Special Report to the Storting Document 4:3 (2018–2019) 68 9.5 Control of the use of security cells and restraint beds Any decision to use coercive measures shall be reported to the prison governor. All use of coercive measures shall be reported to the Directorate of the Norwegian Correctional Service on an annual basis. Stays in a security cell shall be reported to the regional level if they exceed three days, and to the Directorate if they exceed six days.208 Use of restraint beds for more than 24 hours shall be reported to the regional level, and any use exceeding 72 hours to the Directorate. Separate logs shall be kept in connection with all use of coercive measures. Up until March 2019, when the ­Directorate issued new guidelines, there were no requirements for the content of these logs. This has now been changed and requirements introduced for logging a number of important circumstances, including clothing, supervision, motivational conversations and ­assessments of whether the measure should be upheld.209 A requirement has also been introduced for logging the times of all supervision by the prison health service. Restraint bed. In some places, we have found, among other things, that information is missing about the times of ­supervision by the health service. This made it difficult for both the prison administration and the supervisory authorities to check whether there had been any ­follow-up of the inmate’s state of health. It was repeatedly found during our visits that administrative decisions to use a security cell lacked a description of other, less intrusive measures that had been tried or assessed.210 In some prisons, the Parliamentary Ombudsman found administrative decision based on grounds that were so inadequate that it is doubtful that they met the stringent requirements for use of security cells that are laid down by law.211 During previous visits to the prisons, the ­Parliamentary Ombudsman has found weaknesses in some of the logs. There are many instances of incomplete ­information. An inmate is for example stated to be lying in the same position, without any indication of whether the inmate has been checked for signs of breathing or movement. Many logs lack a description of how the inmate has been followed up in conversations. This is particularly important in order to be able to assess how the inmate is being followed up, how much human contact the inmate has and what systematic efforts are made to bring the stay in the security cell to an end. In addition to weaknesses in content, we have found major inadequacies in the logging system. Among other things, we have found that several different logging systems are used. The most common form appears to be that supervision is logged in a separate book on a running basis. In several prisons, we have found that pages have been torn out from the log books, or that there are no page numbers so that it is difficult to determine whether the log is complete. In some places, it is uncertain how many days the inmate has spent in the security cell, because there is no information about when the stay was terminated. We have also seen logs according to which all supervision has been carried out every half hour on the minute. This leaves the ­impression that the log was written in advance and does not reflect actual supervision. 208 Directorate of the Norwegian Correctional Service (2002). Guidelines to the Execution of Sentences Act and its Regulations, section 38.7. Revised version of 15 March 2019. 209 Directorate of the Norwegian Correctional Service (2002). Guidelines to the Execution of Sentences Act and its Regulations, section 38.7.4. Revised version of 15 March 2019. 210 The grounds for this must be stated in the decision, see the Public Administration Act Section 25. 211 The Parliamentary Ombudsman’s report after its visits to Bergen 2–8 May 2018, p. 16; and to Trondheim Prison 17–19 March 2015, p. 8. 69 In some instances, the log is kept on loose sheets of ­paper that are put into a binder together with the relevant administrative decision. In other places, the log is kept as a Word document for the duration of the stay and converted into a PDF file when the stay is terminated. Some prisons keep supervision protocols in the ­Correctional Service’s computer system KOMPIS. This means that neither the prison nor the supervisory ­council will have access to the information once the inmate leaves or is transferred from the prison in question. In one prison, some staff logged supervision in KOMPIS, while others logged supervision in a separate book; some logged it in both places, though the information was not always the same.212 All the solutions have obvious weaknesses as regards the possibility of internal and external supervision. Among other things, the Parliamentary Ombudsman has pointed out that: ‘Use of a security cell or restraint bed should be logged in such a way as to ensure correct and ­complete ­documentation and prevent any subsequent ­corrections’.213 Inadequate procedures and unclear systems for keeping records have made the prisons’ control of their own practices difficult, and made it impossible for super­visory bodies and the Parliamentary ­Ombudsman to fully carry out their control functions. In several prisons visited by the Parliamentary Ombudsman, the supervision protocols showed that inmates had been placed in security cells without an administrative decision. In other places, records of security cell usage have been missing. This is a very serious matter. It is the ­responsibility of central government authorities to put into place a verifiable system for keeping records of the use of security cells. Main findings The use of security cells and restraint beds entails a high degree of sensory deprivation and a risk of harmful effects on health. Our findings include several examples of situations where the use of security cells can have extreme negative impact, even after a relatively short amount of time. They also show that it takes much too long before inmates receive help to get out of unhygienic and degrading conditions. Findings from our visits show that inmates do not receive the help they need from the prison to bring their confinement to an end as soon as possible. Inmates can thus be held in security cells for longer than strictly necessary. Training and guidelines are lacking on how staff should follow up inmates to ensure maximum consideration for their welfare and as brief a stay as possible. It is particularly censurable that individuals in an acute life crisis and who wish to harm themselves or take their own life are placed in a security cell without ­satisfactory follow-up. Informed by research, our findings show that the risk of harm is greater where inmates are confined on account of a mental crisis. Inmates who are strapped to a restraint bed risk being traumatised in an acute life crisis. Close follow-up is therefore needed. Despite the risk of both mental and physical harm associated with the use of security cells and restraint beds, the Execution of Sentences Act and its Regulations do not contain any requirements for staff supervision in connection with such measures.214 Inadequate procedures and unclear systems for keeping records make the prisons’ control of their own practices difficult, and prevents supervisory bodies and the Parliamentary Ombudsman from carrying out their control functions. 212 The Directorate of the Norwegian Correctional Service’s updated guidelines require that records shall be quality-assured by the prison governor and that the log shall be scanned and saved in Doculive, the Correctional Service’s archiving system. It is not yet clear whether the new guidelines will address the concerns that the Parliamentary Ombudsman has had about record-keeping. 213 The Parliamentary Ombudsman’s reports after visits to Telemark Prison, Skien branch 2–4 June 2015 p. 12, and Ullersmo Prison 29–31 August 2017, p. 21. 214 The guidelines as amended in March 2019 now require continuous supervision by prison officers. See Directorate of the Norwegian Correctional Service (2002). Guidelines to the Execution of Sentences Act and its Regulations, section 38.7. Revised version of 15 March 2019.