Bedroom Monitoring System Faced Consent Failures in Essex Mental Health Wards
A bedroom surveillance system designed to track mental health inpatients more closely has come under scrutiny after an inquiry heard that patients who objected to being watched could wait up to 72 hours for a decision on whether the cameras would be switched off. In practice, the system was running automatically on admission in some wards at Essex Partnership University NHS Foundation Trust, with consent treated as implied rather than explicitly obtained. That approach has now been conceded as wrong by the trust, which introduced an opt-in policy in August 2026.
The case has exposed a basic tension in modern inpatient psychiatry: technology can reduce risk, but it can also turn a bedroom into a watched space where autonomy feels conditional. In a mental health unit, where trust, dignity and de-escalation are part of treatment, consent cannot be treated as a bureaucratic afterthought.
Oxevision Used Infrared Sensors and Bedroom Cameras to Track Patients
The system at the centre of the inquiry, Oxevision, uses infrared sensors and cameras in patients’ bedrooms to capture images and measure pulse and breathing rates. It does not record sound, and staff monitor the feeds on tablets and computers. LIO Health, the company behind the technology, said the system was intended to support patient safety but must not replace in-person checks or the presence of staff on wards.
That safeguard matters because inpatient mental health care depends on human observation, not simply data points. A pulse reading or a motion alert can inform a response, but it cannot replace the clinical judgment that comes from seeing posture, affect, breathing pattern, agitation or withdrawal in context. When technology starts to substitute for bedside review, the risk is not only technical error but a gradual thinning of clinical presence.
The inquiry heard that NHS England identified 57 incidents nationally in which staff used the system’s 15-second video function instead of checking patients in person. The company said that use went against its guidance. A monitoring tool built to supplement care can quickly become a shortcut when wards are understaffed, overstretched or under pressure to prove surveillance is happening.
Regulator Raised Safety Concerns Over Claims and Performance
NHS England raised serious patient safety concerns about Oxevision with the Medicines and Healthcare products Regulatory Agency in 2021. Those concerns included the way the system was being used and claims about what it could do. Dr Robert Read, the MHRA’s deputy director for innovative devices, said the manufacturer had claimed Oxevision reduced self-harm. Clinical evidence requested by the regulator was not provided, and the claim was removed.
False pulse readings were also reported to the British Standards Institution in 2023. In a device that is marketed around safety and continuous vigilance, even a small pattern of inaccurate physiological data can erode confidence in the whole system. A wrong pulse value is not a harmless software glitch if it shapes escalation decisions, staffing responses or the sense that a patient is stable when they are not.
Laura Cozens, head of patient safety at LIO Health, accepted that more research was needed. She said that research should happen while Oxevision is being used so patients do not miss out on any potential benefits. That is a defensible argument only if the safeguards are real, the consent is explicit, and the technology is not allowed to outpace the evidence supporting it.
Patients Described Safety Benefits, but Consent Remained the Core Problem
Essex Partnership University NHS Foundation Trust said feedback from trialling the technology found 75% of patients thought it made them safer and 70% believed it improved their wellbeing. The trust spent a net total of £4.3m on Oxevision between 2020 and 2025, covering installation, software fees and training. EPUT and LIO Health said the system had helped staff intervene during incidents and save lives.
Those claims sit beside a more uncomfortable clinical question: whether a patient can genuinely feel safe while being watched in their bedroom without clear, informed agreement. In mental health care, safety and coercion can sit close together. A patient may accept monitoring because the ward is unfamiliar or because declining it seems futile. That is not the same thing as freely given consent.
Zephan Trent, the trust’s executive director of strategy, transformation and digital, told the inquiry that consent to use the system should not have been implied at admission and should have been explicit informed consent. The phrasing is direct because the issue is direct. If a patient enters a ward and automatically enters a monitored room, the default has shifted away from choice and toward assumption.
A National Debate Over Digital Surveillance in Inpatient Care
The company said the technology was used in about half of NHS mental health inpatient units, which makes the inquiry’s findings relevant well beyond Essex. When a device becomes widespread before its use is standardized, the same system can function as a clinical aid in one place and as a quasi-obligatory surveillance tool in another. The difference lies not in the hardware but in the governance around it.
NHS England said responsibility for using the technology rested with individual trusts. That decentralization leaves patients vulnerable to uneven practice, especially when policies on consent, access, override and monitoring are not clearly aligned. A system installed for safety can create a new category of harm if patients are effectively unable to decline it in the moment they need privacy most.
The Lampard Inquiry is examining the deaths of more than 2,000 people who were inpatients in mental health units in Essex in the 24 years to the end of 2023. The current hearing is focused on possible interim recommendations, and families and other core participants are due to make submissions at a virtual hearing on Tuesday. Baroness Lampard will then consider whether measures are needed to protect current patients.
For now, the unresolved issue is not whether monitoring technology can ever be useful. It is whether a system built to observe vulnerable patients can be governed with enough transparency, consent and bedside discipline to avoid becoming another layer of institutional control.