Safewards: 10 years on and what have we got?

On Thursday 5 September 2013, Len Bowers, Professor of Psychiatric Nursing at King’s College London, stood at the lectern in a hall at Oxford University to present the results of the cluster randomised controlled trial of “Safewards,” a nurse-led intervention designed to reduce conflict and containment on acute mental health wards.

In the audience, sat Ben Hannigan, a mental health nurse academic who later wrote on his blog:

Len Bowers used his Thursday keynote to share, for the first time anywhere, results from his Safewards trial. Len is a genuinely world leading researcher, and Safewards is a big and important study with seriously major implications for policy, services, education, education and practice. Take note, inpatient mental health nurses: the findings from this one are coming your way. https://benhannigan.com/2013/09/

Two weeks before, Len Bowers did not know whether he had results worth sharing. Having arranged to launch the outcome of this important study at the UK’s foremost mental health nursing research conference, the trial statisticians were still checking and double checking their independent analysis. Len did not have the all-important numbers for the presentation slides.

Safewards was the culmination of 20 years work trying to understand and explain the variation between wards in the rates of conflict behaviours and use of containment measures, both of which cause harm, injury, and expense. The work had involved numerous reviews of literature, qualitative interview studies, quantitative surveys, intervention and longitudinal studies, a multivariate cross-sectional mixed methods study with 128 wards across England, design and testing of new measures, and national and international collaborations.

These studies challenged any idea that patient conflict and staff containment were solely due to patient acuity. Multiple factors, such as managing the physical environment, or staff not explaining the ward structure to patients, were also contributors to increased conflict. The researchers also realised the patterns of conflict behaviour in patients corelated with the use of containment methods by staff. While patient conflict could lead to staff containment, it was also the case that staff use of containment could lead to patient conflict.

The outcome was an evidence-based, theoretical model of conflict and containment on acute mental health wards and the complex relationship between them (Bowers, 2014). The Safewards Model highlights six factors that can generate “flashpoints” – precursors or triggers for conflict and/or containment. These six factors are: the staff team, the physical environment, factors outside hospital, the patient community, patient characteristics, and the regulatory framework. The model hypothesised that wards that reduced and or managed flashpoints better had less conflict and containment.

This dynamic model systematically showed how understanding its central principles can inform strategies that promote the safety of patients and staff, resulting in healthier work and recovery environments.

Informed by the model, approximately 300 interventions to reduce or manage flashpoints, were identified, rated, and compared by researchers. A facilitated group of service users and carers with experience of inpatient care (called Service User and carer Group Advising on Research (SUGAR)) (Simpson et al., 2013) and expert professionals were consulted before selecting 16 potential interventions for pilot testing. Following a promising pilot, the wards involved advised on 10 interventions they felt were both practical and achievable within the average acute ward setting.

Researchers tested the 10 Safewards interventions in a pragmatic cluster randomised controlled trial in 31 wards at 15 hospitals. It demonstrated that Safewards produced a 15% decrease in the rate of conflict and a 26% decrease in the rate of containment (Bowers et al., 2015). By lessening the incidents of both conflict and containment, Safewards benefits both staff and patients by making their ward safer for both groups.

Len Bowers now had his results, they were definitely worth sharing, and Safewards was announced. Following the successful launch, in December 2013, the Safewards website www.safewards.net went live, detailing the model and interventions and advice on implementation. Dissemination through the website and a range of social media platforms made Safewards freely accessible with no registration requirement. This minimised barriers and maximised take-up. This also means no records of the many organisations and wards that have implemented Safewards through the free material. However, the Safewards team are aware of implementation in multiple state and private sector service providers across the UK, with individual services reporting reductions in assaults, seclusion, restraints and rapid tranquilisation (KCL, 2024).

Safewards was further recognised when recommended in the National Institute for Health and Care Excellence (2015) Guideline NG10: “Violence and aggression. Short-term management in mental health, health, and community settings.” In this guideline Safewards was identified as one of only two conflict and containment management interventions with good evidence to recommend their use. Similarly, the UK’s Department of Health (2014) recommended that all mental healthcare providers should consider the Safewards interventions for their services and, on the back of their inspecting of mental health services, the Care Quality Commission (CQC) endorsed it as a “good initiative to embrace a culture of safety” (CQC, 2017).

Implementing Safewards internationally

The easy availability of the model and guidelines on implementation meant that Safewards was swiftly adopted in numerous countries across the world. In the state of Victoria, Australia, $2.4 million was invested in Safewards implementation, initially in 58 mental health units, with a second phase that included emergency departments and acute medical units. The intervention had a large impact on reducing seclusion rates and an accompanying implementation evaluation identified key factors that facilitate or hinder implementation (Fletcher et al., 2017; Fletcher et al., 2021). Also in Australia, a study in one large metropolitan local health district in New South Wales resulted in reductions of conflict and containment of 23.0% and 12.0%, respectively (Dickens et al., 2020). Following the Victoria investment, Safewards has been taken up across Australia in a variety of mental health and general healthcare settings.

In Denmark, Safewards has been implemented in around half of all mental health units (including the Faroe Isles), reducing forced sedation and mechanical restraints. Additionally, Safewards has been implemented in Canada, Holland, Germany, Austria, Switzerland, Sweden, Norway, Finland, Malta, Ireland, The Czech Republic, Slovakia, and Spain. In Germany, Austria and Switzerland, a network of Safewards implementors has driven the uptake across multiple services. This has also been the case in Sweden and the Czech Republic.

In Tasmania, Finland, and Belgium Safewards interventions are either recommended or are being actively implemented in hospitals to drive quality improvements in mental health inpatient care. In 2019, the Ministry of Health of the Republic of Indonesia sent a team of doctors and nurses to the UK to learn about Safewards for use in their services.

Many of these international colleagues have contributed to the Safewards website and training materials being translated into German, Turkish, Dutch, Finnish, Danish, Polish, Czech, and Spanish, with thousands of international visitors to the website every year. In 2023, the British Council provided funding for the material to be translated into Chinese as part of a larger implementation project.

Literature reviews

The international evidence base for Safewards has proliferated and three reviews have now been published. Finch et al.’s (2021) systematic review included 13 studies and concluded that Safewards generally has a positive effect and can help reduce conflict and containment in acute ward settings, whilst recognising the need for more high-quality research. Ward-Stockham et al. (2022) identified 14 studies in a scoping review that evaluated effectiveness as well as staff and service users’ perceptions, and found that Safewards improved cohesion, therapeutic relationships, and ward atmosphere.

Mullen et al. (2022) conducted an integrative literature review of Safewards within inpatient and forensic mental health units and included 19 papers. They concluded that Safewards can be effective in reducing containment and conflict within inpatient mental health and forensic mental health units, although this outcome varied across the literature. This review also revealed the limitations of fidelity measures, the importance of involving staff in implementation and the need for greater inclusion of patient perspectives.

Further developments

Safewards has been taken up across a range of services in multiple countries. However, in doing so, it has also moved from its original context – acute inpatient mental healthcare.

A recent project commissioned by NHS England to explore the possible use of Safewards in Children and Young People’s (CYP) Mental Health services demonstrated that the interventions were both acceptable and adaptable (Simpson et al., 2024). Work co-ordinated by Dr Tessa Maguire and her team in Australia has shown comparable results in forensic and secure mental health services and they have published adaptations for such settings (Maguire et al., 2018, 2022).

Safewards is now being used in other health contexts – both mental health and general health settings, and there is a growing need for further research to explore if the use of Safewards produces similar outcomes, in terms of reduction of conflict and containment, as seen in acute inpatient mental health.

There is also a need to gather more evidence on the experience and use of Safewards from the patient perspective. While there are a few quality critiques from Canada (Mendolia et al., 2017) and Australia (Kennedy et al., 2019) as well as adaptions co-designed with the Māori community in New Zealand (Knauf et al., 2023), there remains much to learn from experts by experience and carer communities.

The Safewards journey is now 10 years running and counting. While there is much to celebrate in these 10 years, there are also questions and challenges that should occupy our thinking for the years to come. Finally, while Safewards indicates that we can actively reduce both conflict and containment, there remains an open question as to what more we can achieve and how best to do this.

Researchers’ neglect of inpatient care

Researchers have tended to give little attention to acute inpatient wards. Perhaps, it’s the challenge of designing and testing complex interventions in potentially volatile and risky environments, or maybe there is the lingering stigma attached to psychiatric hospitals, their patients and even staff.

Two studies published in JMH identified the nature, purpose, aims and functions of acute psychiatric wards and suggested that admissions to acute inpatient mental health wards typically result from severe acute mental health problems coupled with one or more of the following factors: high risk to self or others, refusal of treatment, life stressors, concerns about deterioration, failure to manage activities of daily living, and need for assessment (Bowers et al., 2009).

Some of the theorised key functions of inpatient mental health care include ensuring safety, providing assessment and psychiatric treatment, rehabilitation, meeting basic care needs, and physical health care. Support may also be provided for other issues, such as financial or housing problems. Key components of inpatient care include: 24-h staff presence, multidisciplinary team input, treatment provision, regular observation, containment, and tolerance of behaviour that would be unacceptable or unmanageable in the community. Inpatient admissions may also provide respite for patients, families, and communities (Bowers et al., 2005).

One relatively new area for research attention in inpatient mental healthcare settings has been the introduction of surveillance technologies. Digital technologies, such as body-worn cameras and vision-based patient monitoring and management (VBPMM) systems appear to be attractive to service managers, perhaps with half an eye on staffing shortages and budgets and may offer some support in close monitoring of patients at risk. One such study in this edition of JMH explores their use in reducing self-harm incidents (Gandhi, 2024). However, these developments are contentious, and various concerns have been raised by patients and staff (Foye, Regan, et al., 2024; Foye, Wilson, et al., 2024; Simpson, 2023).

At the heart of many concerns is the possible impact on the human rights of patients under the Human Rights Act (HRA). The HRA applies to all patients under our care and includes rights such as the right to be free from inhuman or degrading treatment, the right to respect for privacy, and the right to be free from discrimination. It is entirely reasonable that being subject to vision monitoring systems may be experienced as dehumanising and to impact on a person’s sense of dignity and privacy (McEntee, 2024). Possible challenges to employing a human rights approach in acute inpatient settings are explored in a new paper in this edition (Davis Le Brun, 2024), which raises important questions for anyone working in this integral part of our mental health services.

There is strong evidence that Safewards can help create calmer, safer environments without the intrusiveness of surveillance technologies and perhaps this approach deserves more attention and investment.

Disclosure statement

Both Alan Simpson and Geoff Brennan continue to be involved in implementation, training and evaluation studies for Safewards. We also manage the Safewards website and social media accounts. We do not have any financial interests.

Additional information

Funding

The author(s) reported there is no funding associated with the work featured in this article.