Psychosocial interventions for conversion and dissociative disorders in adults
Abstract
Background
Conversion and dissociative disorders are conditions where people experience unusual neurological symptoms or changes in awareness or identity. However, symptoms and clinical signs cannot be explained by a neurological disease or other medical condition. Instead, a psychological stressor or trauma is often present. The symptoms are real and can cause significant distress or problems with functioning in everyday life for the people experiencing them.
Objectives
To assess the beneficial and harmful effects of psychosocial interventions of conversion and dissociative disorders in adults.
Search methods
We conducted database searches between 16 July and 16 August 2019. We searched Cochrane Central Register of Controlled Trials (CENTRAL), MEDLINE, Embase, and eight other databases, together with reference checking, citation searching and contact with study authors to identify additional studies.
Selection criteria
We included all randomised controlled trials that compared psychosocial interventions for conversion and dissociative disorders with standard care, wait list or other interventions (pharmaceutical, somatic or psychosocial).
Data collection and analysis
We selected, quality assessed and extracted data from the identified studies. Two review authors independently performed all tasks. We used standard Cochrane methodology. For continuous data, we calculated mean differences (MD) and standardised mean differences (SMD) with 95% confidence interval (CI). For dichotomous outcomes, we calculated risk ratio (RR) with 95% CI. We assessed and downgraded the evidence according to the GRADE system for risk of bias, imprecision, indirectness, inconsistency and publication bias.
Main results
We included 17 studies (16 with parallel‐group designs and one with a cross‐over design), with 894 participants aged 18 to 80 years (female:male ratio 3:1).
The data were separated into 12 comparisons based on the different interventions and comparators. Studies were pooled into the same comparison when identical interventions and comparisons were evaluated. The certainty of the evidence was downgraded as a consequence of potential risk of bias, as many of the studies had unclear or inadequate allocation concealment. Further downgrading was performed due to imprecision, few participants and inconsistency.
There were 12 comparisons for the primary outcome of reduction in physical signs.
Inpatient paradoxical intention therapy compared with outpatient diazepam: inpatient paradoxical intention therapy did not reduce conversive symptoms compared with outpatient diazepam at the end of treatment (RR 1.44, 95% CI 0.91 to 2.28; 1 study, 30 participants; P = 0.12; very low‐quality evidence).
Inpatient treatment programme plus hypnosis compared with inpatient treatment programme: inpatient treatment programme plus hypnosis did not reduce severity of impairment compared with inpatient treatment programme at the end of treatment (MD –0.49 (negative value better), 95% CI –1.28 to 0.30; 1 study, 45 participants; P = 0.23; very low‐quality evidence).
Outpatient hypnosis compared with wait list: outpatient hypnosis might reduce severity of impairment compared with wait list at the end of treatment (MD 2.10 (higher value better), 95% CI 1.34 to 2.86; 1 study, 49 participants; P < 0.00001; low‐quality evidence).
Behavioural therapy plus routine clinical care compared with routine clinical care: behavioural therapy plus routine clinical care might reduce the number of weekly seizures compared with routine clinical care alone at the end of treatment (MD –21.40 (negative value better), 95% CI –27.88 to –14.92; 1 study, 18 participants; P < 0.00001; very low‐quality evidence).
Cognitive behavioural therapy (CBT) compared with standard medical care: CBT did not reduce monthly seizure frequency compared to standard medical care at end of treatment (RR 1.56, 95% CI 0.39 to 6.19; 1 study, 16 participants; P = 0.53; very low‐quality evidence). CBT did not reduce physical signs compared to standard medical care at the end of treatment (MD –4.75 (negative value better), 95% CI –18.73 to 9.23; 1 study, 61 participants; P = 0.51; low‐quality evidence). CBT did not reduce seizure freedom compared to standard medical care at end of treatment (RR 2.33, 95% CI 0.30 to 17.88; 1 trial, 16 participants; P = 0.41; very low‐quality evidence).
Psychoeducational follow‐up programmes compared with treatment as usual (TAU): no study measured reduction in physical signs at end of treatment.
Specialised CBT‐based physiotherapy inpatient programme compared with wait list: no study measured reduction in physical signs at end of treatment.
Specialised CBT‐based physiotherapy outpatient intervention compared with TAU: no study measured reduction in physical signs at end of treatment.
Brief psychotherapeutic intervention (psychodynamic interpersonal treatment approach) compared with standard care: brief psychotherapeutic interventions did not reduce conversion symptoms compared to standard care at end of treatment (RR 0.12, 95% CI 0.01 to 2.00; 1 study, 19 participants; P = 0.14; very low‐quality evidence).
CBT plus adjunctive physical activity (APA) compared with CBT alone: CBT plus APA did not reduce overall physical impacts compared to CBT alone at end of treatment (MD 5.60 (negative value better), 95% CI –15.48 to 26.68; 1 study, 21 participants; P = 0.60; very low‐quality evidence).
Hypnosis compared to diazepam: hypnosis did not reduce symptoms compared to diazepam at end of treatment (RR 0.69, 95% CI 0.39 to 1.24; 1 study, 40 participants; P = 0.22; very low‐quality evidence).
Outpatient motivational interviewing (MI) and mindfulness‐based psychotherapy compared with psychotherapy alone: psychotherapy preceded by MI might decrease seizure frequency compared with psychotherapy alone at end of treatment (MD 41.40 (negative value better), 95% CI 4.92 to 77.88; 1 study, 54 participants; P = 0.03; very low‐quality evidence).
The effect on the secondary outcomes was reported in 16/17 studies. None of the studies reported results on adverse effects. In the studies reporting on level of functioning and quality of life at end of treatment the effects ranged from small to no effect.
Authors' conclusions
The results of the meta‐analysis and reporting of single studies suggest there is lack of evidence regarding the effects of any psychosocial intervention on conversion and dissociative disorders in adults. It is not possible to draw any conclusions about potential benefits or harms from the included studies.
PICO
Plain language summary
Therapeutic and social interventions for conversion and dissociative disorders
The aim of this review is to provide a better understanding of what is an effective and useful intervention (treatment) for people with conversion disorders and dissociative disorders. The interventions we look at are non‐medical. Instead they concern therapy or social interventions.
Background
Conversion and dissociative disorders are conditions where people experience unusual neurological symptoms (relating to the nerves and nervous system) or changes in awareness or identity. Neurological disease or other medical conditions cannot explain these clinical signs; often a psychological (affecting or arising in the mind) stressor or trauma is present. The symptoms are real and can cause distress or problems with functioning in everyday life for the people experiencing them.
This review seeks to help these patients, as well as the clinicians, policy makers and healthcare services working with these disorders.
Review question
What is the evidence for psychosocial (relating to social factors and individual thought and behaviour) intervention of conversion and dissociative disorders?
Search date
We searched medical databases between 16 July and 16 August 2019.
Results of search
We read 3048 summaries of articles, resulting in 17 studies that met our criteria for the conditions, the groups of people, the interventions and the types of studies that are the focus of this review.
The 17 studies had 894 participants, and each study was relatively small.
More studies are under way, and we will include them in updates of this review.
Study characteristics
The studies took place in nine different countries worldwide, with adults aged 18 to 80 years, who had a diagnosis of conversion or dissociative disorder for any length of time. Some studies were conducted in either psychiatric or neurological settings. Some included people already in hospital, some included people attending outpatient clinics.
The interventions were all psychosocial, meaning that they focused on psychological or social interventions such as therapy, hypnosis or simply teaching people about their illness. The number of sessions varied.
The included studies all compared the intervention to a control group to see if the interventions made any difference. The control groups received a different psychosocial intervention, medication or the care that people would normally get if they had the same condition but were not part of a research study.
The primary outcome we looked for was a reduction in physical signs.
Key results
We investigated the effect of different types of psychosocial interventions, ranging from hypnosis to behavioural therapy. None of the studies were conducted to a high enough standard to be able to say anything conclusive about the evidence of the results.
There was a reduction in physical signs at the end of treatment for three interventions.
Hypnosis reduced the severity of impairment compared to people on a wait list for treatment; behavioural therapy, given on top of routine care to inpatients, reduced the number of weekly seizures (fits) and symptom severity compared with people receiving routine care alone; and psychotherapy preceded by motivational interviewing (a talking therapy that attempts to move an person away from a state of indecision or uncertainty to positivity) compared with psychotherapy alone reduced seizure frequency.
Quality of the evidence
Most of the included studies had methodological flaws and the quality of evidence used to assess the effectiveness of the different treatments was judged as low or very low. Due to this low‐quality evidence, we cannot say how reliable the results are.
Conclusion
The results of the meta‐analysis and reporting of single studies suggest that there is lack of evidence regarding the effects of any psychosocial intervention of conversion and dissociative disorders in adults. Therefore, it is not possible to draw any conclusions about potential benefits or harms from the included studies.
However, the review shows that research in this area is possible.
Authors' conclusions
Summary of findings
| Paradoxical intention therapy compared with diazepam for conversion disorder | ||||||
| Patient or population: people with conversion disorder according to DSM‐IV or ICD‐10 criteria Settings: outpatient and inpatient Intervention: paradoxical intention therapy Comparison: diazepam over 45 days | ||||||
| Outcomes | Illustrative comparative risks* (95% CI) | Relative effect | No of participants | Certainty of the evidence | Comments | |
|---|---|---|---|---|---|---|
| Assumed risk | Corresponding risk | |||||
| Diazepam | Paradoxical intention therapy | |||||
| Reduction in physical signs (number of patients without any conversive attacks in last 2 weeks) End of treatment | Study population | RR 1.44 (0.91 to 2.28) | 30 | ⊕⊝⊝⊝ | Paradoxical intention therapy may have no effect on physical signs at end of treatment. | |
| 600 per 1000 | 864 per 1000 | |||||
| Level of functioning | — | — | — | — | — | No studies assessed this outcome. |
| Quality of life | — | — | — | — | — | No studies assessed this outcome. |
| Adverse events | — | — | — | — | — | No studies assessed this outcome. |
| *The basis for the assumed risk (e.g. the median control group risk across studies) is provided in footnotes. The corresponding risk (and its 95% confidence interval) is based on the assumed risk in the comparison group and the relative effect of the intervention (and its 95% CI). CI: confidence interval; DSM‐IV:Diagnostic and Statistical Manual of Mental Disorders 4th Edition; ICD‐10:International Classification of Diseases, Tenth Revision; RR: risk ratio. | ||||||
| GRADE Working Group grades of evidence | ||||||
| aDowngraded two levels due to imprecision (wide confidence intervals; based on one study with few patients). | ||||||
| Hypnosis + treatment as usual compared with treatment as usual for conversion disorder | ||||||
| Patient or population: people with conversion disorder according to DSM‐IV or ICD‐10 criteria Settings: inpatient Intervention: hypnosis + TAU Comparison: TAU | ||||||
| Outcomes | Illustrative comparative risks* (95% CI) | Relative effect | No of participants | Certainty of the evidence | Comments | |
|---|---|---|---|---|---|---|
| Assumed risk | Corresponding risk | |||||
| TAU | Hypnosis + TAU | |||||
| Reduction in physical signs (Severity of impairment) Measured by the VRMC scale (higher is better) Range: 1–7 End of treatment | The mean reduction in physical signs in the control group was 5.9 | MD 0.49 lower (1.28 lower to 0.30 higher) | — | 45 | ⊕⊝⊝⊝ | Hypnosis + TAU may have no effect on physical signs at end of treatment |
| Level of functioning | — | — | — | — | — | No studies assessed this outcome |
| Quality of life | — | — | — | — | — | No studies assessed this outcome |
| Adverse events | — | — | — | — | — | No studies assessed this outcome |
| *The basis for the assumed risk (e.g. the median control group risk across studies) is provided in footnotes. The corresponding risk (and its 95% confidence interval) is based on the assumed risk in the comparison group and the relative effect of the intervention (and its 95% CI). CI: confidence interval; DSM‐IV:Diagnostic and Statistical Manual of Mental Disorders 4th Edition; ICD‐10:International Classification of Diseases, Tenth Revision; MD: mean difference; TAU: treatment as usual; VRMC: The Video Rating Scale for Motor Conversion Symptoms. | ||||||
| GRADE Working Group grades of evidence | ||||||
| aDowngraded one level due to high risk of bias. | ||||||
| Hypnosis compared with wait list for conversion disorder | ||||||
| Patient or population: people with conversion disorder according to DSM‐IV or ICD‐10 criteria Settings: outpatient Intervention: hypnosis Comparison: wait list | ||||||
| Outcomes | Illustrative comparative risks* (95% CI) | Relative effect | No of participants | Certainty of the evidence | Comments | |
|---|---|---|---|---|---|---|
| Assumed risk | Corresponding risk | |||||
| Wait list | Hypnosis | |||||
| Reduction in physical signs (Severity of impairment) Measured by the VRMC scale (higher is better) Range: 1–7 End of treatment | The mean reduction in physical signs in the control group was 3.8 | MD 2.10 higher (1.34 higher to 2.86 higher) | — | 49 | ⊕⊕⊝⊝ | Hypnosis may have little effect on reduction in physical signs at end of treatment |
| Level of functioning | — | — | — | — | — | No studies assessed this outcome |
| Quality of life | — | — | — | — | — | No studies assessed this outcome |
| Adverse events | — | — | — | — | — | No studies assessed this outcome |
| *The basis for the assumed risk (e.g. the median control group risk across studies) is provided in footnotes. The corresponding risk (and its 95% confidence interval) is based on the assumed risk in the comparison group and the relative effect of the intervention (and its 95% CI). CI: confidence interval; DSM‐IV:Diagnostic and Statistical Manual of Mental Disorders 4th Edition; ICD‐10:International Classification of Diseases, Tenth Revision; MD: mean difference; VRMC: Video Rating Scale for Motor Conversion Symptoms. | ||||||
| GRADE Working Group grades of evidence | ||||||
| aDowngraded one level due to high risk of bias. | ||||||
| Behavioural therapy + TAU compared with TAU for conversion disorder | ||||||
| Patient or population: people with conversion disorder according to DSM‐IV or ICD‐10 criteria Settings: inpatient Intervention: behavioural therapy + TAU Comparison: TAU | ||||||
| Outcomes | Illustrative comparative risks* (95% CI) | Relative effect | No of participants | Certainty of the evidence | Comments | |
|---|---|---|---|---|---|---|
| Assumed risk | Corresponding risk | |||||
| TAU | Behavioural therapy + TAU | |||||
| Reduction in physical signs Number of weekly seizures, as assessed by daily self‐reported treatment diary End of treatment | The mean reduction in physical signs in the control group was 27.8 | MD 21.40 lower (27.88 lower to 14.92 lower) | — | 18 | Very lowa,b ⊕⊝⊝⊝ | Behavioural therapy + TAU may have little effect on reduction in physical signs at end of treatment. |
| Reduction in physical signs (symptom severity) Measured by Clinical Global Impression scale (lower is better) Range: 1–7 End of treatment | The mean reduction in physical signs in the control group was 4.48 | MD 2.90 lower (3.41 lower to 2.39 lower) | — | 90 | Very lowa,b ⊕⊝⊝⊝ | Behavioural therapy + TAU may have little effect on reduction in physical signs at end of treatment. |
| Level of functioning | — | — | — | — | — | No studies assessed this outcome. |
| Quality of life | — | — | — | — | — | No studies assessed this outcome. |
| Adverse events | — | — | — | — | — | No studies assessed this outcome. |
| *The basis for the assumed risk (e.g. the median control group risk across studies) is provided in footnotes. The corresponding risk (and its 95% confidence interval) is based on the assumed risk in the comparison group and the relative effect of the intervention (and its 95% CI). CI: Confidence interval; DSM‐IV:Diagnostic and Statistical Manual of Mental Disorders 4th Edition; ICD‐10:International Classification of Diseases, Tenth Revision; MD: mean difference. | ||||||
| GRADE Working Group grades of evidence | ||||||
| aDowngraded two levels due to high risk of bias. | ||||||
| Cognitive behavioural therapy compared with standard medical care for conversion disorder | ||||||
| Patient or population: people with conversion disorder according to DSM‐IV or ICD‐10 criteria Settings: outpatient Intervention: cognitive behavioural therapy Comparison: standard medical care | ||||||
| Outcomes | Illustrative comparative risks* (95% CI) | Relative effect | No of participants | Certainty of the evidence | Comments | |
|---|---|---|---|---|---|---|
| Assumed risk | Corresponding risk | |||||
| Standard medical care | Cognitive behavioural therapy | |||||
| Reduction in physical signs Reduction in monthly seizure frequency as assessed by a daily self‐reported seizure diary End of treatment | Study population | RR 1.56 (0.39 to 6.19) | 16 | ⊕⊝⊝⊝ | Cognitive behavioural therapy may have little or no effect on reducing physical signs at end of treatment. | |
| 286 per 1000 | 446 per 1000 | |||||
| Reduction in physical signs Monthly seizure frequency as assessed by a daily self‐reported seizure diary (lower is better) End of treatment | The mean reduction in physical signs in the control group was 6.75 | MD –4.75 lower (18.73 lower to 9.23 higher) | — | 61 | ⊕⊕⊝⊝ | Cognitive behavioural therapy may have little or no effect on reducing physical signs at end of treatment. |
| Reduction in physical sign Seizure freedom as assessed by a daily self‐reported seizure diary End of treatment | Study population | RR 2.33 (0.30 to 17.88) | 16 | ⊕⊝⊝⊝ | Cognitive behavioural therapy may have little or no effect on reducing physical signs at end of treatment. | |
| 143 per 1000 | 333 per 1000 | |||||
| Level of functioning As measured by the GAF (range 0–100) scale and the WSAS scale (0–40) (lower is better) End of treatment | — | SMD 0.44 higher (1.69 lower to 2.57 higher) I2 = 91% | — | 74 | ⊕⊝⊝⊝ | Cognitive behavioural therapy may have little or no effect on level of functioning at end of treatment. |
| Quality of life As assessed by QOLIE31 (higher is better) Range: 15–97 End of treatment | The mean quality of life in the control group was 9.7 | MD 11.20 higher (7.98 lower to 30.38 higher) | — | 16 | ⊕⊝⊝⊝ | Cognitive behavioural therapy may have little or no effect on quality of life at end of treatment. |
| Adverse events | — | — | — | — | — | No studies assessed this outcome. |
| *The basis for the assumed risk (e.g. the median control group risk across studies) is provided in footnotes. The corresponding risk (and its 95% confidence interval) is based on the assumed risk in the comparison group and the relative effect of the intervention (and its 95% CI). CI: confidence interval; DSM‐IV:Diagnostic and Statistical Manual of Mental Disorders 4th Edition; GAF: global assessment of functioning; ICD‐10:International Classification of Diseases, Tenth Revision; MD: mean difference; QOLIE31: quality of life in epilepsy inventory; RR: Risk Ratio; SMD: standardised mean difference; WSAS: work and social adjustment scale. | ||||||
| GRADE Working Group grades of evidence | ||||||
| aDowngraded one level due to high risk of bias. | ||||||
| Systematic follow‐up programme compared with TAU for conversion disorder | ||||||
| Patient or population: people with conversion disorder according to DSM‐IV or ICD‐10 criteria Settings: outpatient Intervention: systematic follow‐up programme Comparison: TAU | ||||||
| Outcomes | Illustrative comparative risks* (95% CI) | Relative effect | No of participants | Certainty of the evidence | Comments | |
|---|---|---|---|---|---|---|
| Assumed risk | Corresponding risk | |||||
| TAU | Systematic follow‐up programme | |||||
| Reduction in physical signs End of treatment | — | — | — | — | — | No studies assessed this outcome at end of treatment. |
| Level of functioning As assessed by WSAS scale (lower is better) Range: 0–40 End of treatment | The mean level of functioning in the control group was 25.52 | MD 7.12 lower (12.47 lower to 1.77 lower) | — | 43 | ⊕⊝⊝⊝ | Psychoeducational follow‐up programme may have little effect on level of functioning at end of treatment. |
| Quality of life | — | — | — | — | — | No studies assessed this outcome. |
| Adverse events | — | — | — | — | — | No studies assessed this outcome. |
| *The basis for the assumed risk (e.g. the median control group risk across studies) is provided in footnotes. The corresponding risk (and its 95% confidence interval) is based on the assumed risk in the comparison group and the relative effect of the intervention (and its 95% CI). CI: confidence interval; DSM‐IV:Diagnostic and Statistical Manual of Mental Disorders 4th Edition; ICD‐10:International Classification of Diseases, Tenth Revision; MD: mean difference; TAU: treatment as usual; WSAS: Work and Social Adjustment Scale. | ||||||
| GRADE Working Group grades of evidence | ||||||
| aDowngraded two levels due to high risk of bias. | ||||||
| Specialised CBT‐based physiotherapy programme compared with wait list for conversion disorder | ||||||
| Patient or population: people with conversion disorder according to DSM‐IV or ICD‐10 criteria Settings: inpatient Intervention: specialised CBT‐based physiotherapy programme Comparison: wait list | ||||||
| Outcomes | Illustrative comparative risks* (95% CI) | Relative effect | No of participants | Certainty of the evidence | Comments | |
|---|---|---|---|---|---|---|
| Assumed risk | Corresponding risk | |||||
| Wait list | Specialised CBT‐based physiotherapy | |||||
| Reduction in physical signs End of treatment | — | — | — | — | — | No studies assessed this outcome. |
| Level of functioning As measured by FIM (higher is better) Range: 18–126 End of treatment | The mean level of functioning in the control group was 80.9 | MD 9.20 higher (6.06 higher to 12.34 higher) | — | 118 | ⊕⊕⊝⊝ | Specialised CBT‐based physiotherapy may slightly improve level of functioning at end of treatment. |
| Quality of life | — | — | — | — | — | No studies assessed this outcome. |
| Adverse events | — | — | — | — | — | No studies assessed this outcome. |
| *The basis for the assumed risk (e.g. the median control group risk across studies) is provided in footnotes. The corresponding risk (and its 95% confidence interval) is based on the assumed risk in the comparison group and the relative effect of the intervention (and its 95% CI). CBT: cognitive behavioural therapy; CI: confidence interval; DSM‐IV:Diagnostic and Statistical Manual of Mental Disorders 4th Edition; FIM: Functional Independence Measure Motor; ICD‐10:International Classification of Diseases, Tenth Revision; MD: mean difference. | ||||||
| GRADE Working Group grades of evidence | ||||||
| aDowngraded one level due to high risk of bias. | ||||||
| Specialised CBT‐based physiotherapy‐led intervention compared with TAU for conversion disorder | ||||||
| Patient or population: people with conversion disorder according to DSM‐IV or ICD‐10 criteria Settings: outpatients at day hospital Intervention: specialised CBT‐based physiotherapy‐led intervention Comparison: TAU | ||||||
| Outcomes | Illustrative comparative risks* (95% CI) | Relative effect | No of participants | Certainty of the evidence | Comments | |
|---|---|---|---|---|---|---|
| Assumed risk | Corresponding risk | |||||
| TAU | Specialised CBT‐based physiotherapy‐led intervention | |||||
| Reduction in physical signs End of treatment | — | — | — | — | — | No studies assessed this outcome at end of treatment. |
| Level of functioning As assessed by WSAS scale (lower is better) Range: 0–40 End of treatment | The mean level of functioning in the control group was 26.9 | MD 7.10 lower (11.40 lower to 2.80 lower) | — | 54 | ⊕⊝⊝⊝ | Specialised CBT‐based physiotherapy intervention may slightly improve level of functioning at end of treatment. |
| Quality of life | — | — | — | — | — | No studies assessed this outcome. |
| Adverse events | — | — | — | — | — | No studies assessed this outcome. |
| *The basis for the assumed risk (e.g. the median control group risk across studies) is provided in footnotes. The corresponding risk (and its 95% confidence interval) is based on the assumed risk in the comparison group and the relative effect of the intervention (and its 95% CI). CBT: cognitive behavioural therapy; CI: confidence interval; DSM‐IV:Diagnostic and Statistical Manual of Mental Disorders 4th Edition; ICD‐10:International Classification of Diseases, Tenth Revision; TAU: treatment as usual; WSAS: Work and Social Adjustment Scale. | ||||||
| GRADE Working Group grades of evidence | ||||||
| aDowngraded one level due to high risk of bias. | ||||||
| Brief psychotherapeutic intervention compared with standard care for conversion disorder | ||||||
| Patient or population: people with conversion disorder according to DSM‐IV or ICD‐10 criteria Settings: outpatients Intervention: brief psychotherapeutic intervention Comparison: standard care | ||||||
| Outcomes | Illustrative comparative risks* (95% CI) | Relative effect | No of participants | Certainty of the evidence | Comments | |
|---|---|---|---|---|---|---|
| Assumed risk | Corresponding risk | |||||
| Standard care | Brief psychotherapeutic intervention | |||||
| Reduction in physical signs As assessed by SDQ‐20 (lower is better) Range: 20–100 End of treatment | Study population | RR 0.12 (0.01 to 2.00) | 19 | ⊕⊝⊝⊝ | Brief psychotherapeutic intervention may have no effect on physical signs at end of treatment. | |
| 400 per 1000 | 48 per 1000 | |||||
| Level of functioning | — | — | — | — | — | No studies assessed this outcome. |
| Quality of life As assessed by SF‐36 (lower is better) Range: 0–100 End of treatment | The mean quality of life in the control group was 50.56 | MD 6.99 lower (28.09 lower to 14.11 higher) | 16 | ⊕⊝⊝⊝ | Brief psychotherapeutic intervention may have little effect on quality of life after end of treatment. | |
| Adverse events | No studies assessed this outcome. | |||||
| *The basis for the assumed risk (e.g. the median control group risk across studies) is provided in footnotes. The corresponding risk (and its 95% confidence interval) is based on the assumed risk in the comparison group and the relative effect of the intervention (and its 95% CI). CI: confidence interval; DSM‐IV:Diagnostic and Statistical Manual of Mental Disorders 4th Edition; ICD‐10:International Classification of Diseases, Tenth Revision; MD: mean difference; RR: Risk Ratio; SDQ‐20: somatoform dissociation questionnaire; SF‐36: 36‐item Short Form. | ||||||
| GRADE Working Group grades of evidence | ||||||
| aDowngraded one level due to high risk of bias. | ||||||
| CBT + APA compared with CBT alone for conversion disorder | ||||||
| Patient or population: people with conversion disorder according to DSM‐IV or ICD‐10 criteria Settings: outpatients Intervention: CBT + APA Comparison: CBT | ||||||
| Outcomes | Illustrative comparative risks* (95% CI) | Relative effect | No of participants | Certainty of the evidence | Comments | |
|---|---|---|---|---|---|---|
| Assumed risk | Corresponding risk | |||||
| CBT | CBT + APA | |||||
| Reduction in physical signs (overall physical impact) As measured by PMDRS, total score (lower is better) Range: 0–128 End of treatment | The mean reduction in physical signs in the control group was 33.2 | MD 5.60 higher (15.48 lower to 26.68 higher) | — | 21 | ⊕⊝⊝⊝ | CBT + APA may have no effect on physical signs at end of treatment. |
| Level of functioning | — | — | — | — | — | No studies assessed this outcome. |
| Quality of life | — | — | — | — | — | No studies assessed this outcome. |
| Adverse events | — | — | — | — | — | No studies assessed this outcome. |
| *The basis for the assumed risk (e.g. the median control group risk across studies) is provided in footnotes. The corresponding risk (and its 95% confidence interval) is based on the assumed risk in the comparison group and the relative effect of the intervention (and its 95% CI). APA: adjunctive physical activity; CBT: cognitive behavioural therapy; CI: confidence interval; DSM‐IV:Diagnostic and Statistical Manual of Mental Disorders 4th Edition; ICD‐10:International Classification of Diseases, Tenth Revision; MD: mean difference; PMDRS: Psychogenic Movement Disorders Rating Scale. | ||||||
| GRADE Working Group grades of evidence | ||||||
| aDowngraded one level due to high risk of bias. | ||||||
| Hypnosis compared with diazepam for conversion disorder | ||||||
| Patient or population: people with conversion disorder according to DSM‐IV or ICD‐10 criteria Settings: emergency unit Intervention: hypnosis Comparison: diazepam | ||||||
| Outcomes | Illustrative comparative risks* (95% CI) | Relative effect | No of participants | Certainty of the evidence | Comments | |
|---|---|---|---|---|---|---|
| Assumed risk | Corresponding risk | |||||
| Diazepam | Hypnosis | |||||
| Reduction in physical signs Number with symptom freedom End of treatment | Study population | RR 0.69 (0.39 to 1.24) | 40 | ⊕⊝⊝⊝ | Hypnosis may have no effect on physical signs at end of treatment. | |
| 650 per 1000 | 448 per 1000 | |||||
| Level of functioning | — | — | — | — | — | No studies assessed this outcome. |
| Quality of life | — | — | — | — | — | No studies assessed this outcome. |
| Adverse events | — | — | — | — | — | No studies assessed this outcome. |
| *The basis for the assumed risk (e.g. the median control group risk across studies) is provided in footnotes. The corresponding risk (and its 95% confidence interval) is based on the assumed risk in the comparison group and the relative effect of the intervention (and its 95% CI). CI: confidence interval; DSM‐IV:Diagnostic and Statistical Manual of Mental Disorders 4th Edition; ICD‐10:International Classification of Diseases, Tenth Revision; RR: risk ratio. | ||||||
| GRADE Working Group grades of evidence | ||||||
| aDowngraded one level due to high risk of bias. | ||||||
| Psychotherapy preceded by motivational interviewing compared with psychotherapy alone for conversion disorder | ||||||
| Patient or population: people with conversion disorder according to DSM‐IV or ICD‐10 criteria Settings: outpatient Intervention: psychotherapy preceded by motivational interviewing Comparison: psychotherapy alone | ||||||
| Outcomes | Illustrative comparative risks* (95% CI) | Relative effect | No of participants | Certainty of the evidence | Comments | |
|---|---|---|---|---|---|---|
| Assumed risk | Corresponding risk | |||||
| Psychotherapy | Psychotherapy preceded by motivational interviewing | |||||
| Reduction in physical signs Decrease in seizure frequency End of treatment | The mean reduction in physical signs in the control group was 34.8 | MD 41.40 higher (4.92 higher to 77.88 higher) | — | 54 | ⊕⊝⊝⊝ | Psychotherapy preceded by motivational interviewing may have little effect on physical signs at end of treatment. |
| Level of functioning | — | — | — | — | — | No studies assessed this outcome. |
| Quality of life As measured by QOLIE10 (lower is better) Range: 10–50 End of treatment | The mean quality of life in the control group was 1.8 | MD 5.40 higher (0.26 higher to 10.54 higher) | — | 47 | ⊕⊝⊝⊝ | Psychotherapy preceded by motivational interviewing may have little effect on quality of life at end of treatment. |
| Adverse events | — | — | — | — | — | No studies assessed this outcome. |
| *The basis for the assumed risk (e.g. the median control group risk across studies) is provided in footnotes. The corresponding risk (and its 95% confidence interval) is based on the assumed risk in the comparison group and the relative effect of the intervention (and its 95% CI). CI: Confidence interval; DSM‐IV:Diagnostic and Statistical Manual of Mental Disorders 4th Edition; ICD‐10:International Classification of Diseases, Tenth Revision; MD: mean difference; QOLIE10: quality of life in epilepsy. | ||||||
| GRADE Working Group grades of evidence | ||||||
| aDowngraded one level due to high risk of bias. | ||||||
Background
Conversion and dissociative disorders have been described in psychology and medicine in a scientific context since Freud, Charcot and Janet (Charcot 1887; Freud 1896; Janet 1907). These early pioneers proposed some of the first theories of both cause and effective treatment.
In the following, we have described the current definitions of these conditions, given a brief overview of the range of different psychosocial interventions that are now used, as well as linked these to the different theories regarding how an intervention might work.
Description of the condition
In this review, we include both dissociative disorders and conversion disorders as defined in the Diagnostic and Statistical Manual of Mental Disorders 5th Edition (DSM‐5) (APA 2013), and the International Classification of Diseases, Tenth Revision (ICD‐10) (WHO 1993). In ICD‐10, conversion disorders are included in the category of dissociative disorders (WHO 1993), whereas DSM‐5 defines the conditions separately (APA 2013).
Dissociation is a collapse of usually integrated functions such as awareness, memory, orientation, or sensory and motor function in response to, or by reactivating, a severe emotional stress or trauma, whereas examples of conversion disorders (in the DSM‐5 definition) are psychogenic non‐epileptic seizures (PNES), paralysis, gait disturbances, anaesthesia, tremor, dystonia (involuntarily muscle contractions, causing repetitive or twisting movements), functional blindness or aphonia (loss of voice due to disturbance of the vocal organs).
DSM‐5 (APA 2013) defines conversion disorder as:
-
one or more symptoms of altered voluntary motor or sensory function;
-
clinical findings that show evidence of incompatibility between the symptoms and recognised neurological or medical conditions;
-
symptoms or deficit that are not better explained by another medical or mental disorder;
-
symptoms or deficit that cause clinically significant distress or impairment in social, occupational or other important areas of functioning or warrants medical evaluation.
In DSM‐5, the dissociative disorders are placed independently and separated from the conversion section and next to, but not as part of, the trauma‐ and stressor‐related disorders, reflecting the close relationship between these diagnostic classes. Dissociative disorders in DSM‐5 include dissociative identity disorder (disruption of identity characterised by two or more distinct personality states), depersonalisation/derealisation disorder (experiences of unreality or detachment from one's mind or self or detachment from one's surroundings) and dissociative amnesia (inability to recall autobiographical information). It may sometimes involve travel or confused wandering away from one's life (dissociative fugue).
Common for these conditions is that a psychological stressor or trauma is often present, but is not a requirement to establish the diagnosis.
The epidemiology of conversion disorder is complicated by case definition, case ascertainment and identifying a suitable study population, but community surveys suggest a minimum prevalence of 50 per 100,000. Deveci 2007 found the prevalence in a city population to be 5.6% and Stone 2009 found conversion symptoms for 18%, among people referred to neurological outpatient clinics in the UK. There is no good evidence to suggest that conversion disorder is becoming less common or that it is more often found in low‐income countries. Conversion symptoms can occur in both men and women although all case series show a predominance of women (Akagi 2002). Several studies have pointed out that childhood trauma, specifically physical or sexual abuse, emotional or physical neglect, and a greater number of stressful life events and traumatic episodes characterise conversion disorders (Kranick 2011; Roelofs 2002; Sar 2009). Scevola 2013 found trauma history in 49% of people with PNES and sexual abuse is particularly prevalent in this group, present in 30% (Stone 2004) to 45% (Selkirk 2008) of people with PNES.
With regard to dissociative disorders, Johnson surveyed a population of 658 people in New York and found a one‐year prevalence of dissociative disorders of 9.1% (conversion disorders not included) (Johnson 2006). A different study likewise found a prevalence of dissociative disorders of 10% (Ross 1991).
In some clinical terminology, the concept of dissociation may be used synonymously with psychogenic dissociation (e.g. amnesia, fugue, stupor, trance or dissociative personality disorder), whereas the concept of conversion may be described clinically as somatoform dissociation. We included both somatoform dissociation (conversion) and psychoform dissociation in this review.
The prognosis for people with conversion disorder is poorly studied, but results from several studies suggest the prospects for immediate recovery are good but a significant number of people will relapse. Factors associated with quick recovery are an acute onset and prompt treatment (Ron 2001). Prognosis may also be influenced by symptom pattern and some studies suggest that non‐epileptic attacks and people presenting with tremor or amnesia have a poorer prognostic outcome than those with hysterical blindness, aphonia and motor disorders (Toone 1990). Trauma in childhood seems to have an impact on the seriousness of the development and progress of conversion disorder (Selkirk 2008; Stone 2004).
Description of the intervention
Psychosocial interventions include all psychological interventions specified in the UK Department of Health review of psychological therapies (Department of Health 2001), social interventions such as social skills training and befriending and packages of interventions that have a psychosocial focus.
Many psychosocial interventions have been developed for conversion and dissociative disorders. Most have been developed over decades based on observations and theory formation in the early years of psychiatry and psychology by pioneers such as Charcot, Freud and Janet (Charcot 1887; Freud 1896; Janet 1907), and later as part of behavioural or cognitive approaches.
Some of the most common psychosocial interventions include the following.
-
Behavioural therapy consists of graded exposure to the body sensations or to situations perceived as threatening in order to reduce the patient’s apprehensive reaction towards them. If trauma is a part of the condition, it may be practised as 'prolonged exposure' (Myers 2017), or be combined with other forms of therapy.
-
Cognitive behavioural therapy (CBT) is a structured, present‐oriented psychotherapy directed towards solving current problems and teaching the patients skills to modify unhelpful thinking and behaviour. One important part of CBT is helping clients change their unhelpful thinking and behaviour, which leads to enduring improvement in their mood and functioning.
-
Psychoeducation is connected to CBT and concerns teaching patients about their condition, thereby empowering them to take an active part in the management of the condition and the recovery from it (Colom 2011; Zhao 2015).
-
Hypnosis is a therapeutic technique in which clinicians make suggestions to people who have undergone a procedure designed to relax them and focus their minds. Hypnosis strengthens the ability to handle emotional stress and it often forms part of a treatment package, for example, together with CBT or eye movement desensitisation reprocessing (EMDR) (Fine 2012; Fine 2001).
-
Psychodynamic andpsychoanalytic therapies consist of a set of psychological therapies, where psychological symptoms can be seen as manifestations of intrapsychic or unconscious conflicts. These therapies use different therapeutic strategies to reveal, interpret and resolve such conflicts. In the field of dissociation or conversion, early life events and trauma experience may be an important part of the therapy.
-
Specialised physiotherapy contains elements of CBT and aims to retrain motor or sensory function by redirecting attention and addressing unhelpful illness beliefs and behaviours. The patient's problems are considered in a broad biopsychosocial framework where symptom‐predisposing, precipitating and perpetuating factors can be addressed within a multidisciplinary environment (Nielsen 2016).
-
Paradoxical intention therapy is where the therapist encourages the patient to engage in the unwanted behaviour, promoting the worsening of the symptoms rather than their removal.
-
Models of dissociation and trauma treatment. In phase‐oriented treatments, patients are working from establishing safety and stability, for example, symptom reduction, emphasised emotion regulation and impulse control (Stage 1), to focus on maintaining stability while exploring trauma narratives and resolving trauma‐related emotions (Stage 2) and finally to emphasise integration and living without reliance on dissociation (Stage 3) (Dorahy 2014; van der Hart 2012).
-
Eye movement desensitisation reprocessing (EMDR) is a structured therapy that encourages the patient to briefly focus on the trauma memory while simultaneously experiencing bilateral stimulation (eye movements), which is associated with a reduction in the vividness and emotion associated with the trauma memories. It is a treatment divided into phases where desensitisation in relation to a trauma is essential.
Psychosocial interventions can be delivered individually or in group formats, sometimes following a very structured manual, at others adapting the intervention to the individual or group in question.
How the intervention might work
Because most theories about the origin of conversion and dissociative disorders are concerned with underlying psychological stressors or trauma, the common hypothesis is that offering treatments that work with and address this in patients' lives might produce a positive treatment effect and a reduction in symptoms.
The basis for how the different methods and procedures in different treatment approaches to conversion and dissociative disorder are usually associated with a specific theoretical grounding, which often present divergent views on aetiology as well as on what they regard as active mediators in treatment. In general, however, for most models, the mechanism of action of therapy is the integration of dissociated parts of the personality or therapeutic work with maladaptive thoughts towards more adaptive and appropriate thoughts. For example, in therapy, one can work on integration of dissociated elements of the self (psychodynamic) or to weaken maladaptive schemas and construct new, more adaptive schemas (in cognitive therapy a schema is an organised pattern of thought and behaviour or a mental structure of preconceived ideas representing some aspect of the world).
It must be pointed out that not just the method but also the relationship and the therapeutic alliance between the patient and the therapist are main factors for a satisfactory outcome of a therapy course (Horvath 1991; Martin 2000).
Cognitive behavioural approaches
These are based on Aaron Beck's personality model, which suggests that the personality is composed of different modes that are collections of schemas responsible for coding cognitive, affective, behavioural and physiological information and for generating response. A mode with associated schemas is activated only when a particular schema for orientation, related to a particular situation or feeling, triggers it. Based on this theory, Kennedy 2013 has developed a model for dissociation.
At level 1, scary stimuli can result in dissociation of the schema for orientation, which results in incorrect integration of incoming information. The level involves 'detachment', that is, a change of consciousness that is a result of the fight/flight/freeze response, which results in depersonalisation or derealisation or 'compartmentalisation' (i.e. where trauma‐related information is partially stored without integration into the normal memory system and thus is inaccessible). However, when situations occur that are similar to the trauma, trauma‐related information can be activated with severe anxiety and discomfort, as in flashbacks in post‐traumatic stress disorder (PTSD). The therapy works by breaking this 'compartmentalisation' and establishing reintegration.
Level 2 describes dissociative compartmentalisation that can occur within the single mode between the various schemas for coding cognitive, affective, behavioural and physiological information. A mode associated with a traumatic experience or an inter/intrapsychic issue can activate a dissociative process between the individual schemas. An effective schema that is unacceptable to consciousness (e.g. one unbearable feeling related to a traumatic event) can be isolated and kept out of consciousness, or a schema of behaviour can be dissociated so that the person experiences losing a physical skill in connection with a trauma or retrauma (PNES, sensory loss, paralysis or gait disturbance), which also falls within the concept of compartmentalisation.
Level 3 describes dissociation between different modes in Beck's personality model. Generally, integrated modes representing the well‐integrated personality may be subject to dissociation upon severe stress or trauma. An example is the dissociative identity disorder or the disintegration occurring in borderline personality disorder. Based on this model, an adapted traditional cognitive treatment plan can be prepared, taking into account the dissociation level, in order to establish reintegration within schemas and modes.
Brown 2013 has introduced The Integrative Cognitive Model, which is based on the individual's way of storing information in memory. During a dissociative process, the patient activates a primary attentional system and the most active hypothesis for an event (e.g. a misinterpretation of a symptom) combined with the corresponding sensory data creates a counterproductive working model or primary representation. This 'rogue' representation is the starting point in the case formulation that is the basis of the treatment plan. In the course of the treatment, a 'socialisation' takes place, which refers to some form of emotional neutralisation in relation to the patient's notions of the condition. Brown emphasises that not all dissociative reactions are the result of traumatic painful memories, but when it is the case, he recommends follow‐up treatment with trauma‐focused cognitive (Ehlers 2000) or metacognitive therapy (Wells 2004).
The structural dissociation and trauma approach
The two Dutch psychologists, Ellert Nijenhuis and Onno van der Hart, are the most prominent in the theory formation and research on dissociation and trauma (Nijenhuis 2009; Nijenhuis 2010; Steele 2009). Based on Janet 1907, they define structural dissociation as a lack of personality integration, manifested by the existence of two or more inadequately integrated and dissociated parts of the personality. Traumatic dissociation leads to fragmentation in the individual's personality (i.e. of the entire dynamic, biopsychosocial system), which constitutes the mental and behavioural conditions of the person. They emphasise that this is fundamentally different from that mentioned in the cognitive model above regarding changes in perception and memory.
Regarding symptom formation, Nijenhuis and van der Hart distinguish between psychoform and somatoform consciousness and identity (e.g. amnesia, fugue, flashbacks and dissociative identity disorder), while they consider somatoform dissociation as the proper term for conversion, more specifically neurological conditions and other conditions involving the body.
The model is based on Janet's phase‐divided treatment. In phase 1, one strives to provide reassurance and the ability to experience bodily reactions; in phase 2, one looks at the integration of the different parts of the personality by working with phobias for traumatic memories and for anxious attachment; and in phase 3, one works with personality reintegration, phobias for ordinary life, change and intimacy (van der Hart 2012).
The psychodynamic approach
There exist only a few guidelines and general recommendations on conducting psychoanalytic and psychodynamic psychotherapy for dissociative and conversion disorders. Some authors describe a broader perspective (Kalogjera‐Sackellares 2004; Kaplan 2014), while Matthews 1997 propose a psychodynamic psychotherapy for people who have experienced severe childhood trauma. Howell 2011 discusses the identification and diagnosis of dissociative identity disorder and outlines a phase‐oriented treatment plan, which includes facilitating a therapeutic relationship, emphasising the multiplicity of transferences (the patient redirects feelings or desires for another person to an entirely different person, frequently the therapist), countertransferences (redirection of a therapist's feelings towards a patient) and the potential enactments of this transference and countertransference.
The comprehensive guidelines for treating dissociative identity disorder in adults proposes a phase‐oriented treatment (Chu 2011). In phase 1, safety, stabilisation and symptom reduction are established; phase 2 involves confronting, working through and integrating traumatic memories; and in phase 3 the objectives are identity integration and rehabilitation. The guideline is referring to individual psychodynamic‐oriented psychotherapy as the most common, often incorporating other techniques as CBT techniques, hypnosis or EMDR.
Significant contributions from contemporary psychoanalysis to the theory and treatment of dissociative disorders concerns attachment and relational theory (Bradfield 2011; Bromberg 2009; Howell 2005).
Dynamic interpersonal therapy for functional somatic disorders is a mentalisation‐based approach (Bateman 2013), which entails a focus on restoring the capacity for stress regulation by fostering the use of more adaptive attachment strategies in response to stress, and recovery of the capacity for (embodied) mentalising (Luyten 2013). Mentalisation is the ability to understand the mental state, of oneself or others, that underlies overt behaviour. Dynamic interpersonal therapy consists of three phases. The first phase focuses on the treatment alliance and the collaborative formulation of a treatment (i.e. formulation of a shared and acceptable illness theory that recognises the complexity of the disorder through consensus). The second phase consists of the working through of interpersonal affective focus (how the patient perceives others in relation to a self‐perception, and the affect that links these two experiences) and consolidation of treatment gains. This is used as a guide to explore the typical interpersonal patterns with the aim to foster patients' capacities to reflect on the bodily self, others and the self‐in‐relation‐to‐others. The final phase focuses on the aims to transfer what one has 'learned' during treatment to the everyday context of the patient to prevent relapses and to foster autonomy and resilience long‐term.
The contextual approach
This treatment model, described by Gold 2009, is not based on a specific theory but involves elements from both cognitive and psychodynamic methods, based on practical clinical experience and empirical evidence. Attachment theory, though, with a specific focus on disorganised attachment and dissociation, has particular theoretical and practical significance in this model (Barach 1991; Liotti 2006; Williams 2019), and the overall approach of the therapy is based on the observation of the absence of reliable attachment resources for these patients. Disorganised attachment is an insecure attachment style, hypothesised to be an outcome of childhood abuse and trauma, and closely related to the development of dissociation. Focus on therapeutic relationship and treatment alliance are also considered essential. The primary objective of the treatment is to improve adaptive functioning, to focus on problems that find expression in the present and teaching the patient skills that can be applied to difficulties as they arise. The treatment strategy is based on the fact that these patients may have difficulty with affect management, and identifying the presence, intensity and type of affect resulting in an associated inappropriate and incomprehensible behaviour.
Affective arousal is so intense that cognitive processing does not sufficiently contribute to behavioural regulation and a result is that the patient experiences the dissociative episode as 'not me' and may feel shame. The intervention concerns fostering capacity for emotional de‐escalation and increasing the cognitive capacities for recognising the presence of affect. Another aspect is to reduce the propensity of chronic high arousal. Gold 2013 recommends training with a log sheet, relaxation techniques and a 'grounding' programme. He states that trauma exposure is inappropriate for these patients.
The neurobiological approach
Neuroimaging techniques have contributed to the understanding of the basic mechanisms of conversion and dissociative disorders and on some points, pathophysiological equivalents exist for the psychological theories mentioned above and have reinforced their treatment strategies (Aybek 2014; Perez 2015; Perez 2017).
Psychoeducation
Psychoeducation concerns teaching patients about their condition. It is connected to CBT in that it is focused on the present and works with empowering patients to engage with their illness in more informed and helpful ways, the theory being that the more informed patients are, the better equipped they will be in dealing with their condition and with working on recovering from it.
Psychoeducation can be performed in many ways, both formally and informally, and can last from one session to more elaborate programmes over longer periods (Colom 2011; Zhao 2015).
Why it is important to do this review
It is apparent that conversion and dissociative disorders cause clinically significant suffering and that developing effective interventions is important. The main purpose of this review is to update the knowledge about treating conversion and dissociative disorders. This is important for people with conversion and dissociative disorder to understand the evidence supporting the treatments offered to them and to help clinicians prescribe effective treatments.
Previous reviews on this theme
The review is an update of the Cochrane Review "Psychosocial interventions for conversion disorder", published in 2005 (Ruddy 2005). This study included conversion disorder motor and sensory symptoms or impairment, that cannot be explained by a neurological cause (conversion disorder in Diagnostic and Statistical Manual of Mental Disorders 4th Edition (DSM‐IV)), and all dissociative states in DSM‐IV and ICD‐10. The review included all randomised controlled trials (RCT) that compared psychosocial interventions for conversion disorder with standard care (SC) or other interventions (biological or psychosocial). Only three studies (119 participants) met the inclusion criteria. One study was concerned with paradoxical injunction therapy and the other two studied the value of hypnosis. All studies were of poor methodological quality and it was, therefore, difficult to place much value on the results. The authors concluded that the use of psychosocial interventions for conversion disorder required more research and it was not possible to draw any conclusions about their potential benefits or harms from the included studies.
The latest Cochrane Review that deals with conditions broadly within the field of conversion disorders is "Psychological and behavioural treatments for adults with non‐epileptic attack disorder" (Martlew 2007; Martlew 2014). In the 2014 update, 12 studies met the inclusion criteria (four RCTs and eight non‐controlled studies). Overall, three examined CBT, two investigated hypnosis, one assessed paradoxical intention and one had a mixed intervention design. They classified two included studies at low risk of bias, one at unclear and nine at high risk of bias. For quality of the evidence (GRADE), six studies were of very low, two were of low and three were of moderate quality. However, most included studies reported improved outcomes for the intervention under investigation. The authors concluded that there is little reliable evidence to support the use of any treatment, including CBT, in the treatment of non‐epileptic seizures.
Since the last Cochrane Review in 2014 (Martlew 2014), one meta‐analysis on PNES has been published. Carlson 2017 synthesised data from 13 studies, of which 2/13 studies used an RCT design. Studies were included if they evaluated the effectiveness of at least one psychological intervention undertaken to lessen the frequency of PNES and using seizure frequency as an outcome measure. They found a moderate to high level of heterogeneity (I2 = 58% to 78%) and in general serious bias regarding incomplete data, primary and secondary outcome measures, intervention application and duration. The findings highlight the potential for psychological interventions as a favourable alternative to the lack of treatment options offered to people with PNES, but brought no new evidence to the field.
Another meta‐analysis from 2014 dealt with the effectiveness of psychotherapy for severe somatoform disorder (Koelen 2014). The authors included prospective studies. The patients had a diagnosis of somatoform disorder, primarily severe conversion and somatisation (diagnosis of hypochondriasis and body dysmorphic disorder was excluded) and received psychotherapy in secondary or tertiary care. The review included 10 randomised and six non‐randomised studies. Study quality was moderately poor (ranging from very poor to very good), measured using the Psychotherapy Quality Rating Scale (Kocsis 2010), and heterogeneity was high. The effect size for physical symptoms was large, for psychological symptoms moderate to large and for functional impairment moderate, but effect sizes were generally lower than those typically found for other mental disorders. Post‐hoc analysis indicated that psychodynamic interventions were more effective in improving functioning than cognitive interventions, although not in improving symptoms and there was no difference in the effectiveness of group versus individual therapy was found.
With regard to physiotherapy for conversion disorder, one systematic review by Nielsen 2013 identified 29 studies evaluating the effect of physical treatment of adults. No RCTs and only one controlled study was described. The Nielsen review does not define one primary outcome, it only defines that they look at the same range of outcomes as used in the individual studies, and that these showed encouraging results with improvement in 60% to 70% of patients. Combining motor relearning with behavioural therapy was the most common approach. They concluded that the evidence to guide physiotherapy treatment for conversion disorder was of low quality. Likewise, there was limited and poor‐quality evidence for the efficacy of physiotherapy management of child and adolescents with conversion disorder (FitzGerald 2015).
Finally, Brand 2009 reviewed 20 empirical reports of treatment for dissociative disorders, mainly dissociative identity disorder.
Collectively, these reports suggest that treatment for dissociative disorders is associated with decreased symptoms of dissociation, depression, PTSD, distress and suicidality. Effect sizes, based on pre/post measures, were in the medium to large range across studies. However, there were significant methodological limitations illustrating a serious lack of well‐designed studies in the treatment of dissociative disorders. The authors proposed good arguments for using case studies instead of RCTs when treating people with severe psychopathology, high comorbidity and the need for long‐term treatment.
Conclusion
To conclude, several reviews and meta‐analyses have been published since the 2005 version of this Cochrane Review. However, none have consistently examined the same diagnostic spectrum (i.e. both conversion and dissociative disorders). In addition, we are following the revised and expanded Cochrane Handbook for Systematic Reviews of Interventions and are updating the methods from the first review (Higgins 2019).
Objectives
To assess the beneficial and harmful effects of psychosocial interventions of conversion and dissociative disorders in adults.
Methods
Criteria for considering studies for this review
Types of studies
We included all RCTs that met our inclusion criteria. We included studies irrespective of language, publication type and publication status.
Types of participants
Participants were adults aged 18 to 80 years, of any gender and nationality. Participants fulfilled the criteria for having a conversion or dissociative disorder according to the DSM‐IV (APA 1994), DSM‐5 (APA 2013), or ICD‐10 (WHO 1993) criteria. As some studies may have been conducted prior to DSM‐IV or ICD‐10, we included all studies where a large majority (80% or greater) of participants fulfilled current diagnostic criteria or any earlier diagnostic equivalent. We consider the use of conversion and dissociative disorders to cover the diagnostic criteria equivalent to ICD‐10 F.44 codes (F44.0 Dissociative amnesia, F44.1 Dissociative fugue, F44.2 Dissociative stupor, F44.3 Trance and possession disorders, F44.4 Dissociative motor disorders, F44.5 Dissociative convulsions, F44.6 Dissociative anaesthesia and sensory loss, F44.7 Mixed dissociative [conversion] disorders, F44.8 Other dissociative [conversion] disorders, F44.9 Dissociative [conversion] disorder, unspecified) and in the DSM system the DSM‐IV codes 300.11 Conversion disorder, 300.12 Dissociative amnesia, 300.13 Dissociative fugue, 300.14 Dissociative identity disorder, 300.6 Depersonalisation disorder and 300.15 Dissociative disorder not otherwise specified.
Studies with participants with any length of illness were included, as well as studies with participants being treated in any intervention setting, as long as the diagnostic criteria were fulfilled.
If the search identified studies where comorbidity occurred, we planned to comment on this and the participants included if they had conversion or dissociative disorder according to the DSM‐IV (APA 1994), DSM‐5 (APA 2013), or ICD‐10 (WHO 1993) criteria.
Types of interventions
We included RCTs that compared a psychosocial (including psychotherapeutic) intervention for conversion or dissociative disorder with another intervention (pharmacological or psychosocial, or mixed) or with SC or wait list controls.
Experimental interventions
Psychosocial interventions
Our understanding of psychosocial interventions follows the traditional use of this term in medical health, to cover interventions of either a psychological nature, a social nature, or both. Often what makes an intervention psychosocial is the theoretical framework it places itself within, giving a particular approach to how that intervention is regarded. Psychosocial interventions are described in much more detail in the Description of the intervention section, and include the following.
-
Behavioural therapy
-
Cognitive behavioural therapy (CBT)
-
Psychoeducation
-
Hypnosis
-
Psychodynamic andpsychoanalytic therapies
-
Specialised physiotherapybased on cognitive behavioural therapy (CBT)
-
Paradoxical intention therapy
-
Models of dissociation and trauma treatment
-
Eye movement desensitisation reprocessing (EMDR)
Controls
Controls could be SC, wait list controls, pharmaceutical interventions or another psychosocial intervention.
-
Pharmacological interventions: any pharmaceutical medication officially recognised by US or European law.
-
SC: standard medical care (SMC) or treatment as usual (TAU) is the care that a person would normally receive had they not been included in the research study. What this contains varies greatly in different settings and we describe this with the studies.
-
Wait list controls: wait list controls are those patients referred to the same intervention, but who have not yet received it due to the allocation in the study. They are usually free to pursue other interventions outside of the study, but will not receive any controlled intervention while waiting. They will receive the same intervention as the active group, only delayed in time.
Comparisons
If possible, we compared the effects of any similar interventions with controls on both primary and secondary outcomes. By 'similar', we mean interventions that broadly speaking fall under the same category of psychosocial interventions listed above, namely the same forms of psychotherapy, interventions consisting of psychoeducation or comparable other psychosocial training. As the field of psychosocial interventions contains a variety of different approaches, the actual decision of what to group into comparisons is based on clinical judgement.
Types of outcome measures
If a study had more than one measure on the same outcome, only data from one measure for each outcome were included in the analysis. When several measures on the same outcome appeared in a study, we discussed this within the review author team and choose one measure based on which measure was most widely used and which we deemed most clinically relevant.
Time points
Studies were included that measured the effects at end of treatment as the main time point, and if available, follow‐up effects, which were then divided into shorter term (up to and including five months after end of treatment) and longer term (from six months after end of treatment).
Primary outcomes
-
Reduction in physical signs.
This would be expected to be seen in the improvement of physical functioning or in reduction of conversion experiences such as seizures or other discrete conversion episodes.
There are no golden standards in scales for this outcome, but possible ways could be using scales measuring physical functioning such as the physical function dimension of 36‐item Short‐Form (SF‐36) (Jenkinson 1996), or by the Somatoform Dissociation Questionnaire (SDQ‐5 and SDQ‐20) (Nijenhuis 1996). It could also be by counting number of seizures daily or weekly, or by using binary outcomes such as improvement or no improvement.
Secondary outcomes
-
Level of functioning: this is a new addition to this updated version of the review and has been added as a secondary outcome because level of functioning is often one of the main ways of assessing a patient's ability to participate in life despite any illness or incapacity. This could be measured using the Global Assessment of Functioning (GAF) scale (APA 1994), the Work and Social Adjustment Scale (WSAS) (Mundt 2002), or similar.
-
Quality of life: this was part of the secondary outcomes in the original review, and is still very relevant in current research. This can be measured using the full version of the SF‐36 (Jenkinson 1996), Quality of Life in Epilepsy‐31 (QOLIE31) (Cramer 1998), or similar.
-
Mental state: this secondary outcome was also used in the original review, though, in this update, we decided to divide it into anxiety and depression where possible, to get a more accurate picture of patients' experiences, resulting in three secondary outcomes: mental state, anxiety and depression. The general component of mental state can be measured using instruments such as the Symptom Check List 90 (SCL‐90) (Derogatis 1977), or the mental component of the 12‐‐item Short Form (SF‐12)/SF‐36 (Jenkinson 1996). Anxiety could be measured using instruments such as Becks Anxiety Index (BAI) (Beck 1988), or the anxiety subscale of the Hospital Anxiety and Depression Scale (HADS) (Zigmond 1983). Depression could be measured using instruments such as Becks Depression Index version II (BDI‐II) (Beck 1996), Hamilton Depression Rating Scale (HDRS) (Hamilton 1960), or by the depression subscale from the HADS (Zigmond 1983).
-
Dropout rate/leaving the study early: this would often be measured by numbers or percentage.
-
Use of health service resources: often measured in how many subsequent visits a patient has to hospital/accident and emergency department or local general practitioner/clinician.
-
Adverse effects: in psychotherapy, the general adverse effects are described by Barlow 2010 and Crawford 2016. Examples could be the risk of psychotic decompensation or self‐injury through psychological intervention of psychologically fragile patients or that participants experience more positive attention from staff or family by being in a study (secondary gains arise), which could be maintaining the symptoms.
Search methods for identification of studies
For this update, the main review author (CG) in co‐operation with an Information Specialist of the Cochrane Common Mental Disorder Group and an Academic Research Librarian of the Research Unit in Region Zealand, revised the search strategies in line with current Cochrane Common Mental Disorder Group practices (Higgins 2019).
Electronic searches
We searched the electronic resources listed below, using the strategy detailed in Appendix 1.
-
Cochrane Central Register of Controlled Trials (CENTRAL) (2019, Issue 7 in the Cochrane Library). Searched on 16 July 2019.
-
MEDLINE via Ovid (from 1946). Searched on 16 July 2019.
-
CINAHL via EBSCO. Searched on 16 July 2019.
-
Embase via Ovid from 1974. Searched on 16 July 2019.
-
PsycINFO via Ovid from 1806. Searched on 16 July 2019.
-
ERIC via EBSCO. Searched on 16 August 2019.
-
Web of Science Core Collection (Thomson Reuters). Searched on 16 July 2019.
We searched online clinical trial registers for ongoing or recently completed studies on 16 July 2019, including the ISRCTN Registry (www.isrctn.com/), US National Institutes of Health Ongoing Trials Register ClinicalTrials.gov (clinicaltrials.gov), the World Health Organization (WHO) International Clinical Trials Registry Platform (who.int/trialsearch/), and the EU Clinical Trials Register (www.clinicaltrialsregister.eu/).
There were no limitations on languages.
Searching other resources
We checked references, screened reference lists, citation searching and contacted relevant study authors to identify additional studies.
Data collection and analysis
We conducted the review according to the Cochrane Handbook for Systematic Reviews of Interventions (Higgins 2019). In the following section, we report only the methods that we were able to use in this update.
Selection of studies
Following the Cochrane Handbook for Systematic Reviews of Interventions (Higgins 2019), the main review author (CG) plus one of the other members of the author team did the initial screening of abstracts using Covidence. This was divided so the main review author (CG) read all abstracts, and three other authors (OJS, RR, US) read a third of the abstracts each. When disagreements arose about inclusion of abstracts, this was discussed between the two voting review authors. In case of no consensus, we consulted a third review author. A record of all papers rejected on the basis of their abstracts were documented in Covidence but are not included in the review.
We retrieved the full papers of all remaining abstracts and other potentially relevant articles identified by the various search strategies (reference checking, personal communications, etc.). All papers in languages other than English were (as far as possible) translated or reviewed by someone who spoke the language. Two review authors independently reviewed all articles, following the same format as for abstracts, with the main review author (CG) reading all full texts, and three other review authors (OJS, RR, US) reading a third each. The process of full‐text screening was done using Covidence, which was also used for documenting the results of each reviewer's judgement.
When disagreements arose about inclusion of a full text, the two voting review authors discussed them, and, if no consensus was reached, a third author was consulted. If there was a lack of clarity about the suitability of inclusion of a study, then we contacted the authors of the article for more information.
Any ongoing studies were followed up by contacting the principal investigators to enquire whether any data were available that could be included in this review.
Where studies had multiple publications, we collated the reports of the same study so that each study, rather than each report, was the unit of interest for the review, and such studies had a single identifier with multiple references
We documented the reasons for exclusion of the full papers in the Characteristics of excluded studies table.
Data extraction and management
Two review authors (CG, HC) completed the extraction of data using Covidence.
They independently extracted data onto a data collection form. We extracted information on title, authors, year of publication, source, setting, country, participant characteristics, diagnosis, comorbidity, study design and methods, interventions, outcomes and relevant information for 'Risk of bias' assessments. We exported data to Review Manager 5 and one review author (OJS) performed data analysis (Review Manager 2014). We resolved differences by discussion. In cases of insufficient data, or where data in the published study reports were unclear, we contacted the study authors requesting them to clarify the missing information (see Dealing with missing data).
We completed a Characteristics of included studies table.
Assessment of risk of bias in included studies
In the original version of this review, methodological quality of included studies were assessed by criteria sent out in the Cochrane Handbook for Systematic Reviews of Interventions current at the time; however, following the publication of the revised and expanded Cochrane Handbook for Systematic Reviews of Interventions (Higgins 2011), we updated our methods accordingly.
Working independently, two review authors (CG and HC) assessed the risk of bias of included studies using the tool described in the Cochrane Handbook for Systematic Reviews of Interventions (Higgins 2011).
We assessed the following items.
-
Sequence generation: was the allocation sequence adequately generated?
-
Allocation concealment: was allocation adequately concealed?
-
Blinding of participants, personnel and outcome assessors for each main outcome or class of outcomes: was knowledge of the allocated intervention adequately prevented during the study?
-
Incomplete outcome data for each main outcome or class of outcomes: were incomplete outcome data adequately addressed?
-
Selective outcome reporting: were reports of the study free of suggestion of selective outcome reporting?
-
Other sources of bias: was the study apparently free of other problems that could put it at a high risk of bias?
We included quotations from the text of included studies and comments on how we assessed the risk of bias in the Characteristics of included studies table. We assigned studies to one of three categories (low risk of bias, uncertain risk of bias and high risk of bias), according to guidelines in the Cochrane Handbook for Systematic Reviews of Interventions (Higgins 2011). We resolved disagreements by discussion. We subsequently included the risk of bias assessment in the GRADE evaluation that was used to collectively assess the certainty of the estimates based on the risk of bias, imprecision, indirection and publication bias.
If disputes arose we achieved resolution after consulting with the third review author (OJS).
Measures of treatment effect
Outcomes were assessed using continuous (e.g. changes on physical function scales); categorical measures (e.g. one of three categories on a quality of life scale, such as 'better', 'worse' or 'no change') or dichotomous measures (e.g. either returned to employment or did not return to employment).
Continuous data
We compared the mean score between the two groups to give a mean difference (MD) and presented this with 95% confidence intervals (CI). We wanted to use the overall MD, where possible, to compare the outcome measures from studies. However, because some of the included studies used different rating scales for measuring the same outcome, we needed to use the standardised mean difference (SMD) in some analyses. We considered a statistical significant SMD effect size of: 0.15 or less to have no clinically meaningful effect; 0.15 to 0.40 to have a clinical meaningful but small effect; 0.40 to 0.75 to have a moderate effect; and greater than 0.75 to have a large treatment effect (Cohen 1988; Thalheimer 2002).
Many rating scales are available to measure outcomes in psychosocial studies. These scales vary in the quality of their validation and reliability. Therefore, if a rating scale's validation had not been published in a peer‐reviewed journal, the data were included but we noted that it was not validated. In addition, it is preferable for the rating scale to be either self‐report or completed by an independent observer or relative. We included studies that used the same rating scales to evaluate the effect on a specific outcome in the same analysis for direct comparison.
Continuous data on clinical and social outcomes are often not normally distributed. To avoid the problems of applying parametric tests to non‐parametric data, we applied the following standards to all data before inclusion: 1. standard deviations and means had to be reported in the paper or be obtainable from the authors, 2. when a scale started from a finite number (such as zero), the standard deviation, when multiplied by two must have been less than the mean (otherwise the mean was unlikely to be an appropriate measure of the centre of the distribution) (Altman 1996). If data were non‐parametric, then we reported this in the text but did not used them in the meta‐analysis.
Dichotomous data
For dichotomous outcomes, we calculated the risk ratio (RR) with its associated 95% CI was estimated.
Unit of analysis issues
We planned to include data from randomised cross‐over studies up to the point of first cross‐over (first period only) (Curtin 2002). We planned to include cluster randomised studies; however, we found none. If we find cluster‐randomised studies for updates of this review, they will be eligible for inclusion. We will anticipate that investigators have presented their results for cluster RCTs after appropriately checking for clustering effects (robust standard errors or hierarchical linear models). If it is unclear whether a cluster‐randomised trial has used appropriate checks for clustering, we will contact the investigators for further information. Where appropriate checks are not used, we will request and re‐analyse individual participant data using multilevel models that check for clustering. Following this, we will analyse effect sizes and standard errors in Review Manager 5 (Review Manager 2014), using the generic inverse method (Higgins 2019).
Dealing with missing data
We conducted analyses 'as reported' for continuous outcomes. We attempted to retrieve any missing data from the study authors. We contacted the authors of three studies with unclear or missing data and requested the necessary data. If data remained unavailable, we tried to estimate the missing data using the available information (e.g. if the standard deviation (SD) was missing, we estimated it from the standard error, if reported). When we were unable to obtain missing data, we conducted analyses using available (incomplete) data.
Assessment of heterogeneity
We assessed clinical heterogeneity by examining variability in the participants, interventions and outcomes described in each study. We assessed methodological heterogeneity by inspecting variability in the design of the studies and statistical heterogeneity by assessing the difference in the studies' intervention effects. We assessed heterogeneity between studies by visual inspection of the forest plot for overlapping CIs, using the Chi2 test for homogeneity with a significance level of α (alpha) = 0.10, and the I2 statistic for quantifying inconsistency (estimating the percentage of variation in effect estimates due to heterogeneity rather than sampling error). We judged I2 values of 0% to 40% to indicate little heterogeneity; 30% to 60%, moderate heterogeneity; 50% to 90%, substantial heterogeneity; and 75% to 100%, considerable heterogeneity. For values including overlapping judgements, the degree of heterogeneity was considered as an interval (e.g. with a heterogeneity of 55% being considered as moderate to substantial heterogeneity) (Higgins 2019).
Assessment of reporting biases
We handled different forms of reporting bias, especially publication bias and outcome reporting bias, according to the recommendations of the Cochrane Handbook for Systematic Reviews of Interventions (Higgins 2019; Section 10.1). If more than 10 studies had been included in a given meta‐analysis, we had planned to draw funnel plots to give a visual assessment on whether effects were associated with the size of the study (Egger 1997). Due to the small number of eligible trials (i.e. fewer than 10 studies for each comparisons), this was not possible.
Data synthesis
We included and analysed studies undertaken in any configuration or setting; for instance, in groups, hospitals, people's homes and clinics. We summarised data in a meta‐analysis when they were available. If clinical heterogeneity was not excessive (e.g. there was similarity in participants' characteristics), we performed statistical analysis in Review Manager 5 (Review Manager 2014), according to recommendations in the Cochrane Handbook for Systematic Reviews of Interventions (Higgins 2019; Section 9.4.1). We synthesised data using final values and the inverse variance method in the meta‐analyses. We generally used the fixed‐effect model because most of the analyses included only one study. When more than one study was included, we used the random‐effects model. A random‐effects model has the assumption that apparent differences between study effects are random, but the estimated difference follows a normal distribution. This method gives more weight to small studies, whereas the fixed‐effect model gives more weight to large studies (Higgins 2019; Section 9.5.4).
Subgroup analysis and investigation of heterogeneity
We planned to conduct subgroup analyses for:
-
different symptoms groups;
-
group versus individual therapy;
-
manual‐driven versus non‐manual‐driven therapies.
We planned to conduct these specific analyses to investigate if treatment effects vary between symptoms groups. We also wished to determine if treatment effects are influenced by how the therapy is provided (group versus individual therapy) or if therapies are based on a manual or not, or both.
Sensitivity analysis
We planned to conduct sensitivity analyses for:
-
attrition rate: more than 50% attrition compared with less than 50% attrition;
-
use of intention to treat analyses: the studies using intention to treat analyses compared with the studies not using intention‐to‐treat analyses;
-
differences between cluster and non‐cluster randomised studies;
-
differences between self‐reported and observer‐reported outcomes.
Summary of findings and assessment of the certainty of the evidence
We constructed 12 'Summary of findings' tables for the comparisons. As the field of psychosocial interventions contains a variety of different approaches we were only able to predefine that, when possible, we would group studies together based on clinical judgement as to which interventions were most alike and create the actual comparisons after identification of the included studies. For the studies identified in this version of the review, the team agreed on the following comparisons:
-
inpatient paradoxical intention therapy versus outpatient diazepam;
-
inpatient treatment programme plus hypnosis versus inpatient treatment programme;
-
outpatient hypnosis versus wait list;
-
behavioural therapy plus routine clinical care versus routine clinical care;
-
CBT versus SMC;
-
psychoeducational follow‐up programmes versus TAU;
-
specialised CBT‐based physiotherapy inpatient programme versus wait list;
-
specialised CBT‐based physiotherapy outpatient intervention versus TAU;
-
brief psychotherapeutic intervention (psychodynamic interpersonal treatment approach) versus SC;
-
CBT plus APA versus CBT alone;
-
hypnosis versus diazepam;
-
psychotherapy preceded by motivational interviewing compared with psychotherapy alone.
We included the primary outcome reduction in physical signs and the secondary outcomes level of functioning and quality of life assessed at the relevant time points.
We used the GRADE approach to assess the certainty of the evidence associated with each of these outcomes (Guyatt 2008). The GRADE approach appraises the certainty of a body of evidence based on the extent to which one can be confident that an estimate of effect or association reflects the item being assessed. Considerations are due to: within‐study risk of bias, directness of evidence, heterogeneity of the data, precision of effect estimates and risk of publication bias (Andrews 2013a; Andrews 2013b; Balshem 2011; Brunetti 2013; Guyatt 2011; Mustafa 2013).
Results
Description of studies
See Characteristics of included studies; Characteristics of excluded studies; and Characteristics of ongoing studies tables for detailed information on the individual studies and Figure 1 for an overview of the search process.
PRISMA flow diagram.
Results of the search
The electronic searches resulted in 3171 references, some of which were duplicates and thus removed. This left 3048 texts to be screened in Covidence, of which 2860 were judged irrelevant on the basis of their abstracts. In the process of obtaining full texts on the relevant studies, we identified one more relevant study that was not found by the electronic search and added it to the full texts in Covidence, giving a total of 168 full‐text papers that were then read and assessed for eligibility.
For planned and ongoing studies, we contacted the relevant authors and requested possible results and timeframes (see Characteristics of ongoing studies table for details). None of these had any results ready to share and we designated seven studies as ongoing at this stage.
In a few otherwise relevant studies it was unclear whether the data could be extracted, in which case, we contacted the authors, all of whom responded and some supplied us with missing data and missing information about methodology (Chen 2018; Thompson 2018; Vermeulen 2018). For the Pleizier 2017 study, our question was with regard to the patient population, which the authors (Vermeulen 2018) confirmed fulfilled the criteria to be included in the review. For two other studies, our queries were with regard to the outcome data. This was not available for the Thompson 2013 study, but Chen 2014 did present data on secondary outcomes. All three studies were then included.
We excluded 147 papers after reading the full texts and contacting the authors (see Characteristics of excluded studies table).
This resulted in 21 full texts being included, seven of which were duplicates of the same study, thus the total number of studies included in the analysis was 17 (14 new, and three studies from the original review; see Characteristics of included studies table for details).
A flowchart diagram of this process can be found in Figure 1.
Included studies
Seventeen studies met the inclusion criteria: Aamir 2012; Ataoglu 2003; Chen 2014; Dallocchio 2016; Drane 2016; Goldstein 2010; Hubschmid 2015; Jordbru 2014; Khattak 2006; LaFrance 2014; Moene 2002; Moene 2003; Mousavi 2008; Nielsen 2017; Pleizier 2017; Thompson 2013; Tolchin 2019. Each study is described in detail in the Characteristics of included studies table.
One cross‐over study was included in the review (Jordbru 2014). We used the data from the first period of this study as recommended by Curtin 2002.
Design
All the 17 studies were RCTs, 16 with parallel‐group designs, and one with a cross‐over design (Jordbru 2014).
Sample sizes
The 17 studies included 894 participants. The studies varied in sizes between 16 participants and 195 participants.
Setting
The studies took place in a variety of settings. Four were conducted in psychiatric inpatient settings (Aamir 2012; Ataoglu 2003; Khattak 2006; Moene 2002); three in psychiatric outpatient settings (Goldstein 2010; Hubschmid 2015; Moene 2003); two in neurology departments, one in day‐patient settings (Nielsen 2017) and one in a general outpatient setting (Pleizier 2017); one was in a veteran medical centre with outpatients (Chen 2014); one in an emergency unit (Mousavi 2008); one was in a hospital rehabilitation clinic with inpatients (Jordbru 2014); one with outpatients in three different academic hospital settings (LaFrance 2014); three originated in an epilepsy unit, two of those mainly taking place at the hospital (Thompson 2013) or in local clinics (Tolchin 2019), and one mainly by telephone (Drane 2016); and one study gave no information on the setting (Dallocchio 2016).
Country
The studies took place in nine countries, with a single study in each of Iran, Italy, Norway, Switzerland and Turkey; two studies from each of Pakistan and the UK, three studies taking place in The Netherlands and four in the USA. Thus, the locations spanned a wide range of countries given the small number of studies included.
Participants
Participants were all adults aged 18–80 years with a confirmed diagnosis of conversion disorder or dissociative disorder. In one study, there were doubts about whether this was the case (Pleizier 2017), but by contacting the author, it was confirmed that the majority of participants did have conversion disorder (Vermeulen 2018). The duration of symptoms varied from a few hours in one study (Mousavi 2008) to several months or years in most other studies. The total number of participants was 894 with 673 women (75%) and 221 men (25%), giving a female:male ratio of 3:1.
Interventions
The interventions were varied and included CBT (Goldstein 2010; LaFrance 2014), hypnosis (Moene 2002; Moene 2003; Mousavi 2008), a psychoeducative focus (Chen 2014; Drane 2016; Pleizier 2017; Thompson 2013), physical rehabilitation within a cognitive behavioural framework (Dallocchio 2016; Jordbru 2014; Nielsen 2017); psychodynamic therapy (Hubschmid 2015); paradoxical intervention (Ataoglu 2003); behavioural therapy (Aamir 2012; Khattak 2006); and motivational interviewing given prior to the therapy (Tolchin 2019).
Controls
Elleven studies received TAU or SMC as controls (Aamir 2012; Chen 2014; Goldstein 2010; Hubschmid 2015; Khattak 2006; LaFrance 2014; Moene 2002; Nielsen 2017; Pleizier 2017; Thompson 2013; Tolchin 2019), two studies had wait list controls (Jordbru 2014; Moene 2003), while two studies had psychopharmaceutical interventions as controls (Ataoglu 2003; Mousavi 2008).
One study had four interventions and no control, only one of which was clearly a psychosocial intervention and one clearly a pharmaceutical intervention, thus for the purpose of using the data in this review, we chose the psychosocial as active intervention and the pharmaceutical as control (Mousavi 2008).
In one study the reported control group was not part of the original randomisation for the study and data for that group thus not eligible for inclusion here (Dallocchio 2016). As that study had two intervention arms where the participants were randomised correctly, we decided to extract data from those instead, using the least invasive intervention as control.
One study had a design where similar interventions were given to both the active and control groups, but with the active group having additional components compared with the controls (Drane 2016).
None of the studies had any restrictions on, or documentation of, whether their control groups received other interventions outside of the study.
Duration of study
The duration of studies varied between a few hours (Mousavi 2008) and more than four months (Goldstein 2010; LaFrance 2014). Interventions for inpatients or day patients tended to be relatively short, lasting five‐days to one week in hospital, some with additional follow‐up afterwards. Interventions given to outpatients of psychoeducative nature tended to have few actual contacts spread out over some months, while therapeutic interventions delivered for outpatients tended to require regular weekly or bi‐weekly visits to an outpatient clinic over 12 weeks.
Outcomes
The studies were primarily clinical, with one feasibility study (Nielsen 2017), which had relevant outcome measures for use of this review.
Rating scales and measurements
Studies used the following scales and measures.
Primary outcomes
For the primary outcome of a reduction in physical signs, several studies used patients seizure diaries as the main measurement, where patients, relatives or staff noted the number of daily or weekly seizures (Aamir 2012; Drane 2016; Goldstein 2010; LaFrance 2014; Mousavi 2008; Tolchin 2019). This would then be used to judge change either as a change in numbers, or percentage when. One study devised its own scale (Drane 2016).
One study measured the primary outcome by looking at the number of participants without conversive attacks in the last two weeks (Ataoglu 2003).
Two studies used a video rating scale developed by the authors, to judge the severity of symptoms (Moene 2002; Moene 2003).
Some studies used different standardised scales to measure reduction in physical signs (Dallocchio 2016; Hubschmid 2015; Khattak 2006; Nielsen 2017; Pleizier 2017). In those studies, the following scales was used:
-
physical component of the SF‐36 (Jenkinson 1996);
-
SDQ‐20 (Nijenhuis 1996);
-
Psychogenic Movement Disorder Scale (PMDRS) total score (Hinson 2005);
-
Clinical Global Impression (CGI) (Guy 1976).
Finally, three studies did not report on our primary outcome (Chen 2014; Jordbru 2014; Thompson 2013).
Secondary outcomes
Studies used the following scales.
Level of functioning:
-
GAF (APA 1994; Jones 1995);
-
Functional Independence Measure (FIM) (McDowell 1996);
-
WSAS (Mundt 2002).
Quality of life:
-
full version SF‐36 (Jenkinson 1996);
-
QOLIE31 (Cramer 1998);
-
Quality of Life in Epilepsy 10 (QOLIE10‐P) (Cramer 1996).
General mental state:
-
SCL‐90 (Derogatis 1977);
-
SF‐36 or SF‐12 Mental Component) (Jenkinson 1996).
Depression:
-
HADS – Depression Component (Zigmond 1983);
-
HRSD (Hamilton 1960);
-
BDI‐II (Beck 1996).
Anxiety:
-
BAI (Beck 1988);
-
Hamilton Rating Scale for Anxiety (HRSA) (Hamilton 1959);
-
Hospital Anxiety and Depression Scale – Anxiety Component (Zigmond 1983).
Use of health service resources:
-
Client Service Receipt Inventory (adjusted);
-
number of general practitioner consultations (Beecham 2019).
Excluded studies
We excluded 147 full‐text articles for the following reasons: 37 studies had a wrong study design, 35 studies had no control group, 23 studies had a wrong population and seven studies had a wrong intervention. Eight texts were overview articles, 33 were reviews, three were letters to the Editor and one was a book chapter.
We provided references for all the excluded full texts together with the reasons in the Characteristics of excluded studies table with excluded studies.
Studies awaiting assessment
There are no studies awaiting assessment.
Ongoing studies
We found seven ongoing studies. Some had very clear and updated information on the study in the ClinicalTrials Register, in which case we have given that information below, while other studies gave less clear information and were explored via personal communication with the principal investigators, in order to obtain all available data. The following projects are all RCTs, either in preliminary phases, still performing data collection, have not yet made their data available or have not yet published their results.
-
Professor Laura Goldstein 2018 is heading a large research project in the UK (CODES), which has been described in various articles (Goldstein 2016; Goldstein 2017; Robinson 2017). Their data and publication is not yet ready (Goldstein 2018), but is likely to be relevant for future updates of this review.
-
Modum Bad 2018, a treatment institution in Norway specialising in trauma, is also undertaking a research project in this area (NCT02450617b), with data collection hopefully being completed in 2019 according to the principal investigator Harald Baekkelund (Modum Bad 2018).
-
Dr Fobian 2018 is principal investigator on a project "Treatment outcomes of CBT for PNES" (NCT02801136), which is due to publish results soon according to the author (Fobian 2018).
-
Dr Meinlschmidt has conducted a study on "Treatment of globus sensations with psychotherapy (NCT01590992), but according to the author no results are available yet (Meinlschmidt 2018).
-
Another study, "The role of the temporo‐parietal junction in functional neurological disorders. A study with mindfulness‐based stress reduction therapy" had received ethical approval and was due to start late 2018 according to the principal investigator Professor Aybek 2018 (Aybek 2018; DRKS00012997).
-
Dr Kim D Bullock is leading a study on "Embodied virtual reality therapy for functional neurological symptom/conversion disorder', which is currently recruiting participants and expects to be complete by January 2022 (NCT02764476).
-
Dr Anna Philine Senf‐Beckenbach is heading a study named "Evaluation of the effect of a psychotherapy program with body movement focus for patients with dissociative seizures." (DRKS00014251).
Risk of bias in included studies
We assessed the risk of bias of each included study using the Cochrane 'Risk of bias' tool (Higgins 2011). A summary of our assessment is displayed in Figure 2 and Figure 3 and the details of the assessment of each study can be found in the Characteristics of included studies table. We assessed most studies to be studies with an overall high risk of bias.
Risk of bias graph: review authors' judgements about each risk of bias item presented as percentages across all included studies.
Risk of bias summary: review authors' judgements about each risk of bias item for each included study.
Allocation
Random sequence generation
We considered random sequence generation at low risk of bias in 13 studies (Aamir 2012; Ataoglu 2003; Chen 2014; Dallocchio 2016; Drane 2016; Goldstein 2010; Hubschmid 2015; Jordbru 2014; LaFrance 2014; Nielsen 2017; Pleizier 2017; Thompson 2013; Tolchin 2019), at high risk in one study (Mousavi 2008), and at unclear risk of bias in three studies (Khattak 2006; Moene 2002; Moene 2003).
Allocation concealment
We considered allocation concealment at low risk of bias in five studies (Goldstein 2010; Hubschmid 2015; Jordbru 2014; Moene 2002; Tolchin 2019), and at unclear risk of bias in 12 studies (Aamir 2012; Ataoglu 2003; Chen 2014; Dallocchio 2016; Drane 2016; Khattak 2006; LaFrance 2014; Moene 2003; Mousavi 2008; Nielsen 2017; Pleizier 2017; Thompson 2013).
Blinding
We considered the method of blinding of participants and personal unclear in nine studies (Aamir 2012; Chen 2014; Drane 2016; Hubschmid 2015; Jordbru 2014; Moene 2002; Mousavi 2008; Pleizier 2017; Thompson 2013), and at high risk of bias in eight studies (Ataoglu 2003; Dallocchio 2016; Goldstein 2010; Khattak 2006; LaFrance 2014; Moene 2003; Nielsen 2017; Tolchin 2019).
We considered the method of blinding of outcome assessment at low risk of bias in five studies (Ataoglu 2003; Dallocchio 2016; LaFrance 2014; Moene 2002; Tolchin 2019), and unclear in eight studies (Aamir 2012; Chen 2014; Drane 2016; Goldstein 2010; Jordbru 2014; Moene 2003; Pleizier 2017; Thompson 2013). In four studies there was a high risk of bias regarding blinding of outcome assessment (Hubschmid 2015; Khattak 2006; Mousavi 2008; Nielsen 2017).
Incomplete outcome data
Eight studies displayed low risk of bias in regards to incomplete data (Ataoglu 2003; Chen 2014; Dallocchio 2016; Goldstein 2010; Hubschmid 2015; Jordbru 2014; Khattak 2006; Pleizier 2017), and six studies displayed high risk of bias (Drane 2016; Moene 2003; Mousavi 2008; Nielsen 2017; Thompson 2013; Tolchin 2019). In three studies, there was lack of information and we could not assess whether the method used to handle missing data was likely to bias the effect estimate (Aamir 2012; LaFrance 2014; Moene 2002).
Selective reporting
Five studies displayed low risk of bias with regard to selective reporting (Goldstein 2010; Hubschmid 2015; LaFrance 2014; Nielsen 2017; Tolchin 2019), and five studies displayed high risk of bias (Chen 2014; Khattak 2006; Moene 2002; Mousavi 2008; Pleizier 2017). In seven studies, it was unclear whether authors reported all predefined outcomes (Aamir 2012; Ataoglu 2003; Dallocchio 2016; Drane 2016; Jordbru 2014; Moene 2003; Thompson 2013).
Other potential sources of bias
In 15 studies there were no other sources of bias (Aamir 2012; Ataoglu 2003; Chen 2014; Dallocchio 2016; Goldstein 2010; Hubschmid 2015; Jordbru 2014; Khattak 2006; LaFrance 2014; Moene 2002; Moene 2003; Mousavi 2008; Nielsen 2017; Pleizier 2017; Thompson 2013). One study measured the primary outcome using a self‐developed scale that was not properly validated (Drane 2016) and this was assessed to likely bias the effect estimate. In one study, it was unclear whether there was any other sources of bias (Tolchin 2019).
Subgroup analysis
We did not perform any subgroup or sensitivity analyses due to lack of data
Sensitivity analysis
We did not perform any sensitivity analyses due to lack of data
Effects of interventions
See: Summary of findings 1 Paradoxical intention therapy compared with diazepam; Summary of findings 2 Hypnosis plus treatment as usual compared with treatment as usual; Summary of findings 3 Hypnosis compared with wait list; Summary of findings 4 Behavioural therapy plus treatment as usual compared with treatment as usual; Summary of findings 5 Cognitive behavioural therapy as compared with standard medical care; Summary of findings 6 Psychoeducational follow‐up programme compared with treatment as usual; Summary of findings 7 Specialised cognitive behavioural therapy‐based physiotherapy compared with waitlist; Summary of findings 8 Specialised cognitive behavioural therapy‐based physiotherapy intervention compared with treatment as usual; Summary of findings 9 Brief psychotherapeutic intervention compared with standard care; Summary of findings 10 Cognitive behavioural therapy plus adjunctive physical activity compared with cognitive behavioural therapy alone; Summary of findings 11 Hypnosis compared with diazepam; Summary of findings 12 Psychotherapy preceded by motivational interviewing compared with psychotherapy alone
We present the results for each of the primary and secondary outcomes connected to the 12 comparisons below. Most are presented as MDs because there was only one study; where there is more than one study and the outcomes were reported on different scales, we present the standardised mean difference (SMD). We considered a statistical significant SMD effect size of: 0.15 or less to have no clinically meaningful effect; 0.15 to 0.40 to have a clinical meaningful but small effect; 0.40 to 0.75 to have a moderate effect; and greater than 0.75 to have a large treatment effect (Cohen 1988; Thalheimer 2002).
We contacted the authors of three studies with unclear or missing data and requested the necessary information. All three replied, which led to one study being included but with no data available (Thompson 2013), one study included in the final analysis on basis of secondary outcomes as no data were available on primary outcomes (Chen 2014), and one study included in the final analysis as it was clarified that the population consisted of people with conversion and dissociative disorders (Pleizier 2017).
We considered most studies at high risk of bias due to systematic errors. However, we used all eligible studies in the meta‐analyses, as it is recommend in the Cochrane Handbook for Systematic Reviews of Interventions to do so when all the studies are assigned the same risk of bias. We took into account our 'Risk of bias' assessment when considering the quality of the evidence using the GRADE approach, to ensure that judgements about risk of bias and other factors affecting the certainty of the evidence are taken into account when interpreting the results of the review (Higgins 2019; Section 8.8.3.1).
Comparison 1: inpatient paradoxical intention therapy compared with outpatient diazepam
One study compared inpatient paradoxical intention therapy with outpatient diazepam (Ataoglu 2003).
1.1 Reduction in physical signs
Inpatient paradoxical intention therapy did not reduce conversive symptoms, measured by number of participants without conversive attacks in the last week, compared with outpatient diazepam et end of treatment (RR 1.44, 95% CI 0.91 to 2.28; 1 study, 30 participants; P = 0.12) (Analysis 1.1).
1.2 Level of functioning
The study did not measure level of functioning.
1.3 Quality of life
The study did not measure quality of life.
1.4 Mental state
1.4.1 Anxiety
Inpatient paradoxical intention therapy might reduce anxiety symptoms at end of treatment, measured by HADS, compared with outpatient diazepam at end of treatment (MD –3.73, 95% CI –6.96 to –0.50; 1 study, 30 participants; P = 0.02) (Analysis 1.2).
1.5 Dropout rate
In Ataoglu 2003, no participants left the study before the end from either group.
1.6 Use of health service resources
The study did not measure use of health service resources.
1.7 Adverse effects
The study did not measure adverse effects.
Comparison 2: inpatient treatment programme plus hypnosis compared with inpatient treatment programme
One study compared inpatient treatment programme plus hypnosis with inpatient treatment programme (Moene 2002).
2.1 Reduction in physical signs
Inpatient treatment programme plus hypnosis did not reduce severity of impairment, measured using a video rating scale, compared with inpatient treatment programme at the end of treatment (MD –0.49, 95% CI –1.28 to 0.30; 1 study, 45 participants; P = 0.23) or at follow‐up (MD 0.13, 95% CI –0.55 to 0.81; 1 study, 45 participants; P = 0.71) (Analysis 2.1).
2.2 Level of functioning
The study did not measure level of functioning.
2.3 Quality of life
The study did not measure quality of life.
2.4 Mental state
Inpatient treatment programme plus hypnosis did not reduce symptoms of mental state, measured using SCL‐90, compared with inpatient treatment programme at the end of treatment (MD 1.42 95% CI –36.02 to 38.86; 1 study, 45 participants; P = 0.94) or at follow‐up (MD –5.97, 95% CI –44.22 to 32.28; 1 study, 45 participants; P = 0.76) (Analysis 2.2).
2.5 Dropout rate
In Moene 2002, 2/26 participants left the treatment programme and hypnosis group compared with 2/23 of the participants in the treatment programme and individual sessions group before the end of treatment (RR 0.88, 95% CI 0.14 to 5.79; 1 study, 49 participants; P = 0.90) (Analysis 2.3).
2.6 Use of health service resources
The study did not measure use of health service resources.
2.7 Adverse effects
The study did not measure adverse effects.
Comparison 3: outpatient hypnosis compared with wait list
One study compared outpatient hypnosis with wait list (Moene 2003).
3.1 Reduction in physical signs
3.1.1 Severity of impairment
Outpatient hypnosis might reduce severity of impairment, measured using a video rating scale, compared with wait list at the end of treatment (MD 2.10, 95% CI 1.34 to 2.86; 1 study, 49 participants; P < 0.00001) (Analysis 3.1).
3.2 Level of functioning
The study did not measure level of functioning.
3.3 Quality of life
The study did not measure quality of life.
3.4 Mental state
The study did not measure mental state.
3.5 Dropout rate
Outpatient hypnosis did not reduce dropout compared with wait list at end of treatment (RR 4.17, 95% CI 0.50 to 34.66; 1 study, 49 participants; P = 0.19) (Analysis 3.2).
3.6 Use of health service resources
The study did not measure use of health service resources.
3.7 Adverse effects
No study measured adverse effects.
Comparison 4: behavioural therapy plus routine clinical care compared with routine clinical care
Two studies compared behavioural therapy plus routine clinical care with routine clinical care (Aamir 2012; Khattak 2006).
4.1 Reduction in physical signs
4.1.1 Number of weekly seizures
Behavioural therapy plus routine clinical care might reduce the number of weekly seizures at the end of treatment, measured using participant's seizure diaries, compared with routine clinical care alone at end of treatment (MD –21.40, 95% CI –27.88 to –14.92; 1 study, 18 participants; P < 0.00001) (Analysis 4.1).
4.1.2 Symptom severity
Behavioural therapy plus routine clinical care might reduce symptom severity at the end of treatment, measured using the CGI scale, compared with routine clinical care alone at end of treatment (MD –2.90 95% CI –3.41 to –2.39; 1 study, 90 participants; P < 0.00001) (Analysis 4.2).
4.2 Level of functioning
Neither study measured level of functioning.
4.3 Quality of life
Neither study measured quality of life.
4.4 Mental state
4.4.1 Anxiety
Behavioural therapy plus routine clinical care might reduce symptoms of anxiety at the end of treatment, measured using Hamilton Anxiety Scale, compared with routine clinical care alone at end of treatment (MD –5.47, 95% CI –7.08 to –3.86; 2 studies, 108 participants; P < 0.00001) (Analysis 4.3).
4.4.2 Depression
Behavioural therapy plus routine clinical might reduce symptoms of depression at follow‐up, measured using Hamilton Depression Scale, compared with routine clinical care alone at follow‐up (MD –4.99, 95% CI –6.36 to –3.62; 2 studies, 108 participants; P < 0.00001) (Analysis 4.4).
4.5 Dropout rate
Behavioural therapy plus routine clinical care was associated with fewer dropouts compared with routine clinical care alone at end of treatment (RR 0.24, 95% CI 0.06 to 0.90; 2 studies, 118 participants; P = 0.04) (Analysis 4.5).
4.6 Use of health service resources
Neither study measure used of health service resources.
4.7 Adverse effects
Neither study measured adverse effects.
Comparison 5: cognitive behavioural therapy compared with standard medical care
Two studies compared CBT with SMC (Goldstein 2010; LaFrance 2014).
5.1 Reduction in physical signs
5.1.1 Monthly seizure frequency (reduction in %)
CBT did not reduce monthly seizure frequency, measured using participants' seizure diaries, compared with SMC at end of treatment (RR 1.56, 95% CI 0.39 to 6.19; 1 study, 16 participants; P = 0.53) (Analysis 5.1).
5.1.2 Monthly seizure frequency (median interquartile range)
CBT did not reduce monthly seizure frequency, measured using participants' seizure diaries, compared with SMC at end of treatment (MD –4.75, 95% CI –18.73 to 9.23; 1 study, 61 participants; P = 0.51) or at follow‐up (MD –3.50, 95% CI –12.69 to 5.69; 1 study, 59 participants; P = 0.46) (Analysis 5.2).
5.1.3 Seizure freedom
CBT did not reduce seizure freedom, measured using participants' seizure diaries, compared with SMC at end of treatment (RR 2.33, 95% CI 0.30 to 17.88; 1 study, 16 participants; P = 0.41) (Analysis 5.3).
5.2 Level of functioning
CBT did not increase level of functioning compared with SMC at end of treatment (SMD 0.44, 95% CI –1.69 to 2.57; 2 studies, 74 participants; P = 0.69) or at follow‐up (SMD –0.63, 95% CI –1.18 to –0.08; 1 study, 53 participants; P = 0.03) (Analysis 5.4).
5.3 Quality of life
CBT did not increase quality of life, measured using QOLIE31, compared with SMC at end of treatment (MD 11.20, 95% CI –7.98 to 30.38; 1 study, 16 participants; P = 0.25) (Analysis 5.5).
5.4 Mental state
5.4.1 Anxiety
CBT did not reduce symptoms of anxiety compared with SMC at end of treatment (SMD –0.31, 95% CI –0.78 to 0.15; 2 studies, 74 participants; P = 0.18) or at follow‐up (SMD –0.31, 95% CI –0.85 to 0.23; 1 study, 53 participants; P = 0.25) (Analysis 5.6).
5.4.2 Depression
CBT did not reduce symptoms of depression compared with SMC at end of treatment (SMD –0.25, 95% CI –0.71 to 0.21; 2 studies, 74 participants; P = 0.28) or at follow‐up (SMD –0.32, 95% CI –0.86 to 0.23; 1 study, 53 participants; P = 0.25) (Analysis 5.7).
5.4.3 Symptoms of general mental state
CBT might reduce negative mental state symptoms, measured using SCL‐90, compared with SMC at end of treatment (MD –70.60, 95% CI –121.59 to –19.61; 1 study, 16 participants; P = 0.007) (Analysis 5.8).
5.5 Dropout rate
CBT did not decrease dropout compared with SMC at end of treatment (RR 0.37, 95% CI 0.02 to 8.01; 2 studies, 83 participants; P = 0.52) or at follow‐up (RR 1.41, 95% 0.25 to 7.87; 1 study, 64 participants; P = 0.70) (Analysis 5.9).
5.6 Use of health service resources
CBT did not decrease primary health service use, measured using numbers of general practitioner consultations, compared with SMC at follow‐up (MD –1.00, 95% CI –3.32 to 1.32; 1 study, 46 participants; P = 0.40) (Analysis 5.10).
5.7 Adverse effects
Neither study measured adverse effects.
Comparison 6: psychoeducational follow‐up programmes compared with treatment as usual
Two studies compared psychoeducational follow‐up programmes with TAU (Drane 2016; Pleizier 2017).
6.1 Reduction in physical signs
6.1.1 Seizure frequency
Psychoeducational follow‐up programmes might reduce seizure frequency, measured using a self‐developed scale, compared with TAU at follow‐up (MD –0.80, 95% CI –1.44 to –0.16; 1 study, 27 participants; P = 0.01) (Analysis 6.1).
6.1.2 Physical symptoms load
Psychoeducational follow‐up programmes did not reduce physical symptoms load, measured using SSF‐36 Physical Component scale, compared with TAU at follow‐up (MD –0.26, 95% CI –3.76 to 3.24; 1 study, 186 participants; P = 0.88) (Analysis 6.2).
6.2 Level of functioning
Psychoeducational follow‐up programmes might improve level of functioning, measured using WSAS, compared with TAU at end of treatment (MD –7.12, 95% CI –12.47 to –1.77; 1 study, 43 participants; P = 0.009), and at follow‐up (MD –6.11, 95% CI –11.67 to –0.55; 1 study, 43 participants; P = 0.03) (Analysis 6.3).
6.3 Quality of life
Psychoeducational follow‐up programmes might improve quality of life, measured using QOLIE10‐P, compared with TAU at follow‐up (MD –9.30, 95% CI –14.06 to –4.54; 1 study, 27 participants; P = 0.0001) (Analysis 6.4).
6.4 Mental state
6.4.1 Anxiety
Psychoeducational follow‐up programmes did not reduce anxiety symptoms, measured using HADS – Anxiety, compared with TAU at follow‐up (MD –0.47, 95% CI –1.67 to 0.73; 1 study, 192 participants; P = 0.44) (Analysis 6.5)
6.4.2 Depression
Psychoeducational follow‐up programmes did not reduce depression symptoms compared with TAU at follow‐up (SMD –0.30, 95% CI –1.08 to 0.48; 2 studies, 219 participants; P = 0.45) (Analysis 6.6).
6.5 Dropout rate
Psychoeducational follow‐up programmes did not decrease dropout compared with TAU at end of treatment (RR 1.76, 95% CI 0.82 to 3.78; 1 study, 64 participants; P = 0.14) or at follow‐up (RR 0.54, 95% CI 0.00 to 70.37; 2 studies, 259 participants; P = 0.81) (Analysis 6.7).
6.6 Use of health service resources
Psychoeducational follow‐up programmes did not decrease number of hospital visits compared with TAU at end of treatment (RR 0.18, 95% CI 0.02 to 1.43; 1 study, 64 participants; P = 0.10) (Analysis 6.8).
6.7 Adverse effects
Neither study measured adverse effects.
Comparison 7: specialised cognitive behavioural therapy‐based physiotherapy inpatient programme compared with wait list
One study compared CBT with wait list (Jordbru 2014).
7.1 Reduction in physical signs
The study did not measure reduction in physical signs.
7.2 Level of functioning
Specialised CBT‐based physiotherapy inpatient programme might improve level of functioning, measured using FIM, compared with wait list at end of treatment (MD 9.20, 95% CI 6.06 to 12.34; 1 study, 118 participants; P < 0.00001 (Analysis 7.1).
7.3 Quality of life
The study did not measure quality of life.
7.4 Mental state
Specialised CBT‐based physiotherapy inpatient programme might improve mental state, measured using SF‐12, compared with wait list at end of treatment (MD 9.10, 95% CI 4.96 to 13.24; 1 study, 118 participants; P < 0.0001) (Analysis 7.2).
7.5 Dropout rate
Specialised CBT‐based physiotherapy inpatient programme did not reduce dropout compared with wait list at end of treatment (RR 0.23, 95% CI 0.03 to 1.97; 1 study, 60 participants; P = 0.18) (Analysis 7.3).
7.6 Use of health service resources
The study did not measure use of health service resources.
7.7 Adverse effects
The study did not measure adverse effects.
Comparison 8: specialised cognitive behavioural therapy‐based physiotherapy outpatient intervention compared with treatment as usual
One study compared specialised CBT‐based physiotherapy outpatient intervention with TAU (Nielsen 2017).
8.1 Reduction in physical signs
8.1.1 Physical symptom load
Specialised CBT‐based physiotherapy outpatient intervention might reduce physical symptom load, measured using SF‐36 – Physical Component), compared with TAU at follow‐up (MD 9.20, 95% CI 4.00 to 14.40; 1 study, 57 participants; P = 0.0005) (Analysis 8.1).
8.2 Level of functioning
Specialised CBT‐based physiotherapy outpatient intervention might improve level of functioning, measured using WSAS, compared with TAU at end of treatment (MD –7.10, 95% CI –11.40 to –2.80; 1 study, 54 participants; P = 0.001) or at follow‐up (MD –6.70, 95% CI –12.07 to –1.33; 1 study, 57 participants; P = 0.01) (Analysis 8.2).
8.3 Quality of life
The study did not measure quality of life.
8.4 Mental state
8.4.1 Anxiety
Specialised CBT‐based physiotherapy outpatient intervention did not reduce symptoms of anxiety, measured using HADS, compared with TAU at end of treatment (MD –0.60, 95% CI –3.05 to 1.85; 1 study, 54 participants; P = 0.63) or at follow‐up (MD –1.00, 95% CI –3.71 to 1.71; 1 study, 57 participants; P = 0.47) (Analysis 8.3).
8.4.2 Depression
Specialised CBT‐based physiotherapy outpatient intervention might reduce symptoms of depression, measured using HADS, compared with TAU at end of treatment (MD –3.60, 95% CI –7.13 to –0.07; 1 study, 54 participants; P =0.05) and at follow‐up (MD –3.20, 95% CI –5.33 to –0.87; 1 study, 57 participants; P = 0.007) (Analysis 8.4).
8.5 Dropout rate
Specialised CBT‐based physiotherapy outpatient intervention did not improve dropout compared with TAU at end of treatment (RR 0.20, 95% CI 0.02 to 1.61; 1 study, 60 participants; P = 0.13) or at follow‐up (RR 0.50, 95% CI 0.05 to 5.22; 1 study, 60 participants; P = 0.56) (Analysis 8.5).
8.6 Use of health service resources
The study did not measure use of health service resources.
8.7 Adverse effects
The study did not measure adverse effects.
Comparison 9: brief psychotherapeutic intervention (psychodynamic interpersonal treatment approach) compared with standard care
One study compared brief psychotherapeutic interventions with SC (Hubschmid 2015).
9.1 Reduction in physical signs
9.1.1 Conversion symptoms
Brief psychotherapeutic interventions did not reduce conversion symptoms, measured using SDQ‐20, compared with SC at end of treatment (RR 0.12, 95% CI 0.01 to 2.00; 1 study, 19 participants; P = 0.14), at four months' follow‐up (RR 1.13, 95% CI 0.08 to 15.19; 1 study, 17 participants; P = 0.93) or at 10 months' follow‐up (RR 0.16, 95% 0.01 to 2.66; 1 study, 15 participants; P = 0.20) (Analysis 9.1).
9.2 Level of functioning
The study did not measure level of functioning.
9.3 Quality of life
Brief psychotherapeutic interventions did not improve quality of life, measured using SF‐36 – Physical Component, compared with SC at end of treatment (MD –6.99, 95% CI –28.09 to 14.11; 1 study, 16, participants; P = 0.52), at four months' follow‐up (MD –19.53, 95% CI –43.91 to 4.85; 1 study, 16 participants; P = 0.12) or at 10 months' follow‐up (MD –11.43, 95% CI –36.16 to 13.30; 1 study, 14 participants; P = 0.37) (Analysis 9.2).
9.4 Mental state
9.4.1 Depression
Brief psychotherapeutic intervention did not reduce depression symptoms, measured using BDI‐II, compared with SC at end of treatment (RR 1.29, 95% CI 0.10 to 17.14; 1 study, 16 participants; P = 0.85), at four months' follow‐up (RR 3.86, 95% CI 0.50 to 29.55; 1 study, 16 participants; P = 0.19) or at 10 months' follow‐up (RR 1.00, 95% CI 0.08 to 13.02; 1 study, 14 participants; P = 1.00) (Analysis 9.3).
9.5 Dropout rate
Brief psychotherapeutic intervention did not reduce dropout compared with SC at end of treatment (RR 1.09, 95% CI 0.18 to 6.48; 1 study, 23 participants; P = 0.92), at four months' follow‐up (RR 1.09, 95% CI 0.28 to 4.32; 1 study, 23 participants; P = 0.90) or at 10 months' follow‐up (RR 0.82, 95% CI 0.23 to 2.87; 1 study, 23 participants; P = 0.75) (Analysis 9.4).
9.6 Use of health service resources
Brief psychotherapeutic intervention did not reduce use of health service compared with SC at end of treatment (MD –0.16 95% CI –1.25 to 0.93; 1 study, 19 participants; P = 0.77) (Analysis 9.5).
9.7 Adverse effects
The study did not measure adverse effects.
Comparison 10: cognitive behavioural therapy plus adjunctive physical activity compared with cognitive behavioural therapy alone
One study compared CBT plus APA with CBT alone (Dallocchio 2016).
10.1 Reduction in physical signs
10.1.1 Overall physical impacts
CBT plus APA did not reduce overall physical impacts, measured using PMDRS total score, compared with CBT alone at end of treatment (MD 5.60, 95% CI –15.48 to 26.68; 1 study, 21 participants; P = 0.60) (Analysis 10.1).
10.2 Level of functioning
The study did not measure level of functioning.
10.3 Quality of life
The study did not measure quality of life.
10.4 Mental state
10.4.1 Anxiety
CBT plus APA did not reduce symptoms of anxiety, measured using BAI, compared with CBT alone at end of treatment (MD –3.40, 95% CI –8.01 to 1.21; 1 study, 21 participants; P = 0.15) (Analysis 10.2).
10.4.2 Depression
CBT plus APA did not reduce symptoms of depression, measured using Hamilton Depression Scale, compared with CBT alone at end of treatment (MD –0.50, 95% CI –3.32 to 2.32; 1 study, 21 participants; P = 0.73) (Analysis 10.3).
10.5 Dropout rate
CBT plus APA did not reduce dropout compared with CBT alone at end of treatment (MD 1.87, 95% CI 0.40 to 8.65; 1 study, 29 participants; P = 0.42) (Analysis 10.4).
10.6 Use of health service resources
The study did not measure use of health service resources.
10.7 Adverse effects
The study did not measure adverse effects.
Comparison 11: hypnosis compared with diazepam
One study compared hypnosis with diazepam (Mousavi 2008).
11.1 Reduction in physical signs
11.1.1 Symptom freedom
Hypnosis did not increase symptom freedom, measured using PMDRS Total Score, compared with diazepam at end of treatment (RR 0.69, 95% CI 0.39 to 1.24; 1 study, 40 participants; P = 0.22) (Analysis 11.1).
11.2 Level of functioning
The study did not measure level of functioning.
11.3 Quality of life
The study did not measure quality of life.
11.4 Mental state
The study did not measure mental state.
11.5 Dropout rate
The study did not measure dropouts.
11.6 Use of health service resources
The study did not measure use of health service resources.
11.7 Adverse effects
The study did not measure adverse effects.
Comparison 12: outpatient motivational interviewing and mindfulness‐based psychotherapy compared with psychotherapy alone
One study compared psychotherapy preceded by MI with psychotherapy alone (Tolchin 2019).
12.1 Reduction in physical signs
12.1.1 Decrease in seizure frequency
Psychotherapy preceded by MI might have a slight effect on physical signs, measured using participants' seizure diaries, compared with psychotherapy alone at end of treatment (MD 41.40, 95% CI 4.92 to 77.88; 1 study, 54 participants; P = 0.03) (Analysis 12.1).
12.2 Level of functioning
The study did not measure level of functioning.
12.3 Quality of life
Psychotherapy preceded by MI might have a slight effect on quality of life compared with psychotherapy alone at end of treatment (MD 5.40, 95% CI 0.26 to 10.54; 1 study, 47 participants; P = 0.04) (Analysis 12.3).
12.4 Mental state
The study did not measure mental state.
12.5 Dropout rate
Psychotherapy preceded by MI did not change dropouts compared with psychotherapy alone at end of treatment (RR 1.60, 95% CI 0.29 to 8.92; 1 study, 60 participants; P = 0.59) (Analysis 12.4).
12.6 Use of health service resources
12.6.1 Change in monthly visits
Psychotherapy preceded by MI did not change use of health service resources compared with psychotherapy alone at end of treatment (MD –0.21, 95% CI –0.55 to 0.13; 1 study, 54 participants; P = 0.23) (Analysis 12.2).
12.7 Adverse effects
The study did not measure adverse effects.
Discussion
We conducted this systematic review to examine the effects of psychosocial interventions of conversion and dissociative disorders in adults. We considered 168 full‐text reports from which we included 17 studies published in 21 articles in this review. We used all studies with eligible data in the meta‐analyses, as recommended in the Cochrane Handbook for Systematic Reviews of Interventions (Higgins 2019).
Summary of main results
In the following, we give a brief summary and present a short discussion on the results found for the primary and secondary outcomes of the review. As the overall GRADE analysis deemed all studies to be of low to very low quality, none of the following is to be taken as recommendations, but as indications of what might have effect, knowing very well that this would need to be replicated by studies of higher quality.
Primary outcomes
Reduction in physical signs
For the primary outcome, only five comparisons showed statistically significant positive effects on reduction in physical signs. These were:
-
hypnosis treatment for outpatients (Comparison 3);
-
behavioural therapy plus routine clinical care for inpatients (Comparison 4);
-
psychoeducational follow‐up programmes for outpatients (Comparison 6);
-
specialised CBT‐based physiotherapy‐led intervention (Comparison 8);
-
psychotherapy preceded by MI (Comparison 12).
Comparison 3 consisted of one study only, which showed that hypnosis compared to wait list reduced the severity of physical impairment at the end of treatment (Moene 2003). Interestingly, hypnosis for inpatients did not have the same effect (Moene 2002), perhaps because for inpatients the hypnosis treatment was an addition to an already full treatment programme, whereas for the outpatients it was the sole treatment. It could be argued that any treatment is better than none, but we cannot know this is the case without considering adverse effects and these were not reported.
Both studies in Comparison 4 had positive results on primary outcomes (Aamir 2012; Khattak 2006). Aamir 2012 showed a statistically significant reduction in the number of weekly seizures at end of treatment, while Khattak 2006 produced a statistically significant reduced symptom severity at end of treatment. It is noticeable that these two studies (of which Aamir 2012 was modelled on the Khattak 2006) are the studies most alike of all our comparisons, and also the only comparison group that had a positive effect on the primary outcome in more than one study. It would be of interest to see if similar results were obtained by new RCTs of a similar design, but with a much larger number of participants and lower risk of bias.
In Comparison 6, the studies showed different results. Drane 2016 showed a statistically significant reduction in seizure frequency compared to TAU at follow‐up, while Pleizier 2017 showed that psychoeducational follow‐up programmes did not reduce physical symptom load compared to TAU at follow‐up. The studies used different measures (a self‐developed scale based on number of seizures versus the SF‐36 Physical Component), but it would still be expected that similar effects would show on both measures for this outcome.
Comparison 8 only had one study, which was a five‐day outpatient specialised physiotherapy‐led intervention compared to TAU (Nielsen 2017). This was conducted as a feasibility study, inspired by the Jordbru 2014 study (which was inpatient cognitive behavioural‐based physiotherapy), but as one was inpatients and the other outpatients, we decided to have the two studies in separate comparisons.
In Comparison 7, Jordbru 2014 found no statistically significant effects on this outcome, while Comparison 8 found a reduction in physical signs at four months' follow‐up. It would be interesting if the authors of Nielsen 2017 decided to conduct a new and more comprehensive RCT, and see if the indication of effect found here could be affirmed.
Comparison 12 consisted of only one study, which showed that intervention led to a decrease in seizure frequency at end of treatment (Tolchin 2019). This study was conducted to investigate whether adherence to treatment would be higher if psychotherapy was preceded by MI immediately after diagnosis. Both groups thus received psychotherapy, but as the active group attended more sessions (i.e. adherence was better), it could be argued as to what might actually be the active mediator in the intervention. Perhaps a new study that compared the effect of the same number of actually received psychotherapy sessions in each arm might give some information on the extent MI or psychotherapy is the active mediator.
Secondary outcomes
The overall findings of the analysis do point towards statistically significant treatment effects of psychosocial interventions on some of the secondary outcomes.
Level of functioning
Three different comparisons showed an effect on level of functioning. Comparison 5 showed that CBT increased level of functioning compared to SMC at follow‐up (Goldstein 2010), while both Comparison 6 (psychoeducational follow‐up programmes versus TAU) (Chen 2014), and Comparison 8 (specialised CBT‐based physiotherapy led intervention compared to TAU) (Nielsen 2017), showed improved level of functioning both at end of treatment and follow‐up.
Quality of life
Comparison 6 found effects on improved quality of life at follow‐up (Drane 2016), and comparison 12 found improved quality of life at end of treatment (Tolchin 2019).
Mental state, anxiety and depression
Two comparisons showed effects on general mental state. Comparison 5 found that CBT compared to SMC reduced symptoms of general mental state at end of treatment (LaFrance 2014), while Comparison 7 showed that specialised CBT‐based physiotherapy in an inpatient programme improved mental state compared to wait list at end of treatment (Jordbru 2014).
Two different comparisons showed effects on anxiety. Comparison 1 showed that inpatient paradoxical intention therapy reduced anxiety symptoms at end of treatment compared to outpatient diazepam (Ataoglu 2003), while Comparison 4 (behavioural therapy plus routine clinical care compared to routine clinical care alone) likewise found effects in reduction in symptoms of anxiety at end of treatment (Aamir 2012; Khattak 2006). Comparison 4 also found effects on reduction in symptoms of depression at follow‐up.
Dropout rate
Only Comparison 4 showed statistical significant reduction in dropout at end of treatment (Aamir 2012; Khattak 2006).
Use of health service resources
We found no statistically significant results for use of health services resources.
Adverse effects
We found no statistically significant results for adverse effects.
Overall completeness and applicability of evidence
The review contains a mixture of strengths and issues in terms of overall completeness and applicability of evidence.
Issues concerning the overall completeness and applicability of the evidence
The review highlights some major issues concerning the overall completeness and applicability of the evidence of the benefits and harms of psychosocial interventions on people diagnosed with conversion or dissociative disorders compared with SC; wait‐list controls; or with another intervention, either pharmaceutical/biological or a different psychosocial intervention.
The impact of the applicability of findings related to the duration of studies, choice of interventions and comparators, diagnostic criteria, issues related to the ratings scales and lack of data of adverse effects.
Duration of studies
The duration of studies was very short, ranging from a few hours to four months. The mean duration of studies was 9.5 weeks (range two hours to 20 weeks). Due to lack of data it was not possible to investigate the long‐term benefits of psychological interventions for conversion and dissociative disorders. Overall, the evidence on long‐term effects on psychosocial interventions for people with conversion and dissociative disorders is lacking, and it is possible that the small, possibly beneficial effects found in this review might be diminished over the time.
Interventions and comparators
The studies in this review compared one type of psychosocial intervention with either another intervention (either pharmacological or psychosocial, or mixed) or with SC/SMC or wait list controls.
We had hoped to have combined those comparators in meta‐analyses; however, this was not possible due to variations in the interventions.
Lack of validated scales
The field of research on conversion disorder, dissociative disorder and related conditions has no clear guidance about the best way to measure the efficacy of treatments. There seem to be a variety of views on this, from researchers developing their own instruments (e.g. Moene 2002 has created VMCR), some studies using scales that are not widely used (e.g. Nielsen 2017, which used PMDRS), while others prefer to simply ask their participants to note the number of seizures in a day, week or month (e.g. Goldstein 2010). Some authors also mentioned that the nature of these disorders makes it difficult to decide on which aspect to focus, be it physical ability or quality of life.
This lack of common measures and lack of commonly used validated scales means it becomes almost impossible to compare the effects of different treatments. The findings of this review highlight the wide variety of different scales used to measure the primary outcome across the studies.
Lack of adverse effects
The studies did not describe adverse effects. As it is unusual for any treatment, psychosocial, medical or otherwise, to have no adverse effects, we would expect some of the studies in the review to have observed adverse effects, even if these are not reported.
Lack of participant involvement
None of the studies included report on whether they involved participants in deciding which outcomes to focus on. This would be a major improvement for future studies, to investigate which outcomes are most important to patients and include those. For this update of the review, we amended the secondary outcomes from the original review to be more in line with the interest of patients, but direct patient involvement could also be improved further in future updates.
Strengths in the overall completeness and applicability of the evidence
Countries and settings
One strength of the review was that the studies were from a wide range of countries and settings, which included low‐, medium‐ and high‐income economies, showing how widespread this condition is, as well as the interest in the psychosocial treatment of it.
Diagnostic criteria
A different strength of the review was that it concerned a diagnostic area usually very consistent in identifying the conditions in it, with only 4% of patients being misdiagnosed (Stone 2005). Furthermore, in a follow‐up study of Stone 2009, only 0.4% had acquired an organic disease diagnosis that was unexpected at initial assessment. These percentages are very low compared with other conditions, both psychiatric and somatic ones.
This means clinicians and policy makers can rest assured that whatever the findings of the review (or future versions of it), it will be applicable to conditions diagnosed in this way.
Quality of the evidence
As assessed by GRADE, the overall quality of the evidence ranged from low to very low. We assessed most of studies to have high risk of bias, which may cause systematic errors, that is overestimation of benefits and underestimation of harms (Higgins 2019). The certainty of the evidence was downgraded as a consequence of potential risk of bias, as many of the studies had unclear or inadequate allocation concealment, making them prone to selection bias. Due to the nature of psychological interventions, maintaining an adequate level of blinding of both personal and participants is often difficult. Despite this, it should be noted that most studies included in this review still managed a considerable level of blinding.
Our results were based on 17 studies with a limited number of participants (894). Due to the heterogeneous reporting of outcomes, it was not possible to combine results into a collective analysis to investigate for common effects across studies. Therefore, the results are based on individual studies with only few participants included. This led to imprecision and inconsistency of the estimates, ultimately resulting in further downgrading of the evidence. We were unable to perform funnel plot or any other analyses to assess whether there was risk of publication bias.
These important methodological limitations have reduced the reliability and robustness of the results in this review. Currently, there is insufficient evidence to draw any conclusions on the effect of any form of psychosocial intervention of conversion and dissociative disorders in adults. Further research may change the estimates of the treatment effect, but such studies ought to be conducted without risk of systematic errors (bias), random errors (play of chance) and design errors (Keus 2010). Overall, we found consistency between conclusions in the previous (Ruddy 2005) and the current review.
To advance this field, there is a need for high‐quality RCTs with large numbers of participants, that include similar reporting of core outcomes related to this disorder.
Potential biases in the review process
We conducted extensive searches of relevant databases. Two review authors independently selected studies for inclusion and extracted data. Disagreements were resolved by discussion with team members. We assessed risk of bias in all studies according to the recommendations provided in Chapter 8 of the Cochrane Handbook for Systematic Reviews of Interventions (Higgins 2011).
Data collection as risk of bias
To our knowledge we did not introduce any bias in the data collection process, as we used a comprehensive search strategy and made all available attempts to retrieve any missing data.
We do not know whether there could be any publication bias, in that it is possible that studies have been conducted but not been published, which would have been of interest to this review.
Deciding on rating scales
As many studies used several scales measuring our predefined outcomes, we made a choice for each study about which one to include. This was based on what the review author team deemed most relevant and of best quality, with reference to those scales highlighted in our Methods section. As the selection process was subjective and not a guideline‐based decision process, this could have introduced bias. We would assume that the various instruments assessing similar aspects would provide similar results if the instruments had been of similar quality. However, this is not necessarily the case as many self‐created and unvalidated rating scales were used.
Most of the 12 comparators had data on our primary outcome: 'Reduction in physical signs', although this was measured in many different ways. There were also data on many of our predefined secondary outcomes, most of them measuring important outcomes for these people. We had six secondary outcomes, which in fact was eight because we decided to divide the mental health outcome into anxiety and depression where possible. This high number of outcomes was chosen due to the clinical importance of these outcomes. However, we are aware that this might be problematic as it increases the likelihood of finding a statistical significant result just by chance (Jakobsen 2014).
Subjective judgements as risk of bias
It is inevitable that there will be certain aspects in the many steps of the process which may have or will have had subjective judgements involved.
For this review, there are a few areas we would like to address.
First, the risk of bias assessment itself will have had some subjective discussion behind the final decisions. Whether to be very strict on a judgement may or may not be clouded by the review authors' backgrounds and daily routines. In the case of this review, the main author is also a clinician, thus has a knowledge of what is possible and desirable in everyday clinical practice. The other review author working on risk of bias has a different background, one of extensive research, and we deliberately chose this combination to minimise the effects of any particular perspective. We followed the guidance of the Cochrane Handbook for Systematic Reviews of Interventions (Higgins 2019), and discussed any discrepancies and found consensus on all items. Hence, we do not judge that our processes would have introduced bias, but want to point out that this could be perceived if one is not aware of the knowledge and experience each review author brings to the process.
Second, in choosing which studies to group to create the possibility for comparisons, a risk of bias is also possible. As there are no guidelines on which comparisons to make, the choice presented in this review was based on our clinical judgement, discussed at length among the review author team to provide some coherence in comparing studies, while also ensuring fair and relevant groupings of studies. This should not have affected judgements about overall risk of bias, as the studies were chosen independently of any results, and the comparisons are made on the primary and secondary outcomes as predefined by this review, rather than by the individual studies. However, it is worth noting that such a decision will most likely depend on the clinical experience of the review authors, when studies are as diverse as the ones included here.
In conclusion, we judge that the review author team and the processes followed will not have introduced bias in the production of this review, but wanted to mention these areas for future updates and other teams to be aware of.
Agreements and disagreements with other studies or reviews
To our knowledge, this update of the previous Cochrane Review from Ruddy 2005 is the most extensive and comprehensive systematic review of RCTs of psychosocial interventions for conversion and dissociative disorders. The 2005 review identified 260 references and only three RCTs (119 participants) met the inclusion criteria. In the current review, we identified 3045 relevant research papers and included 17 RCTs (894 participants).
The review authors' conclusions in the 2005 study was that "If psychosocial interventions are available for people with conversion disorder then at present they should be viewed as experimental with slight evidence in favour of help rather than harm" and "It is unclear what effects these interventions and other psychosocial interventions have on social functioning, interpersonal relationships, quality of life or satisfaction with care."
It was reasonable to presume with more studies since 2005 and that there may be a better evidence base for drawing conclusions about treatment of these conditions. Unfortunately, despite there being some tentative conclusions about effectiveness in the current review, the overall findings are inconclusive because of the poor quality of the studies and because there was little opportunity to combine results from studies given a wide variety of different psychosocial interventions.
The results were assessed to be of very low to low quality of evidence measured by GRADE and most studies had a high risk of bias. Therefore, we likewise found no evidence of any conclusive benefits of any psychosocial intervention for conversion and dissociative disorders in adults. However, comparing the two reviews, we can see some improvement in the method and quality of studies in this area and find it encouraging to know that potentially high‐quality studies will appear soon (see 'Ongoing studies' and Characteristics of ongoing studies table).
Several other reviews and meta‐analyses have been published since the previous version of our review. However, none of them have consistently examined the same diagnostic spectrum (i.e. both conversion and dissociation). Our results confirm the findings of Martlew 2014 who concluded that there is little reliable evidence to support the use of any specific treatment for non‐epileptic seizures. They used the same methods as us, but found only four RCTs, which we included in our review.
In a review of Carlson 2017 the results of the analyses indicated that psychological interventions for non‐epileptic seizures may lead to greater rates of seizure reduction and seizure freedom compared to those who do not receive psychotherapy. When compared with the existing evidence, the results of the meta‐analyses indicated that psychological interventions may yield greater rates of seizure reduction (82%) and seizure freedom (47%) compared with those who do not receive psychotherapy (14% to 23%). However, they found there was a high risk of bias and a there was no proper analysis of the quality of evidence. They only found two RCTs, both are included in our review.
Some non‐randomised studies found a medium‐ to large‐effect size of psychotherapy and physical treatment for a range of psychological symptoms, physical symptoms and functional impairment (Brand 2009; Koelen 2014; Nielsen 2013). However, these studies had methodological limitations.
A more detailed description of these studies can be found in the 'Previous reviews on this theme' in the Why it is important to do this review section.
PRISMA flow diagram.
Risk of bias graph: review authors' judgements about each risk of bias item presented as percentages across all included studies.
Risk of bias summary: review authors' judgements about each risk of bias item for each included study.
Comparison 1: Inpatient paradoxical intention therapy versus outpatient diazepam, Outcome 1: Reduction in physical signs: no conversion symptoms in last 2 weeks
Comparison 1: Inpatient paradoxical intention therapy versus outpatient diazepam, Outcome 2: Mental state – anxiety (Hamilton)
Comparison 1: Inpatient paradoxical intention therapy versus outpatient diazepam, Outcome 3: Dropout rate
Comparison 2: Inpatient treatment programme plus hypnosis versus inpatient treatment programme, Outcome 1: Reduction in physical signs: severity of impairment (VRMC)
Comparison 2: Inpatient treatment programme plus hypnosis versus inpatient treatment programme, Outcome 2: Mental state (SCL‐90)
Comparison 2: Inpatient treatment programme plus hypnosis versus inpatient treatment programme, Outcome 3: Dropout rate
Comparison 3: Outpatient hypnosis versus wait list, Outcome 1: Reduction in physical signs: severity of impairment (VRMC)
Comparison 3: Outpatient hypnosis versus wait list, Outcome 2: Dropout rate
Comparison 4: Behavioural therapy plus routine clinical care versus routine clinical care, Outcome 1: Reduction in physical signs: number of weekly fits
Comparison 4: Behavioural therapy plus routine clinical care versus routine clinical care, Outcome 2: Reduction in physical signs: symptom severity (Clinical Global Impression CGI)
Comparison 4: Behavioural therapy plus routine clinical care versus routine clinical care, Outcome 3: Mental state – anxiety (HADS)
Comparison 4: Behavioural therapy plus routine clinical care versus routine clinical care, Outcome 4: Mental state – depression (HADS)
Comparison 4: Behavioural therapy plus routine clinical care versus routine clinical care, Outcome 5: Dropout rate
Comparison 5: Cognitive behavioural therapy versus standard medical care, Outcome 1: Reduction in physical signs: monthly seizure frequency (reduction in %)
Comparison 5: Cognitive behavioural therapy versus standard medical care, Outcome 2: Reduction in physical signs: monthly seizure frequency
Comparison 5: Cognitive behavioural therapy versus standard medical care, Outcome 3: Reduction in physical sign: seizure freedom
Comparison 5: Cognitive behavioural therapy versus standard medical care, Outcome 4: Level of functioning
Comparison 5: Cognitive behavioural therapy versus standard medical care, Outcome 5: Quality of life (QOLIE31)
Comparison 5: Cognitive behavioural therapy versus standard medical care, Outcome 6: Mental state – anxiety
Comparison 5: Cognitive behavioural therapy versus standard medical care, Outcome 7: Mental state – depression
Comparison 5: Cognitive behavioural therapy versus standard medical care, Outcome 8: Mental state (SCL‐90)
Comparison 5: Cognitive behavioural therapy versus standard medical care, Outcome 9: Dropout rate
Comparison 5: Cognitive behavioural therapy versus standard medical care, Outcome 10: Use of health services (number of general practitioner consultations)
Comparison 6: Psychoeducational follow‐up programmes versus treatment as usual, Outcome 1: Reduction in physical signs: seizure frequency (self‐made scale)
Comparison 6: Psychoeducational follow‐up programmes versus treatment as usual, Outcome 2: Reduction in physical signs: physical symptom load (SF‐36 – physical)
Comparison 6: Psychoeducational follow‐up programmes versus treatment as usual, Outcome 3: Level of functioning (WSAS)
Comparison 6: Psychoeducational follow‐up programmes versus treatment as usual, Outcome 4: Quality of life (QOLIE10‐P)
Comparison 6: Psychoeducational follow‐up programmes versus treatment as usual, Outcome 5: Mental state – anxiety (HADS)
Comparison 6: Psychoeducational follow‐up programmes versus treatment as usual, Outcome 6: Mental state – depression
Comparison 6: Psychoeducational follow‐up programmes versus treatment as usual, Outcome 7: Dropout rate
Comparison 6: Psychoeducational follow‐up programmes versus treatment as usual, Outcome 8: Use of health services (number hospital visits)
Comparison 7: Specialised cognitive behavioural therapy‐based physiotherapy inpatient programme versus wait list, Outcome 1: Level of functioning (Functional Independence Measure Motor (FIM))
Comparison 7: Specialised cognitive behavioural therapy‐based physiotherapy inpatient programme versus wait list, Outcome 2: Mental state (SF‐12)
Comparison 7: Specialised cognitive behavioural therapy‐based physiotherapy inpatient programme versus wait list, Outcome 3: Dropout rate
Comparison 8: Specialised cognitive behavioural therapy‐based physiotherapy outpatient intervention compared to treatment as usual (TAU), Outcome 1: Reduction in physical signs: physical symptom load (SF‐36 – Physical Component)
Comparison 8: Specialised cognitive behavioural therapy‐based physiotherapy outpatient intervention compared to treatment as usual (TAU), Outcome 2: Level of functioning (WSAS)
Comparison 8: Specialised cognitive behavioural therapy‐based physiotherapy outpatient intervention compared to treatment as usual (TAU), Outcome 3: Mental state – anxiety (HADS)
Comparison 8: Specialised cognitive behavioural therapy‐based physiotherapy outpatient intervention compared to treatment as usual (TAU), Outcome 4: Mental state – depression (HADS)
Comparison 8: Specialised cognitive behavioural therapy‐based physiotherapy outpatient intervention compared to treatment as usual (TAU), Outcome 5: Dropout rate
Comparison 9: Brief psychotherapeutic intervention (psychodynamic interpersonal treatment approach) versus standard care, Outcome 1: Reduction in physical signs: conversions symptoms (SDQ20)
Comparison 9: Brief psychotherapeutic intervention (psychodynamic interpersonal treatment approach) versus standard care, Outcome 2: Quality of life (SF‐36)
Comparison 9: Brief psychotherapeutic intervention (psychodynamic interpersonal treatment approach) versus standard care, Outcome 3: Mental state – depression (BDI‐II)
Comparison 9: Brief psychotherapeutic intervention (psychodynamic interpersonal treatment approach) versus standard care, Outcome 4: Dropout rate
Comparison 9: Brief psychotherapeutic intervention (psychodynamic interpersonal treatment approach) versus standard care, Outcome 5: Use of health services (emergency department visits)
Comparison 10: Cognitive behavioural therapy plus adjunctive physical activity versus cognitive behavioural therapy alone, Outcome 1: Reduction in physical signs: overall physical impact (Psychogenic Movement Disorder Scale (PMDRS)
Comparison 10: Cognitive behavioural therapy plus adjunctive physical activity versus cognitive behavioural therapy alone, Outcome 2: Mental state – anxiety (BAI)
Comparison 10: Cognitive behavioural therapy plus adjunctive physical activity versus cognitive behavioural therapy alone, Outcome 3: Mental state – depression (Hamilton)
Comparison 10: Cognitive behavioural therapy plus adjunctive physical activity versus cognitive behavioural therapy alone, Outcome 4: Dropout rate
Comparison 11: Hypnosis versus diazepam, Outcome 1: Reduction in physical signs: symptom freedom
Comparison 12: Outpatient motivational interviewing and mindfulness‐based psychotherapy compared with psychotherapy alone, Outcome 1: Reduction in physical signs: decrease in seizure frequency
Comparison 12: Outpatient motivational interviewing and mindfulness‐based psychotherapy compared with psychotherapy alone, Outcome 2: Changes in monthly visits
Comparison 12: Outpatient motivational interviewing and mindfulness‐based psychotherapy compared with psychotherapy alone, Outcome 3: Quality of life
Comparison 12: Outpatient motivational interviewing and mindfulness‐based psychotherapy compared with psychotherapy alone, Outcome 4: Dropout rate
| Paradoxical intention therapy compared with diazepam for conversion disorder | ||||||
| Patient or population: people with conversion disorder according to DSM‐IV or ICD‐10 criteria Settings: outpatient and inpatient Intervention: paradoxical intention therapy Comparison: diazepam over 45 days | ||||||
| Outcomes | Illustrative comparative risks* (95% CI) | Relative effect | No of participants | Certainty of the evidence | Comments | |
|---|---|---|---|---|---|---|
| Assumed risk | Corresponding risk | |||||
| Diazepam | Paradoxical intention therapy | |||||
| Reduction in physical signs (number of patients without any conversive attacks in last 2 weeks) End of treatment | Study population | RR 1.44 (0.91 to 2.28) | 30 | ⊕⊝⊝⊝ | Paradoxical intention therapy may have no effect on physical signs at end of treatment. | |
| 600 per 1000 | 864 per 1000 | |||||
| Level of functioning | — | — | — | — | — | No studies assessed this outcome. |
| Quality of life | — | — | — | — | — | No studies assessed this outcome. |
| Adverse events | — | — | — | — | — | No studies assessed this outcome. |
| *The basis for the assumed risk (e.g. the median control group risk across studies) is provided in footnotes. The corresponding risk (and its 95% confidence interval) is based on the assumed risk in the comparison group and the relative effect of the intervention (and its 95% CI). CI: confidence interval; DSM‐IV:Diagnostic and Statistical Manual of Mental Disorders 4th Edition; ICD‐10:International Classification of Diseases, Tenth Revision; RR: risk ratio. | ||||||
| GRADE Working Group grades of evidence | ||||||
| aDowngraded two levels due to imprecision (wide confidence intervals; based on one study with few patients). | ||||||
| Hypnosis + treatment as usual compared with treatment as usual for conversion disorder | ||||||
| Patient or population: people with conversion disorder according to DSM‐IV or ICD‐10 criteria Settings: inpatient Intervention: hypnosis + TAU Comparison: TAU | ||||||
| Outcomes | Illustrative comparative risks* (95% CI) | Relative effect | No of participants | Certainty of the evidence | Comments | |
|---|---|---|---|---|---|---|
| Assumed risk | Corresponding risk | |||||
| TAU | Hypnosis + TAU | |||||
| Reduction in physical signs (Severity of impairment) Measured by the VRMC scale (higher is better) Range: 1–7 End of treatment | The mean reduction in physical signs in the control group was 5.9 | MD 0.49 lower (1.28 lower to 0.30 higher) | — | 45 | ⊕⊝⊝⊝ | Hypnosis + TAU may have no effect on physical signs at end of treatment |
| Level of functioning | — | — | — | — | — | No studies assessed this outcome |
| Quality of life | — | — | — | — | — | No studies assessed this outcome |
| Adverse events | — | — | — | — | — | No studies assessed this outcome |
| *The basis for the assumed risk (e.g. the median control group risk across studies) is provided in footnotes. The corresponding risk (and its 95% confidence interval) is based on the assumed risk in the comparison group and the relative effect of the intervention (and its 95% CI). CI: confidence interval; DSM‐IV:Diagnostic and Statistical Manual of Mental Disorders 4th Edition; ICD‐10:International Classification of Diseases, Tenth Revision; MD: mean difference; TAU: treatment as usual; VRMC: The Video Rating Scale for Motor Conversion Symptoms. | ||||||
| GRADE Working Group grades of evidence | ||||||
| aDowngraded one level due to high risk of bias. | ||||||
| Hypnosis compared with wait list for conversion disorder | ||||||
| Patient or population: people with conversion disorder according to DSM‐IV or ICD‐10 criteria Settings: outpatient Intervention: hypnosis Comparison: wait list | ||||||
| Outcomes | Illustrative comparative risks* (95% CI) | Relative effect | No of participants | Certainty of the evidence | Comments | |
|---|---|---|---|---|---|---|
| Assumed risk | Corresponding risk | |||||
| Wait list | Hypnosis | |||||
| Reduction in physical signs (Severity of impairment) Measured by the VRMC scale (higher is better) Range: 1–7 End of treatment | The mean reduction in physical signs in the control group was 3.8 | MD 2.10 higher (1.34 higher to 2.86 higher) | — | 49 | ⊕⊕⊝⊝ | Hypnosis may have little effect on reduction in physical signs at end of treatment |
| Level of functioning | — | — | — | — | — | No studies assessed this outcome |
| Quality of life | — | — | — | — | — | No studies assessed this outcome |
| Adverse events | — | — | — | — | — | No studies assessed this outcome |
| *The basis for the assumed risk (e.g. the median control group risk across studies) is provided in footnotes. The corresponding risk (and its 95% confidence interval) is based on the assumed risk in the comparison group and the relative effect of the intervention (and its 95% CI). CI: confidence interval; DSM‐IV:Diagnostic and Statistical Manual of Mental Disorders 4th Edition; ICD‐10:International Classification of Diseases, Tenth Revision; MD: mean difference; VRMC: Video Rating Scale for Motor Conversion Symptoms. | ||||||
| GRADE Working Group grades of evidence | ||||||
| aDowngraded one level due to high risk of bias. | ||||||
| Behavioural therapy + TAU compared with TAU for conversion disorder | ||||||
| Patient or population: people with conversion disorder according to DSM‐IV or ICD‐10 criteria Settings: inpatient Intervention: behavioural therapy + TAU Comparison: TAU | ||||||
| Outcomes | Illustrative comparative risks* (95% CI) | Relative effect | No of participants | Certainty of the evidence | Comments | |
|---|---|---|---|---|---|---|
| Assumed risk | Corresponding risk | |||||
| TAU | Behavioural therapy + TAU | |||||
| Reduction in physical signs Number of weekly seizures, as assessed by daily self‐reported treatment diary End of treatment | The mean reduction in physical signs in the control group was 27.8 | MD 21.40 lower (27.88 lower to 14.92 lower) | — | 18 | Very lowa,b ⊕⊝⊝⊝ | Behavioural therapy + TAU may have little effect on reduction in physical signs at end of treatment. |
| Reduction in physical signs (symptom severity) Measured by Clinical Global Impression scale (lower is better) Range: 1–7 End of treatment | The mean reduction in physical signs in the control group was 4.48 | MD 2.90 lower (3.41 lower to 2.39 lower) | — | 90 | Very lowa,b ⊕⊝⊝⊝ | Behavioural therapy + TAU may have little effect on reduction in physical signs at end of treatment. |
| Level of functioning | — | — | — | — | — | No studies assessed this outcome. |
| Quality of life | — | — | — | — | — | No studies assessed this outcome. |
| Adverse events | — | — | — | — | — | No studies assessed this outcome. |
| *The basis for the assumed risk (e.g. the median control group risk across studies) is provided in footnotes. The corresponding risk (and its 95% confidence interval) is based on the assumed risk in the comparison group and the relative effect of the intervention (and its 95% CI). CI: Confidence interval; DSM‐IV:Diagnostic and Statistical Manual of Mental Disorders 4th Edition; ICD‐10:International Classification of Diseases, Tenth Revision; MD: mean difference. | ||||||
| GRADE Working Group grades of evidence | ||||||
| aDowngraded two levels due to high risk of bias. | ||||||
| Cognitive behavioural therapy compared with standard medical care for conversion disorder | ||||||
| Patient or population: people with conversion disorder according to DSM‐IV or ICD‐10 criteria Settings: outpatient Intervention: cognitive behavioural therapy Comparison: standard medical care | ||||||
| Outcomes | Illustrative comparative risks* (95% CI) | Relative effect | No of participants | Certainty of the evidence | Comments | |
|---|---|---|---|---|---|---|
| Assumed risk | Corresponding risk | |||||
| Standard medical care | Cognitive behavioural therapy | |||||
| Reduction in physical signs Reduction in monthly seizure frequency as assessed by a daily self‐reported seizure diary End of treatment | Study population | RR 1.56 (0.39 to 6.19) | 16 | ⊕⊝⊝⊝ | Cognitive behavioural therapy may have little or no effect on reducing physical signs at end of treatment. | |
| 286 per 1000 | 446 per 1000 | |||||
| Reduction in physical signs Monthly seizure frequency as assessed by a daily self‐reported seizure diary (lower is better) End of treatment | The mean reduction in physical signs in the control group was 6.75 | MD –4.75 lower (18.73 lower to 9.23 higher) | — | 61 | ⊕⊕⊝⊝ | Cognitive behavioural therapy may have little or no effect on reducing physical signs at end of treatment. |
| Reduction in physical sign Seizure freedom as assessed by a daily self‐reported seizure diary End of treatment | Study population | RR 2.33 (0.30 to 17.88) | 16 | ⊕⊝⊝⊝ | Cognitive behavioural therapy may have little or no effect on reducing physical signs at end of treatment. | |
| 143 per 1000 | 333 per 1000 | |||||
| Level of functioning As measured by the GAF (range 0–100) scale and the WSAS scale (0–40) (lower is better) End of treatment | — | SMD 0.44 higher (1.69 lower to 2.57 higher) I2 = 91% | — | 74 | ⊕⊝⊝⊝ | Cognitive behavioural therapy may have little or no effect on level of functioning at end of treatment. |
| Quality of life As assessed by QOLIE31 (higher is better) Range: 15–97 End of treatment | The mean quality of life in the control group was 9.7 | MD 11.20 higher (7.98 lower to 30.38 higher) | — | 16 | ⊕⊝⊝⊝ | Cognitive behavioural therapy may have little or no effect on quality of life at end of treatment. |
| Adverse events | — | — | — | — | — | No studies assessed this outcome. |
| *The basis for the assumed risk (e.g. the median control group risk across studies) is provided in footnotes. The corresponding risk (and its 95% confidence interval) is based on the assumed risk in the comparison group and the relative effect of the intervention (and its 95% CI). CI: confidence interval; DSM‐IV:Diagnostic and Statistical Manual of Mental Disorders 4th Edition; GAF: global assessment of functioning; ICD‐10:International Classification of Diseases, Tenth Revision; MD: mean difference; QOLIE31: quality of life in epilepsy inventory; RR: Risk Ratio; SMD: standardised mean difference; WSAS: work and social adjustment scale. | ||||||
| GRADE Working Group grades of evidence | ||||||
| aDowngraded one level due to high risk of bias. | ||||||
| Systematic follow‐up programme compared with TAU for conversion disorder | ||||||
| Patient or population: people with conversion disorder according to DSM‐IV or ICD‐10 criteria Settings: outpatient Intervention: systematic follow‐up programme Comparison: TAU | ||||||
| Outcomes | Illustrative comparative risks* (95% CI) | Relative effect | No of participants | Certainty of the evidence | Comments | |
|---|---|---|---|---|---|---|
| Assumed risk | Corresponding risk | |||||
| TAU | Systematic follow‐up programme | |||||
| Reduction in physical signs End of treatment | — | — | — | — | — | No studies assessed this outcome at end of treatment. |
| Level of functioning As assessed by WSAS scale (lower is better) Range: 0–40 End of treatment | The mean level of functioning in the control group was 25.52 | MD 7.12 lower (12.47 lower to 1.77 lower) | — | 43 | ⊕⊝⊝⊝ | Psychoeducational follow‐up programme may have little effect on level of functioning at end of treatment. |
| Quality of life | — | — | — | — | — | No studies assessed this outcome. |
| Adverse events | — | — | — | — | — | No studies assessed this outcome. |
| *The basis for the assumed risk (e.g. the median control group risk across studies) is provided in footnotes. The corresponding risk (and its 95% confidence interval) is based on the assumed risk in the comparison group and the relative effect of the intervention (and its 95% CI). CI: confidence interval; DSM‐IV:Diagnostic and Statistical Manual of Mental Disorders 4th Edition; ICD‐10:International Classification of Diseases, Tenth Revision; MD: mean difference; TAU: treatment as usual; WSAS: Work and Social Adjustment Scale. | ||||||
| GRADE Working Group grades of evidence | ||||||
| aDowngraded two levels due to high risk of bias. | ||||||
| Specialised CBT‐based physiotherapy programme compared with wait list for conversion disorder | ||||||
| Patient or population: people with conversion disorder according to DSM‐IV or ICD‐10 criteria Settings: inpatient Intervention: specialised CBT‐based physiotherapy programme Comparison: wait list | ||||||
| Outcomes | Illustrative comparative risks* (95% CI) | Relative effect | No of participants | Certainty of the evidence | Comments | |
|---|---|---|---|---|---|---|
| Assumed risk | Corresponding risk | |||||
| Wait list | Specialised CBT‐based physiotherapy | |||||
| Reduction in physical signs End of treatment | — | — | — | — | — | No studies assessed this outcome. |
| Level of functioning As measured by FIM (higher is better) Range: 18–126 End of treatment | The mean level of functioning in the control group was 80.9 | MD 9.20 higher (6.06 higher to 12.34 higher) | — | 118 | ⊕⊕⊝⊝ | Specialised CBT‐based physiotherapy may slightly improve level of functioning at end of treatment. |
| Quality of life | — | — | — | — | — | No studies assessed this outcome. |
| Adverse events | — | — | — | — | — | No studies assessed this outcome. |
| *The basis for the assumed risk (e.g. the median control group risk across studies) is provided in footnotes. The corresponding risk (and its 95% confidence interval) is based on the assumed risk in the comparison group and the relative effect of the intervention (and its 95% CI). CBT: cognitive behavioural therapy; CI: confidence interval; DSM‐IV:Diagnostic and Statistical Manual of Mental Disorders 4th Edition; FIM: Functional Independence Measure Motor; ICD‐10:International Classification of Diseases, Tenth Revision; MD: mean difference. | ||||||
| GRADE Working Group grades of evidence | ||||||
| aDowngraded one level due to high risk of bias. | ||||||
| Specialised CBT‐based physiotherapy‐led intervention compared with TAU for conversion disorder | ||||||
| Patient or population: people with conversion disorder according to DSM‐IV or ICD‐10 criteria Settings: outpatients at day hospital Intervention: specialised CBT‐based physiotherapy‐led intervention Comparison: TAU | ||||||
| Outcomes | Illustrative comparative risks* (95% CI) | Relative effect | No of participants | Certainty of the evidence | Comments | |
|---|---|---|---|---|---|---|
| Assumed risk | Corresponding risk | |||||
| TAU | Specialised CBT‐based physiotherapy‐led intervention | |||||
| Reduction in physical signs End of treatment | — | — | — | — | — | No studies assessed this outcome at end of treatment. |
| Level of functioning As assessed by WSAS scale (lower is better) Range: 0–40 End of treatment | The mean level of functioning in the control group was 26.9 | MD 7.10 lower (11.40 lower to 2.80 lower) | — | 54 | ⊕⊝⊝⊝ | Specialised CBT‐based physiotherapy intervention may slightly improve level of functioning at end of treatment. |
| Quality of life | — | — | — | — | — | No studies assessed this outcome. |
| Adverse events | — | — | — | — | — | No studies assessed this outcome. |
| *The basis for the assumed risk (e.g. the median control group risk across studies) is provided in footnotes. The corresponding risk (and its 95% confidence interval) is based on the assumed risk in the comparison group and the relative effect of the intervention (and its 95% CI). CBT: cognitive behavioural therapy; CI: confidence interval; DSM‐IV:Diagnostic and Statistical Manual of Mental Disorders 4th Edition; ICD‐10:International Classification of Diseases, Tenth Revision; TAU: treatment as usual; WSAS: Work and Social Adjustment Scale. | ||||||
| GRADE Working Group grades of evidence | ||||||
| aDowngraded one level due to high risk of bias. | ||||||
| Brief psychotherapeutic intervention compared with standard care for conversion disorder | ||||||
| Patient or population: people with conversion disorder according to DSM‐IV or ICD‐10 criteria Settings: outpatients Intervention: brief psychotherapeutic intervention Comparison: standard care | ||||||
| Outcomes | Illustrative comparative risks* (95% CI) | Relative effect | No of participants | Certainty of the evidence | Comments | |
|---|---|---|---|---|---|---|
| Assumed risk | Corresponding risk | |||||
| Standard care | Brief psychotherapeutic intervention | |||||
| Reduction in physical signs As assessed by SDQ‐20 (lower is better) Range: 20–100 End of treatment | Study population | RR 0.12 (0.01 to 2.00) | 19 | ⊕⊝⊝⊝ | Brief psychotherapeutic intervention may have no effect on physical signs at end of treatment. | |
| 400 per 1000 | 48 per 1000 | |||||
| Level of functioning | — | — | — | — | — | No studies assessed this outcome. |
| Quality of life As assessed by SF‐36 (lower is better) Range: 0–100 End of treatment | The mean quality of life in the control group was 50.56 | MD 6.99 lower (28.09 lower to 14.11 higher) | 16 | ⊕⊝⊝⊝ | Brief psychotherapeutic intervention may have little effect on quality of life after end of treatment. | |
| Adverse events | No studies assessed this outcome. | |||||
| *The basis for the assumed risk (e.g. the median control group risk across studies) is provided in footnotes. The corresponding risk (and its 95% confidence interval) is based on the assumed risk in the comparison group and the relative effect of the intervention (and its 95% CI). CI: confidence interval; DSM‐IV:Diagnostic and Statistical Manual of Mental Disorders 4th Edition; ICD‐10:International Classification of Diseases, Tenth Revision; MD: mean difference; RR: Risk Ratio; SDQ‐20: somatoform dissociation questionnaire; SF‐36: 36‐item Short Form. | ||||||
| GRADE Working Group grades of evidence | ||||||
| aDowngraded one level due to high risk of bias. | ||||||
| CBT + APA compared with CBT alone for conversion disorder | ||||||
| Patient or population: people with conversion disorder according to DSM‐IV or ICD‐10 criteria Settings: outpatients Intervention: CBT + APA Comparison: CBT | ||||||
| Outcomes | Illustrative comparative risks* (95% CI) | Relative effect | No of participants | Certainty of the evidence | Comments | |
|---|---|---|---|---|---|---|
| Assumed risk | Corresponding risk | |||||
| CBT | CBT + APA | |||||
| Reduction in physical signs (overall physical impact) As measured by PMDRS, total score (lower is better) Range: 0–128 End of treatment | The mean reduction in physical signs in the control group was 33.2 | MD 5.60 higher (15.48 lower to 26.68 higher) | — | 21 | ⊕⊝⊝⊝ | CBT + APA may have no effect on physical signs at end of treatment. |
| Level of functioning | — | — | — | — | — | No studies assessed this outcome. |
| Quality of life | — | — | — | — | — | No studies assessed this outcome. |
| Adverse events | — | — | — | — | — | No studies assessed this outcome. |
| *The basis for the assumed risk (e.g. the median control group risk across studies) is provided in footnotes. The corresponding risk (and its 95% confidence interval) is based on the assumed risk in the comparison group and the relative effect of the intervention (and its 95% CI). APA: adjunctive physical activity; CBT: cognitive behavioural therapy; CI: confidence interval; DSM‐IV:Diagnostic and Statistical Manual of Mental Disorders 4th Edition; ICD‐10:International Classification of Diseases, Tenth Revision; MD: mean difference; PMDRS: Psychogenic Movement Disorders Rating Scale. | ||||||
| GRADE Working Group grades of evidence | ||||||
| aDowngraded one level due to high risk of bias. | ||||||
| Hypnosis compared with diazepam for conversion disorder | ||||||
| Patient or population: people with conversion disorder according to DSM‐IV or ICD‐10 criteria Settings: emergency unit Intervention: hypnosis Comparison: diazepam | ||||||
| Outcomes | Illustrative comparative risks* (95% CI) | Relative effect | No of participants | Certainty of the evidence | Comments | |
|---|---|---|---|---|---|---|
| Assumed risk | Corresponding risk | |||||
| Diazepam | Hypnosis | |||||
| Reduction in physical signs Number with symptom freedom End of treatment | Study population | RR 0.69 (0.39 to 1.24) | 40 | ⊕⊝⊝⊝ | Hypnosis may have no effect on physical signs at end of treatment. | |
| 650 per 1000 | 448 per 1000 | |||||
| Level of functioning | — | — | — | — | — | No studies assessed this outcome. |
| Quality of life | — | — | — | — | — | No studies assessed this outcome. |
| Adverse events | — | — | — | — | — | No studies assessed this outcome. |
| *The basis for the assumed risk (e.g. the median control group risk across studies) is provided in footnotes. The corresponding risk (and its 95% confidence interval) is based on the assumed risk in the comparison group and the relative effect of the intervention (and its 95% CI). CI: confidence interval; DSM‐IV:Diagnostic and Statistical Manual of Mental Disorders 4th Edition; ICD‐10:International Classification of Diseases, Tenth Revision; RR: risk ratio. | ||||||
| GRADE Working Group grades of evidence | ||||||
| aDowngraded one level due to high risk of bias. | ||||||
| Psychotherapy preceded by motivational interviewing compared with psychotherapy alone for conversion disorder | ||||||
| Patient or population: people with conversion disorder according to DSM‐IV or ICD‐10 criteria Settings: outpatient Intervention: psychotherapy preceded by motivational interviewing Comparison: psychotherapy alone | ||||||
| Outcomes | Illustrative comparative risks* (95% CI) | Relative effect | No of participants | Certainty of the evidence | Comments | |
|---|---|---|---|---|---|---|
| Assumed risk | Corresponding risk | |||||
| Psychotherapy | Psychotherapy preceded by motivational interviewing | |||||
| Reduction in physical signs Decrease in seizure frequency End of treatment | The mean reduction in physical signs in the control group was 34.8 | MD 41.40 higher (4.92 higher to 77.88 higher) | — | 54 | ⊕⊝⊝⊝ | Psychotherapy preceded by motivational interviewing may have little effect on physical signs at end of treatment. |
| Level of functioning | — | — | — | — | — | No studies assessed this outcome. |
| Quality of life As measured by QOLIE10 (lower is better) Range: 10–50 End of treatment | The mean quality of life in the control group was 1.8 | MD 5.40 higher (0.26 higher to 10.54 higher) | — | 47 | ⊕⊝⊝⊝ | Psychotherapy preceded by motivational interviewing may have little effect on quality of life at end of treatment. |
| Adverse events | — | — | — | — | — | No studies assessed this outcome. |
| *The basis for the assumed risk (e.g. the median control group risk across studies) is provided in footnotes. The corresponding risk (and its 95% confidence interval) is based on the assumed risk in the comparison group and the relative effect of the intervention (and its 95% CI). CI: Confidence interval; DSM‐IV:Diagnostic and Statistical Manual of Mental Disorders 4th Edition; ICD‐10:International Classification of Diseases, Tenth Revision; MD: mean difference; QOLIE10: quality of life in epilepsy. | ||||||
| GRADE Working Group grades of evidence | ||||||
| aDowngraded one level due to high risk of bias. | ||||||
| Outcome or subgroup title | No. of studies | No. of participants | Statistical method | Effect size |
| 1.1 Reduction in physical signs: no conversion symptoms in last 2 weeks Show forest plot | 1 | 30 | Risk Ratio (IV, Fixed, 95% CI) | 1.44 [0.91, 2.28] |
| 1.1.1 End of treatment | 1 | 30 | Risk Ratio (IV, Fixed, 95% CI) | 1.44 [0.91, 2.28] |
| 1.2 Mental state – anxiety (Hamilton) Show forest plot | 1 | 30 | Mean Difference (IV, Fixed, 95% CI) | ‐3.73 [‐6.96, ‐0.50] |
| 1.2.1 End of treatment | 1 | 30 | Mean Difference (IV, Fixed, 95% CI) | ‐3.73 [‐6.96, ‐0.50] |
| 1.3 Dropout rate Show forest plot | 1 | 30 | Risk Ratio (M‐H, Fixed, 95% CI) | Not estimable |
| Outcome or subgroup title | No. of studies | No. of participants | Statistical method | Effect size |
| 2.1 Reduction in physical signs: severity of impairment (VRMC) Show forest plot | 1 | Mean Difference (IV, Fixed, 95% CI) | Subtotals only | |
| 2.1.1 End of treatment | 1 | 45 | Mean Difference (IV, Fixed, 95% CI) | ‐0.49 [‐1.28, 0.30] |
| 2.1.2 Follow‐up | 1 | 45 | Mean Difference (IV, Fixed, 95% CI) | 0.13 [‐0.55, 0.81] |
| 2.2 Mental state (SCL‐90) Show forest plot | 1 | Mean Difference (IV, Fixed, 95% CI) | Subtotals only | |
| 2.2.1 End of treatment | 1 | 45 | Mean Difference (IV, Fixed, 95% CI) | 1.42 [‐36.02, 38.86] |
| 2.2.2 Follow‐up | 1 | 45 | Mean Difference (IV, Fixed, 95% CI) | ‐5.97 [‐44.22, 32.28] |
| 2.3 Dropout rate Show forest plot | 1 | 49 | Risk Ratio (M‐H, Fixed, 95% CI) | 0.88 [0.14, 5.79] |
| Outcome or subgroup title | No. of studies | No. of participants | Statistical method | Effect size |
| 3.1 Reduction in physical signs: severity of impairment (VRMC) Show forest plot | 1 | 49 | Mean Difference (IV, Fixed, 95% CI) | 2.10 [1.34, 2.86] |
| 3.1.1 End of treatment | 1 | 49 | Mean Difference (IV, Fixed, 95% CI) | 2.10 [1.34, 2.86] |
| 3.2 Dropout rate Show forest plot | 1 | Risk Ratio (IV, Fixed, 95% CI) | Subtotals only | |
| 3.2.1 End of treatment | 1 | 49 | Risk Ratio (IV, Fixed, 95% CI) | 4.17 [0.50, 34.66] |
| Outcome or subgroup title | No. of studies | No. of participants | Statistical method | Effect size |
| 4.1 Reduction in physical signs: number of weekly fits Show forest plot | 1 | 18 | Mean Difference (IV, Fixed, 95% CI) | ‐21.40 [‐27.88, ‐14.92] |
| 4.1.1 End of treatment | 1 | 18 | Mean Difference (IV, Fixed, 95% CI) | ‐21.40 [‐27.88, ‐14.92] |
| 4.2 Reduction in physical signs: symptom severity (Clinical Global Impression CGI) Show forest plot | 1 | 90 | Mean Difference (IV, Fixed, 95% CI) | ‐2.90 [‐3.41, ‐2.39] |
| 4.2.1 End of treatment | 1 | 90 | Mean Difference (IV, Fixed, 95% CI) | ‐2.90 [‐3.41, ‐2.39] |
| 4.3 Mental state – anxiety (HADS) Show forest plot | 2 | 108 | Mean Difference (IV, Random, 95% CI) | ‐5.47 [‐7.08, ‐3.86] |
| 4.3.1 End of treatment | 2 | 108 | Mean Difference (IV, Random, 95% CI) | ‐5.47 [‐7.08, ‐3.86] |
| 4.4 Mental state – depression (HADS) Show forest plot | 2 | 108 | Mean Difference (IV, Random, 95% CI) | ‐4.99 [‐6.44, ‐3.53] |
| 4.4.1 Follow‐up | 2 | 108 | Mean Difference (IV, Random, 95% CI) | ‐4.99 [‐6.44, ‐3.53] |
| 4.5 Dropout rate Show forest plot | 2 | Risk Ratio (IV, Random, 95% CI) | Subtotals only | |
| 4.5.1 End of treatment | 2 | 118 | Risk Ratio (IV, Random, 95% CI) | 0.24 [0.06, 0.90] |
| Outcome or subgroup title | No. of studies | No. of participants | Statistical method | Effect size |
| 5.1 Reduction in physical signs: monthly seizure frequency (reduction in %) Show forest plot | 1 | Risk Ratio (IV, Fixed, 95% CI) | Subtotals only | |
| 5.1.1 End of treatment | 1 | 16 | Risk Ratio (IV, Fixed, 95% CI) | 1.56 [0.39, 6.19] |
| 5.2 Reduction in physical signs: monthly seizure frequency Show forest plot | 1 | Mean Difference (IV, Fixed, 95% CI) | Subtotals only | |
| 5.2.1 End of treatment | 1 | 61 | Mean Difference (IV, Fixed, 95% CI) | ‐4.75 [‐18.73, 9.23] |
| 5.2.2 Follow‐up | 1 | 59 | Mean Difference (IV, Fixed, 95% CI) | ‐3.50 [‐12.69, 5.69] |
| 5.3 Reduction in physical sign: seizure freedom Show forest plot | 1 | Risk Ratio (IV, Fixed, 95% CI) | Subtotals only | |
| 5.3.1 End of treatment | 1 | 16 | Risk Ratio (IV, Fixed, 95% CI) | 2.33 [0.30, 17.88] |
| 5.4 Level of functioning Show forest plot | 2 | Std. Mean Difference (IV, Random, 95% CI) | Subtotals only | |
| 5.4.1 End of treatment | 2 | 74 | Std. Mean Difference (IV, Random, 95% CI) | 0.44 [‐1.69, 2.57] |
| 5.4.2 Follow‐up | 1 | 53 | Std. Mean Difference (IV, Random, 95% CI) | ‐0.63 [‐1.18, ‐0.08] |
| 5.5 Quality of life (QOLIE31) Show forest plot | 1 | 16 | Mean Difference (IV, Fixed, 95% CI) | 11.20 [‐7.98, 30.38] |
| 5.5.1 End of treatment | 1 | 16 | Mean Difference (IV, Fixed, 95% CI) | 11.20 [‐7.98, 30.38] |
| 5.6 Mental state – anxiety Show forest plot | 2 | Std. Mean Difference (IV, Random, 95% CI) | Subtotals only | |
| 5.6.1 End of treatment | 2 | 74 | Std. Mean Difference (IV, Random, 95% CI) | ‐0.31 [‐0.78, 0.15] |
| 5.6.2 Follow‐up | 1 | 53 | Std. Mean Difference (IV, Random, 95% CI) | ‐0.31 [‐0.85, 0.23] |
| 5.7 Mental state – depression Show forest plot | 2 | Std. Mean Difference (IV, Random, 95% CI) | Subtotals only | |
| 5.7.1 End of treatment | 2 | 74 | Std. Mean Difference (IV, Random, 95% CI) | ‐0.25 [‐0.71, 0.21] |
| 5.7.2 Follow‐up | 1 | 53 | Std. Mean Difference (IV, Random, 95% CI) | ‐0.32 [‐0.86, 0.23] |
| 5.8 Mental state (SCL‐90) Show forest plot | 1 | 16 | Mean Difference (IV, Fixed, 95% CI) | ‐70.60 [‐121.59, ‐19.61] |
| 5.8.1 End of treatment | 1 | 16 | Mean Difference (IV, Fixed, 95% CI) | ‐70.60 [‐121.59, ‐19.61] |
| 5.9 Dropout rate Show forest plot | 2 | Risk Ratio (IV, Random, 95% CI) | Subtotals only | |
| 5.9.1 End of treatment | 2 | 83 | Risk Ratio (IV, Random, 95% CI) | 0.37 [0.02, 8.01] |
| 5.9.2 Follow‐up | 1 | 64 | Risk Ratio (IV, Random, 95% CI) | 1.41 [0.25, 7.87] |
| 5.10 Use of health services (number of general practitioner consultations) Show forest plot | 1 | 46 | Mean Difference (IV, Fixed, 95% CI) | ‐1.00 [‐3.32, 1.32] |
| 5.10.1 Follow‐up | 1 | 46 | Mean Difference (IV, Fixed, 95% CI) | ‐1.00 [‐3.32, 1.32] |
| Outcome or subgroup title | No. of studies | No. of participants | Statistical method | Effect size |
| 6.1 Reduction in physical signs: seizure frequency (self‐made scale) Show forest plot | 1 | 27 | Mean Difference (IV, Fixed, 95% CI) | ‐0.80 [‐1.44, ‐0.16] |
| 6.1.1 Follow‐up | 1 | 27 | Mean Difference (IV, Fixed, 95% CI) | ‐0.80 [‐1.44, ‐0.16] |
| 6.2 Reduction in physical signs: physical symptom load (SF‐36 – physical) Show forest plot | 1 | 186 | Mean Difference (IV, Fixed, 95% CI) | ‐0.26 [‐3.76, 3.24] |
| 6.2.1 Follow‐up | 1 | 186 | Mean Difference (IV, Fixed, 95% CI) | ‐0.26 [‐3.76, 3.24] |
| 6.3 Level of functioning (WSAS) Show forest plot | 1 | Mean Difference (IV, Fixed, 95% CI) | Subtotals only | |
| 6.3.1 End of treatment | 1 | 43 | Mean Difference (IV, Fixed, 95% CI) | ‐7.12 [‐12.47, ‐1.77] |
| 6.3.2 Follow‐up | 1 | 43 | Mean Difference (IV, Fixed, 95% CI) | ‐6.11 [‐11.67, ‐0.55] |
| 6.4 Quality of life (QOLIE10‐P) Show forest plot | 1 | 27 | Mean Difference (IV, Fixed, 95% CI) | ‐9.30 [‐14.06, ‐4.54] |
| 6.4.1 Follow‐up | 1 | 27 | Mean Difference (IV, Fixed, 95% CI) | ‐9.30 [‐14.06, ‐4.54] |
| 6.5 Mental state – anxiety (HADS) Show forest plot | 1 | 192 | Mean Difference (IV, Fixed, 95% CI) | ‐0.47 [‐1.67, 0.73] |
| 6.5.1 Follow‐up | 1 | 192 | Mean Difference (IV, Fixed, 95% CI) | ‐0.47 [‐1.67, 0.73] |
| 6.6 Mental state – depression Show forest plot | 2 | 219 | Std. Mean Difference (IV, Random, 95% CI) | ‐0.30 [‐1.08, 0.48] |
| 6.6.1 Follow‐up | 2 | 219 | Std. Mean Difference (IV, Random, 95% CI) | ‐0.30 [‐1.08, 0.48] |
| 6.7 Dropout rate Show forest plot | 2 | Risk Ratio (IV, Random, 95% CI) | Subtotals only | |
| 6.7.1 End of treatment | 1 | 64 | Risk Ratio (IV, Random, 95% CI) | 1.76 [0.82, 3.78] |
| 6.7.2 Follow‐up | 2 | 259 | Risk Ratio (IV, Random, 95% CI) | 0.54 [0.00, 70.37] |
| 6.8 Use of health services (number hospital visits) Show forest plot | 1 | Risk Ratio (IV, Fixed, 95% CI) | Subtotals only | |
| 6.8.1 End of treatment | 1 | 64 | Risk Ratio (IV, Fixed, 95% CI) | 0.18 [0.02, 1.43] |
| Outcome or subgroup title | No. of studies | No. of participants | Statistical method | Effect size |
| 7.1 Level of functioning (Functional Independence Measure Motor (FIM)) Show forest plot | 1 | 118 | Mean Difference (IV, Fixed, 95% CI) | 9.20 [6.06, 12.34] |
| 7.1.1 End of treatment | 1 | 118 | Mean Difference (IV, Fixed, 95% CI) | 9.20 [6.06, 12.34] |
| 7.2 Mental state (SF‐12) Show forest plot | 1 | 118 | Mean Difference (IV, Fixed, 95% CI) | 9.10 [4.96, 13.24] |
| 7.2.1 End of treatment | 1 | 118 | Mean Difference (IV, Fixed, 95% CI) | 9.10 [4.96, 13.24] |
| 7.3 Dropout rate Show forest plot | 1 | Risk Ratio (IV, Fixed, 95% CI) | Subtotals only | |
| 7.3.1 End of treatment | 1 | 60 | Risk Ratio (IV, Fixed, 95% CI) | 0.23 [0.03, 1.97] |
| Outcome or subgroup title | No. of studies | No. of participants | Statistical method | Effect size |
| 8.1 Reduction in physical signs: physical symptom load (SF‐36 – Physical Component) Show forest plot | 1 | 57 | Mean Difference (IV, Fixed, 95% CI) | 9.20 [4.00, 14.40] |
| 8.1.1 Follow‐up | 1 | 57 | Mean Difference (IV, Fixed, 95% CI) | 9.20 [4.00, 14.40] |
| 8.2 Level of functioning (WSAS) Show forest plot | 1 | Mean Difference (IV, Fixed, 95% CI) | Subtotals only | |
| 8.2.1 End of treatment | 1 | 54 | Mean Difference (IV, Fixed, 95% CI) | ‐7.10 [‐11.40, ‐2.80] |
| 8.2.2 Follow‐up | 1 | 57 | Mean Difference (IV, Fixed, 95% CI) | ‐6.70 [‐12.07, ‐1.33] |
| 8.3 Mental state – anxiety (HADS) Show forest plot | 1 | Mean Difference (IV, Fixed, 95% CI) | Subtotals only | |
| 8.3.1 End of treatment | 1 | 54 | Mean Difference (IV, Fixed, 95% CI) | ‐0.60 [‐3.05, 1.85] |
| 8.3.2 Follow‐up | 1 | 57 | Mean Difference (IV, Fixed, 95% CI) | ‐1.00 [‐3.71, 1.71] |
| 8.4 Mental state – depression (HADS) Show forest plot | 1 | Mean Difference (IV, Fixed, 95% CI) | Subtotals only | |
| 8.4.1 End of treatment | 1 | 54 | Mean Difference (IV, Fixed, 95% CI) | ‐3.60 [‐7.13, ‐0.07] |
| 8.4.2 Follow‐up | 1 | 57 | Mean Difference (IV, Fixed, 95% CI) | ‐3.20 [‐5.53, ‐0.87] |
| 8.5 Dropout rate Show forest plot | 1 | Risk Ratio (IV, Fixed, 95% CI) | Subtotals only | |
| 8.5.1 End of treatment | 1 | 60 | Risk Ratio (IV, Fixed, 95% CI) | 0.20 [0.02, 1.61] |
| 8.5.2 Follow‐up | 1 | 60 | Risk Ratio (IV, Fixed, 95% CI) | 0.50 [0.05, 5.22] |
| Outcome or subgroup title | No. of studies | No. of participants | Statistical method | Effect size |
| 9.1 Reduction in physical signs: conversions symptoms (SDQ20) Show forest plot | 1 | Risk Ratio (IV, Fixed, 95% CI) | Subtotals only | |
| 9.1.1 End of treatment | 1 | 19 | Risk Ratio (IV, Fixed, 95% CI) | 0.12 [0.01, 2.00] |
| 9.1.2 Follow‐up | 1 | 17 | Risk Ratio (IV, Fixed, 95% CI) | 1.12 [0.08, 15.19] |
| 9.1.3 Follow‐up 10 months | 1 | 15 | Risk Ratio (IV, Fixed, 95% CI) | 0.16 [0.01, 2.66] |
| 9.2 Quality of life (SF‐36) Show forest plot | 1 | Mean Difference (IV, Fixed, 95% CI) | Subtotals only | |
| 9.2.1 End of treatment | 1 | 16 | Mean Difference (IV, Fixed, 95% CI) | ‐6.99 [‐28.09, 14.11] |
| 9.2.2 Follow‐up 4 months | 1 | 16 | Mean Difference (IV, Fixed, 95% CI) | ‐19.53 [‐43.91, 4.85] |
| 9.2.3 Follow‐up 10 months | 1 | 14 | Mean Difference (IV, Fixed, 95% CI) | ‐11.43 [‐36.16, 13.30] |
| 9.3 Mental state – depression (BDI‐II) Show forest plot | 1 | Risk Ratio (IV, Fixed, 95% CI) | Subtotals only | |
| 9.3.1 End of treatment | 1 | 16 | Risk Ratio (IV, Fixed, 95% CI) | 1.29 [0.10, 17.14] |
| 9.3.2 Follow‐up 4 months | 1 | 16 | Risk Ratio (IV, Fixed, 95% CI) | 3.86 [0.50, 29.55] |
| 9.3.3 Follow‐up 10 months | 1 | 14 | Risk Ratio (IV, Fixed, 95% CI) | 1.00 [0.08, 13.02] |
| 9.4 Dropout rate Show forest plot | 1 | Risk Ratio (IV, Fixed, 95% CI) | Subtotals only | |
| 9.4.1 End of treatment | 1 | 23 | Risk Ratio (IV, Fixed, 95% CI) | 1.09 [0.18, 6.48] |
| 9.4.2 Follow‐up 4 months | 1 | 23 | Risk Ratio (IV, Fixed, 95% CI) | 1.09 [0.28, 4.32] |
| 9.4.3 Follow‐up 10 months | 1 | 23 | Risk Ratio (IV, Fixed, 95% CI) | 0.82 [0.23, 2.87] |
| 9.5 Use of health services (emergency department visits) Show forest plot | 1 | 19 | Mean Difference (IV, Fixed, 95% CI) | ‐0.16 [‐1.25, 0.93] |
| 9.5.1 End of treatment | 1 | 19 | Mean Difference (IV, Fixed, 95% CI) | ‐0.16 [‐1.25, 0.93] |
| Outcome or subgroup title | No. of studies | No. of participants | Statistical method | Effect size |
| 10.1 Reduction in physical signs: overall physical impact (Psychogenic Movement Disorder Scale (PMDRS) Show forest plot | 1 | 21 | Mean Difference (IV, Fixed, 95% CI) | 5.60 [‐15.48, 26.68] |
| 10.1.1 End of treatment | 1 | 21 | Mean Difference (IV, Fixed, 95% CI) | 5.60 [‐15.48, 26.68] |
| 10.2 Mental state – anxiety (BAI) Show forest plot | 1 | 21 | Mean Difference (IV, Fixed, 95% CI) | ‐3.40 [‐8.01, 1.21] |
| 10.2.1 End of treatment | 1 | 21 | Mean Difference (IV, Fixed, 95% CI) | ‐3.40 [‐8.01, 1.21] |
| 10.3 Mental state – depression (Hamilton) Show forest plot | 1 | 21 | Mean Difference (IV, Fixed, 95% CI) | ‐0.50 [‐3.32, 2.32] |
| 10.3.1 End of treatment | 1 | 21 | Mean Difference (IV, Fixed, 95% CI) | ‐0.50 [‐3.32, 2.32] |
| 10.4 Dropout rate Show forest plot | 1 | Risk Ratio (IV, Fixed, 95% CI) | Subtotals only | |
| 10.4.1 End of treatment | 1 | 29 | Risk Ratio (IV, Fixed, 95% CI) | 1.87 [0.40, 8.65] |
| Outcome or subgroup title | No. of studies | No. of participants | Statistical method | Effect size |
| 11.1 Reduction in physical signs: symptom freedom Show forest plot | 1 | Risk Ratio (IV, Fixed, 95% CI) | Subtotals only | |
| 11.1.1 End of treatment | 1 | 40 | Risk Ratio (IV, Fixed, 95% CI) | 0.69 [0.39, 1.24] |
| Outcome or subgroup title | No. of studies | No. of participants | Statistical method | Effect size |
| 12.1 Reduction in physical signs: decrease in seizure frequency Show forest plot | 1 | 54 | Mean Difference (IV, Fixed, 95% CI) | 41.40 [4.92, 77.88] |
| 12.1.1 End of treatment | 1 | 54 | Mean Difference (IV, Fixed, 95% CI) | 41.40 [4.92, 77.88] |
| 12.2 Changes in monthly visits Show forest plot | 1 | 54 | Mean Difference (IV, Fixed, 95% CI) | ‐0.21 [‐0.55, 0.13] |
| 12.2.1 End of treatment | 1 | 54 | Mean Difference (IV, Fixed, 95% CI) | ‐0.21 [‐0.55, 0.13] |
| 12.3 Quality of life Show forest plot | 1 | 47 | Mean Difference (IV, Fixed, 95% CI) | 5.40 [0.26, 10.54] |
| 12.3.1 End of treatment | 1 | 47 | Mean Difference (IV, Fixed, 95% CI) | 5.40 [0.26, 10.54] |
| 12.4 Dropout rate Show forest plot | 1 | 60 | Risk Ratio (M‐H, Fixed, 95% CI) | 1.60 [0.29, 8.92] |
| 12.4.1 End of treatment | 1 | 60 | Risk Ratio (M‐H, Fixed, 95% CI) | 1.60 [0.29, 8.92] |
