Self-harm is defined as “intentional self-poisoning or injury, irrespective of the apparent purpose” (National Institute for Health and Care Excellence, 2022, p. 81). Rates of self-harm have become more prevalent, with 25.7% of women, and 9.7% of men, reporting self-harm across their life (McManus et al., 2016). Differential rates of self-harm have been reported for those who identify as LGBTQ+ (Liu et al., 2019), those from non-white ethnic backgrounds (Bhui et al., 2007) and those from deprived socio-economic backgrounds (Geulayov et al., 2022). Self-harm is also linked with suicidality (Geulayov et al., 2019). Overall, self-harm is a complex and life-threatening phenomenon; one that warrants significant attention.
The present study, by Stevens and colleagues (2024), seeks to identify whether a brief text message intervention could aid in reducing the number of times that people need to return to hospital for self-harm treatment. The randomised controlled trial (RCT), by Steven and colleagues (2024), is the first to explore this type of intervention on a broader population in Australia. Previous research has been focused on specific populations, like military personnel, therefore, this study has the potential to contribute to the interventions’ generalisability to a wider group of people.
Could brief text messages reduce the number of times people need to re-present to hospital for self-harm treatment?
Methods
Stevens et al., (2024) conducted a parallel RCT which compared participants who received treatment as usual (TAU), with participants who received TAU, plus nine brief short message service (SMS) messages over 12 months, after their discharge from three hospitals in New South Wales, Australia.
The SMS intervention included three different messages sent to participants on a rotating schedule. All SMS messages were personalised with the participants name, included an expression of concern for participant wellbeing, and included details of available local mental health services and crisis lines. The messages were developed by consultants with lived experience.
Once enrolled, participants were stratified by first and subsequent self-harm presentation as there was significant variability amongst participants. After stratification, participants were then randomly assigned and enrolled by clinicians into either the TAU or TAU plus SMS condition. The allocation sequence was concealed.
The study also interestingly used a single-consent Zelen design (Zelen, 1979) which has been discussed in previous a previous blog by Lucy Maconick (2022). The Zelen design is a modification to a usual RCT, whereby participants in the intervention group only are asked to provide consent after randomisation. As the study used a Zelen design, this meant that the participants were not blinded to being in the SMS condition.
Results
The study enrolled 431 participants into the TAU group and 373 participants into the TAU plus SMS group. There was a larger proportion of females (n=520) to males (n=284). The study had main three primary outcomes, and two secondary outcomes.
Primary outcomes
- The number of times that people repeated self-harm, and needed hospital treatment for this, reduced significantly at 12 and 24 months, after receiving TAU plus the SMS intervention, with a relative risk reduction (RRR) of 22%. The authors state that at the 24-month follow-up, this 22% equates to 123 less people re-presenting for self-harm treatment in the SMS group.
- Females had higher rates of re-presentation for self-harm, but the effect of the brief text message was significant and larger for females compared to males, where there was no difference between TAU and TAU plus intervention regarding the frequency of repeat, hospital treated, self-harm.
- There was no difference, between TAU and TAU plus intervention, in the time between when participants first presented to hospital with self-harm, to when they presented to hospital again for self-harm, over the 24-month follow-up period.
Secondary outcomes
- There was no difference between TAU and TAU plus intervention regarding the proportion of participants who repeated self-harm or not, over the 24-month follow-up period of the study, meaning the intervention did not reduce the number of times people repeated self-harm.
- Throughout the study, over the 24-month period, there were a total of 16 deaths.
- The study also analysed suicide rates amongst the 16 deaths and sadly found that there were 4 suicides across the 24-month period. The authors stated that the suicides occurred in the TAU group only.
People presented fewer times to hospital for treatment of their repeat self-harm at 12 and 24 months, after receiving the brief text message intervention.
Conclusions
Overall, this study shows that a brief text message intervention could reduce the number of times that people return to hospital for treatment of repeat self-harm. The authors also state that “the 22% reduction in repetition of hospital-treated self-harm was clinically meaningful” (Stevens et al., 2024, p. 106). This is an important finding, given how easily implementable such an intervention could be, as it is brief and automated. However, the authors argue that it is likely that more studies are needed to establish further efficacy and economic feasibility of the SMS intervention.
The intervention helped to reduce repeat self-harm re-presentations to hospital, but further research is needed to ensure its efficacy and affordability.
Strengths and limitations
The RCT, by Stevens et al., (2024), was the first to explore a brief SMS intervention in a population beyond a narrow demographic. By exploring a wider population, this study provided evidence that the intervention could have efficacy in hospital settings, where a large proportion of the population seek self-harm treatment. As this was an RCT, the study was randomised, which is another strength, as this helps to improve internal validity. It was also good to see that messages included in the intervention were created by people with lived mental health experience, likely improving their appropriateness.
However, the study used non-standardised TAU condition. Although this could be argued as more accurate, as there is variability in clinical settings, it could confound results. Witt et al. (2018) state that non-standard TAU is a significant source of heterogeneity in studies that investigate self-harm interventions, and it is imperative to have clarity regarding what constitutes TAU. Although the authors did detail what TAU was likely to consist of, they stated that it varied, which could have influenced outcomes. For example, if one participant received a care plan that did not include psychological therapies, compared to another participant that did receive psychological therapies, this could have caused confounding variables.
Another main limitation was the single-consent Zelen design. Although the Zelen design can improve drop-out rates, as participants may be happier to be allocated to the treatment condition (Homer, 2002), the single-consent Zelen design asked for consent in the treatment group only, which the authors state led to increased attrition rates, which could have led to attrition bias. Participants in the treatment condition were also aware of their allocation, which could have altered participant behaviour, ultimately impacting the validity of the study. Additionally, the study did not make clear whether the researchers were blinded, so this may have further impacted the findings.
Although an interesting addition to research, the study has left some concerns and questions regarding validity.
Implications for practice
Self-harm is a significant burden on the wellbeing of the Australian population (Australian Institute of Health and Welfare, 2019), therefore the implications of the reduction in repeat self-harm event rates are significant for patients, services and practitioners.
Qualitative research has demonstrated that brief SMS interventions may be helpful to people who are self-harming by fostering a sense of feeling supported through caring messages, and information about services, if urgent mental health support is needed (Duan et al., 2020). This approach to patient aftercare therefore may reduce demand on stretched Australian emergency departments (Australian Institute of Health and Welfare, 2023), as messages encourage patients to use support helplines for preventative self-harm care. However, prior to implementation, further research must explore why patients are re-presenting less, and whether crisis lines, or supportive messages, are a casual mechanism by which re-presentation rates are reduced. Knowing the exact casual mechanisms will ensure interventions are efficacious, ease demand, and achieve the best outcomes in terms of self-harm reduction.
As the UK experiences the same difficulties as Australia, like high demand for urgent services (Pines et al., 2011), and increases in self-harm (McManus et al., 2016), the same implications apply. As the NHS moves towards providing Enhanced Primary Care Services, which include mental health services for people with more complex presentations, such an intervention may offer practitioners an easily implementable tool which could help reduce escalation to secondary care and improve mental health literacy; both outcomes have the potential to enhance patient safety and experience in primary care, especially for more at-risk patients. Further research is needed to explore an SMS intervention in primary care settings to determine the efficacy of the intervention.
Finally, the cost-effectiveness of the intervention must be examined. There is evidence to suggest that digital health interventions can be cost-effective (Gentili et al., 2022), which could make an SMS intervention feasible in the NHS, without adding significant financial burden. In the Talking Therapies service that I work within, the service has an established automated SMS service, which may be the case across other services, so it could be easily implementable with little additional resources.
Overall, a brief SMS intervention could be implemented within mental health services to reduce burden on urgent services and improve the wellbeing of the recipients, however, more research is needed regarding mechanisms of change, generalisability and economic viability before adoption.
Further research should explore the cost-effectiveness of the intervention, to support the viability.
Statement of interests
No conflicts of interest to declare.
Links
Primary paper
Stevens, G. J., Sperandei, S., Carter, G. L., Munasinghe, S., Hammond, T. E., Gunja, N., de la Riva, A., Brakoulias, V., & Page, A. (2024). Efficacy of a short message service brief contact intervention (SMS-SOS) in reducing repetition of hospital-treated self-harm: randomised controlled trial. The British Journal of Psychiatry, 224(3), 106–113. https://doi.org/10.1192/bjp.2023.152
Other references
Australian Institute of Health and Welfare. (2019). The health impact of suicide and self-inflicted injuries in Australia. https://www.aihw.gov.au/reports/burden-of-disease/health-impact-suicide-self-inflicted-injuries-2019/contents/about
Australian Institute of Health and Welfare. (2023). Emergency department care. https://www.aihw.gov.au/reports-data/myhospitals/sectors/emergency-department-care
Bhui, K., McKenzie, K., & Rasul, F. (2007). Rates, risk factors & methods of self harm among minority ethnic groups in the UK: a systematic review. BMC Public Health, 7(1), 336. https://doi.org/10.1186/1471-2458-7-336
Duan, S., Wang, H., Wilson, A., Qiu, J., Chen, G., He, Y., Wang, Y., Ou, J., & Chen, R. (2020). Developing a Text Messaging Intervention to Reduce Deliberate Self-Harm in Chinese Adolescents: Qualitative Study. JMIR MHealth and UHealth, 8(6), e16963. https://doi.org/10.2196/16963
Gentili, A., Failla, G., Melnyk, A., Puleo, V., Tanna, G. L. Di, Ricciardi, W., & Cascini, F. (2022). The cost-effectiveness of digital health interventions: A systematic review of the literature. Frontiers in Public Health, 10. https://doi.org/10.3389/fpubh.2022.787135
Geulayov, G., Casey, D., Bale, E., Brand, F., Clements, C., Farooq, B., Kapur, N., Ness, J., Waters, K., Patel, A., & Hawton, K. (2022). Socio-economic disparities in patients who present to hospital for self-harm: patients’ characteristics and problems in the Multicentre Study of Self-harm in England. Journal of Affective Disorders, 318, 238–245. https://doi.org/10.1016/j.jad.2022.08.106
Geulayov, G., Casey, D., Bale, L., Brand, F., Clements, C., Farooq, B., Kapur, N., Ness, J., Waters, K., Tsiachristas, A., & Hawton, K. (2019). Suicide following presentation to hospital for non-fatal self-harm in the Multicentre Study of Self-harm: a long-term follow-up study. The Lancet Psychiatry, 6(12), 1021–1030. https://doi.org/10.1016/S2215-0366(19)30402-X
Homer, C. S. E. (2002). Using the Zelen design in randomized controlled trials: debates and controversies. Journal of Advanced Nursing, 38(2), 200–207. https://doi.org/10.1046/j.1365-2648.2002.02164.x
Liu, R. T., Sheehan, A. E., Walsh, R. F. L., Sanzari, C. M., Cheek, S. M., & Hernandez, E. M. (2019). Prevalence and correlates of non-suicidal self-injury among lesbian, gay, bisexual, and transgender individuals: A systematic review and meta-analysis. Clinical Psychology Review, 74, 101783. https://doi.org/10.1016/j.cpr.2019.101783
McManus, S., Bebbington, P., Jenkins, R., & Brugha, T. (2016). Mental Health and Wellbeing in England: the Adult Psychiatric Morbidity Survey 2014. https://openaccess.city.ac.uk/id/eprint/23646/1/
National Institute for Health and Care Excellence. (2022). Self-harm: assessment, management and preventing recurrence NICE guideline [NG225]. https://www.nice.org.uk/guidance/ng225
Pines, J. M., Hilton, J. A., Weber, E. J., Alkemade, A. J., Al Shabanah, H., Anderson, P. D., Bernhard, M., Bertini, A., Gries, A., Ferrandiz, S., Kumar, V. A., Harjola, V.-P., Hogan, B., Madsen, B., Mason, S., Öhlén, G., Rainer, T., Rathlev, N., Revue, E., … Schull, M. J. (2011). International Perspectives on Emergency Department Crowding. Academic Emergency Medicine, 18(12), 1358–1370. https://doi.org/10.1111/j.1553-2712.2011.01235.x
Witt, K., de Moraes, D. P., Salisbury, T. T., Arensman, E., Gunnell, D., Hazell, P., Townsend, E., van Heeringen, K., & Hawton, K. (2018). Treatment as usual (TAU) as a control condition in trials of cognitive behavioural-based psychotherapy for self-harm: Impact of content and quality on outcomes in a systematic review. Journal of Affective Disorders, 235, 434–447. https://doi.org/10.1016/j.jad.2018.04.025
Zelen, M. (1979). A New Design for Randomized Clinical Trials. New England Journal of Medicine, 300(22), 1242–1245. https://doi.org/10.1056/NEJM197905313002203
