Introduction
Mental health legislation that permits admission and treatment without consent under certain circumstances is a long-standing feature of psychiatry, and sometimes involves restrictive practices such as physical restraint, mechanical restraint, and seclusion (Kelly, Reference Kelly2025). There is a long-standing consensus that involuntary admission, treatment without consent, and other restrictive practices (e.g. physical restraint, seclusion) should be minimised and, ideally, eliminated both in Ireland (Mental Health Commission, 2014) and internationally (World Health Organization, 2012; Gill et al. Reference Gill, Drew, Rodrigues, Muhsen, Morales Cano, Savage, Pathare, Allan, Galderisi, Javed, Herrman and Funk2024).
Rates of involuntary admission vary internationally. In 2019, one cross-jurisdictional study found that the median rate of involuntary hospitalisation was 106·4 per 100,000 population per year, with Austria having the highest rate (282 per 100,000 population per year) and Italy the lowest (14·5 per 100,000 population per year) (Sheridan Rains et al. Reference Sheridan Rains, Zenina, Dias, Jones, Jeffreys, Branthonne-Foster, Lloyd-Evans and Johnson2019). Ireland’s rate (57.5 per 100,000 population per year) was below the median and approximately half that of the United Kingdom jurisdictions (100.8 per 100,000 population per year). This is likely attributable to a combination of legislative, service-related, and social differences between the jurisdictions, including the explicit exclusion of personality disorder as a sole reason for involuntary admission in Ireland (but not in England and Wales) (Conlan-Trant & Kelly, Reference Conlan-Trant and Kelly2022). In 2023, there were 15,631 admissions to Irish psychiatric units and hospitals, of which 16% were involuntary, indicating a continued reduction in involuntary admissions, to 49.9 per 100,000 population in 2023 (Daly et al. Reference Daly, Lovett and Lynn2024).
Rates of restrictive practices such as physical restraint and seclusion also vary between jurisdictions. In 2024, one paper which examined worldwide data found that rates of coercive practices in mental healthcare (i.e. where such restrictive practices can be used in inpatient psychiatry units, psychiatric hospitals, and other permitted mental healthcare settings) were “highly variable” across jurisdictions, with population rates of physical restraint and seclusion varying “by a factor of more than 100 between countries” (Savage et al. Reference Savage, Lepping, Newton-Howes, Arnold, Staggs, Kisely, Hasegawa, Reid and Noorthoorn2024; p.4). The median rate of physical restraint in the countries studied was 57 episodes of physical restraint per 100,000 population per year, with the highest rate in England (110 per 100,000 population per year) and the lowest in Germany (0.2 per 100,000 population per year); Ireland’s rate (82 per 100,000 population per year) was above the median (57 per 100,000 population per year). Internationally, the median rate of seclusion was 48 episodes of seclusion per 100,000 population per year, with the highest rate in Japan (190 per 100,000 population per year) and the lowest in Wales (1.0 per 100,000 population per year); Ireland’s rate (38 per 100,000 population per year) was below the median (48 per 100,000 population per year). Rates of all coercive interventions were highest in Japan, which does not track physical restraint (making that median less reliable).
In Ireland, the Mental Health Commission collects data about the use of restrictive practices in “approved centres,” which are inpatient psychiatry facilities that treat voluntary and involuntary patients under Ireland’s Mental Health Act, 2001; that is, private psychiatry hospitals and public psychiatry inpatient services across Ireland’s community health organisations (CHOs) which serve discrete catchment populations.Footnote 1
The Mental Health Commission defines “physical restraint” as “the use of physical force (by one or more persons) for the purpose of preventing the free movement of a person’s body when the person poses an immediate threat of serious harm to self or others” (Mental Health Commission, 2022a; p. 7). In 2023, 879 patients experienced physical restraint in Ireland a total of 2,572 times (Mental Health Commission, 2024). This was a substantial reduction from 2018, when 1,207 patients experienced physical restraint a total of 5,665 times (Mental Health Commission, 2019).
The Mental Health Commission defines “seclusion” as “the placing or leaving of a person in any room, at any time, day or night, such that the person is prevented from leaving the room by any means” (Mental Health Commission, 2022b; p. 9). In 2023, 473 patients were placed in seclusion in Ireland a total of 895 times (Mental Health Commission, 2024). This was a substantial reduction from 2018, when 760 patients were placed in seclusion a total of 1,799 times (Mental Health Commission, 2019).
The Mental Health Commission defines “mechanical means of bodily restraint” as “the use of devices or bodily garments for the purpose of preventing or limiting the free movement of a person’s body” (Mental Health Commission, 2022c; p. 10). In 2023, just “two approved centres reported the use of mechanical restraint for immediate threat of serious harm to self or others,” with “a total duration of 17 hours and 46 minutes (≈ nine episodes) in 2023, compared to 39 hours (≈ 20 episodes) in 2022” (Mental Health Commission, 2024; p. 24). Given that “the frequency of mechanical restraint as a restrictive practice in approved centres remains low” in Ireland compared to other jurisdictions, mechanical restraint is not considered further in this paper.
Notwithstanding these reductions in restrictive practices and Ireland’s comparatively low rates of involuntary admissions and restriction, there is considerable variation in the use of restrictive measures within countries. We set out to examine rates of use of physical restraint and seclusion across Ireland’s acute adult approved centres in order to identify (a) challenges with the nature and quality of data that are publicly available; (b) potential lessons for Ireland and other jurisdictions from comparisons across approved centres (to whatever extent such comparisons are possible and informative), and (c) suggestions for future data collection and research.
Methods
This paper uses only data that were already in the public domain. We extracted the number of episodes of physical restraint and seclusion in each approved centre in 2023 from the Mental Health Commission’s report on The Use of Restrictive Practices in Approved Centres: Seclusion, Mechanical Restraint, and Physical Restraint (Mental Health Commission, 2024). We included only acute adult mental health admission units and excluded continuing care, rehabilitation and forensic units, as well as centres for children and adolescents. We accessed additional information about the type of service provided in each approved centre through the Mental Health Commission website.Footnote 2
We used the Health Research Board’s Annual Report on the Activities of Irish Psychiatric Units and Hospitals 2023 (Daly et al. Reference Daly, Lovett and Lynn2024) to identify each approved centre’s catchment population (based on the Central Statistics Office’s Census of Population 2022).Footnote 3 Given that two acute adult approved centres cover the single catchment of Dublin North Central, their numbers of episodes of physical restraint and seclusion were combined to produce a rate for the catchment-area population. We could thus compare restrictive practice rates across public acute approved centres, giving rates per 100,000 population per year.
To generate a national rate of restrictive practices in adult acute units per 100,000 population, we added psychiatric intensive care units and private and independent approved centres (which do not have set catchment populations) to the dataset. Forensic services, continuing care facilities, rehabilitation units, and centres for children and adolescents remained excluded.
Data were stored and described using Microsoft Excel. This was a descriptive study of public data so analytic statistics were not appropriate. Approved centres which reported total numbers less than 5 were estimated at 2 for the purpose of calculations. The population of Ireland (5,149,139 people) was obtained from the Central Statistics Office’s Census of Population 2022.
Results
Rates of physical restraint varied 16-fold between approved centres, ranging from 116 episodes per 100,000 population per year to 7 per 100,000 population (Table 1). When all types of adult acute units were included, a total of 2,017 episodes of physical restraint were recorded, yielding a national rate of 39 episodes per 100,000 population in 2023.
Table 1. Rates of physical restraint and seclusion in inpatient psychiatry facilities (“approved centres”) in Ireland (2023)
aNumbers <5 were designated as 2 for the purpose of calculation.
bComprises both acute and continuing care beds.
c HSE (Health Service Executive)-funded public beds in St John of God Hospital, Dublin.
CHO: Community health organisation (i.e. part of Ireland’s HSE public mental health service covering a discrete geographical area).
N/A: Not applicable.
Eight approved centres did not have a seclusion room or report episodes of seclusion. Among approved centres that used seclusion, rates varied 19-fold, ranging from 38 episodes per 100,000 population to 2 per 100,000 population. When all types of adult acute units were included, a total of 756 episodes of seclusion were recorded, yielding a national rate of 15 episodes per 100,000 population. There was no significant difference between the mean physical restraint rate in approved centres with seclusion rooms and those without (34 episodes of physical restraint per 100,000 population per year versus 31 episodes of physical restraint per 100,000 population per year, respectively).
Discussion
Rates of physical restraint vary 16-fold and rates of seclusion vary 19-fold across Ireland’s inpatient psychiatry facilities. The highest rate of physical restraint recorded in an Irish approved centre (116 per 100,000 population per year) is comparable with the highest rate internationally (110 per 100,000 population per year in England), and the lowest rate in Ireland (7 per 100,000 population per year) is higher than the lowest internationally (0.2 per 100,000 population per year in Germany) (Savage et al. Reference Savage, Lepping, Newton-Howes, Arnold, Staggs, Kisely, Hasegawa, Reid and Noorthoorn2024). Ireland’s national rate of physical restraint (39 per 100,000 population per year) is below the international median (57 per 100,000 population per year).
The highest rate of seclusion recorded in an Irish approved centre (38 per 100,000 population per year) is lower than the highest rate internationally (190 per 100,000 population per year in Japan), and the lowest rate in Ireland (2 per 100,000 population per year) is slightly higher than the lowest rate internationally (1 per 100,000 population per year in Wales) (Savage et al. Reference Savage, Lepping, Newton-Howes, Arnold, Staggs, Kisely, Hasegawa, Reid and Noorthoorn2024). Ireland’s national rate of seclusion (15 per 100,000 population per year) is below the international median (48 per 100,000 population per year).
Overall, there are within-country variations in rates of physical restraint and seclusion in Ireland, but these are of a lesser magnitude than between-country variations, based on worldwide data (Savage et al. Reference Savage, Lepping, Newton-Howes, Arnold, Staggs, Kisely, Hasegawa, Reid and Noorthoorn2024). In general, Ireland’s national rates of restrictive practices are lower than those of other jurisdictions, consistent with Ireland’s low rate of involuntary admission (Sheridan Rains et al. Reference Sheridan Rains, Zenina, Dias, Jones, Jeffreys, Branthonne-Foster, Lloyd-Evans and Johnson2019; Conlan-Trant & Kelly, Reference Conlan-Trant and Kelly2022; Daly et al. Reference Daly, Lovett and Lynn2024).
This study’s total episodes of physical restraint (2,017) and seclusion (756) differed from those reported by the Mental Health Commission for 2023 (2,572 and 895 respectively) (Mental Health Commission, 2024) due to our study excluding continuing care, rehabilitation, forensic, and child and adolescent facilities. This allowed us to determine a rate of restrictive practice in acute adult psychiatric units only and allowed comparisons with similar international figures which used similar criteria when possible (Savage et al. Reference Savage, Lepping, Newton-Howes, Arnold, Staggs, Kisely, Hasegawa, Reid and Noorthoorn2024). Our rates per 100,000 are less than the Irish rates presented by Savage and colleagues, who explored worldwide data, because they studied data from 2020 (Savage et al. Reference Savage, Lepping, Newton-Howes, Arnold, Staggs, Kisely, Hasegawa, Reid and Noorthoorn2024) and we used data from 2023. The persistent fall in physical restraint and seclusion in Ireland in recent years is consistent with the Mental Health Commission’s Seclusion and Restraint Reduction Strategy (Mental Health Commission, 2014; Lucey et al. Reference Lucey, Kiernan, Farrelly, Downey and Stepala2025).
In this study we used catchment population as a denominator, rather than bed numbers or admission numbers. There are other ways of presenting these figures which would yield different information; for example, analysing the numbers of patients who experience physical restraint or seclusion (rather than the number of episodes), using bed numbers or bed-days as the denominator, or calculating the aggregate duration of all physical restraint and seclusion episodes (in minutes) per 100,000 population. We chose to use episodes of physical restraint and seclusion with catchment-area population as the denominator in order to provide a more robust evaluation of restrictive practices in the populations served and to optimise comparability with published work (Savage et al. Reference Savage, Lepping, Newton-Howes, Arnold, Staggs, Kisely, Hasegawa, Reid and Noorthoorn2024).
As the Mental Health Commission points out, “approved centres vary in size, bed capacity, admission pathways and type of service delivered. Therefore, any attempt at comparative analysis between approved centres, types of service or geographical areas should be qualified, and should be undertaken cautiously” (Mental Health Commission, 2024; p. 12). The Mental Health Commission adds that “the variation between services can be due to a number of factors including differing practices and cultures” and “the range of de-escalation techniques available to, and employed within, a service.” Future data collection could usefully record the de-escalation options available to staff in different approved centres.
The same consideration applies to “ward design factors, such as the availability of intensive care and low-stimulus facilities, and the ward environment (décor, milieu, comfort),” “staff numbers (including lower nurse-patient ratios), skills mix, experience and training,” and “changes in service provision within an approved centre over time,” which the Mental Health Commission also highlights as potentially relevant. Future work could also seek to link rates of restrictive practices in approved centres with these factors, as well as access to psychiatric intensive care units.
The role of clinical demand is especially difficult to evaluate across centres, but differences in restrictive practices might relate to differing clinical characteristics of patients, due, for example, to age distribution, social deprivation, or urbanicity, which is associated with increased rates of restrictive practices internationally (Husum et al. Reference Husum, Bjørngaard, Finset and Ruud2010) and in our data. The Mental Health Commission suggests that “variations in the prevalence and acuity of mental illness, including the number of emergency and involuntary admissions,” might be relevant, along with “services in some areas treating more acute residents,” “the use of sedating psychotropic medication,” and “the frequent or prolonged seclusion or restraint of one resident, which could result in distorted figures.”
We found that the mean physical restraint rate for approved centres with seclusion rooms (34 episodes of physical restraint per 100,000 population per year) was similar to that of approved centres without seclusion rooms (31 episodes of physical restraint per 100,000 population per year). This casts doubt on the common idea that physical restraint is commenced at an earlier point on sites without seclusion rooms as a preventative measure in these settings.
Overall, it appears likely that a combination of factors contributes to variations in rates of use of restrictive practices across approved centres. The association with urbanicity suggests that factors which are commonly associated with urban settings play a key role, potentially including increased rates of substance misuse, downward social drift, and forensic issues. It is possible that the cooccurrence of several of these risk factors in the same person has an effect that is more than additive, especially in terms of physical restraint. Among the six approved centres with the highest rates of physical restraint in the country, five are in Dublin, with Tallaght University Hospital showing the lowest rate of physical restraint among these five Dublin approved centres (Table 1).
This paper has several limitations. It is based on publicly available data which were collected for other purposes and is therefore subject to variable data collection practices at different sites. Ireland’s catchment-area admission system is relatively robust, but some patients are admitted to approved centres outside their catchment areas (e.g. private or independent hospitals or other approved centres if their local one is full). While it is hoped that these factors are minor and largely even out, they might affect comparisons between certain approved centres (e.g. in 2023, there were temporary closures of beds in St James’s Hospital, Dublin and Bantry General Hospital, Cork). In addition, certain areas have a variable number of non-acute facilities (e.g. continuing care and rehabilitation units) and our comparison focused on acute beds only. Future work might usefully apply a broader lens to these issues.
Conclusions
This paper set out to examine rates of use of physical restraint and seclusion across Ireland’s acute adult approved centres in order to identify (a) challenges with the nature and quality of data that are publicly available; (b) potential lessons for Ireland and other jurisdictions from comparisons across approved centres (to whatever extent such comparisons are possible and informative), and (c) suggestions for future data collection and research.
Regarding (a), challenges with the available data, it is clear that limitations on existing data do not permit full and detailed explanation of variations in rates of restrictive practices across approved centres. More fine-grained data collection would help to clarify the potential roles of other patient factors (e.g. presentations, diagnoses, treatments), environmental factors (e.g. resources in different approved centres, physical infrastructure, staff training), and background risk factors (e.g. social deprivation, community supports). Hopefully, data collection will continue to improve over future years in order to facilitate such deeper, multi-layered analysis.
Regarding (b), potential lessons for Ireland and other jurisdictions, we found that there are within-country variations in rates of physical restraint and seclusion in Ireland, but these are of a lesser magnitude than between-country variations. Overall, Ireland’s rates of restrictive practices are lower than those in other jurisdictions, consistent with Ireland’s low rate of involuntary admission. For both Ireland and other jurisdictions, it is useful to note that, even with low rates of involuntary admission and restrictive practices, within-country variations are still present. Strategies to reduce and hopefully eliminate the need for these practices in the future will need to be tailored to local settings for optimal impact, given these notable variations in rates across different inpatient settings.
Regarding (c), future research, ongoing data collection could usefully link service resourcing and other local characteristics (e.g. urbanicity) to rates of restrictive practices, in order to better understand variations between and within countries. Patient and family perspectives can help to contextualise these practices in a broader way. For example, previous research has shown that perceived coercion at time of admission, assessed in retrospect by the patient, is more closely associated with involuntary status and symptoms of mental illness than it is with subsequent formal coercive practices, such as seclusion and restraint (O’Callaghan et al. Reference O’Callaghan, Plunkett and Kelly2021). The use of restrictive practices in other settings, such as police custody, prisons, and private homes, also merits study in order to gain a full understanding of the use of these practices in mental illness.
Finally, while Ireland has relatively low rates of involuntary admission and coercive practices, continued focus and research are essential if we are to maintain these low rates, improve on them, and – ideally – eliminate the need for these practices entirely in the future.
Acknowledgements
The authors are very grateful to the editor and reviewers for their comments and suggestions.
Financial support
This research received no specific grant from any funding agency, commercial, or not-for-profit sectors.
Competing interests
The authors declare none.
Ethical standards
The authors assert that all procedures contributing to this work comply with the ethical standards of the relevant national and institutional committee on human experimentation and with the Helsinki Declaration of 1975, as revised in 2008. The authors assert that ethical approval for the publication of this paper was not required by their local ethics committee. This paper uses only data that were already in the public domain.
Open access
