradict the meta‐analysis findings.

Appendices

Appendix 1. Preliminary MEDLINE (Ovid) search strategy

Ovid MEDLINE(R) ALL <1989 to January 13, 2023>

1 Eye Movement Desensitization Reprocessing/

2 (eye movement desensitisation or eye movement desensitization or EMDR).mp. [mp=title, book title, abstract, original title, name of substance word, subject heading word, floating sub‐heading word, keyword heading word, organism supplementary concept word, protocol supplementary concept word, rare disease supplementary concept word, unique identifier, synonyms]

3 1 or 2

4 exp Cardiovascular Diseases/

5 (cardiovascular or cvd).tw.

6 exp Myocardial Ischemia/

7 ((myocardial adj infarct$) or mi).tw.

8 Defibrillators, Implantable/

9 (heart adj (attack$ or infarc$ or isch?emia$)).tw.

10 exp Coronary Artery Bypass/

11 (Coronary Artery Bypass or cabg).tw.

12 Angioplasty, Transluminal, Percutaneous Coronary/

13 Percutaneous Transluminal Coronary.tw.

14 (pctci or ptca).tw.

15 (coronary artery disease or cad or acute coronar$ or acs).tw.

16 exp stroke/

17 Stroke$.tw.

18 cerebrovascular.tw.

19 ((brain or vascular or lacunar or venous or cerebral or isch?emic) adj2 (accident$ or infarct$ or event$ or attack$)).tw.

20 (cva or cvas).tw.

21 Ischemic Attack, Transient/

22 (transient adj3 isch?emi$).tw.

23 (tia or tias).tw.

24 exp Neoplasms/

25 (cancer$ or neoplas$ or tumo?r$ or oncolog$).tw.

26 exp Organ Transplantation/

27 ((organ$ or heart$ or kidney$ or lung$ or pancrea$) adj (transplant$ or graft$)).tw.

28 exp HIV Infections/

29 (hiv or Human immunodeficiency virus or aids or Acquired Immune Deficiency Syndrome).tw.

30 exp Asthma/

31 asthma$.tw.

32 acute disease/ or chronic disease/ or critical illness/

33 ((health or medical or physical$ or critical$) adj (ill$ or condition$ or disease$)).tw.

34 exp Intensive Care Units/

35 exp Critical Care/

36 Critical Illness/

37 exp Sepsis/

38 Shock, Septic/

39 Anaphylaxis/

40 anaphyla$.tw.

41 (sepsis or septic$).tw.

42 exp Emergency Service, Hospital/

43 emergency department.tw.

44 (intensive care or icu$).tw.

45 critical care.mp. [mp=title, book title, abstract, original title, name of substance word, subject heading word, floating sub‐heading word, keyword heading word, organism supplementary concept word, protocol supplementary concept word, rare disease supplementary concept word, unique identifier, synonyms]

46 4 or 5 or 6 or 7 or 8 or 9 or 10 or 11 or 12 or 13 or 14 or 15 or 16 or 17 or 18 or 19 or 20 or 21 or 22 or 23 or 24 or 25 or 26 or 27 or 28 or 29 or 30 or 31 or 32 or 33 or 34 or 35 or 36 or 37 or 38 or 39 or 40 or 41 or 42 or 43 or 44 or 45

47 3 and 46

Contributions of authors

AB: conceptualisation, data curation, formal analysis, funding acquisition, project administration, visualisation, methodology, writing ‐ original draft, review and editing.

DSB: conceptualisation, supervision, validation, writing ‐ review and editing.

NP: conceptualisation, supervision, validation, writing ‐ review and editing.

HM: data curation, methodology, formal analysis, writing ‐ review and editing.

NH: formal analysis, writing ‐ review and editing.

SC: formal analysis, writing ‐ review and editing.

MPWG: conceptualisation, supervision, validation, writing ‐ review and editing.

RC: conceptualisation, formal analysis, supervision, validation, visualisation, writing ‐ original draft, review and editing.

All authors reviewed, edited, and approved the final version of the protocol.

Sources of support

Internal sources

• David Baldwin, UK

  University of Southampton, Southern Health NHS Foundation Trust

• Natalie Pattison, UK

  University of Hertfordshire, East and North Herts NHS Trust

• Helen Moyses, UK

  University of Southampton

• Samuele Cortese, UK

  University of Southampton

• Mike Grocott, UK

  MPWG is Director and his salary is part‐funded by NIHR Southampton Biomedical Research Centre.

• Rebecca Cusack, UK

  University Hospital Southampton, University of Southampton

External sources

• Andrew Bates, UK

  Andrew Bates is funded by a National Institute for Health and Care Research (NIHR) (Clinical Doctoral Research Fellowship) for this research project. This article presents independent research funded by the NIHR. The views expressed are those of the author and not necessarily those of the NIHR, NHS, or the Department of Health and Social Care. The funder had no role in the design, conduct, or publication of the review, protocol, or update.

• Mike Grocott, UK

  MG is an NIHR Senior Investigator, and is an elected council member of the Royal College of Anaesthetists. The funder had no role in the design, conduct, or publication of the review, protocol, or update.

• Nathan Huneke, UK

  NH is funded by a Medical Research Council Fellowship.

Declarations of interest

AB: reports being the investigator of a study that is eligible for inclusion in the work; the study was funded by the National Institute for Health and Care Research (NIHR).

DSB: reports being the investigator of a study that is eligible for inclusion in the work; the study was funded by the NIHR. DSB is president of the British Association of Psychopharmacology and Editor‐in‐Chief of Human Psychopharmacology, for which he receives a personal editorial honorarium.

NP: reports being the investigator of a study that is eligible for inclusion in the work; the study was funded by the NIHR.

HM: declares no conflicts of interest.

NH: reports having been awarded a fellowship by the Medical Research Council to research the mechanisms of the placebo response in anxiety disorders.

SC: declares honoraria from the British Association of Psychopharmacology, Canadian ADHD Resource Alliance, Association for Child and Adolescent Mental Health (ACAMH), and Medice UK. He has served on the advisory board of the ACAMH. He has served as deputy editor of Evidence‐Based Mental Health (now BMJ Mental Health), associate editor of Child and Adolescent Mental Health, and on the editorial board of the Journal of Child Psychology and Psychiatry. He declares grants from the NIHR and European Research Agency.

MPWG: reports honoraria from AstraZeneca; personal payment. Reports advisory board fees and research grant award from Edwards Lifesciences; personal payment and paid to institution. Reports consultancy fee from South West Sensor Ltd; paid to institution. Reports consultancy fee from Sphere Medical; personal payment. Reports being the investigator of a study that is eligible for inclusion in the work; the study was funded by NIHR. Reports declaring an opinion on the topic of the review in print and lectures. Reports an affiliation to an organisation that has declared an opinion on the topic.

RC: reports being the investigator of a study that is eligible for inclusion in the work; the study was funded by the NIHR.

New