Included studies

38 qualitative studies reported in 39 articles were identified for this review (Alexander 2019, Bailey 2019, Bergmans 2009, Biddle 2020, Brown 2013, Bywaters 2002, Chan 2017, Cooper 2011, Crona 2017, Cutcliffe 2006, Dunkley 2018, Fogarty 2018, Frey 2018, Frost 2016, Haberstroh 2012, Heredia Montesinos 2019, Holliday 2015, Holliday 2018, Holm 2011, Horrocks 2005, Hume 2007, Idenfors 2015a, Idenfors 2015b, Kelada 2018, Klineberg 2013, Lewis 2016, Long 2016, McGill 2019, Owens 2016, Peterson 2015, Rissanen 2009, River 2018, Rivlin 2013, Strike 2006, Vatne 2018, Ward 2013, Weber 2002, Williams 2018, Wong 2015). Two articles (Idenfors 2015a, Idenfors 2015b) reported results from the same study at different time points (initial interview and at 6 months follow-up).

The included studies are summarised in Table 2.

The studies were carried out in the following countries: UK (Alexander 2004, Bailey 2019, Biddle 2020, Bywaters 2002, Cooper 2011, Cutcliffe 2006, Dunkley 2018, Holliday 2018, Horrocks 2005, Hume 2007, Klineberg 2013, Long 2016, Owens 2016, Rivlin 2013, Ward 2013); Australia (Fogarty 2018, Frost 2016, McGill 2019, River 2018); Canada (Bergmans 2009, Chan 2017, Lewis 2016, Strike 2006); Finland (Rissanen 2009); Germany (Heredia Montesinos 2019); New Zealand (Peterson 2015); Norway (Holm 2011, Vatne 2018); Sweden (Crona 2017, Idenfors 2015a, Idenfors 2015b); USA (Brown 2013, Frey 2018, Haberstroh 2012, Holliday 2015, Weber 2002, Williams 2018, Wong 2015). One article reported study cohorts from Australia, Belgium and USA (Kelada 2018).

The following settings were represented in the included studies: community (Alexander 2004, Frey 2018, Holliday 2018, Kelada 2018, Long 2016, McGill 2019, Peterson 2015, Williams 2018); e-community or online group (Frost 2016, Haberstroh 2012, Lewis 2016); health care (including: emergency [emergency department (Cooper 2011, Holliday 2015, Horrocks 2005, Idenfors 2015a, Idenfors 2015b, Owens 2016); emergency psychiatric services (Cutcliffe 2006)]; inpatient (Hume 2007, Weber 2002); outpatient (Dunkley 2018, Heredia Montesino 2019, Holm 2011, Strike 2006, Vatne 2018); primary care (Bailey 2019)); educational [including: university (Brown 2013); secondary school (Klineberg 2013)]; prison [including: adult male prison (Rivlin 2013); adult female prison (Ward 2013)]. Three studies represented mixed settings, including: community and emergency department (Biddle 2020); community and outpatient (River 2018); emergency department and other health setting (Wong 2015). Six studies did not explicitly report the setting and it was unclear from the information reported (Bergmans 2009, Bywaters 2002, Chan 2017, Crona 2017, Fogarty 2018, Rissanen 2009).

The studies included people who have self-harmed in the following age groups: adults (aged 18 years-plus: Alexander 2004, Bergmans 2009, Biddle 2020, Brown 2013, Cooper 2011, Crona 2017, Cutcliffe 2006, Dunkley 2018, Fogarty 2018, Frey 2018, Frost 2016, Haberstroh 2012, Heredia Montesino 2019, Holm 2011, Horrocks 2005, Hume 2007, Lewis 2016, Long 2016, McGill 2019, Owens 2016, Peterson 2015, River 2018, Rivlin 2013, Strike 2006, Vatne 2018, Ward 2013, Weber 2002, Williams 2018, Wong 2015); adolescent and adults (age 16 to 25 years: Bailey 2019, Bywaters 2002, Holliday 2015, Idenfors 2015a, Idenfors 2015b, Kelada 2018; age 12 to 21 years: Rissanen 2009; age <20 to 58 years: Chan 2017); adolescents (age 11 to 17 years: Holliday 2018, Klineberg 2013).

Reporting of ethnicity and race information in general was poor: 20 studies did not report ethnicity/ race information (Bergmans 2009, Biddle 2020, Chan 2017, Cooper 2011, Crona 2017, Cutcliffe 2006, Dunkley 2018, Fogarty 2018, Holliday 2015, Holm 2011, Horrocks 2005, Hume 2007, Idenfors 2015a, Idenfors 2015b, Long 2016, Peterson 2015, Rissanen 2009, River 2018, Strike 2006, Vatne 2018, Ward 2013). One study reported that all participants were women of Turkish descent but did not otherwise report ethnicity/race information (Heredia Montesinos 2019). Three studies only partially reported ethnicity/ race information (Frost 2016 [12 participants were from an ethnic minority group but the rest of participants’ (the majority’s) ethnicity/ race information was not reported]; Kelada 2018 [ethnicity/ race information was only reported for one of the included countries; of these participants, the majority were Caucasian or of white ethnic origin]; McGill 2019 [2 participants from the total cohort were from an ethnic minority group but the rest of participants’ (the majority’s) ethnicity/ race information was not reported]). Fourteen studies reported ethnicity/ race information (Alexander 2004, Bailey 2019, Brown 2013, Bywaters 2002, Frey 2018, Haberstroh 2012, Holliday 2018, Klineberg 2013, Lewis 2016, Owens 2016, Rivlin 2013, Weber 2002, Williams 2018, Wong 2015). Of these studies, most reported all or a majority of participants were Caucasian or of white ethnic origin (Alexander 2004, Bailey 2019, Brown 2013, Bywaters 2002, Frey 2018, Haberstroh 2012, Holliday 2018, Lewis 2016, Owens 2016, Rivlin 2013, Weber 2002, Williams 2018); 3 studies included mostly or all ethnic minority groups (Heredia Montesinos 2019 [women of Turkish descent in Germany], Klineberg 2013 [study conducted in a UK secondary school with White British & White other, Bangladeshi, Pakistani, Indian and Sri Lankan Tamil, Black including British and African, mixed ethnicity including White & Black African, African & Asian, White & Black Caribbean, White & Oriental Asian, Pakistani and Asian British students], Wong 2015 [Chinese immigrants in USA]).

See the literature search strategy in appendix B and study selection flow chart in appendix C.

Excluded studies

Studies not included in this review with reasons for their exclusions are provided in appendix J.

Included studies

A single economic search was undertaken for all topics included in the scope of this guideline but no economic studies were identified which were applicable to this review question. See the literature search strategy in appendix B and economic study selection flow chart in appendix G.

Excluded studies

Economic studies not included in the guideline economic literature review are listed, and reasons for their exclusion are provided in appendix J.

Economic

No economic studies were identified which were applicable to this review question.

The outcomes that matter most

The aim of this review question was to identify the information and support needs of people who have self-harmed. As a result, the views of people who have self-harmed were considered the most important for this question. The committee suggested potential themes which may have arisen from the evidence such as Information content and Information format, but did not want to constrain the question; therefore, any views and preferences about information and support needs regarded as useful/ not useful or important/ not important by the population were included.

The quality of the evidence

When assessed using GRADE CERQual methodology the evidence was found to range in quality from very low to high quality, with most of the evidence being of moderate quality. The recommendations were drafted mostly based on the evidence but in some parts supplemented accordingly with the committee’s own expertise.

In most cases, the evidence was downgraded due to poor applicability where the themes were not based on any research from a UK context, or due to methodological problems that may have had an impact on the findings (for example due to ethical issues, lack of discussion of author reflexivity, and/ or bias arising through study design, recruitment or data collection processes). Some downgrading for adequacy occurred when the richness or quantity of the data was low, for example when the understanding of the theme would benefit from more specific or informative data. Another issue resulting in downgrading was incoherence within the findings.

Despite the range of quality of evidence, the committee agreed that most themes identified were representative of their own knowledge and experience, including those of low or very low quality. The committee also discussed the fact that many of the themes identified in this review were also found in the review on the information and support needs of family members and carers, and agreed that this showed some of the themes were widely experienced by different populations. For this reason, the committee considered all evidence when drafting the recommendations, supplementing any poor quality data with their own expertise when necessary.

Benefits and harms

There was evidence from the sub-theme ‘Network – family and/ or friends’ that people who had self-harmed considered family/ carers to be important sources of information and support, and therefore wanted certain information to be shared with family members in order to reduce stigma around the topic and enable family/ carers to provide adequate support. However, evidence from the sub-theme ‘Ensuring a safe environment’ also showed that people who had self-harmed valued anonymity and privacy when seeking information. Committee members were concerned about the potential distress caused to the service user by sharing information on self-harm with family or carers, and compared this with the potential benefits of doing so as identified by the evidence from the review on involving family members and carers (Evidence Report D). The committee agreed that information should be shared with family/ carers but only if appropriate, according to the service user’s individual circumstances.

The sub-theme ‘Include individual in conversations with agencies’ showed that people who had self-harmed felt it was important that they were included in conversations about self-harm, and the committee agreed that information sharing would promote the involvement of patients in their own care. The recommendations about information content were based on the evidence from sub-themes identified under the theme ‘Information content’, in which participants identified a wide range of useful information content, such as general information about self-harm and its management (from the sub-themes ‘General information’, ‘Understanding self-harm and why people self-harm’ and ‘Management of self-harm’), as well as specific advice for what to do in certain circumstances and where to find additional support and information (from the sub-themes ‘Referral and follow-up support needs’, ‘Routes to professional care’ and ‘How to access support’). In particular, the committee agreed that information should be provided to people who had self-harmed and their family members and carers about the impact of stigma, based on evidence from the sub-theme ‘Address stigma’ from this review, and the sub-theme ‘Addressing stigma’ from the review on the information and support needs of family members and carers. The committee agreed that addressing how stigma surrounding self-harm can negatively affect people would provide a number of benefits, including preparing people for how to combat stigma, and enabling family members and carers to provide better support not clouded by inaccurate myths surrounding self-harm. Many of the identified themes were consistent with evidence on the information needs of family members and carers, as identified in the sub-themes under ‘Information content’ in Evidence Report B.

There was evidence from sub-themes identified under the themes ‘Source of information/ support’ and ‘Type of support’ that people who had self-harmed considered multiple different sources of information and support useful, including healthcare professionals, peers, social groups, local services, school staff, and other third party groups such as police. The importance of content provided by other people who had self-harmed was highlighted by the sub-theme ‘Lived experience content’, which showed that people who had self-harmed felt that sharing experiences could provide hope and promote recovery. Additionally, people who had self-harmed who were being discharged into the community wanted reassurances of a continuation of support in community settings and thought that practical help with other factors such as finances were helpful, as evidenced in the sub-themes ‘Support for discharge to community’ and ‘Vocational or practical support’, respectively. The committee agreed this evidence showed that healthcare professionals needed to provide information about how to access further information and support from a multitude of sources, especially when people who had self-harmed were not currently in a healthcare setting. However, there was conflicting evidence from the sub-theme ‘Address limitations of signposting’ that being pointed to other services or sources of information and support was a barrier to seeking help. The committee discussed the potential risk that signposting people elsewhere could devalue their current needs, and agreed, based on evidence from the sub-theme ‘Address need in crisis’, that having access to local services, including out-of-hours services, would in fact reduce the likelihood that people would be left without crisis support when they need it. The committee also agreed that providing information about available local and online support and how to access it would overall help to remove barriers to help-seeking rather than reaffirm them, through the provision of further sources of information and support.

There was evidence from the sub-theme ‘Self-help services, materials or activities’ that people who had self-harmed thought that self-help materials and information were useful. The committee discussed whether providing these would discourage people from seeking help or support when it was needed, but agreed that information about approaches to self-care could empower individuals to manage their urges to self-harm and promote autonomy. However, the committee agreed it was important to ensure people are still encouraged to seek help and access services. They agreed, based on their knowledge and experience, that the provision of unhelpful information about managing self-harm alone could make people feel dismissed or hopeless. Therefore, any information about self-care should encourage help-seeking and include information about its importance. The committee discussed the importance of providing information about responding to injuries caused by self-harm but were concerned this, too, could discourage help-seeking. They agreed that information about responding to injuries could include wound management to reduce rates of high-risk injuries or infection, but that information on injury response should also include how and when to seek care for injuries caused by self-harm.

There was evidence that family members and carers had further information and support needs in addition to those identified by people who had self-harmed, which were individual to their own experiences. People who had self-harmed said it would be helpful if family members and carers were provided with information about how to recognise potential self-harm in the sub-theme ‘education and training support need’, which was supplemented by evidence from family members and carers in the sub-theme ‘how to recognise potential self-harm’ in Evidence Report B. The committee decided family members and carers should be provided this information when appropriate so they could provide effective care and support to people who had self-harmed outside of healthcare settings. The rest of the recommendations on specific information and support for family members and carers were based on the evidence as outlined in Evidence Report B.

The recommendations about the principles of information sharing were based on the evidence. There was evidence from the sub-themes ‘Individualised approaches to care’ and ‘Content is fit for purpose’ that people who had self-harmed felt it was important for the provision of information to be personalised and appropriate for the service user’s individual needs. Based on this evidence as well as the committee’s own knowledge and experience, the committee decided that it was important to emphasise that information provided to service users should be tailored to their specific circumstances. They agreed it was important that the provision of any information and support be considered with this principle in mind to avoid potential harm arising from giving too much or too little information inappropriately to the person who had self-harmed, and to avoid contradicting any existing care plan. Based on their experience, the committee also agreed that information may need to be further tailored if the person has any co-existing health conditions, neurodevelopmental conditions or a learning disability to ensure it is person centred. Evidence from the sub-themes ‘positive communication’ and ‘positive emotional support’ from this review, and the sub-theme ‘positive emotional support’ from the review on the information and support needs of family members and carers showed that sensitivity, positivity, encouragement, respect and consideration for individual circumstances when communicating were important to both people who had self-harmed and their family members and carers, who felt these factors would improve the quality of care. The provision of information in this manner was seen as therapeutic in itself. The sub-themes identified under the theme ‘Information format’ showed that people who had self-harmed also thought it would be helpful if information and support was provided in different formats, including online, through direct contact and in written documents. The committee agreed that any information provided should be accessible to all people to ensure everyone had equal access to resources, and that this was in line with the NICE guidelines on service user experience in adult NHS mental health (CG136), patient experience in adult NHS mental health services (QS15) and babies, children and young people’s experience of healthcare (NG204), which also have further recommendations relating to information and support and were therefore appropriate to refer to. The committee agreed that the guidance sufficiently covered a number of topics regarding how information and support should be provided, including accessibility, and therefore decided that these recommendations should be followed because they would improve the standard of care.

Evidence from the sub-theme ‘Minority support - cultural sensitivity, language’ showed that access to information and support that was sensitive to their cultural background was important to people who had self-harmed, which required staff to be culturally competent and able to recognise when the person’s background may intersect with their self-harm. The committee discussed the fact that this included religious, racial, cultural, sexual orientation, gender identity, educational, health, and economic factors, and agreed that these were likely to influence how a person would access or react to care. For example, discrimination based on other factors may intersect with discrimination regarding self-harm to result in increased difficulties accessing services, and cultural pressures may also influence how comfortable people feel about seeking support. The committee therefore agreed that any provision of information and support should be adapted for situations when a person’s self-harm or care may be influenced by these factors.

Evidence from the sub-theme ‘Sufficient time for assessment/ consultation’ showed that people thought it was important they be provided the time to discuss self-harm without being rushed, which could feel impersonal. This evidence aligned with the sub-theme ‘Making time for the patient’ from the review on non-specialist staff skills (Evidence Report R), and the committee used it to inform the recommendation that staff training should address how to deal with time constraints. The committee also discussed the sub-theme ‘Education and training support need’ and used this evidence when developing recommendations about staff training. The sub-theme ‘Support interventions’ showed that people who have self-harmed generally felt that psychosocial interventions were a source of support, which the committee kept in mind when drafting the intervention recommendations, although these recommendations were drafted largely based on the effectiveness evidence.

Cost effectiveness and resource use

The committee noted that no relevant published economic evaluations had been identified and no additional economic analysis had been undertaken in this area. Therefore, they based the recommendations on the evidence, their knowledge and experience, and on existing NICE guidance. They recognised that there is a moderate variation across the NHS in responding to information and support needs of people who have self-harmed. However, it was pointed out how the recommended adjustments that promote information or enhance support to people who had self-harmed would have a minimal cost impact to the NHS in terms of extra healthcare professionals’ time. This may be offset by better health outcomes by improving the care and quality of life of people who have self-harmed.

Excluded qualitative studies

Please note that the current search was undertaken with the search for review question B (What are the information and support needs of the families and carers of people who have self-harmed?) and the list of excluded studies only lists the 76 studies that were excluded for both reviews in contrast to the 89 excluded studies specified in the PRISMA diagram (Appendix C). This is because routing used in EPPI-Reviewer to separate the results of review questions A and B (for which a combined search was performed) resulted in EPPI-Reviewer being unable to generate the PRISMA diagram in the usual format, with the excluded studies for review questions A and B separated. The (89-76 =) 13 studies not listed in the excluded studies tables are studies that met the inclusion criteria for review question B. There were 3 studies that were included in both review question A and B.

Excluded economic studies