Effectiveness of psychological interventions for carers of patients with severe mental illness (SMI): a systematic review

Abstract

Objectives

Interventions for carers of patients with severe mental illness (SMI) are effective in improving patient outcomes. This review examined the effectiveness of psychological interventions or support designed to help carers of patients with SMI.

Design

A systematic review of randomised controlled trials (RCTs) was conducted. Study selection, data abstraction and risk of bias assessments were independently conducted and duplicated. The reporting of this review follows the Preferred Reporting Items for Systematic Reviews and Meta-Analyses (PRISMA) guidelines.

Data sources

CINAHL, CENTRAL, EMBASE, Medline and PsycINFO were searched from inception to 30 April 2024.

Eligibility criteria

Only studies using an RCT were considered. All other research designs were excluded. The included population was carers over 18 caring for a patient with SMI. Any psychological, psychosocial or psychoeducational interventions were included. Patients under 18, diagnosed with dementia or learning disabilities comorbidity, were excluded.

Data extraction and synthesis

Study characteristics and outcome data (mean and SD) for each included study were collected. Review Manager (RevMan) was used to manage the data. Meta-analysis was not considered appropriate due to the heterogeneity of the included studies. The results were presented as a narrative synthesis.

Results

This review included 3869 participants from 32 RCTs and demonstrated a beneficial effect of psychological interventions for carers. Psychoeducation was widely used and the most evaluated intervention for carers.

Conclusions

Psychological interventions for carers are beneficial in helping to reduce negative outcomes and enhance positive outcomes. Future research could focus on exploring the impact of interventions on less evaluated outcomes such as depressive and anxiety symptoms.

PROSPERO registration number

CRD42021233181.

STRENGTHS AND LIMITATIONS OF THIS STUDY.

• This is an up-to-date systematic review of various psychological interventions offered to the carers of patients with severe mental illness.

• Eight different outcomes were assessed to capture the wider impact on carers.

• There are evaluations of short-term, medium-term and long-term benefits of interventions.

• There were challenges in completing the meta-analysis due to data heterogeneity.

Introduction

An estimated 1.5 million people care for someone with mental ill health in the UK.¹ The importance of carers in supporting and caring for people with severe mental illness (SMI) has long been recognised. For example, the cost of schizophrenia and psychosis to society in England was estimated to be nearly £11.8 billion, but families who were carers save the public purse £1.24 billion per year. However, they perceived they did not receive any support for their own needs.²

The ‘Triangle of Care’ emphasised the importance of collaborative working practices among patients, professionals, and carers. The importance of the involvement of carers in the treatment of psychosis was highlighted by the National Institute for Health and Care Excellence guidance for psychosis and schizophrenia,³ which recommended offering family therapy/intervention as part of the treatment.

Research has shown that support from carers plays a crucial role in improving the outcomes of patients with SMI.^{4 5} Family interventions for people with SMI can potentially decrease the probability of relapse and improve medication adherence.⁶

Many studies and reviews about SMI focus on patient outcomes, such as relapse rates or symptom reduction. However, the impact of interventions for carers also needed to be reviewed. This was essential as carers provided most of the support and contact for people with mental health difficulties outside of a hospital setting. Identifying the effects of interventions on carers’ outcomes was essential to determine which interventions could be effectively used in clinical practice. It was likely that when carers were supported, they could offer better support to mentally unwell patients, positively impacting patient outcomes and reducing the overall burden on the healthcare system.

In 2015, a systematic review was conducted to examine the effectiveness of different psychological interventions on the caregiving experience of adults with SMI.⁷ The review also looked at secondary outcomes, such as the quality of life of carers, satisfaction with services and psychological distress. The literature search for their review ended in 2013.

This systematic review provided an up-to-date evaluation of the effectiveness of various psychological interventions offered to the carers of patients with SMI. Eight different outcomes (burden, satisfaction, stigma, coping, psychological morbidity, depressive symptoms, anxiety symptoms and well-being) were used to gain a more comprehensive understanding of the impact of psychological interventions on carers. In this systematic review, results have been analysed at multiple time points (1–36 months) to evaluate the short-term, medium-term and long-term effects of these interventions on carers of patients with SMI.

This review aimed to examine the effectiveness of any psychological or psychoeducational intervention or support designed to help carers of patients with SMI. It was envisaged that the findings of this systematic review would help identify areas for future development to improve the quality of support available for carers.

Method

Registration details

The review protocol was registered on the PROSPERO database on 16 April 2021, available at https://www.crd.york.ac.uk/prospero/display_record.php?RecordID=233181.

Search strategy and selection criteria

A systematic search for all relevant literature was conducted using the Preferred Reporting Items for Systematic Reviews and Meta-Analyses (PRISMA)⁸ search strategy of the databases, including CINAHL, EMBASE, Medline, PsycINFO and CENTRAL, from inception to 30 April 2024. All studies were peer-reviewed and limited to English only. Reference lists of all included studies were searched for additional studies.

Combinations of keywords and index terms were used to describe the population, condition and intervention: (intervention OR support OR program* OR psychoeducation) AND (care* OR famil* OR relative*) AND (mental and illness OR psychiatric and illness OR psychiatric and patient OR schizophreni* OR depression OR bipolar) AND (stigma OR psychological and morbidity OR depression OR anxiety OR distress OR well being or wellbeing OR caregiver or reward OR coping OR satisfaction OR burden) AND (randomised OR randomised OR controlled OR RCT). The details of each search string can be viewed in online supplemental material appendix 1. Only studies using a randomised controlled trial (RCT) were considered. The investigators needed to be explicit about the type of randomisation used to allocate participants; all other research designs were excluded (see online supplemental appendix material 2).

Titles and abstracts of all publications identified in the initial search were screened by three reviewers (SA, JT and HS) to identify studies that met the inclusion criteria. Two reviewers (SA, JT, HS, DM or CK) independently screened the full text of each paper, and any disagreements were resolved through a third reviewer (DM or CK). Authors were contacted to help clarify any uncertainties regarding the conduct of their study or the data presented in their paper.

Risk of bias

The Cochrane Risks of Bias tool⁹ was used to assess six domains: selection bias (random sequence of generation and allocation concealment), performance bias (blinding of participants and personnel), detection bias (blinding of outcome assessment), attrition bias (incomplete outcome data) and reporting bias (selective outcome reporting).

All reviewers (SA, JT, HS, DM and CK) independently assessed studies, and a team meeting was held to agree on an overall risk of bias rating. The risk of bias was rated as ‘high risk’ (seriously weakens confidence in the results), ‘low risk’ (unlikely to alter the results) or ‘unclear risk’. Studies with four or more ‘high-risk’ markers were excluded from the review. The risk of bias table for all included studies has been provided as ‘online supplemental appendix material 3’.

Data items

Participants: The population of interest was carers aged over 18 who care for a patient with SMI. A carer was defined as a family member, relative, partner or friend. Diagnostic and Statistical Manual of Mental Disorders Version 4 Text Revision (DSM-IV-TR), International Classification of Diseases Version 10 (ICD-10) or clinical assessment determined the patient’s diagnosis. Patients aged under 18, with a diagnosis of dementia or any learning disabilities comorbidity, were excluded.

Interventions: These were any psychological, psychosocial or psychoeducational carer intervention. Interventions could be delivered with or without the patient(s) as participants. Interventions were categorised into five groups defined below:

1. Psychoeducation (PE): Psychoeducational interventions provided information, educational materials or feedback/advice.¹⁰ The intervention could be delivered in primary or secondary care settings or within family homes. They could also be delivered through the post, email, face-to-face lectures or through information published online. The number of sessions could vary and were usually time-limited.

2. Family focused support (FFS): This intervention used family/peer support to give advice and support. The sessions were unstructured, with the discussions examining shared experiences and giving and receiving help founded on key principles of respect, shared responsibility and mutual agreement on what was helpful. Underpinning this was understanding another’s situation empathically through the shared experience of emotional and psychological pain.¹¹

3. Professional-led support (PLS): PLS groups commonly used content based on research evidence.¹² They were usually unstructured and could be either time-limited or ongoing, with carers coming to the sessions for the period they needed. The groups were designed to help families get information about mental health, medication, self-care, mutual assurance, communication, problem-solving and advocacy.

4. Skills development/therapy (SDT): This category included studies that focused on developing new skills and new ways of thinking with difficulties arising from caring for a family member with mental health difficulties (either directly or indirectly). As such, the interventions were structured and concentrated on specific difficulties identified by the families and/or professionals. Two main types of skills development/training were identified,¹³ as described below:

   1. Interventions included developing coping skills aimed to increase family members’ abilities to deal with acute symptoms, stress or persistent symptoms. These interventions incorporated a broad array of illness management strategies, including relapse prevention, stress management, coping strategies, goal-setting and problem-solving.

   2. Cognitive–behavioural approaches included teaching coping skills and modifying dysfunctional beliefs and behaviours about the illness, the family member with mental health problems, the self and the environment.

      In addition, any family therapy approach was included. The main principle of family therapy was that all members of a system (family) were influential in maintaining an equilibrium that allowed the family to function. Family therapy saw the family group as the unit for study and treatment. Therapists saw the problems of individuals in the context of the environment, especially the family. The therapy concentrated on the interpersonal processes rather than individual problems within the family.¹⁴

5. Multimodal approach (MMA): MMA intervention involved using multiple modes (ie, information sharing, skills training and formal support) and their intermodal relationships contributed to meaning-making.¹⁵ In the one study using this approach, two modes were used:

   1. A 13-section toolkit with flexible use depending on the individual needs of the relative. The sections included information and advice regarding what is psychosis?; managing positive symptoms; dealing with crises; managing stress—thinking differently; treatment options; the future and a resource directory.

   2. Each participant was supported in using the toolkit by an early intervention service support worker. Support was offered by email or telephone (relatives’ preference) for a maximum of 1 hour per week over 6 months. Support was targeted at helping relatives identify the key difficulties they faced and guiding them to find the most relevant information and strategies in the toolkit.

Comparator: The comparator was treatment as usual (TAU). Any pharmacological or non-psychological treatment intervention was excluded. No comparisons were made between different psychological interventions.

Outcome: Eight carer outcomes were included in this review: burden, satisfaction, stigma, coping, psychological morbidity, depressive symptoms, anxiety symptoms and well-being. Postintervention outcomes were recorded at six time points: 1 month, 3 months, 6 months, 12 months, 24 months and 36 months. For the purpose of this paper, up to 6 months was considered a short-term effect, over 6 months and up to 12 months was medium-term effect, and over 12 months was a long-term effect.

Study types: RCTs for psychological interventions for one or more carers of a patient diagnosed with SMI.

Settings: The patient could be an inpatient or outpatient.

Data extraction

Study characteristics and outcome data (mean and SD) for each included study were collected by three reviewers (SA, JT and HS) independently using the data extraction tool developed for this study (see online supplemental appendix material 4 and online supplemental appendix material 5). Disagreements were resolved through discussions with team members (DM and CK).

The named outcome measures screened for carers’ mental well-being inclusive of depressive and anxiety symptoms. Included studies used validated self-reported outcome measures (eg, General Health Questionnaire-28), Family Burden Interview Schedule (FBIS), Depression, Anxiety and Stress Scale-21), Brief Symptom Inventory-18). ‘Online supplemental appendix material 5’ contains outcome measures and data collection tools for all included studies.

Review Manager (RevMan V.5.4 for Mac)¹⁶ software was used to manage the data.

Data synthesis

Meta-analysis was initially planned but was not considered appropriate due to the heterogeneity of the included studies. The reasons for heterogeneity were due to the differences in the patient populations (inpatient and outpatient services), diagnostic groups, design of interventions, and use of outcome measures. Consequently, the results are presented as a narrative synthesis.

Patient and public involvement

Patients and the public were not involved in this research.

Results

The five databases yielded a total of 10 504 hits. The flow diagram is shown in figure 1.

Figure 1. PRISMA flow diagram.⁸ PRISMA, Preferred Reporting Items for Systematic Reviews and Meta-Analyses; RCT, randomised controlled trial; TAU, treatment as usual.

A total of 91 potentially relevant articles were selected for further examination against the inclusion criteria. Of these, 59 articles did not meet the inclusion criteria due to; not being an RCT, the randomisation process not explicit, the patient population not having a psychiatric diagnosis (ie, diagnosis of dementia), data on carers was not available, outcomes of interest not reported, dual diagnosis of learning disability and carers of patients under the age of 18. The remaining 32 studies were included in this review.

The studies involved 3869 carers with a median sample size of 120 (ranging from 32 to 800).

Five studies included carers from inpatient units, 24 from outpatients and 2 studies included from both settings. 24 studies included carers of patients diagnosed with schizophrenia or schizoaffective disorder, 2 studies defined SMI as a psychotic disorder and 3 studies included patients with bipolar disorder. Diagnoses of undefined mental illness, schizophrenia and mood disorder, recent-onset psychosis and moderate depression were each used in one study.

Interventions

Table 1 provides information about the overall findings of the interventions (see online supplemental appendix material 6 for more in-depth finding breakdown).

Table 1. Overall findings of the systematic review.

Intervention	Number of studies reporting intervention (N=32)	Favourable results out of total number of results (percentage)	Statistically significant favourable results (percentage)
Psychoeducation	21	63/67 (94.0%)	37/63 (58.7%)
Family-focused support	7	24/28 (85.7%)	9/24 (37.5%)
Profession-led support	2	3/8 (37.5%)	1/3 (33.3%)
Skills development/therapy	6	15/15 (100%)	11/15 (73%)
Multimodal approach	2	10/11 (91%)	5/10 (50%)

Psychoeducation

This systematic review showed that PE was an effective psychological intervention for the carers of patients with SMI. It was the most offered and evaluated intervention (21 studies). A total of 2454 carers were recruited. 17 studies were conducted in outpatient services,417,32 3 in inpatient services33,35 and 1 study in both settings.³⁶

21 studies used this approach. All 8 outcomes of interest were reported: burden (17 studies), satisfaction (6 studies), stigma (3 studies), coping (2 studies), psychological morbidity (2 studies), depressive symptoms (1 study), anxiety symptoms (2 studies) and well-being (2 studies).

In total, there were 67 results. 63 results (94.0%) favoured the intervention group, with 37 (58.7%) statistically significant results.

Burden: 17 studies analysed the effect of PE on burden, which effectively reduced burden in the short, medium and long term. Burden was the most recorded outcome (33 results), with at least one result reported at each of the 6 time points. PE reduced burden in 31 results (91%), with 21 (68%) statistically significant results. At 6 months, one study³¹ recorded a ‘no effect’ result, and one study²⁵ showed results favouring the control group. Neither results were statistically significant.

Satisfaction: PE effectively improved satisfaction in the short, medium and long term. Six studies analysed the effectiveness of PE on satisfaction.^{17 18 20 25 27 34} 12 results were reported with time points ranging from 1 month to 24 months. Nine out of 12 (75%) results favoured the intervention, with 3 findings statistically significant. Two results showed no effect. These were at 1 month¹⁸ (1 out of 3) and 6 months²⁵ (1 out of 5).

Stigma: PE was effective in reducing stigma in the short and medium term. Three studies analysed the effectiveness of PE on stigma.^{28 31 37},^{28 31} Six results were recorded (one at 1 month, two at 3 months, two at 6 months and one at 12 months) and favoured the intervention group. All the results were statistically significant.

Coping: PE effectively improved coping in the short and medium term. Two studies reported four results for coping.^{20 28} The effectiveness was evaluated at 1 month, 6 months and 12 months time points. Findings showed that all results favoured the intervention group. One result found at 6 months was statistically significant.²⁸

Psychological morbidity: PE was effective in reducing psychological morbidity in the short, medium and long term. Psychological morbidity was measured in two PE interventions,^{24 28} and three results were reported. Three findings were reported at 6 months, 12 months and 24 months. All findings favoured the intervention group, with the finding at 6 months statistically significant.

Depressive symptoms: PE was effective in improving depressive symptoms in the short and medium term. Depressive symptoms were measured in one study²⁶ with two results (6 and 12 months). Both results favoured the intervention group, with one finding at 12 months which was statistically significant.

Anxiety symptoms: PE was effective in improving anxiety symptoms in the short and medium term. Two studies measured the effectiveness of PE on anxiety symptoms.^{26 31} Four results were reported at 3, 6 and 12 months, and all results favoured the intervention. Of these, three were statistically significant at 3 months and two studies at 6 months.

Well-being: PE was effective in improving well-being in the medium term. Two studies examined well-being after receiving PE.^{26 31} In total, three results were reported: two results at 6 months and one at 12 months. All findings favoured the intervention group, of which two findings were statically significant (76%) at 6 months and 12 months.²⁶

Family-focused support

Seven studies evaluated the effectiveness of FFS on all outcomes except well-being. A total of 800 carers were included. Five studies were conducted in outpatient services18,2038 39; one study was based on inpatient services⁴⁰ and one study was conducted in both settings.³⁶

Overall, there were a total of 28 results. 24 results (85.7%) favoured the intervention group, with nine (37.5%) statistically significant results.

Four studies^{19 20 39 40} assessed effectiveness on burden, two studies for satisfaction,^{18 20} one study on stigma,³⁴ three studies looked at coping,^{20 38 39} two studies on psychological morbidity,^{38 39} one study each for depressive symptoms³⁹ and anxiety symptoms.³⁹

Burden: FFS effectively reduced burden in the short, medium and long term. Eight results were recorded at various time points ranging from 1 month to 36 months. Six out of eight results favoured the FFS group, and 50% of the results were statistically significant. One result each at 3 months³⁹ and 1 month⁴⁰ favoured the control group, though neither was statistically significant.

Satisfaction: FFS was effective in improving satisfaction in the short, medium and long term. Two studies^{18 20} analysed the effectiveness of FFS on satisfaction. Six results were reported with time points ranging from 1 month to 24 months. All results (100%) were in favour of the intervention group (FFS) in improving satisfaction, with two findings (at 12 months and 24 months) statistically significant.

Stigma: FFS was effective in reducing stigma in the short term. One study³⁶ evaluated stigma. There was one follow-up at 1 month, and this finding was statistically significant and favoured the intervention group.

Coping: FFS was effective in improving coping in the short and medium term. Three studies^{20 38 39} evaluated coping. Seven results were reported at time points ranging from 1 month to 12 months. Six results favoured the intervention group. No results were statistically significant. One study²⁰ at 1 month showed no effect (1 out of 2 results).

Psychological morbidity: FFS was effective in reducing psychological morbidity in the short term. Two studies^{38 39} evaluated psychological morbidity. A total of four results were reported from 1 month to 6 months. All findings favoured the intervention group, with two statistically significant.

Depressive and anxiety symptoms: FFS was effective in improving depressive and anxiety symptoms in the short term. One study³⁹ evaluated depressive and anxiety symptoms outcomes. The data were collected at 3 months. Both results favoured the intervention but were not statistically significant.

Professional-led support

Two studies used PLS to evaluate four outcomes: burden, satisfaction, stigma and psychological morbidity.^{41 42} A total of 113 carers were recruited to the studies and cared for inpatients or outpatients. Eight results were reported, with three results (37.5%) in favour of the intervention group, with one result (33.3%) being statistically significant.

Burden: PLS was effective in reducing burden in the long term but not effective in the medium term. One study⁴¹ evaluated burden at two time points (12 months and 24 months). The result at 12 months was in favour of TAU although not statistically significant, whereas the result at 24 months was in favour of the intervention group and was statistically significant.

Satisfaction: PLS was effective in improving satisfaction in the short term. One study⁴² evaluated satisfaction at two time points (3 months and 6 months). Both results showed improvement in satisfaction in the intervention group, although neither result was statistically significant.

Stigma: PLS was not effective in reducing stigma in the short term. One study⁴² evaluated stigma at 3 months and 6 months. Both results favoured TAU but were not statistically significant.

Psychological morbidity: PLS was not found to be effective in reducing psychological morbidity in the short term. One study⁴² evaluated psychological morbidity. Two results were reported at 3 months and 6 months. Both results favoured TAU, with one statistically significant result at 3 months.

Skills development/therapy

Six studies2324 43,46 evaluated the effectiveness of SDT on all eight outcomes. A total of 419 carers were recruited. The patients were all from an outpatient service, except one study recruited from an inpatient service. Data were collected at all time points.

There were 15 results; all (100%) favoured the intervention group, with 11 (73%) statistically significant results.

Three studies^{23 24 44} evaluated burden, with one study each for satisfaction,⁴⁶ stigma,⁴⁶ coping,⁴⁴ psychological morbidity,²⁴ depressive symptoms,⁴³ anxiety symptoms⁴³ and well-being.⁴⁵

Burden: SDT was effective in reducing burden in the short, medium and long term. A total of six results on burden were recorded at all time points (1 month to 36 months). All results favoured the intervention group (reduction in burden), with four (67%) statistically significant.

Satisfaction and stigma: SDT effectively improved satisfaction and reduced stigma in the short term. One study⁴⁶ evaluated satisfaction and stigma at 3 months. These results favoured the intervention group and were statistically significant.

Coping: SDT was effective in improving coping in the short term. One study⁴⁴ evaluated SDT on coping. A total of two results measured at 1 month and 6 months were in favour of the intervention group and were statistically significant.

Psychological morbidity: SDT was effective in reducing psychological morbidity in the medium and long term. One study²⁴ evaluated SDT on psychological morbidity. A total of two results were reported at 12 months and 24 months. Both results were in favour of the intervention group. The result at 24 months was statistically significant.

Depressive and anxiety symptoms: SDT was effective in improving depressive and anxiety symptoms in the short term. One study⁴³ evaluated SDT on depressive and anxiety symptoms at 1 month. Both were in favour of the intervention group and were statistically significant.

Well-being: SDT was effective in improving well-being in the medium term. One study⁴⁵ evaluated well-being at 12 months. The result favoured the intervention group, though it was not statistically significant.

Multimodal approach

Two studies^{47 48} used this approach to evaluate five outcomes: satisfaction, coping, psychological morbidity, anxiety symptoms and well-being. A total of 903 carers participated. The outcomes were recorded at the 3 months and 6 months time points.

There were 11 results, with 10 results (91%) favouring the intervention group. Five results (50%) were statistically significant, indicating that MMA effectively improved ’carers’ outcomes.

Satisfaction: MMA was effective in improving satisfaction in the short term. One study⁴⁸ evaluated satisfaction at 3 months and 6 months. Both results favoured the intervention group and were statistically significant.

Coping: MMA was effective in improving coping in the short term. One study⁴⁷ evaluated coping at 6 months. This result was in favour of the intervention group and was statistically significant.

Psychological morbidity: MMA was effective in reducing psychological morbidity in the short term. Two studies^{47 48} evaluated psychological morbidity at 3 months and 6 months. Three results were reported. All results favoured the intervention group but were not statistically significant.

Anxiety symptoms: MMA was effective in improving anxiety symptoms in the short term. One study⁴⁸ evaluated anxiety symptoms at both 3 months and 6 months time points. There were two results reported. Both results favoured the intervention group, and the 6-month result was statistically significant.

Well-being: MMA was effective in improving well-being in the short term. Both studies^{47 48} evaluated well-being at 3 months and 6 months. Three results were in favour of the intervention group, and the result at 6 months was statistically significant.

Table 2 provides the information on the overall results of the included interventions on the outcomes measures.

Table 2. Overall effect on outcomes.

Outcomes	Favourable/total(n/N—statistically significant)	Not favourable/total(n/N—statistically significant)
Burden	44/48(29/44)	4/48(0/4)
Satisfaction	21/23(8/21)	2/23(0/2)
Stigma	8/10(8/8)	2/7(0/2)
Coping	10/15(0/10)	5/15(3/5)
Psychological morbidity	11/15(5/11)	4/15(1/4)
Depressive symptoms	4/4(4/4)	0/4
Anxiety symptoms	8/8(5/8)	0/8
Well-being	7/9(3/7)	2/9(0/2)

Note: n=number of significant interventions, N=total of interventions.

Discussion

Carer involvement has been shown to reduce morbidity and mortality of patients with SMI.^{49 50} It has been observed that improving family coping skills could help families provide support to individuals with mental illness and improve patient outcomes.¹³

This systematic review aimed to evaluate the effectiveness of psychological interventions for the carers of patients with SMI. The reviewers followed the PICOS guidance and principles to conduct this systematic review, and the risk of bias was used to determine the validity of the included articles. The overall findings of this review indicated that psychological interventions benefit carers.¹³

This review showed that the most commonly used intervention offered to carers of patients with SMI was PE, which was evaluated in 21 out of 32 studies. 91% of the results favoured the intervention group. Similar findings have been concluded in published scientific research: carers of patients with SMI experienced high levels of burden, emotional distress and lack of social support and providing support and education to carers from mental health professionals improved their ability to manage the patient’s symptoms and improve their quality of life.⁵¹ The advantages of PE include that it can be offered in both inpatient and outpatient settings. The delivery of PE was flexible and did not require a specifically trained professional group, making it a valuable and cost-effective intervention that could be offered to the carers at any stage of the ’patient’s care pathway.

The second most widely used intervention offered to carers was FFS. According to the findings from this review, FFS was 100% effective in improving satisfaction and reducing stigma, psychological morbidity and anxiety symptoms. Family members or carers provide emotional and social support to patients and act as advocates for them. Providing focused support to carers was likely to increase their involvement in the care and treatment of patients and improve their understanding of their loved one’s mental health condition, thereby helping to improve their satisfaction and reduce stigma. The importance of involving carers in the care of patients with SMI has been highlighted, and it was concluded that PE programmes and family therapy improved patient outcomes.⁵²

There were mixed findings on the effectiveness of PLS as an intervention for the carers of patients with SMI. Only two included studies in this review evaluated the effectiveness of PLS. It has been recognised that mental health professionals need to provide carers with practical and emotional support and involve them in the treatment process. This systematic review showed that PLS provided carers with support and validation, making them feel heard and involved in the care of their loved ones, thus improving their satisfaction.

The effectiveness of SDT as an intervention showed 100% results in favour of the intervention group. The results indicated that SDT effectively improves carers’ outcomes: burden, coping, psychological morbidity and well-being. It was noteworthy that due to the limited number of studies, the generalisation of these findings could be limited.

MMA intervention provided flexibility in terms of availability and cost-effectiveness. The intervention was offered online and could be accessed by carers at a convenient time, without the direct involvement of a facilitator at a specific time. There were advantages in offering this intervention; however, it was worth noting that only two studies were developed and conducted by the same lead researcher. Therefore, there may be limitations in generalising the results of this intervention.

This systematic review showed that a wide variety of interventions have been offered to the carers of patients with SMI. Five categories based on the underlying themes were used to group these interventions (table 1).

In general, the support provided to carers was effective in improving the outcome measures. The studies included in this systematic review used a range of standardised approaches such as manual, protocol, toolkit, package, booklet, programme and manualised sessions. Using a standardised approach made it possible to ensure the intervention was delivered consistently, which helped to maintain its validity and ensure adherence.

The results of this systematic review were strongly in favour of the effectiveness of PE in reducing burden on the carers of patients with SMI. The findings supported the use of PE for the carers, highlighting the flexibility of this intervention and its cost-effectiveness.

PE was an approach that provided helpful information to carers to assist them in caring for patients with SMI. Psychoeducational interventions for carers effectively improved patient compliance, reduced relapse rates³ and reduced carers’ global morbidities and burden.⁵²

PE was found to be effective in improving the outcomes of carers. By investing in and delivering PE, mental health services could reduce carers’ burden and improve the experience of carers, which was likely to have positive outcomes for the patients. It could also help reduce the financial burden on the healthcare system.

According to this review (as shown in table 2), burden was the most frequently evaluated outcome. This was supported by structured assessment tools, such as the Family Burden Interview Schedule (FBIS)⁵³ and Burden Assessment Schedule (BAS),⁵⁴ developed for this purpose. This review found that some outcomes that could be important for carers, such as depressive and anxiety symptoms, were not extensively evaluated. Depressive and anxiety symptoms are common mental health issues that affect many people, with around 17% of adults in England having a common mental disorder.⁵⁵ It was possible that previous research has not focused on the effectiveness of psychological interventions for depressive and anxiety symptoms in carers, as these conditions could also be treated with medication. Alternatively, it could be that the treatment focus was primarily on patients, and carers were expected to seek support and treatment for their mental health problems.

The COVID-19 pandemic has had a devastating effect on a global scale, leading to severe implications for healthcare services worldwide. The use of digital platforms within healthcare systems has improved during the pandemic. A future-oriented approach to organisation, planning and development of healthcare services might consider modernisation and adaption through digital technologies.

Strengths

This review comprehensively searched the current evidence base for interventions for carers of patients with SMI. It included various carers’ outcomes and examined older and newer interventions, providing a platform for future research.

Limitations

A limited number of studies have evaluated several interventions, making it challenging to generalise findings.

Conclusion

In conclusion, this systematic review found various psychological interventions were offered to the carers of patients with SMI. Among the available interventions, PE emerged as the most researched intervention for carers. A significantly higher number of results from the included studies favoured the intervention group over TAU. Based on the strength of the available evidence from this systematic review, it could be concluded that the results for PE meet the generalisability threshold.

Despite including more studies and a more extended period for inclusion of published literature, this systematic review found that several interventions for carers of patients with SMI were evaluated by a single study. Similar findings were also noted in a systematic review and meta-analysis.⁷

This review showed that, despite their high prevalence in the general population, depressive symptoms, anxiety symptoms and psychological morbidity were relatively less evaluated outcomes. This provided a basis for future research in these areas.

Data availability statement

All data relevant to the study are included in the article or uploaded as online supplemental information.