New inspection report of facilities used to hold immigration detainees following arrest and before transfer to residential detention
His Majesty's Inspectorate of Prisons for England and Wales (HMI Prisons) this week published an inspection report of short-term holding facilities (STHFs) attached to immigration reporting centres.
You can download the 35-page report here.
The inspection, carried out by HMI Prisons between 20 and 29 October 2025, examined all 10 holding rooms across the UK: Capital Building in Liverpool, Dallas Court in Salford, Drumkeen House in Belfast, Eaton House in Hounslow, Loughborough, Festival Court in Glasgow, Ruskin Square in Croydon, Sandford House in Solihull, Vulcan House in Sheffield, and Waterside Court in Leeds. The facilities are used to hold immigration detainees following arrest or reporting and before transfer to residential detention.
While HMI Prisons found that some progress had been made since its last inspections in identifying risks and vulnerability among detainees (particularly through the work of the national Detention Gatekeeper), it concluded that there was still not a sharp enough focus on safeguarding, either in decisions to detain or in making use of available support mechanisms.
The inspection identified nine key concerns, two of which were designated as priorities requiring immediate attention. Inspectors found that, in some cases, the Home Office took insufficient or no account of known vulnerabilities when deciding whether to detain an individual, and that healthcare provision was inadequate, with no routine health screening and delays in access to medication.
HMI Prisons says identification and consideration of vulnerability were not always appropriate or effective. Although 7.6% of detainees held in the six months to August 2025 had been assessed at higher levels of risk in detention, this was lower than in immigration removal centres and did not always reflect individual circumstances. Detention paperwork in some cases failed to specify known vulnerabilities, and arrest teams did not consistently pass on relevant information to the national Detention Gatekeeper, whose role is to prevent inappropriate detention.
The report describes the Detention Gatekeeper as playing a productive role overall, but highlights serious failures in some cases. In one instance, the Gatekeeper had refused to authorise detention for a woman who disclosed sexual abuse and torture, yet local immigration staff detained her, during which she self-harmed; she was released later that day, and the Home Office subsequently recognised her as a victim of modern slavery. In other cases, the Gatekeeper authorised detention without fully considering previous risk assessments or conclusive decisions identifying individuals as vulnerable.
Staff training in adult safeguarding was in place, but HMI Prisons found it was not tailored to the immigration context. Some detainee custody officers were unfamiliar with the Home Office policy on adults at risk, and most had little knowledge of the National Referral Mechanism. Use of vulnerable adult warning forms was inconsistent and appeared to understate the level of need, with significant disparities between sites.
Personal safety incidents, while relatively few, exposed further weaknesses. Seven self-harm incidents had been recorded in the previous year, though inspectors found that not all incidents were logged. In one case, a detainee with a history of mental ill health was not initially subject to safeguarding measures despite clear indicators of risk, later claiming to have swallowed a blade. In another, a detainee found in severe distress after attempting to ligature was transferred hours later without medical assistance, on the basis that healthcare would be available more quickly at his destination. HMI Prisons said lessons were not always learnt from such incidents.
Alongside safeguarding, inspectors raised concerns about detainees' legal rights and the length of detention. The average detention lasted four hours and 43 minutes, but some people were held for almost 13 hours and the average length of detention was rising. Women were held for nearly an hour longer on average than men, and there was wide variation between centres. Delays were often caused by late movement orders or transport arrangements, resulting in detainees being transferred late at night after being held since the morning.
Of those detained in a recent three-month period, nearly three-quarters were ultimately released rather than removed. The report notes: "In all, 1,187 detainees had been detained in the three months to the end of June 2025. The great majority were transferred to an IRC for removal. At the time of the inspection five remained in detention. Of the rest, 850 (72%) had ultimately been released and 332 (28%) removed."
While detainees were told they could contact legal representatives, access to advice was uneven. Lists of solicitors were displayed, but some did not offer immigration advice and some STHFs listed no providers offering free legal aid. Detention paperwork was issued in English only, with interpretation used to explain it where necessary.
The report also questions the proportionality of restraint practices. Although staff said force was rarely used, inspectors found that handcuffs were almost always applied when escorting detainees to vans, even in secure areas and in circumstances where arrest teams had not used restraints earlier in the day. Risk assessments were often treated as a formality, and some use of force incidents were not properly recorded or scrutinised.