Long-term segregation in mental health hospitals is framed as containment, not care
Long-term segregation in mental health hospitals in England is used to manage people judged to pose a sustained risk of harm to others. In practice, the qualitative evidence collected across the HOPE(S) programme portrays a restrictive intervention that many participants experienced as profoundly harmful, dehumanising, and stripped of therapeutic purpose. The study drew on perspectives from people with lived experience, family members, HOPE(S) practitioners, clinical staff, commissioners, and regulators, giving the findings unusual breadth across the care system rather than a single professional vantage point.
A central message runs through the interviews and analysis with little ambiguity: long-term segregation was not understood by most participants as treatment. It was described instead as isolation that eroded dignity, increased distress, and deepened already complex clinical and social vulnerability. In a setting where psychiatric care should reduce suffering, the experience of prolonged segregation often did the opposite.
The timing of this work matters because the HOPE(S) programme was introduced nationally for 2022 to 2025 to reduce, and ultimately end, the use of long-term segregation for autistic people, individuals with learning disabilities, and children and young people. The study examined whether that policy ambition aligned with what patients, relatives, and staff were actually seeing on the ground.
Four themes capture the human cost of isolation
Reflexive thematic analysis produced four interrelated themes: dehumanisation and erosion of personhood; safeguarding and systemic failure; psychological and relational harm; and loss of hope and systemic inertia. Each theme describes a different layer of damage, yet all point to the same underlying problem: segregation can become a structural answer to clinical complexity when services lack safer alternatives.
Dehumanisation surfaced as more than emotional discomfort. Participants described loss of ordinary personhood, with people subjected to conditions that made them feel invisible, controlled, and reduced to the risk they were thought to represent. In a psychiatric system, that kind of reduction has direct consequences. Once a person is seen primarily through the lens of risk containment, opportunities for rapport, rehabilitation, and meaningful observation shrink quickly.
Safeguarding and systemic failure formed another recurring thread. The accounts suggested that long-term segregation often persists not because it works, but because services struggle to manage escalation, staffing pressure, environmental limitations, and fragmented responsibility. In that context, segregation can become a default response when the system cannot provide individualized support, sensory adjustment, or skilled de-escalation. The result is a form of institutional inertia in which restrictive practice outlives any credible therapeutic rationale.
Psychological injury, broken relationships, and the collapse of hope
The psychological burden described in this study was severe. Participants linked segregation to distress, fear, deterioration in mental well-being, and damage to trust between patients and services. Family members and staff also described relational rupture, with prolonged isolation separating people from the relationships that ordinarily buffer psychiatric crisis.
A single sentence captures the clinical reality: prolonged isolation did not stabilize people for many participants, it injured them.
That injury is not abstract. Human beings regulate distress through connection, predictability, and a sense of agency. When those elements are removed for long periods, especially in people already experiencing mental health needs, autism, or learning disability, the risk of worsening agitation, despair, and mistrust rises rather than falls. The study’s qualitative evidence aligns with that biological and psychological plausibility, even without pretending to measure symptom trajectories or event rates.
Loss of hope emerged as a separate but related harm. Once segregation becomes prolonged, patients can stop expecting change, families can lose confidence in services, and staff may normalize a practice they would otherwise question. Systemic inertia then feeds itself. The longer the practice continues, the harder it becomes to imagine a different model, even when the ethical case for change is overwhelming.
Rights violations sit at the center of the analysis
The findings were interpreted through a human rights lens and were judged to reflect breaches of the Convention on the Rights of Persons with Disabilities and the Convention on the Rights of the Child. That rights-based framing is not decorative language. It shifts the issue from service preference to legal and ethical obligation.
Long-term segregation collides with core rights principles because it can deny dignity, participation, and equitable treatment while exposing people to harm in an environment they cannot meaningfully leave. For children and young people, the stakes are even sharper. A restrictive measure that can shape development, attachment, and trust is difficult to reconcile with any child-centered model of care, especially when it is deployed for extended periods rather than brief emergency stabilization.
The study positions this practice as incompatible with a therapeutic mental healthcare system that claims to protect vulnerable people. The fact that participants across multiple stakeholder groups converged on similar descriptions strengthens the ethical weight of the findings. This was not simply a dispute over terminology. It was a shared account of suffering and exclusion.
HOPE(S) points toward rights-based alternatives, but the system must move
The HOPE(S) model was introduced to reduce and ultimately end long-term segregation for autistic people, individuals with learning disabilities, and children and young people. In this qualitative evaluation, the need for such reform appeared urgent rather than aspirational. Participants repeatedly described the absence of safer, more responsive alternatives as a driver of continued segregation.
Rights-based, trauma-informed care offers a different logic. Instead of controlling behavior through prolonged isolation, it aims to understand distress, adapt environments, support communication, and build relational safety. Those approaches require staffing, training, clinical leadership, and a willingness to treat distress as meaningful communication rather than as a reason to withdraw a person from care.
The study does not leave room for comfort with the status quo. It presents long-term segregation as a practice that causes profound psychological harm, strips away dignity, and sits uneasily with human rights standards. The unanswered operational question is not whether alternatives are needed, but whether hospitals, commissioners, and regulators will implement them with enough consistency to make segregation obsolete rather than merely discouraged.