Hostname: page-component-54dcc4c588-nx7b4 Total loading time: 0 Render date: 2025-09-24T13:47:56.648Z Has data issue: false hasContentIssue false
  • English
  • Français

Umbrella review of psychosocial and ward-based interventions to reduce self-harm and suicide risks in in-patient mental health settings

Published online by Cambridge University Press:  08 September 2025

Leah Quinlivan Open the ORCID record for Leah Quinlivan [Opens in a new window] ,
Jodie Westhead Open the ORCID record for Jodie Westhead [Opens in a new window] ,
Jane Graney ,
Fanyi Su ,
Sarah Steeg Open the ORCID record for Sarah Steeg [Opens in a new window] ,
Emma Nielsen ,
Eloise Curtis ,
Ellie Wildbore ,
Faraz Mughal Open the ORCID record for Faraz Mughal [Opens in a new window]  and
Rachel Elliott
...Show all authors
Leah Quinlivan*
Affiliation:
Division of Psychology and Mental Health, University of Manchester, UK Manchester Academic Health Science Centre, University of Manchester, UK National Institute for Health and Care Research (NIHR GM PSRC) Greater Manchester Patient Safety Research Collaboration, University of Manchester, UK
Jodie Westhead
Affiliation:
Division of Psychology and Mental Health, University of Manchester, UK Manchester Academic Health Science Centre, University of Manchester, UK National Institute for Health and Care Research (NIHR GM PSRC) Greater Manchester Patient Safety Research Collaboration, University of Manchester, UK
Jane Graney
Affiliation:
Division of Psychology and Mental Health, University of Manchester, UK Manchester Academic Health Science Centre, University of Manchester, UK National Institute for Health and Care Research (NIHR GM PSRC) Greater Manchester Patient Safety Research Collaboration, University of Manchester, UK
Fanyi Su
Affiliation:
Manchester Centre for Health Economics, University of Manchester, UK
Sarah Steeg
Affiliation:
Division of Psychology and Mental Health, University of Manchester, UK Manchester Academic Health Science Centre, University of Manchester, UK National Institute for Health and Care Research (NIHR GM PSRC) Greater Manchester Patient Safety Research Collaboration, University of Manchester, UK
Emma Nielsen
Affiliation:
National Institute for Health and Care Research (NIHR GM PSRC) Greater Manchester Patient Safety Research Collaboration, University of Manchester, UK
Eloise Curtis
Affiliation:
National Institute for Health and Care Research (NIHR GM PSRC) Greater Manchester Patient Safety Research Collaboration, University of Manchester, UK
Ellie Wildbore
Affiliation:
National Institute for Health and Care Research (NIHR GM PSRC) Greater Manchester Patient Safety Research Collaboration, University of Manchester, UK
Faraz Mughal
Affiliation:
National Institute for Health and Care Research (NIHR GM PSRC) Greater Manchester Patient Safety Research Collaboration, University of Manchester, UK School of Medicine, Keele University, UK College of Medicine and Health, University of Birmingham, UK
Rachel Elliott
Affiliation:
School of Medicine, Keele University, UK
Roger T. Webb
Affiliation:
Division of Psychology and Mental Health, University of Manchester, UK Manchester Academic Health Science Centre, University of Manchester, UK National Institute for Health and Care Research (NIHR GM PSRC) Greater Manchester Patient Safety Research Collaboration, University of Manchester, UK
Nav Kapur
Affiliation:
Division of Psychology and Mental Health, University of Manchester, UK Manchester Academic Health Science Centre, University of Manchester, UK National Institute for Health and Care Research (NIHR GM PSRC) Greater Manchester Patient Safety Research Collaboration, University of Manchester, UK Mersey Care NHS Foundation Trust, Manchester, UK
*
Correspondence: Leah Quinlivan. Email: leah.quinlivan@manchester.ac.uk
Rights & Permissions [Opens in a new window]

Abstract

Background

Understanding what psychosocial interventions can reduce self-harm and suicide within in-patient mental health settings can be challenging, due to clinical demands and the large volume of published reviews.

Aims

To summarise evidence from systematic reviews on psychosocial and ward-level interventions (excluding environmental modifications) for self-harm and suicide that may enhance patient safety in in-patient mental health settings.

Method

We systematically searched Medline, Embase, CINAHL, PsycINFO and CDSR (2013–2023) for systematic reviews on self-harm and suicide prevention interventions that included in-patient data. Review quality was assessed using AMSTAR-2, primary study overlap via an evidence matrix, and evidence strength evaluated (GRADE algorithm). Findings were narratively synthesised, with input from experts-by-experience throughout (PROSPERO ID: CRD42023442639).

Results

Thirteen systematic reviews (seven meta-analyses, six narrative), comprising over 160 000 participants, were identified. Based on quantitative reviews, cognitive–behavioural therapy reduces repeat self-harm by follow-up, and dialectical behaviour therapy decreases the frequency of self-harm. Narrative review evidence suggested that post-discharge follow-up, as well as system and ward-based interventions (e.g. staff training) may reduce suicide and/or self-harm. However, review quality varied, patient involvement was lacking and methodological quality of trials informing reviews was predominately low. Overlap was slight (covered area 12.4%).

Conclusions

The effectiveness of interventions to prevent self-harm and suicide in in-patient settings remains uncertain due to variable quality reviews, evidence gaps, poor methodological quality of primary studies and a lack of pragmatic trials and co-production. There is an urgent need for better, co-designed research within in-patient mental health settings.

Keywords

Self-harmpsychosocial interventionssuicidetherapeutic interventionsumbrella review

Information

Type
Review
Information
BJPsych Open , Volume 11 , Issue 5 , September 2025 , e196
Check for updates
Creative Commons
Creative Common License - CCCreative Common License - BY
This is an Open Access article, distributed under the terms of the Creative Commons Attribution licence (https://creativecommons.org/licenses/by/4.0/), which permits unrestricted re-use, distribution and reproduction, provided the original article is properly cited.
Copyright
© The Author(s), 2025. Published by Cambridge University Press on behalf of Royal College of Psychiatrists

Suicide and self-harm are global public health priorities, with an estimated 700 000 lives lost to suicide and at least 14 million episodes of self-harm occurring annually.1,Reference Moran, Chandler, Dudgeon, Kirtley, Knipe and Pirkis2 In the UK, over 69 000 individuals died by suicide between 2011 and 2021.3 Approximately one-third of these individuals were mental health patients, and 6% of these deaths occurred during in-patient care for patients aged 16 years and older.3,Reference Hunt, Baird, Turnbull, Ibrahim, Shaw and Appleby4 Patients discharged from hospital have an elevated suicide risk.Reference Hunt, Baird, Turnbull, Ibrahim, Shaw and Appleby4,Reference Musgrove, Carr, Kapur, Chew-Graham, Mughal and Ashcroft5 Around 12% of patient deaths occur within three months post-discharge, and risk of suicide is particularly high within the first two weeks.3–Reference Musgrove, Carr, Kapur, Chew-Graham, Mughal and Ashcroft5 Self-harm, defined as self-injury or self-poisoning irrespective of suicidal intent,6 is strongly associated with suicide and is therefore a major patient safety concern in mental health services.Reference Quinlivan, Littlewood, Webb and Kapur7 Approximately 20% of individuals presenting to hospital following an episode of self-harm are admitted to in-patient mental health services worldwide.Reference Witt, McGill, Leckning, Hill, Davies and Robinson8 In the UK, 76% of mental health in-patients who died by suicide had a history of self-harm.Reference Hunt, Baird, Turnbull, Ibrahim, Shaw and Appleby4 Timely, evidence-based and compassionately delivered interventions are essential to prevent self-harm repetition and suicide.Reference Awenat, Peters, Gooding, Pratt, Shaw-Núñez and Harris9

Mental health in-patient suicide rates have reduced since 2009, but progress in England has stagnated since 2016.Reference Hunt, Baird, Turnbull, Ibrahim, Shaw and Appleby4 Structural changes, including enhanced ward safety and ligature point removal, may have contributed to improved patient safetyReference Hunt, Baird, Turnbull, Ibrahim, Shaw and Appleby4 and evidence beyond physical safety measures may further reduce suicide rates in this setting. Psychosocial interventions that involve structured, non-pharmacological treatments with a psychological or social focus, can mitigate self-harm and suicide risk,Reference Mann, Michel and Auerbach10,Reference Hawton, Witt, Salisbury, Arensman, Gunnell and Hazell11 but access to therapeutic therapies remains limited.Reference Staniszewska, Mockford, Chadburn, Fenton, Bhui and Larkin12,Reference Evlat, Wood and Glover13 Evaluating the suitability and effectiveness of self-harm and suicide prevention interventions for in-patient care is further complicated by clinical demands, information overload and limited evidence translation into practice.Reference Choi and Kang14,Reference Boudreaux, Larkin, Vallejo Sefair, Ma, Li and Ibrahim15 Umbrella reviews provide an opportunity to synthesise broad systematic review evidence to guide clinical practice.Reference Fusar-Poli and Radua16

Our objective was to summarise the evidence from systematic reviews on psychosocial and ward-level interventions for preventing self-harm and suicide that may enhance patient safety for adults (aged 16 years and older) in in-patient mental health settings. We sought to evaluate the quality and relevance of this systematic review evidence for mental health in-patient settings to inform practice.

Method

This study is reported according to the Preferred Reported Items for Overviews of Reviews (PRIOR) guideline,Reference Gates, Gates, Pieper, Fernandes, Tricco and Moher17 and was registered on the International Prospective Register of Systematic Reviews (PROSPERO) (registration number: CRD42023442639; 5.06.23). Minor changes included utilising the Adapted Algorithm for GRADE,Reference Pollock, Farmer, Brady, Langhorne, Mead and Mehrholz18 in place of Guyatt et alReference Guyatt, Oxman, Akl, Kunz, Vist and Brozek19 to evaluate the methodological quality and certainty of the evidence.

Inclusion and exclusion criteria

Reviews were eligible for inclusion if they were: (a) peer reviewed systematic reviews; (b) they included at least one primary study that evaluated psychosocial interventions based in mental health in-patient settings (wards or post-discharge services); (c) reported data for adults aged 18 or over, or composite results for adults and adolescents. Our included outcomes were consistent with clinical guidelines for self-harm,6 and included any self-harm, or self-injury irrespective of suicidal intent (See Table 1). Suicidal ideation, while important, was excluded due to our focus on behaviour.Reference Mann, Michel and Auerbach10,Reference Kapur, Cooper, O’Connor and Hawton20 We did not have restrictions on study designs, comparators, or psychiatric diagnoses, but we prioritised reviews which focused on interventions which went beyond the purely environmental such as the removal of ligature points. Exclusion criteria included: (a) suicidal ideation (as the composite or main outcome); (b) studies that only reported data for children and adolescents; (c) studies based in prisons and other custodial criminal justice institutions, as well as educational, community or voluntary settings; (d) theoretical and opinion-based reviews, letters, commentaries, non-systematic reviews and reviews of qualitative research; and (e) review articles that were not translated to English.

Table 1 PICOa (population, intervention, comparison and outcomes) criteria

a. PICO categories were based on clinical guidelines for self-harm.6

Search strategy and selection criteria

Search strategy details (e.g. terms, reasons for exclusion and additional references) are in the Supplementary materials available at https://doi.org/10.1192/bjo.2025.10811. We searched Embase, PsycInfo, MEDLINE, CINAHL and the Cochrane Database of Systematic Reviews (CDSR) from January 2013 to December 2023, using broad search term strings to capture reviews in this area. Search terms were developed with a specialist librarian and content experts, and adapted from the clinical guidelines and Cochrane reviews of interventions for self-harm.6,Reference Witt, Hetrick, Rajaram, Hazell, Salisbury and Townsend21 We used forward and backward citation chaining to supplement database searches. Two reviewers (L.Q., J.W.) independently reviewed all the titles, abstracts and full texts of potentially eligible studies, F.M. checked a random 10%, and J.G. cross-checked 100% of the data extracted. Disagreements were resolved via consensus and discussion with senior authors (R.T.W., N.K., R.E.), the wider team (F.M., J.W., F.S.) and our PPIE group (MS4MH-R) members.

Data extraction

Data extraction was performed in duplicate (L.Q., J.W.) using a standardised form, verified (J.G., F.M.) and reviewed by a multidisciplinary team. We extracted study characteristics, including author details, population, author-defined interventions, comparison, and outcome (PICO), methods, evaluation of bias, heterogeneity, GRADE assessment and results.

Quality assessment

Two researchers (L.Q., J.W.) independently evaluated the methodological quality of systematic reviews using the Assessment of Multiple Systematic Reviews Tool (AMSTAR-2).Reference Shea, Reeves, Wells, Thuku, Hamel and Moran22 Each of the 16 items is evaluated as either positive (yes), negative (no), or partial positive. Based on seven ‘critical’ and nine ‘non-critical’ domains, we classified reviews into ‘high’, ‘moderate’, ‘low’ and ‘critically low’ quality. Using stringent criteria, reviews with a partial yes, or that did not report data for the AMSTAR 2Reference Shea, Reeves, Wells, Thuku, Hamel and Moran22 item were classified as ‘No’ for domain classifications. In accordance with AMSTAR-2 guidance, reviews that listed justifications for excluded references in a summary format, rather than as a list of individual citations, were marked as negative. As part of evidence evaluation, we tabulated and synthesised data for the adapted GRADE algorithmReference Pollock, Farmer, Brady, Langhorne, Mead and Mehrholz18 and other key methodological data. For the adapted GRADE algorithm, quantitative reviews received downgrades based on the assessment of methodological quality, via heterogeneity, number of participants, risk of bias and items from AMSTAR.Reference Pollock, Farmer, Brady, Langhorne, Mead and Mehrholz18,Reference Shea, Reeves, Wells, Thuku, Hamel and Moran22 Reviews were rated as providing a high level of evidence if they received zero downgrades, moderate if one or two downgrades, low if three or four downgrades and very low if the review received five or six downgrades.Reference Pollock, Farmer, Brady, Langhorne, Mead and Mehrholz18 We also report any use of GRADE evaluationsReference Guyatt, Oxman, Akl, Kunz, Vist and Brozek19 in reviews. We based our conclusions on the combined evidence from quantitative and narrative reviews.

Overlap analysis

Overlap analysis was conducted to examine the degree to which the same primary studies were included in multiple systematic reviews. We estimated the degree of the primary studies, pairwise, overlap in the reviews (fraction of evidence synthesised in two or more reviews) via the covered area (formula: N/rc) and corrected covered area (CCA; formula: (N – r)/((r × c) – r)) using the open-access Graphical Representation of Overlap for OVErviews open-access tool.Reference Bracchiglione, Meza, Bangdiwala, Guzmán, Urrutia and Bonfill23,Reference Pieper, Antoine, Mathes, Neugebauer and Eikermann24 N is the total number of publications, r is the total number of rows (unique primary studies) and cca is the total number of columns (the number of included reviews). The degrees of overlap observed were categorised as slight (CCA 0–5%), moderate (CCA 6–10%), high (CCA 11–15%) and very high overlap (CCA > 15%).Reference Bracchiglione, Meza, Bangdiwala, Guzmán, Urrutia and Bonfill23,Reference Pieper, Antoine, Mathes, Neugebauer and Eikermann24 One study was removed from overlap analyses due to insufficient information for the extraction of primary studies’ references, which may result in an underestimation of the true degree of overlap.Reference Fox, Huang, Guzmán, Funsch, Cha and Ribeiro25

Synthesis methods

The purpose of this umbrella review was to descriptively synthesise the systematic review evidence for interventions that may be helpful in preventing self-harm and/or suicide for in-patient mental health settings. Given the high degree of heterogeneity across the set of reviews, we used a systematic approach to narratively describe and synthesise the data in tables and groups.Reference Popay, Roberts, Sowden, Petticrew, Arai and Rodgers26 We grouped systematic review results into those with quantitative (meta-analysis) and narrative synthesis analysis. We summarised the findings from the two groups separately and reported the results and outcomes as described by the review. We reported detailed results for interventions with evidence of efficacy, but also reported contradictory findings and summarised those without. For quantitative reviews, we reported important parameters for significant results including pooled odds ratios, mean differences, 95% CI and means and s.d.s where available. We reported the I 2 statistic as a measure of between-study variation and heterogeneity.Reference Deeks, Higgins, Altman, Higgins, Thomas, Chandler, Cumpston, Li and Page27 The outcomes were classified into self-harm, attempted suicide and suicide, based on the information available in the publication. Interventions are reported as defined by the systematic reviews (Table 2). We provide detailed information on reviews, including interventions, control groups, study designs and outcomes in Table 4, and additional detailed results are in the Supplementary materials 2, Tables 1 and 3.

Results

Fig. 1 summarises the results of the search, which yielded 1116 studies, of which, 1041 were excluded at the title and abstract screening stage. We screened 74 for full text eligibility and identified 23 additional studies through manual searches. In total, after full text screening and stratification, 31 reviews included relevant data for healthcare settings, and 13 met our inclusion criteria for mental health in-patient settings (see Fig. 1 for flow chart).

Fig. 1 Flow diagram of included studies. CDSR, Cochrane Database of Systematic Reviews.

Evaluation of bias in reviews

Five studies used the Cochrane Risk of Bias Tool,Reference Witt, Hetrick, Rajaram, Hazell, Salisbury and Townsend21,Reference Hawton, Witt, Salisbury, Arensman, Gunnell and Hazell28–Reference Yiu, Rowe and Wood31 four used another method of evaluating bias (e.g. the Joanna Briggs Institute appraisal tool),Reference Fox, Huang, Guzmán, Funsch, Cha and Ribeiro25,Reference Nawaz, Reen, Bloodworth, Maughan and Vincent32–Reference Wand, Browne, Jessop and Peisah35 and three reviews did not report informationReference Mann, Michel and Auerbach10,Reference DeCou, Comtois and Landes36,Reference Luxton, June and Comtois37 (see Table 2). The predominant concerns for bias included participant attrition and incomplete outcome data. Bias due to the absence of blinding was common but is challenging given the nature of psychosocial interventions.

Assessment of the reviews’ methodological quality and certainty of the evidence

AMSTAR-2 assessments for the included reviews are presented in Table 2. The methodological quality varied widely across reviews: two reviews were evaluated as providing a ‘high quality’Reference Witt, Hetrick, Rajaram, Hazell, Salisbury and Townsend21,Reference Hawton, Witt, Salisbury, Arensman, Gunnell and Hazell28 summary of the results and available data, with the remainder rated as ‘critically low’ quality due to more than one critical flaw (e.g. lack of pre-registration and reference list with justifications for each excluded study) (Table 2). According to the adapted GRADE algorithm,Reference Pollock, Farmer, Brady, Langhorne, Mead and Mehrholz18 two reviewsReference Witt, Hetrick, Rajaram, Hazell, Salisbury and Townsend21,Reference Hawton, Witt, Salisbury, Arensman, Gunnell and Hazell28 provided a moderate level of evidence. The risk of bias for included trials in all the reviews resulted in ‘downgrades’ for quality, which reduced the certainty of evidence. For example, Witt et alReference Witt, Hetrick, Rajaram, Hazell, Salisbury and Townsend21 conducted a high-quality systematic review, but the primary trials were reported as ‘low quality’ for cognitive-behaviour therapy, moderate certainty for emotion-regulation therapy and ‘very low quality’ for dialectical-behaviour therapy versus treatment as usual. Additional information that contributed to the methodological evaluation of reviews is presented in Table 3.

Table 2 AMSTAR-2Reference Shea, Reeves, Wells, Thuku, Hamel and Moran22 ratings across the 13 systematic reviews

NA, not applicable; PICO, patient, intervention, comparison, outcome.

a. Critical flaws AMSTAR ratings:Reference Shea, Reeves, Wells, Thuku, Hamel and Moran22 judgements are made on an evaluation of critical and non-critical weaknesses. High: zero or one non-critical weakness; moderate: more than one non-critical weakness; low: one critical flaw without non-critical weaknesses; critically low: more than one critical flaw, with or without non-critical weaknesses.

Table 3 Quality evaluation for included reviews (Adapted Algorithm for GRADE,Reference Pollock, Farmer, Brady, Langhorne, Mead and Mehrholz18 AMSTAR-2,Reference Shea, Reeves, Wells, Thuku, Hamel and Moran22 risk of bias, heterogeneity, generalisability)

Adapted GRADE scoring: reviews are rated as providing a high level of evidence if they have received zero downgrades, moderate if one or two downgrades, low if three or four downgrades and very low if the review receives five or six downgrades (not applicable for narrative reviews).

CT, controlled trials; CCT, controlled cohort studies; CBT, cognitive behaviour therapy; CAT, cognitive analytic therapy; CPDpd, Cognitive behaviour therapy adapted for people who have received a personality disorder diagnosis; CROB, Cochrane Risk of Bias tool; EUC, enhanced usual care; MBT, mentalisation-based therapy; DBT, dialectical behaviour therapy; RCT, randomised controlled trials; ITS, interrupted times series designs; Obs., observational study design; PP, pre–post; QED, quasi-experimental designs; RoB, risk of bias; TAU, treatment as usual.; CROB, Cochrane Risk of Bias tool.

Characteristics of included studies

Characteristics of the included reviews (e.g. including interventions, outcomes, comparators, bias, heterogeneity) are presented in Tables 3 and 4 and are summarised here. Additional information is presented in the Supplementary material. In total, 13 reviews evaluated interventions that included in-patient mental health settings. The systematic reviews were published between 2013 and 2022. The 949 primary studies were published between 1970 and 2021, inclusive, with an approximate total of over 160 000 participants studied in the reported primary research. Overall, most primary studies were conducted in Western Europe and North America. Five reviews reported data for adults,Reference Witt, Hetrick, Rajaram, Hazell, Salisbury and Townsend21,Reference Hawton, Witt, Salisbury, Arensman, Gunnell and Hazell28,Reference Sobanski, Josfeld, Peikert and Wagner30,Reference Yiu, Rowe and Wood31,Reference Wand, Browne, Jessop and Peisah35 six reported composite data for adults and adolescentsReference Mann, Michel and Auerbach10,Reference Fox, Huang, Guzmán, Funsch, Cha and Ribeiro25,Reference Hou, Wang, Guo, Zhang, Liu and Qi29,Reference Nawaz, Reen, Bloodworth, Maughan and Vincent32,Reference Donker, Calear, Grant, Van Spijker, Fenton and Hehir34,Reference DeCou, Comtois and Landes36 and two reviews had insufficient reporting for age.Reference Rozek, Baker, Rugo, Steigerwald, Sippel and Holliday33,Reference Luxton, June and Comtois37 Wand et alReference Wand, Browne, Jessop and Peisah35 evaluated interventions for older adults. The percentage of female participants in the included primary studies ranged from 6 to 98%, and 6/13 reviews had insufficient reporting for gender. Seven reviews conducted meta-analyses,Reference Witt, Hetrick, Rajaram, Hazell, Salisbury and Townsend21,Reference Fox, Huang, Guzmán, Funsch, Cha and Ribeiro25,Reference Hawton, Witt, Salisbury, Arensman, Gunnell and Hazell28–Reference Yiu, Rowe and Wood31,Reference DeCou, Comtois and Landes36 and six narratively summarised the data.Reference Nawaz, Reen, Bloodworth, Maughan and Vincent32–Reference Wand, Browne, Jessop and Peisah35,Reference Luxton, June and Comtois37 Most reviews evaluated cross-setting interventions that included in-patient settings and two reviews specifically focused on in-patient settings.Reference Nawaz, Reen, Bloodworth, Maughan and Vincent32,Reference Rozek, Baker, Rugo, Steigerwald, Sippel and Holliday33 Six reviews evaluated post-discharge interventions as part of their overall review,Reference Mann, Michel and Auerbach10,Reference Witt, Hetrick, Rajaram, Hazell, Salisbury and Townsend21,Reference Hou, Wang, Guo, Zhang, Liu and Qi29,Reference Sobanski, Josfeld, Peikert and Wagner30,Reference Wand, Browne, Jessop and Peisah35,Reference DeCou, Comtois and Landes36,Reference Kapur, Gorman, Quinlivan and Webb38 six evaluated ward-based interventionsReference Witt, Hetrick, Rajaram, Hazell, Salisbury and Townsend21,Reference Sobanski, Josfeld, Peikert and Wagner30,Reference Rozek, Baker, Rugo, Steigerwald, Sippel and Holliday33,Reference Kapur, Gorman, Quinlivan and Webb38 and three had insufficient reporting as regards the timing of the intervention.Reference Fox, Huang, Guzmán, Funsch, Cha and Ribeiro25,Reference Donker, Calear, Grant, Van Spijker, Fenton and Hehir34,Reference DeCou, Comtois and Landes36

Table 4 Study characteristics for included reviews (e.g. included study dates, age, gender intervention details, controls, follow-up, designs, outcomes)

CT, controlled trials; CBT, cognitive behaviour therapy; EUC, enhanced usual care; MBT, mentalisation-based therapy; DBT, dialectical behaviour therapy; RCT, randomised controlled trials; ITS, interrupted times series designs; Obs, observational study design; PP, pre–post; QED, quasi-experimental designs; TAU, treatment as usual; NSSI, non-suicidal self-injury; PD, patients diagnosed with personality disorder; CAMS, collaborative assessment and management of suicidality.

Quantitative reviews

Cognitive–behavioural therapy (CBT)

In an updated Cochrane review of psychosocial interventionsns for self-harm,Reference Hawton, Witt, Salisbury, Arensman, Gunnell and Hazell11,Reference Hawton, Witt, Salisbury, Arensman, Gunnell and Hazell28 Witt et alReference Witt, Hetrick, Rajaram, Hazell, Salisbury and Townsend21 found that CBT reduced self-harm repetition compared with treatment as usual by post-intervention (odds ratio 0.35, 95% CI [0.12 to 1.02]; N = 238; k = 4; I 2 = 0%), 6-month (odds ratio 0.52, 95% CI [0.38 to 0.70]; N = 1260; k = 12; I 2 = 2%) and 12-month follow up (odds ratio 0.81, 95% CI [0.66 to 0.99]; N = 2458; k = 9; I 2 = 0%). The evidence suggested that CBT reduced the frequency of self-harm repetition at six- (mean difference −0.71, 95% CI [−1.32 to −0.11]; N = 118; k = 4; I 2 = 0%) and 12-month follow-up (mean difference 1.18, s.d. = 4.22, n = 40 versus mean difference 4.58; s.d. 8.37; n = 33; mean difference –3.40, 95% CI [−6.54 to −0.26]; N = 73; k = 1; I 2 not applicable), but not the post-intervention assessment. Using the GRADE criteria,Reference Guyatt, Oxman, Akl, Kunz, Vist and Brozek19 Witt et alReference Witt, Hetrick, Rajaram, Hazell, Salisbury and Townsend21 rated the quality of evidence as ‘low certainty’.

Sobanski et alReference Sobanski, Josfeld, Peikert and Wagner30 found that pooled interventions for patients who attempted suicide resulted in significantly fewer episodes of suicidal behaviour compared with controls (relative risk 0.66; 95% CI [0.48, 0.90]; Z = 2.63, p < 0.01; odds ratio 0.56, 95% CI [0.36–0.84], p < 0.01, k = 18, I 2 = 51%). In separate intervention analyses, Sobanksi et alReference Sobanski, Josfeld, Peikert and Wagner30 found significant treatment effects for CBT compared with treatment as usual (relative risk 0.66; 95% CI [0.48–0.90]; Z = 2.61, p = 0.009; odds ratio 0.53, 95% CI [0.34–0.83]; k = 10, p = 0.005, I 2 = 28%) and psychodynamic interventions (mentalisation-based therapy (MBT), brief psychodynamic interpersonal therapy) in reducing suicide re-attempts frequencies (relative risk 0.21; 95% CI [0.08–0.57]; Z = 3.08, p = 0.002; odds ratio 0.17, 95% CI [0.06–0.45]; k = 2, p < 0.0004, I 2 = 30%). However, treatment effects for CBT were only significant for longer follow-up for (>12 months) (relative risk 0.60, Z = 2.38, p = 0.02). However, in a meta-analysis that evaluated pooled interventions (e.g. post-admission CBT, dialectical beahviour therapy (DBT), insight-oriented therapy, gratitude diaries) based in psychiatric in-patient settings, Yiu et alReference Yiu, Rowe and Wood31 found no significant differences between treatment and control conditions for suicide attempts (relative risk 0.92; 95% CI [0.41–2.06]; Z = 0.18, k = 10, p = 0.86, I 2 = 0%).

Dialectical behaviour therapy (DBT)

Evidence from three meta-analyses indicated that DBT may be effective in reducing the frequency of self-harm.Reference Witt, Hetrick, Rajaram, Hazell, Salisbury and Townsend21,Reference Fox, Huang, Guzmán, Funsch, Cha and Ribeiro25,Reference Hawton, Witt, Salisbury, Arensman, Gunnell and Hazell28 Witt et alReference Witt, Hetrick, Rajaram, Hazell, Salisbury and Townsend21 found evidence of beneficial treatment effects for DBT in reducing the frequency of repeated self-harm by post-intervention follow-up (mean difference −5.00, 95% CI [−8.92 to −1.08]; N = 659; k = 7; I 2 = 49%). Using a composite outcome that included suicide attempts, non-suicidal self-injury, self-harm and suicide, DeCou et alReference DeCou, Comtois and Landes36 found positive treatment effects for DBT in reducing ‘self-directed violence’ compared with controls (weighted mean effect size, d = −0.324, 95% CI [−0.471 to −0.176], k = 15, I 2 = 45.48%).Reference DeCou, Comtois and Landes36

Other interventions (MBT, emotion-based regulation psychotherapy, social support interventions)

Based on one trial rated as providing high certainty evidence, Witt et alReference Witt, Hetrick, Rajaram, Hazell, Salisbury and Townsend21 found evidence to suggest that mentalisation-based therapy may reduce self-harm repetition (18/71 versus 31/63; odds ratio 0.35, 95% CI [0.17 to 0.73]; N = 134; k = 1; I 2 = not applicable) and the frequency of repetition (mean difference 0.38, s.d. = 0.38, n = 71 versus mean 1.66, s.d. = 2.87, N = 63; mean difference -1.28, 95% CI [−2.01 to −0.55]; N = 134; k = 1; I 2 = not applicable). Based on moderate certainty evidence (GRADE)Reference Guyatt, Oxman, Akl, Kunz, Vist and Brozek19, Witt et alReference Witt, Hetrick, Rajaram, Hazell, Salisbury and Townsend21 found positive treatment effects from group-based emotional regulation psychotherapy for reducing repeat self-harm, but not for the frequency of repetition (odds ratio 0.34, 95% CI [0.13 to 0.88]; N = 83; k = 2; I 2 = 0%).

Hou et alReference Hou, Wang, Guo, Zhang, Liu and Qi29 found evidence to suggest that social support interventions, defined as having at least one intervention component that promoted social support/connectedness, or decreased social isolation/feelings of loneliness reduced deaths by suicide (pooled relative risk 0.48, 95% CI [0.27 to 0.85], k = 10, p = 0.01). Social support interventions had greater benefit for reducing suicide, when delivered face-to-face, for people who had attempted suicide, but not for other delivery methods or populations (relative risk 0.24, 95% CI [0.10 to 0.58]).Reference Hou, Wang, Guo, Zhang, Liu and Qi29

Narrative reviews

CBT and DBT

Consistent with the meta-analyses,Reference Witt, Hetrick, Rajaram, Hazell, Salisbury and Townsend21,Reference Hawton, Witt, Salisbury, Arensman, Gunnell and Hazell28 the results from two narrative reviews indicated beneficial effects for CBT in reducing suicide attempts.Reference Mann, Michel and Auerbach10,Reference Rozek, Baker, Rugo, Steigerwald, Sippel and Holliday33 Rozek et alReference Rozek, Baker, Rugo, Steigerwald, Sippel and Holliday33 evaluated psychotherapies to address co-occurring suicidal thoughts and behaviours and post-traumatic stress disorder (PTSD). Suicide-specific treatments in this review,Reference Rozek, Baker, Rugo, Steigerwald, Sippel and Holliday33 significantly reduced suicidal behaviour outcomes and PTSD symptoms. Three narrative reviews suggested that DBT was beneficial in reducing self-harm and suicide attempts.Reference Mann, Michel and Auerbach10,Reference Nawaz, Reen, Bloodworth, Maughan and Vincent32,Reference Rozek, Baker, Rugo, Steigerwald, Sippel and Holliday33 In their evaluation of interventions to reduce self-harm on in-patient wards, Nawaz et alReference Nawaz, Reen, Bloodworth, Maughan and Vincent32 found that DBT was the most frequently implemented and effective intervention to reduce self-harm. Additional supportive evidence was ascertained for the systems training emotional predictability and problem-solving therapy (STEPPs) intervention in reducing hospital admissions for self-harm for patients diagnosed with borderline personality disorder.Reference Nawaz, Reen, Bloodworth, Maughan and Vincent32

Post-discharge follow-up contacts

Luxton et alReference Luxton, June and Comtois37 indicated that repeated follow-up contact (postcards/ telephone contact) for patients discharged from hospital may reduce repeat suicide attempts (3 studies) and suicide (2 studies). Other studies in this review reported inconclusive results or did not demonstrate any preventative effects. In their review of evidence-based interventions for suicide prevention, Mann et alReference Mann, Michel and Auerbach10 also suggested that post-discharge follow-up contact (e.g. brief contact, enhanced follow-up, caring texts) reduced suicidal behaviour. Wand et alReference Wand, Browne, Jessop and Peisah35 suggested a comprehensive aftercare programme for older adults may be beneficial in reducing suicide. However, the strength of evidence was poor, with significant methodological limitations, heterogeneity and small absolute risk reductions.Reference Wand, Browne, Jessop and Peisah35

System, staff training or ward-level interventions

Mann et alReference Mann, Michel and Auerbach10 suggested system-level changes may be effective in reducing suicide via evidence from two UK studies evaluating the implementation of evidence-based recommendations (e.g. improved depression management, low staff turnover, continuity of care) in mental health services. Nawaz et alReference Nawaz, Reen, Bloodworth, Maughan and Vincent32 found evidence to suggest that mixed interventions that combine therapeutic and ward-based approaches significantly reduced self-harm (2 studies). Ward-based interventions to prevent self-harm were inconclusive, with three studies showing reductions in self-harm, and three that did not. Staff training that included the provision of additional nurses on two acute wards, assistance with implementation of changes according to a model of conflict and containment (one study), and problem-solving training (one study), significantly reduced self-harm on in-patient wards. Nawaz et alReference Nawaz, Reen, Bloodworth, Maughan and Vincent32 found evidence to suggest that combinations of a therapeutic approach and ward-based changes also reduced self-harm (2 studies).Reference Nawaz, Reen, Bloodworth, Maughan and Vincent32 However, most of this evidence was based on weaker pre–post designs with small sample sizes, with complex poorly defined interventions.Reference Nawaz, Reen, Bloodworth, Maughan and Vincent32

Non-significant findings for interventions

Several reviews evaluated interventions, but did not find statistically significant treatment effects for reducing self-harm, attempted suicide, or suicide.Reference Witt, Hetrick, Rajaram, Hazell, Salisbury and Townsend21,Reference Fox, Huang, Guzmán, Funsch, Cha and Ribeiro25,Reference Hou, Wang, Guo, Zhang, Liu and Qi29–Reference Donker, Calear, Grant, Van Spijker, Fenton and Hehir34 These interventions are listed in Table 5.

Table 5 Non-significant treatment effects for interventions in included systematic reviews

Review overlap: fraction of evidence synthesised in two or more reviews

Figure 2 presents the Graphical Representation of Overlap for OVErviews (GROOVE) heat map for the primary study overlap analysis.Reference Bracchiglione, Meza, Bangdiwala, Guzmán, Urrutia and Bonfill23 The Covered Area for the reviews was 12.4% and the Corrected Covered Area was 4.4%, indicating a slight degree of overlap for the overall review.Reference Bracchiglione, Meza, Bangdiwala, Guzmán, Urrutia and Bonfill23 Some pairs of reviews had ‘very high’, or ‘moderate’ overlap. For example, Sobanksi et alReference Sobanski, Josfeld, Peikert and Wagner30 had very high overlap with Hawton et alReference Hawton, Witt, Salisbury, Arensman, Gunnell and Hazell28 and Luxton et alReference Luxton, June and Comtois37 had high overlap with Hou et alReference Hou, Wang, Guo, Zhang, Liu and Qi29 As expected, Witt et alReference Witt, Hetrick, Rajaram, Hazell, Salisbury and Townsend21 had high overlap with Hawton et alReference Hawton, Witt, Salisbury, Arensman, Gunnell and Hazell28 given this was an updated review of the work. Yiu et alReference Yiu, Rowe and Wood31 had moderate overlap with Nawaz et alReference Nawaz, Reen, Bloodworth, Maughan and Vincent32 Rozek et alReference Rozek, Baker, Rugo, Steigerwald, Sippel and Holliday33 and Sobanski et alReference Sobanski, Josfeld, Peikert and Wagner30

Fig. 2 Heat map for primary study overlap analysis for citations for self-harm, suicide and suicidal behaviour.

Discussion

We synthesised evidence from 13 systematic reviews assessing the efficacy and/or the effectiveness of self-harm and suicide prevention interventions. Our aim was to evaluate the quality and relevance of this evidence for reducing the likelihood of self-harm and/or suicide in mental health in-patient settings. Overall, most quantitative and narrative reviews suggested support for CBT and DBT in reducing self-harm and suicide attempts, but evidence for reducing suicide deaths was limited. Narrative reviews highlighted promising interventions, including post-discharge follow-up, implementing evidence-based recommendations,Reference Mann, Michel and Auerbach10 ward-based changes (e.g. additional nurses, increased access to therapeutic activities) and staff training as part of broader interventions.Reference Nawaz, Reen, Bloodworth, Maughan and Vincent32 However, their real-world applicability and effectiveness in mental health in-patient settings is less clear due to differences in patient populations, high heterogeneity in the synthesis of interventions and a lack of pragmatic trials and co-production.

CBT demonstrated efficacy in reducing repeat self-harm and suicide attempts, particularly with longer follow-up.Reference Mann, Michel and Auerbach10,Reference Witt, Hetrick, Rajaram, Hazell, Salisbury and Townsend21,Reference Sobanski, Josfeld, Peikert and Wagner30 DBT was associated with a reduction in the frequency of repeat self-harm, but not in reducing the likelihood of repetition.Reference Mann, Michel and Auerbach10,Reference Witt, Hetrick, Rajaram, Hazell, Salisbury and Townsend21,Reference Fox, Huang, Guzmán, Funsch, Cha and Ribeiro25 However, these findings should be interpreted cautiously, as only a small subset of primary studies included in these reviews originated from in-patient settings. For example, Witt et alReference Witt, Hetrick, Rajaram, Hazell, Salisbury and Townsend21 included only six trials from in-patient settings. CBT may also be more suitable to community settings, whereas DBT could be more effective in reducing self-harm on in-patient wards.Reference Nawaz, Reen, Bloodworth, Maughan and Vincent32 However, the evidence base for DBT includes relatively weak trials with specific populations.Reference Witt, Hetrick, Rajaram, Hazell, Salisbury and Townsend21 We know little of the effectiveness for adapted interventions as they are used in mental health in-patient settings.Reference Nawaz, Reen, Bloodworth, Maughan and Vincent32

Other reviews had conflicting results on the effectiveness of CBT for reducing suicide attempts.Reference Fox, Huang, Guzmán, Funsch, Cha and Ribeiro25,Reference Yiu, Rowe and Wood31,Reference Donker, Calear, Grant, Van Spijker, Fenton and Hehir34 Yiu et alReference Yiu, Rowe and Wood31 found no statistically significant evidence supporting psychosocial interventions compared with treatment as usual in reducing suicide attempts for mental health in-patients, and no studies evaluating self-harm as an outcome. Lack of evidence for effectiveness and conflicting findings may be due to the review design (e.g. heterogeneous intervention types, duration, follow-up and comparator treatments). The review evidence also highlights the importance of tailoring interventions to clinical need. For example, one systematic review found no significant effect of CBT in preventing suicide attempts among patients diagnosed with schizophrenia spectrum disorders and psychosis.Reference Donker, Calear, Grant, Van Spijker, Fenton and Hehir34 Evidence from another review suggested that brief CBT for suicide prevention may be effective in reducing co-occurring trauma symptoms and suicide attempts.Reference Rozek, Baker, Rugo, Steigerwald, Sippel and Holliday33

Quality of the evidence

Methodological limitations across both primary studies and systematic reviews emphasise the need for improved quality research and reporting. Only two reviews met ratings for ‘high quality’ according to the AMSTAR-2Reference Shea, Reeves, Wells, Thuku, Hamel and Moran22 criteria.Reference Witt, Hetrick, Rajaram, Hazell, Salisbury and Townsend21,Reference Hawton, Witt, Salisbury, Arensman, Gunnell and Hazell28 Only two reviews, led by the same team, evaluated the certainty of the evidence for clinical practice.Reference Witt, Hetrick, Rajaram, Hazell, Salisbury and Townsend21,Reference Hawton, Witt, Salisbury, Arensman, Gunnell and Hazell28 Witt et alReference Witt, Hetrick, Rajaram, Hazell, Salisbury and Townsend21 rated evidence quality as low to moderate for CBT-based psychotherapy, moderate for emotion-regulation therapy and very low quality for DBT versus treatment as usual.

Strengths and limitations

Clinical demands, and the large volume of publications on self-harm and suicide prevention interventions, may reduce the likelihood of evidence translation into mental health in-patient settings. Our evidence synthesis of 13 systematic reviews highlights important evidence gaps that lay the foundation for future research. We provide detailed evidence evaluations and summaries to support knowledge mobilisation and translation to high-demand clinical practice. We did not conduct any meta-analyses, due to the high degree of heterogeneity among the included reviews and the potential for misleading conclusions. Restricting our analyses to meta-analyses of controlled trials, may have resulted in the exclusion of potentially promising system-level interventions. However, we provide important statistical parameters, detailed results and methodological details in our results. We focused on self-harm and suicide because they are key patient safety outcomes.Reference Kapur, Gorman, Quinlivan and Webb38 Other outcomes, including quality of life, functioning and mental health symptoms are also important,Reference Owens, Fox, Redwood, Davies, Foote and Salisbury39 but were beyond the scope of this umbrella review.

Although we used a published search strategy6 and broad approach, we may have missed some published reviews, including those published in countries where English is not widely used. We utilised robust methodological assessments, including AMSTAR-2Reference Shea, Reeves, Wells, Thuku, Hamel and Moran22 and the adapted GRADE algorithm,Reference Pollock, Farmer, Brady, Langhorne, Mead and Mehrholz18 and provide a transparent detailed evaluation of the evidence base to inform clinical practice. However, umbrella reviews are subject to multiple sources of bias, variable reporting and heterogeneity in primary studies and reviews. Included studies varied greatly in methodological robustness, and ranged from observational designs to randomised controlled trials, with many subject to bias (e.g. attrition, reporting). Nearly half of the reviews lacked sufficient gender-related data, while other reviews typically reported binary data (male/female). Reporting for other protected characteristics, such as ethnicity, physical disabilities, neurodivergence, as well as socio-economic position, was largely absent, possibly due to deficiencies in the primary study reporting.

The reporting quality for age varied across reviews, which is a limitation of the evidence base. Several reviews reported composite data combining children/adolescents with adults,Reference Fox, Huang, Guzmán, Funsch, Cha and Ribeiro25,Reference Hou, Wang, Guo, Zhang, Liu and Qi29,Reference Donker, Calear, Grant, Van Spijker, Fenton and Hehir34,Reference DeCou, Comtois and Landes36 while others reported age inadequately.Reference Rozek, Baker, Rugo, Steigerwald, Sippel and Holliday33,Reference DeCou, Comtois and Landes36 Descriptions of ward settings were inadequately reported in most systematic reviews, limiting the ability to assess the effectiveness of interventions in specific in-patient care settings (e.g. intensive versus acute). To optimise the evaluation of interventions in mental health care, future primary research and systematic reviews should provide granular and detailed information on ward type and in-patient setting (e.g. acute wards, psychiatric intensive care units, forensic wards). While we excluded research based in custodial criminal justice settings, one review(Reference Nawaz, Reen, Bloodworth, Maughan and Vincent32) reported interventions that were evaluated in forensic wards. We do not know if these wards were included in other reviews due to poor reporting. These systematic limitations in the evidence base highlight key biases that should inform future research.

There was no patient and public involvement and engagement reported in reviews or primary studies, which is a substantial limitation. However, we integrated lived experience perspectives throughout this umbrella review process, ensuring experiential evaluation and relevance to real-world clinical practice. Our research team consisted of people with lived experience and a diverse, multidisciplinary group of health services researchers, clinicians and methodologists, which enriched our evidence synthesis and our interpretation of it. We excluded reviews of qualitative research, which is a limitation. However, our aim was to summarise the effectiveness of interventions, based on systematic review evidence. We have completed a lived experience commentary alongside this review, to enrich our summary of quantitative reviews.

Comparisons with other research

Our findings are consistent with previous research emphasising the need for better quality intervention trials.Reference Witt, Hetrick, Rajaram, Hazell, Salisbury and Townsend21 Our synthesis supports conclusions that CBT-based interventions have the strongest evidence base for reducing repeat self-harm, while DBT may be more effective for decreasing the frequency of self-harm repetition.Reference Awenat, Peters, Gooding, Pratt, Shaw-Núñez and Harris9,Reference Knipe, Padmanathan, Newton-Howes, Chan and Kapur40–Reference Turecki, Brent, Gunnell, O’Connor, Oquendo and Pirkis41 Our conclusions align with those reported from other reviews.Reference Awenat, Peters, Gooding, Pratt, Shaw-Núñez and Harris9,Reference Witt, Hetrick, Rajaram, Hazell, Salisbury and Townsend21 Although psychosocial interventions show promise in reducing self-harm and suicide, methodological limitations in primary studies and insufficient inclusion of lived-experience involvement weaken the strength of the evidence base.

Research on developing psychological interventions for in-patient wards and to prevent self-harm and suicide is rapidly expanding.Reference Berry, Raphael, Wilson, Bucci, Drake and Edge42 Recent evidence from a randomised controlled trial with 200 mental health in-patient participants, found evidence to suggest that adding brief CBT to treatment as usual significantly reduced post-discharge 6-month suicide reattempts.Reference Diefenbach, Lord, Stubbing, Rudd, Levy and Worden43 Consistent with Rozek et al,Reference Rozek, Baker, Rugo, Steigerwald, Sippel and Holliday33 a high-quality reviewReference van Ballegooijen, Rawee, Palantza, Miguel, Harrer and Cristea44 found that both direct and indirect suicide prevention interventions reduced suicide attempts. Hajek Gross et alReference Hajek Gross, Oehlke, Prillinger, Goreis, Plener and Kothgassner45 found no significant effect for mentalisation therapy in reducing self-harm repetition compared with controls, which contrasts with the preliminary evidence cited in Witt et al.Reference Witt, Hetrick, Rajaram, Hazell, Salisbury and Townsend21 Pre–post studies in this review suggested a reduction in self-harm frequency, with longer treatment durations yielding greater effects.Reference Hajek Gross, Oehlke, Prillinger, Goreis, Plener and Kothgassner45 Future, co-designed, qualitative trial research in this area may provide important insights into intervention development for self-harm and suicide prevention in mental health in-patient settings.

Clinical implications

In-patient mental health settings are a key setting for suicide prevention.Reference Hunt, Baird, Turnbull, Ibrahim, Shaw and Appleby4 Efforts to reduce self-harm and suicide in this setting, have predominantly focused on environmental adaptations, including ligature removal or restrictive practice.Reference Awenat, Peters, Gooding, Pratt, Shaw-Núñez and Harris9 In the UK, in-patient suicide rates have remained static since 2016,Reference Hunt, Baird, Turnbull, Ibrahim, Shaw and Appleby4 highlighting the urgent need to consider interventions that may improve patient safety in this setting. Access to therapeutic interventions and care may improve patient safety and experiences, and reduce suicide rates. However, evidence from this umbrella review suggests an urgent need to develop self-harm and suicide prevention interventions that are feasible and acceptable for mental health in-patient settings. Implementation barriers include poor fidelity to interventions, inadequate staff training and the challenge of adapting interventions to high-demand ward environments and acute patient crises.Reference Evlat, Wood and Glover13,Reference Csipke, Williams, Rose, Koeser, McCrone and Wykes46,Reference Berry, Handerer, Bucci, Penn, Morley and Raphael47 .

Interventions may be more widely implemented if developed collaboratively with staff and patients as part of quality improvement efforts.Reference Kapur, Tham, Turnbull, Richards, Rodway and Clements48,49 Embedding lived/living experience perspectives throughout all stages, from study design to implementation and evaluation, may enhance intervention relevance and acceptability. Weak evidence for psychosocial interventions does not necessarily indicate a lack of clinical benefit, but may reflect the omission of patient-centred outcomes. Reductions in self-harm may not always align with patient priorities, and interventions might provide benefits in broader areas, including general functioning, social participation and engagement with services.Reference Owens, Fox, Redwood, Davies, Foote and Salisbury39

As the intervention evidence-base continues to develop for in-patient settings, immediate steps can be taken to reduce self-harm and suicide and improve patient experience.49 The UK National Health Service has introduced co-produced standards of care for in-patient mental healthcare,49 emphasising equity, trauma-informed practice, autism-informed approaches and cultural competence (see Table 6 for a summary of the Culture of Care Standards core commitments). Future psychosocial or system-level interventions should be compassionate, patient-centred and aligned to these standards to ensure clinical relevance.

Table 6 The Culture of Care co-produced standards for in-patient care, summarised and adapted from NHS England49

Adapted from: https://www.england.nhs.uk/long-read/culture-of-care-standards-for-mental-health-inpatient-services/.

Supplementary material

The supplementary material is available online at https://doi.org/10.1192/bjo.2025.10811

Data availability

The authors confirm that the data supporting the findings of this study are available within the article and/or its supplementary materials.

Acknowledgements

We are grateful to our patient and carer advisory members (Mutual Support for Mental Health-Research, MS4MH-R) for their support in developing the review questions, forms, interpreting results and implications. Additional thanks go to Dan and Sadika from MS4MH-R for highlighting critical gaps in access to psychological therapies for self-harm and/or suicide prevention.

Author contributions

N.K. and R.T.W. are co-leads for the National Institute for Health and Care Research (NIHR) Greater Manchester Patient Safety Research Collaboration (GM PSRC) Preventing Self-Harm and Suicide Theme and acquired funding for the work. The research idea was developed from previous research, senior expertise and lived experience over inequities in access to aftercare. J.W. led the systematic search strategy, subsequent reference management and flow chart, under the supervision of R.T.W. and N.K. J.W. extracted data; J.G. and F.M. cross-checked data. L.Q. managed and led the project, extracted and cleaned data, led analyses, PPIE and wrote the first draft. Members of our PPIE group (MS4MH-R), R.E., F.S., F.M., J.W., R.T.W. and N.K. contributed to discussions over inclusion/exclusion, quality and content. S.S. supported with content and methodological expertise. E.C. and E.W. are lived-experience partners for our quality improvement in-patient Culture of Care Programme. E.N. is a self-harm and suicide prevention researcher and lived-experience collaborator for the in-patient quality improvement Culture of Care Programme. Our partners provided ongoing lived-experience insights into research, in-patient mental healthcare and interventions for self-harm and suicide prevention, and contributed towards the interpretation, conclusions and implications. E.N., E.C. and E.W. have also completed a lived-experience viewpoint commentary based on this review. All authors contributed to subsequent drafts and approved the submitted version.

Funding

This paper presents independent research funded by the NIHR GM PSRC (Grant Reference Number NIHR204295). R.T.W. is supported by the NIHR Manchester Biomedical Research Centre (Grant Reference Number NIHR203308). F.M., Doctoral Fellow, NIHR 300957, is funded by the National Institute for Health and Care Research (NIHR) for this study. F.M. is supported by the NIHR GM PSRC (NIHR204295). The views expressed in this article are those of the authors and not necessarily those of National Institute for Health and Care Excellence (NICE), NIHR, National Health Service (NHS) or the UK Department of Health and Social Care.

Declaration of interest

N.K. is a member of the Department of Health and Social Care (England) National Suicide Prevention Advisory Group. He has chaired and contributed to various committees for NICE developing guidelines for suicide prevention, depression and the management of self-harm. He is supported by Mersey Care NHS Foundation Trust as well as the University of Manchester. F.M. was a member of the 2022 self-harm NICE guideline development committee. L.Q., J.W., J.G., S.S., E.N., E.W., E.C., R.E. and R.T.W. declare no conflict of interests.

Patient and public involvement

Research ideas for the grant application were developed together with our PPIE group (MS4MH-R). Our patient and carer advisory members (MS4MH-R) were involved throughout the research process. Our panel collaboratively developed the initial research questions, reviewed protocol and data extraction plans, results and contributed to interpretations, conclusions, implications and dissemination plans. People with lived experience in this area contributed to drafts, are co-authors and have completed a lived-experience commentary based on this review. There was PPIE input into our dissemination plan, which includes communicating key findings to relevant patient groups, carers and mental health services.

Consent statement

This is an overview of published systematic reviews, therefore no patient consent was required.

References

World Health Organization (WHO). Suicide. WHO, 2025 (https://www.who.int/news-room/fact-sheets/detail/suicide).Google Scholar
Moran, P, Chandler, A, Dudgeon, P, Kirtley, OJ, Knipe, D, Pirkis, J, et al. The Lancet Commission on self-harm. Lancet 2024; 404: 1445–92.10.1016/S0140-6736(24)01121-8CrossRefGoogle ScholarPubMed
NCISH. Annual Report: UK Patient and General Population Data, 2011–2021. University of Manchester, 2024.Google Scholar
Hunt, IM, Baird, A, Turnbull, P, Ibrahim, S, Shaw, J, Appleby, L, et al. Psychiatric in-patient care in England: as safe as it can be? An examination of in-patient suicide between 2009 and 2020. Psychol Med 2024; 54: 1702–8.10.1017/S0033291723003628CrossRefGoogle Scholar
Musgrove, R, Carr, MJ, Kapur, N, Chew-Graham, CA, Mughal, F, Ashcroft, DM, et al. Suicide and other causes of death among working-age and older adults in the year after discharge from in-patient mental healthcare in England: matched cohort study. Br J Psychiatry 2022; 221: 468–75.10.1192/bjp.2021.176CrossRefGoogle ScholarPubMed
National Institute for Clinical Excellence. Self-harm: Assessment, Management, and Preventing Recurrence. NICE Guidelines, NG225, 2022.Google Scholar
Quinlivan, L, Littlewood, DL, Webb, RT, Kapur, N. Patient safety and suicide prevention in mental health services: time for a new paradigm? J Ment Health 2020; 29: 1–5.10.1080/09638237.2020.1714013CrossRefGoogle ScholarPubMed
Witt, K, McGill, K, Leckning, B, Hill, NTM, Davies, BM, Robinson, J, et al. Global prevalence of psychiatric in- and out-patient treatment following hospital-presenting self-harm: a systematic review and meta-analysis. eClinicalMedicine 2023; 65: 102295.10.1016/j.eclinm.2023.102295CrossRefGoogle ScholarPubMed
Awenat, YF, Peters, S, Gooding, PA, Pratt, D, Shaw-Núñez, E, Harris, K, et al. A qualitative analysis of suicidal psychiatric inpatients views and expectations of psychological therapy to counter suicidal thoughts, acts and deaths. BMC Psychiatry 2018; 18: 334.10.1186/s12888-018-1921-6CrossRefGoogle ScholarPubMed
Mann, JJ, Michel, CA, Auerbach, RP. Improving suicide prevention through evidence-based strategies: a systematic review. Am J Psychiatry 2021; 178: 611–24.10.1176/appi.ajp.2020.20060864CrossRefGoogle ScholarPubMed
Hawton, K, Witt, KG, Salisbury, TLT, Arensman, E, Gunnell, D, Hazell, P, et al. Psychosocial interventions for self-harm in adults. Cochrane Library, 12 May 2016 (https://www.cochranelibrary.com/cdsr/doi/10.1002/14651858.CD012189/full).CrossRefGoogle Scholar
Staniszewska, S, Mockford, C, Chadburn, G, Fenton, SJ, Bhui, K, Larkin, M, et al. Experiences of in-patient mental health services: systematic review. Br J Psychiatry 2019; 214: 329–38.10.1192/bjp.2019.22CrossRefGoogle ScholarPubMed
Evlat, G, Wood, L, Glover, N. A systematic review of the implementation of psychological therapies in acute mental health inpatient settings. Clin Psychol Psychother 2021; 28: 1574–86.10.1002/cpp.2600CrossRefGoogle ScholarPubMed
Choi, GJ, Kang, H. Introduction to umbrella reviews as a useful evidence-based practice. J Lipid Atheroscler 2023; 12: 3–11.10.12997/jla.2023.12.1.3CrossRefGoogle ScholarPubMed
Boudreaux, ED, Larkin, C, Vallejo Sefair, A, Ma, Y, Li, YF, Ibrahim, AF, et al. Effect of an emergency department process improvement package on suicide prevention: the ED-SAFE 2 Cluster Randomized Clinical Trial. JAMA Psychiatry 2023; 80: 665–74.10.1001/jamapsychiatry.2023.1304CrossRefGoogle ScholarPubMed
Fusar-Poli, P, Radua, J. Ten simple rules for conducting umbrella reviews. BMJ Ment Health 2018; 21: 95–100.Google ScholarPubMed
Gates, M, Gates, A, Pieper, D, Fernandes, RM, Tricco, AC, Moher, D, et al. Reporting guideline for overviews of reviews of healthcare interventions: development of the PRIOR statement. BMJ 2022; 378: e070849.10.1136/bmj-2022-070849CrossRefGoogle ScholarPubMed
Pollock, A, Farmer, SE, Brady, MC, Langhorne, P, Mead, GE, Mehrholz, J, et al. An algorithm was developed to assign GRADE levels of evidence to comparisons within systematic reviews. J Clin Epidemiol 2016; 70: 106–10.10.1016/j.jclinepi.2015.08.013CrossRefGoogle ScholarPubMed
Guyatt, G, Oxman, AD, Akl, EA, Kunz, R, Vist, G, Brozek, J, et al. GRADE guidelines: 1. Introduction – GRADE evidence profiles and summary of findings tables. J Clin Epidemiol 2011; 64: 383–94.10.1016/j.jclinepi.2010.04.026CrossRefGoogle ScholarPubMed
Kapur, N, Cooper, J, O’Connor, RC, Hawton, K. Non-suicidal self-injury v. attempted suicide: new diagnosis or false dichotomy? Br J Psychiatry 2013; 202: 326–8.10.1192/bjp.bp.112.116111CrossRefGoogle ScholarPubMed
Witt, KG, Hetrick, SE, Rajaram, G, Hazell, P, Salisbury, TLT, Townsend, E, et al. Psychosocial interventions for self-harm in adults. Cochrane Library, 22 Apr 2021 (https://www.cochranelibrary.com/cdsr/doi/10.1002/14651858.CD013668.pub2/full).CrossRefGoogle Scholar
Shea, BJ, Reeves, BC, Wells, G, Thuku, M, Hamel, C, Moran, J, et al. AMSTAR 2: a critical appraisal tool for systematic reviews that include randomised or non-randomised studies of healthcare interventions, or both. BMJ 2017; 358: j4008.10.1136/bmj.j4008CrossRefGoogle ScholarPubMed
Bracchiglione, J, Meza, N, Bangdiwala, S, Guzmán, EN, Urrutia, G, Bonfill, X, et al. Graphical representation of overlap for OVErviews. Res Synth Methods 2022; 13: 381–8.10.1002/jrsm.1557CrossRefGoogle ScholarPubMed
Pieper, D, Antoine, SL, Mathes, T, Neugebauer, EAM, Eikermann, M. Systematic review finds overlapping reviews were not mentioned in every other overview. J Clin Epidemiol 2014; 67: 368–75.10.1016/j.jclinepi.2013.11.007CrossRefGoogle Scholar
Fox, KR, Huang, X, Guzmán, EM, Funsch, KM, Cha, CB, Ribeiro, JD, et al. Interventions for suicide and self-injury: a meta-analysis of randomized controlled trials across nearly 50 years of research. Psychol Bull 2020; 146: 1117–45.10.1037/bul0000305CrossRefGoogle ScholarPubMed
Popay, J, Roberts, H, Sowden, A, Petticrew, M, Arai, L, Rodgers, M, et al. Guidance on the conduct of narrative synthesis in systematic reviews. A product from the ESRC methods programme Version 1, 2006: b92.Google Scholar
Deeks, JJ, Higgins, JPT, Altman, DG. Analysing data and undertaking meta-analyses. In Cochrane Handbook for Systematic Reviews of Interventions version 6.0 (updated July 2019) (eds Higgins, PT, Thomas, J, Chandler, J, Cumpston, M, Li, T, Page, MJ, et al.): Ch. 10. Cochrane, 2019 (www.training.cochrane.org/handbook).10.1002/9781119536604.ch10CrossRefGoogle Scholar
Hawton, K, Witt, KG, Salisbury, TLT, Arensman, E, Gunnell, D, Hazell, P, et al. Psychosocial interventions following self-harm in adults: a systematic review and meta-analysis. Lancet Psychiatry 2016; 3: 740–50.10.1016/S2215-0366(16)30070-0CrossRefGoogle ScholarPubMed
Hou, X, Wang, J, Guo, J, Zhang, X, Liu, J, Qi, L, et al. Methods and efficacy of social support interventions in preventing suicide: a systematic review and meta-analysis. BMJ Ment Health 2022; 25: 29–35.Google ScholarPubMed
Sobanski, T, Josfeld, S, Peikert, G, Wagner, G. Psychotherapeutic interventions for the prevention of suicide re-attempts: a systematic review. Psychol Med 2021; 51: 2525–40.10.1017/S0033291721003081CrossRefGoogle ScholarPubMed
Yiu, HW, Rowe, S, Wood, L. A systematic review and meta-analysis of psychosocial interventions aiming to reduce risks of suicide and self-harm in psychiatric inpatients. Psychiatry Res 2021; 305: 114175.10.1016/j.psychres.2021.114175CrossRefGoogle ScholarPubMed
Nawaz, RF, Reen, G, Bloodworth, N, Maughan, D, Vincent, C. Interventions to reduce self-harm on in-patient wards: systematic review. BJPsych Open 2021; 7: e80.10.1192/bjo.2021.41CrossRefGoogle ScholarPubMed
Rozek, DC, Baker, SN, Rugo, KF, Steigerwald, VL, Sippel, LM, Holliday, R, et al. Addressing co-occurring suicidal thoughts and behaviors and posttraumatic stress disorder in evidence-based psychotherapies for adults: a systematic review. J Trauma Stress 2022; 35: 729–45.10.1002/jts.22774CrossRefGoogle ScholarPubMed
Donker, T, Calear, A, Grant, JB, Van Spijker, B, Fenton, K, Hehir, KK, et al. Suicide prevention in schizophrenia spectrum disorders and psychosis: a systematic review. BMC Psychol 2013; 1: 1–10.10.1186/2050-7283-1-6CrossRefGoogle ScholarPubMed
Wand, AP, Browne, R, Jessop, T, Peisah, C. A systematic review of evidence-based aftercare for older adults following self-harm. Aust N Z J Psychiatry 2022; 56: 1398–420.10.1177/00048674211067165CrossRefGoogle ScholarPubMed
DeCou, CR, Comtois, KA, Landes, SJ. Dialectical behavior therapy is effective for the treatment of suicidal behavior: a meta-analysis. Behav Ther 2019; 50: 60–72.10.1016/j.beth.2018.03.009CrossRefGoogle ScholarPubMed
Luxton, DD, June, JD, Comtois, KA. Can postdischarge follow-up contacts prevent suicide and suicidal behavior? Crisis 2013; 34: 32–41.10.1027/0227-5910/a000158CrossRefGoogle ScholarPubMed
Kapur, N, Gorman, LS, Quinlivan, L, Webb, RT. Mental health services: quality, safety and suicide. BMJ Qual Saf 2022; 31: 419–22.10.1136/bmjqs-2021-013532CrossRefGoogle ScholarPubMed
Owens, C, Fox, F, Redwood, S, Davies, R, Foote, L, Salisbury, N, et al. Measuring outcomes in trials of interventions for people who self-harm: qualitative study of service users’ views. BJPsych Open 2020; 6: e22.10.1192/bjo.2019.93CrossRefGoogle ScholarPubMed
Knipe, D, Padmanathan, P, Newton-Howes, G, Chan, LF, Kapur, N. Suicide and self-harm. Lancet Psychiatry 2022; 399: 1903–16.10.1016/S0140-6736(22)00173-8CrossRefGoogle ScholarPubMed
Turecki, G, Brent, DA, Gunnell, D, O’Connor, RC, Oquendo, MA, Pirkis, J, et al. Suicide and suicide risk. Nat Rev Dis Primers 2019; 5: 74.10.1038/s41572-019-0121-0CrossRefGoogle ScholarPubMed
Berry, K, Raphael, J, Wilson, H, Bucci, S, Drake, RJ, Edge, D, et al. A cluster randomised controlled trial of a ward-based intervention to improve access to psychologically-informed care and psychological therapy for mental health in-patients. BMC Psychiatry 2022; 22: 82.10.1186/s12888-022-03696-7CrossRefGoogle ScholarPubMed
Diefenbach, GJ, Lord, KA, Stubbing, J, Rudd, MD, Levy, HC, Worden, B, et al. Brief cognitive behavioral therapy for suicidal inpatients: a randomized clinical trial. JAMA Psychiatry 2024; 81: 1177–86.10.1001/jamapsychiatry.2024.2349CrossRefGoogle ScholarPubMed
van Ballegooijen, W, Rawee, J, Palantza, C, Miguel, C, Harrer, M, Cristea, I, et al. Suicidal ideation and suicide attempts after direct or indirect psychotherapy: a systematic review and meta-analysis. JAMA Psychiatry 2025; 82: 31–7.10.1001/jamapsychiatry.2024.2854CrossRefGoogle ScholarPubMed
Hajek Gross, C, Oehlke, SM, Prillinger, K, Goreis, A, Plener, PL, Kothgassner, OD. Efficacy of mentalization-based therapy in treating self-harm: a systematic review and meta-analysis. Suicide Life Threat Behav 2024; 54: 317–37.10.1111/sltb.13044CrossRefGoogle ScholarPubMed
Csipke, E, Williams, P, Rose, D, Koeser, L, McCrone, P, Wykes, T, et al. Following the Francis report: investigating patient experience of mental health in-patient care. Br J Psychiatry 2016; 209: 35–9.10.1192/bjp.bp.115.171124CrossRefGoogle ScholarPubMed
Berry, K, Handerer, F, Bucci, S, Penn, G, Morley, H, Raphael, J, et al. Ensuring that psychological interventions are delivered as intended on mental health inpatient wards. Br J Clin Psychol 2024; 64: 371–84.10.1111/bjc.12510CrossRefGoogle ScholarPubMed
Kapur, N, Tham, SG, Turnbull, P, Richards, N, Rodway, C, Clements, C, et al. Quality improvement for suicide prevention and self-harm intervention: addressing the implementation gap and saving lives. Lancet Psychiatry 2025; 12: 303–9.10.1016/S2215-0366(24)00442-5CrossRefGoogle ScholarPubMed
NHS England. Culture of Care Standards for Mental Health Inpatient Services. NHS England, 2024 (https://www.england.nhs.uk/long-read/culture-of-care-standards-for-mental-health-inpatient-services).Google Scholar
Figure 0

Table 1 PICOa (population, intervention, comparison and outcomes) criteria

Figure 1

Fig. 1 Flow diagram of included studies. CDSR, Cochrane Database of Systematic Reviews.

Figure 2

Table 2 AMSTAR-222 ratings across the 13 systematic reviews

Figure 3

Table 3 Quality evaluation for included reviews (Adapted Algorithm for GRADE,18 AMSTAR-2,22 risk of bias, heterogeneity, generalisability)

Figure 4

Table 4 Study characteristics for included reviews (e.g. included study dates, age, gender intervention details, controls, follow-up, designs, outcomes)

Figure 5

Table 5 Non-significant treatment effects for interventions in included systematic reviews

Figure 6

Fig. 2 Heat map for primary study overlap analysis for citations for self-harm, suicide and suicidal behaviour.

Figure 7

Table 6 The Culture of Care co-produced standards for in-patient care, summarised and adapted from NHS England49

Supplementary material: File

Quinlivan et al. supplementary material

Quinlivan et al. supplementary material
Download Quinlivan et al. supplementary material(File)
File 29.1 KB
Submit a response

eLetters

No eLetters have been published for this article.