Introduction
The coercion of people with mental health problems and/or intellectual disabilities (IDs) is highly undesirable as a potential contravention of human rights (UN-CRPD 2006). A global effort to minimise reliance on such practices (WHO 2022) and a willingness to learn from services worldwide have successfully reduced such practices (Gooding and McSherry 2018).
Internationally, coercion is defined as physically restraining someone, taking control of the person, and other mechanical means of restraint to stop a seriously challenging behaviour. The term ‘challenging behaviour’ is divided into ‘less demanding’ and ‘more demanding’ challenging behaviour. In the former, measures such as the following must be taken to solve the problem: extra staffing, extra solid equipment, or separating the person from things or people. The displayed behaviour can be challenging for the person and for others. The latter has the same challenges as the former, but in addition, at least one of the following is present: the behaviour is displayed daily; it prevents the person from participating in activities; physical intervention must be taken to prevent the behaviour; and/or the behaviour causes significant damage to the person or to others (Holden 2009). Challenging behaviour is often communication of expressed distress, a reaction to something not being right. When such distress is met with coercive measures, it can lead to increased resistance or powerlessness and passivity or surrender, which is often considered positive, but only treats the symptom that something is wrong and not the cause itself (Henriksen 2022). It often turns out that a lack of communicative skills triggers challenging behaviour as well as a lack of influence in areas that are important to the person themselves (Holden and Gitlesen 2006).
In Norway, in addition to physical and mechanical coercive measures, limited self-determination and autonomy are included in the concept of coercion. Coercion can be used not only when serious challenging behaviour is displayed but also when such behaviour, in the long run, can lead to significant damage.
International research has focused on the prevalence rate, the concept, and preventive efforts concerning restraint in ID settings and the characteristics associated with its use (Fitton and Jones 2020; Luiselli 2009; Lundstrom et al. 2011; Sturmey 2017). Romijn and Fredriks (2012) explored guidelines and practices for reducing coercive use against people with IDs in the United States, the UK, and Australia and found that decisions on coercive restraint are internationally guided by the criterion of ultimum remedium (as a last resort). Gaskin et al. (2013) reviewed articles on initiatives to reduce the use of restraint and seclusion against people with ID, and all the studies they found focused on physical or mechanical restraint reduction. Restraint was used to manage problems of agitation, aggression, and self-harm. Overall, research on the concept of coercion in the population of persons with IDs is more aimed at physical and mechanical restraint to gain control, with less emphasis on restricted self-determination (Luiselli, Sperry, & Draper 2015).
Ratifying the UN Convention on the Rights of Persons with Disabilities (CRPD) entails greater awareness and legal development regarding such rights (Helsedirektoraratet 2021). Norway ratified the CRPD in 2013, and the Norwegian Directorate of Health published a national guideline for health and care services for people with IDs in 2021. The guideline details that high-quality services require competence among employees, interdisciplinary collaboration, collaboration with specialist health services, ethical reflection, the facilitation of self-determination, and the follow-up of people with IDs who refuse assistance. The prevention of challenging behaviour is also highlighted. The guideline refers to a person-centred approach, where autonomy and self-determination are key components for the quality of life (Helsedirektoraratet 2021).
Subjection to coercion clearly violates the fundamental human right to self-determination. There is a tie between self-determination and the quality of life for persons with IDs (Emerson 2021; Lachapelle et al. 2005; Schalock, Bonham, & Verdugo 2008; Wehmeyer 2020). Self-determination is a theoretically based principle that guides reflection on decisions in everyday life, such as what and when one wants to eat, rest, be active, and so on. However, it is also about learning to act on one’s terms based on informed choices and being the one who actively influences and acts as a causal agent in one’s life (Shogren et al. 2015). Self-determination is one out of eight dimensions in the quality-of-life principles, a value upon which services are based, and an outcome for measuring service and support. Self-determination is driven by the person’s preferences and interests and serves to enhance their quality of life (Wehmeyer 2020).
The use of formally approved coercion has been increasing in Norway. It rose by 94% (from 835 to 1,662 persons) in the period 2012–2020. From 2019 to 2020 alone, there was an increase of 15% (from 1,410 to 1,662 persons) (Helsetilsynet 2020). Although the increase contrasts with the purpose of the Health and Care Services Act, we have no evidence that more coercion is used in Norway than in other Western countries (Søndenaa, Dragsten & Whittington 2015). The increasing use of coercion, however, is worrying and problematic in services for people with IDs as it can come in conflict with adopted rules for human dignity and professional ideals (Helsedirektoratet 2015).
Proceedings in the Decision-Making Process
In Norway, the municipality writes individual coercive decisions and sends them to the region’s state administrator for approval. When coercive decisions are drawn up, the municipality is required to take guidance from the region’s habilitation service, which is a specialist health service. The habilitation service must independently give its opinion and recommendation on the coercive decision to the state administrator (Helsedirektoratet 2015).
The state administration considers all aspects of the decision and reviews the recommendation of the municipality and the habilitation service. Approval for certain types of intervention is reviewed in advance of the implementation of the coercive decision, and at each revision point, approval can be given so that the intervention can be continued or increased or the service can be ordered to implement a less serious type of coercion instead (Helsedirektoratet 2015). Thus, the level of coercion can be maintained, increased, or decreased.
The state administrator often approves the decision one year at a time as the behaviour that leads to significant damage is recurring. The coercive measures described in the decisions are carried out in the municipality services, provided for adults in their private homes. The decisions themselves deal with either (1) planned damage prevention measures in recurring emergency situations (serious injury to oneself or others, serious material damage where employees must physically take control, or behaviour that violates the person’s integrity and loss of social respect, for example, taking control of mobile and computer equipment when such use leads to significant damage) or (2) measures to cover the person’s basic needs for food and drink, clothing, rest, sleep, hygiene, and personal safety (Helsedirektoratet 2015). Examples of basic needs include preventing access to assets, notification systems to detect if the person leaves the apartment, or regulated access to food to prevent or limit significant damage. The documentation underpinning these decisions can provide insight into factors associated with decisions to reduce coercion. Thus, an analysis of general trends in the decisions to reduce coercion over time may suggest ways to change practice that accelerate a shift toward a zero-coercion environment in this and wider services. This analysis has been conducted below.
The document records the challenging behaviour that is to be managed, the trigger(s) of the behaviour, and the consequence of the behaviour in the form of significant damage to oneself and others. The person’s everyday situation must also be documented, with an emphasis on participation and self-determination. The documents must also emphasise prevention with a requirement to consider ‘solutions other than the use of coercion’; this requirement can only be waived in special cases, and a justification must then be given. In addition, coercion can only be used when it is professionally and ethically justifiable (Helse og omsorgstjenesteloven 2011), and the potential damage and enforcement of the coercive measure must be evidence based (Helsedirektoratet 2015).
The aim of this study, therefore, was to examine which factors were associated with approval decisions that reduced the level of coercion compared with decisions that maintained the previous level.
Methods
Design and Sample
The study had a cross-sectional design comparing associations between several potential predictor (independent) variables and a single outcome (dependent) variable (reduced vs. maintained coercion level).
The sample consisted of all (n = 120) approvals by the state administrator in one region of Norway (Midt-Norge) in 2020. The participant inclusion criteria were as follows: (1) having a formal diagnosis of ID; (2) having a decision on coercion as part of the services in the municipality; and (3) not actively withholding consent (either the client or their representative).
Documentation
Data from 16 items on the coercion approval form for each decision were extracted for analysis. These items were derived from the national template for coercive decisions, including the template’s headings and subtitles (Helsedirektoratet 2017), the circular IS10/2015 (Helsedirektoratet 2015), and political guidelines (Helsedirektoratet 2021) to specify the best standards for high quality in services and documentation. The items (see Table 2 below) covered a general assessment of the person’s life situation as well as a specific assessment of the challenging behaviour and the significant damage that was to be prevented. The forms are completed by care staff, ideally in consultation with the client and/or their representative. Responses on the form are either free text or pre-structured. Both response types were scored dichotomously (yes/no).
These 16 items were then grouped into four domains for further analysis: problem formulation (five items: 1, 2, 3, 4, and 5); monitoring plan (three items: 6, 7, and 12); client involvement (six items: 8, 9, 10, 11, 13 and 16); and prevention (two items: 14 and 15). All the data were extracted by the first author from electronic files and manually loaded into an SPSS (version 27) file for analysis.
Analysis
There were no missing data on any variable. Analysis was conducted using crosstabulation with the chi-square and p-value testing of associations between the 16 predictor variables and the outcome (McHugh 2012). The four domains derived from the 16 variables were analysed using independent-sample t-tests.
Ethics
The documents constitute clinical data routinely collected to monitor healthcare quality (i.e., the municipalities’ coercive decision journals and the habilitation service’s statement on the municipality’s coercive decision). The decision records are stored in the state administrator’s data archives and require approval from both the Regional Ethics Committee (REK) and the state governor to gain access (REK ref: 203815). The study also required consent from the owner of the records for access. Information about the study, including the opportunity to opt out from participating, was sent to the client’s appointed guardian, who assessed the consent competence of the participant. Of the 132 eligible participants, 12 did not give consent for access to participate, so 120 participants were included. The coercive decisions were read online from the secure server by the first author where they are stored, and all the extracted data were de-identified and anonymised before the analyses.
Results
The sample comprised seventy-three (60.8%) men and forty-seven women (39.2%). The mean age was 40.71 years (SD = 14.3; range 17–79). The proportion of participants in each ID level group was as follows: 45 (37.5%) with mild or moderate IDs, 40 (33.3%) with more severe IDs, and 35 (29.2%) at unspecified levels.
Twenty cases (16.7%) involved a decision to reduce coercion from the 2019 level (‘altered decisions’ below), and the remaining 100 (83.3%) involved maintaining the same level of coercion (‘non-altered decisions’ below). No significant differences were found between the two decision groups in terms of gender, age, or ID level (see Table 1 below).
Table 1
Descriptions of the Two Groups — Non-Altered Decisions (n = 100) and Altered Decisions (n = 20).
| NON-ALTERED DECISIONS | ALTERED DECISIONS | CHI-SQUARE VALUE | P-VALUE | |
|---|---|---|---|---|
| Mild and moderate ID | 38 (38%) | 7 (35%) | 1.71 | NS |
| More severe ID | 31 (31%) | 9 (45%) | ||
| Unspecified | 31 (31%) | 4 (20%) | ||
| Gender | M/F 61/39 | M/F 12/8 | 0.93 | NS |
| Age | Mean = 40.8 (SD = 14.4) | Mean = 40.2 (SD = 14.4) | Student t-test = –0.19 | NS |
The 16 predictor items are compared across decision groups in Table 2 below.
Table 2
Descriptions of the Two Groups—Non-Altered Decisions (n = 100) and Altered Decisions (n = 20)—and Presentation of the Best Practice Items.
| BEST PRACTICE ITEM | DOMAIN | ALL DECISIONS N = 120 | NON-ALTERED DECISIONS N = 100 | ALTERED DECISIONS N = 20 | CHI-SQUARE | P-VALUE |
|---|---|---|---|---|---|---|
| 1. A description of the challenging behaviour is present. | 1 | 117 (97.5%) | 97 (97%) | 20 (100%) | 0.61b | 1.00 |
| 2. A description of what triggers the challenging behaviour is present. | 1 | 98 (81.7%) | 78 (78%) | 20 (100%) | 5.40a | 0.02 |
| 3. A description of the damage caused by the challenging behaviour is present. | 1 | 115 (95.8%) | 95 (95%) | 20 (100%) | 1.04b | 0.59 |
| 4. Evidence-based assessment of the challenging behaviour is present. | 1 | 109 (90.8%) | 89 (89%) | 20 (100%) | 2.42a | 0.21 |
| 5. A precise description of the significant damage that is to be prevented is present. | 1 | 66 (55%) | 56 (56%) | 10 (50%) | 0.24 | 0.62 |
| 6. A personnel plan for follow-up of the client during the new decision period is present. | 2 | 73 (60.8%) | 58 (58%) | 15 (75%) | 2.02 | 0.15 |
| 7. Personnel guidance related to the implementation of the coercive measure is provided. | 2 | 37 (30.8%) | 29 (29%) | 8 (40%) | 0.95 | 0.33 |
| 8. The client participates in designing the plan for implementing the coercive measure. | 3 | 17 (14.2%) | 14 (14%) | 3 (15%) | 0.01a | 1.00 |
| 9. An individual plan is present. | 3 | 54 (45%) | 40 (40%) | 14 (70%) | 6.06 | 0.014 |
| 10. The client participates in describing her/his life situation. | 3 | 52 (43%) | 37 (37%) | 15 (75%) | 9.80 | 0.002 |
| 11. A topic about self-determination is provided in the decision. | 3 | 43 (35.8%) | 31 (31%) | 12 (60%) | 6.09 | 0.014 |
| 12. Recommended guidelines provided by the specialist services during the new decision period are provided. | 2 | 69 (57.5%) | 56 (56%) | 12 (60%) | 0.11 | 0.74 |
| 13. The user’s life situation is described. | 4 | 69 (57.5%) | 53 (53%) | 16 (80%) | 4.97 | 0.026 |
| 14. Other preventive solutions to be used before coercion are described. | 4 | 53 (44.2%) | 41 (41%) | 12 (60%) | 2.44 | 0.118 |
| 15. Other preventive solutions which have been tested in previous decisions are described. | 4 | 83 (69.2%) | 67 (67%) | 16 (80%) | 1.32 | 0.250 |
| 16. The guardian is active in the discussion about the new coercion decision. | 3 | 60 (50%) | 48 (48%) | 12 (60%) | 3.84 | 0.46 |
[i] a. One cell (25.0%), with an expected count less than five. The min. expected count is 3.67 (Fisher’s exact test).
b. Two cells (50.0%), with an expected count less than five. The min. expected count is 0.50 (Fisher’s exact test).
The scores for each of the four domains are presented in Table 3 below.
Discussion
One in six (20 out of 120) decisions involved reduced coercion when comparing 2020 with the preceding year. According to the Health and Care Services Act, the legislation aims to prevent persons with IDs from exposing themselves or others to significant harm as well as to prevent and limit the use of coercion. The services must be provided with respect and integrity and, as far as possible, in accordance with the person’s right to self-determination (Helse- og omsorgstjenesteloven 2011).
The five (out of 16) items that differed significantly between the two groups could be seen as best practices in designing a system for reviewing and reducing coercion in this population. Statistically significant differences were found for items 2 (‘A description of what triggers the challenging behaviour is present’), 9 (‘An individual plan is present’), 10 (‘The client participated in describing her/his life situation’), 11 (‘Evidence of self-determination is provided’), and 13 (‘The user’s life situation is described’).
A description of what triggers the challenging behaviour (item 2), finding strategies to prevent it, and training in the implementation of coercive measures are all necessary to ensure ethical and professional assessments before, during, and after the event. High levels of qualified staff increase awareness of triggers relevant to the person who is liable to coercion through understanding underlying needs that are not being met (Luiselli, Sperry & Draper 2015). To reduce challenging behaviour, it is therefore necessary to draw attention to people’s environmental conditions and well-being, including their physical and mental health, and to treat people with dignity and respect (Friedman 2020). Knowing what triggers the behaviour allows services to use that knowledge in providing improved support (National Collaborating Centre for Mental Health 2015) and is linked to moving an individual away from coercive practices.
The individual or personal plan (IP; item 9) is based on the person’s life situation and needs so that the habilitation consists of coordinated, coherent, and knowledge-based measures. The IP is a document that coordinates interdisciplinary and interprofessional collaboration with a long-term perspective (Helsedirektoraratet 2021). All people receiving health and caring services in Norway have the right to have an IP regardless of any ID (Forskrift om habilitering og rehabilitering 2018). Interestingly, the group that changed decisions into less intrusive measures here had active IPs more often than those who remained at the same level.
Participation in describing one’s life situation (item 10) also differed significantly between the two groups. Comprehensive documentation of a broad range of areas (leisure, home life, education, work/activity, spiritual, and cultural factors) (Dean et al. 2016) and actively seeking participation and self-determination are important not only to meet statutory requirements for case proceedings but also to stimulate a search for other non-coercive solutions (Dean et al. 2016; Helsedirektoratet 2015; Helsedirektoraratet 2021; NOU 2016:17 2016). Decisions with reduced coercive measures accomplished this by providing a more comprehensive description of the person’s life situation (item 13).
Self-determination (item 11) has been an important principle for problematising the use of coercive measures in Norwegian health and care services. This study found a significantly higher recognition of self-determination where the decisions had been changed to less intrusive. Although research on coercion in the population with ID often deals with physical and mechanical measures (Luiseelli, Sperry & Draper 2015), we know relatively little about limited self-determination as a contributing factor to the use of coercion. Skarstad (2018) found that Norway violates the CRPD’s right to self-determination for people with IDs who have coercive decisions. Also, while most of the people with IDs are declared not competent in coercive assessment, they are excluded when decisions about their own lives are being made beyond the use of coercion itself (Skarstad 2018). Research on participation and self-determination refers to increased problem-solving skills, goal achievement, self-regulation, and better tolerance for changes in the environment (Vicente et al. 2020). These improvements can then lead to less challenging behaviour, which, in turn, leads to less use of coercion and restraint (Ellingsen and Berge 2015). An example of self-determination, participation, and problem solving is a quote from a decision: ‘from physical restraint to locking doors, which the user himself thinks is better’. The person avoids the discomfort of being physically stopped, and the change can be related to the findings of Holden (2009) and Ellingsen (2015), where the situation goes from ‘more’ to ‘less’ challenging behaviour and less use of restraint.
The grouping of the items into four domains (problem formulation, monitoring plan, user involvement/participation, and prevention) enabled other characteristics of decisions to be examined. It showed that overall, user involvement/participation was significantly higher in the decisions with less intrusive measures compared to those where the coercion level was maintained. Recent guidelines emphasise the importance of user involvement and participation as a principle (Helsedirektoraratet 2021; UN-CRPD 2006), and this finding provides evidence that such a principle may have concrete beneficial consequences in terms of better care and respect for human rights. The World Health Organization (2019) focuses on influencing policy and building the knowledge and skills to implement person-centred approaches. This is what is required to provide high-quality care and support and to prevent harmful practices (WHO. 2019). Real participation—where the person can express preferences in aspects such as accommodation, living, work, leisure, and cultural activities, or can at least have an advocate who can articulate those preferences—is having the opportunity to actively influence one’s own life (Soresi, Nota, & Wehmeyer 2011). Coercive decisions that have participation and self-determination as core themes can be recognised as indicating a professional approach in the everyday life of the user/client, and one can also assume that the services are striving to be truly person-centred (Kjellevold 2014), which likely enables positive relationships (based on trust and affection) and a reduction in challenging behaviour (Larue et al. 2018).
Strengths and Limitations
A strength of the study is that details of real-world coercive decisions were analysed, and since the study included all 120 decisions in which consent was provided over one year, there was no selection bias. All items on the included forms were completed, so there were also no missing data.
On the other hand, it must be acknowledged that the documents are highly political, as they are scrutinised by powerful independent authorities. So rhetorical methods may well have been used to justify the practice and adapted to the requirements for documentation; one can consider that someone carries out such documentation. Also, there was great variation in terms of detail in the documentation and lacking documentation, even where it is specified that the information must be documented as a legal requirement. This may indicate that the practice of ‘embellishing’ the content is not necessarily widespread.
Furthermore, the content of the coercive decision document template is not standard across the municipalities examined here. The length of the documents varied from 7 to 40 pages of text, including appendices. The headings of the text material provide some information about the content, but often the content overlaps across these headings. Finally, the interpretation of the free text items to categorise their content was inevitably a somewhat subjective process, which raises validity issues mitigated by the coding criteria used. In this study, for item 10 (‘The client participated in describing her/his life situation’), the criterion was whether the person’s opinion, participation, or influence was expressed regarding their life situation. The remaining items were no/yes values for whether they occurred in the documentation.
Conclusion
This study aimed to examine which factors were associated with approval for decisions that reduced the level of coercion compared with decisions that maintained the previous level.
It has been found that a minority of decisions made in one year were changed to be less intrusive and that decisions to reduce coercion differed from decisions to maintain it in various ways. The reduction decisions placed more emphasis on describing triggers of the challenging behaviour, involving service users through participation in their everyday lives, providing supplementary documentation about the person’s general life, having an IP, and considering the principal self-determination. The findings also correspond to the professional and political guidelines that are assigned to high-quality services, where it is emphasised that education in health and social sciences, especially among disability nurses, is of substantial importance following up these guidelines.
Incorporating these five elements into practice and in decision-making procedures may contribute to a shift away from the routine deployment of coercive practices in ID services around the world.
Acknowledgements
We are grateful for the support from the Stat Administrator in Midt-Norway in providing data for the study.
Funding Information
The study is funded by the Department of Mental Healt, Norwegian University of Science and Technology (NTNU), Trondheim, Norway.
Competing Interests
The authors have no competing interests to declare.
Author Contributions
MR, RW, and ES contributed to the design of the study, critically revised the manuscript, gave final approval, and agree to be accountable for all aspects of the work ensuring integrity and accuracy.
MR is first author, collected the data, analysed, and drafted the manuscript.