How clinicians make decisions about CTOs in ACT: a qualitative study

In Norway, as in many other Western countries, there has been a substantial reduction in the number of inpatient beds and a move towards community-based services, referred to as a ‘deinstitutionalization’ of psychiatric care [1, 2]. As part of this process of deinstitutionalization, the primary locus of treatment of severe mental illness (SMI) shifted from hospitals to the community. Different legal mechanisms, such as community treatment orders (CTOs), have been used to compel treatment adherence in more than 75 jurisdictions worldwide [3]. Besides mandating patients with SMI to adhere to treatment, the CTO regime also allows for a rapid recall to hospital for its enforcement. In Norway, CTOs have been widely used since the implementation of the Mental Health Care Act (MHA) in 1961 [4]. Despite lack of robust evidence of effectiveness [5], Norwegian figures suggest that more than a third of all patients discharged from involuntary admissions in 2014 were placed on CTOs [6]. Although two national action plans to reduce the use of coercion have been launched, the involuntary admission and CTO rates in Norway have remained relatively high (61/100,000) compared to other Western countries since recording of data started in 2007 [7].

Decisions regarding compulsory interventions in general adult psychiatry are made by psychiatrists and authorized clinical psychologists [Responsible Clinicians (RCs)]. The legal criteria for involuntary hospitalization and CTOs, which are identical, are based on the presence of severe mental illness (SMI) [8]. CTOs have been referred to as a less restrictive treatment option than involuntary hospitalization, either to improve or restore patients’ health or prevent significant deterioration in the very near future (‘treatment criterion’) [9]. Commitment may also be used if the person is likely to pose an obvious and serious risk to his/her own life or health or that of others (‘dangerousness criterion’). Additionally, voluntary treatment must have been attempted (or obviously futile), and unless the ‘dangerousness’ criterion is met, compulsory mental health care must clearly appear to be the best option for the patient. RCs are required to conduct assessments every 3 months to consider whether the legal criteria are met. Decisions regarding involuntary hospitalization and CTOs do not include medication, and the responsible psychiatrist can initiate a separate medication order. Commitment decisions may be appealed to an independent Supervisory Commission, which conducts documentary controls every third month. If extended commitment after the initial 12 months is requested, the Supervisory Commission makes an independent review.

There is an ongoing discussion as to whether the mandatory element of the CTO produces greater clinical benefits for patients than offering them the same package of services on a voluntary basis [5, 10, 11]. International and Norwegian studies show that the use of CTOs is often justified on the basis of patients’ lack of illness insight [12], and that they are mainly used to provide support and treatment following involuntary hospital admissions [12–14]. Studies show that patients have mixed views of coercion in general and CTOs in particular, while psychiatrists and patients’ relatives are more positive [15–17]. Stensrud et al. [18] found that RCs in Norwegian services were worried about relapse, and therefore reluctant to make dynamic adjustments, even when patients were stable over time. Another Norwegian study [19] shows great variance in how RCs consider clinical and functional improvement. The duration of CTOs seems to depend on the RCs’ attitudes and opinions [13, 14, 19]. Although most clinicians would probably support a shift towards a more person-centred and collaborative decision-making approach, some studies have shown that many clinicians tend to use a more paternalistic type of argumentation in clinical decision-making situations regarding patients with SMI [18, 20, 21].

In a report from 2008, it was estimated that approximately 4000 people with SMI were not receiving appropriate mental health care [22]. In 2009 the Norwegian health authorities decided to fund the piloting of assertive community treatment (ACT) teams to provide services to this target group. The ACT model involves multidisciplinary teams with shared caseload responsibility and a low client–staff ratio (1:10), that provided flexible and intensive home-based support. Prior to this pilot, Norway had limited specialist ambulant community treatment outreach services.

Although ACT has been criticized for being paternalistic and coercive [23], studies show that patients generally are more satisfied with ACT than standard community mental health services [24, 25]. ACT targets patients with SMI with co-occurring substance abuse and poor social functioning, who are often difficult to engage in more traditional services. Some RCTs show that ACT improves outcomes, including reduced hospitalization, improved housing stability and treatment retention [26].

The research-based evaluation of the 12 first Norwegian ACT teams assessed clinical outcomes from 142 patients at baseline and after 2 years follow-up. Although there was no change in the overall number of admissions, there was a 50% reduction in both involuntary admissions and bed-days [27]. Among the 142 included patients, 32% were subject to CTOs at intake during the teams’ first year of operation.

The ACT model has a strong focus on promoting patients’ autonomy and recovery [28, 29], and, as we have shown previously, the enactment of CTOs typically involves competing priorities and role tensions [30]. Some studies show that assertive outreach teams reported using less intrusive approaches than other community mental health services [31, 32]. Studies have found wide variation in the use of CTOs [14, 33, 34], and we need more knowledge about responsible clinicians’ judgments and CTO follow-up decisions. The aim of this study was to explore how RCs within an ACT setting reason and make decisions about the continued use of CTOs, recall to hospital and the discontinuation of CTOs.

In our analysis, we identified three main categories that reflected the overall finding ‘feeling more confident and secure through shared responsibility’: (1) CTO as a tool for achieving patient stability and safety, (2) CTO as a tool for containing dangerousness and preventing harm, and (3) CTO and ACT allowing for more nuanced judgments and reduced coercion. We discuss each in turn.

This study illuminates the RCs’ reasoning surrounding the use of CTOs within an ACT setting, including decisions about their continued use, their termination, and recall to hospital, at a time when ACT represented a relatively new approach. The participants emphasized that CTO decisions involved tensions and challenging professional judgments. Although the use of CTOs was mainly founded on patients’ clinical needs, CTOs were also presented as a tool to contain dangerousness. The participants stressed that being part of a multidisciplinary team with shared caseload, frequent patient contact and comprehensive service provision allowed for more nuanced judgments and increased flexibility.

CTOs were typically described as a useful tool to ensure that patients remain in treatment, mainly to help patients achieve stability and remain safe, which the participants said often takes a long time. Patients’ symptom severity, lack of illness insight, co-occurring substance abuse, frequent readmissions and a history of deliberate self-harm or violence, and the likely consequences of patients’ decisions were presented as decisive factors in CTO decisions.

Norwegian and international studies show that CTOs are often founded on patients’ lack of insight, to prevent psychotic relapse [12], which often has a disruptive effect on people’s quality of life and their capacity for independence. A Norwegian study showed that CTO practices vary [19] and that some RCs only met patients at yearly reviews, knew little about the content of local care services and the impact of CTOs on patients’ everyday lives. Some studies suggest that clinicians’ narrow understanding of ‘lack of insight’ and that a one-sided focus on compliance with medication may impede patients’ recovery process [18, 37, 38]. Our data show that the ACT providers perceived that the multidisciplinary nature of their work allowed for frequent patient contact, coordinated support, relationship building and more flexible treatment options. As a specific component of clinical care, the ACT model’s focus on everyday activities and improving patients’ lives was seen as pivotal to improve patients’ well-being and to gradually help patients work toward greater independence and self-sufficiency [30]. Further, the ACT team’s long-term treatment perspective and the ways in which the teams often managed to gradually involve patients in treatment decisions and joint crisis planning were considered as important in reducing the use of coercion [39].

ACT targets patients with co-occurring substance abuse and low social functioning who have not remained in care. While some studies describe the implementation of CTOs as a complement to involuntary inpatient care [19, 40, 41], the participants in our study presented the use of CTOs combined with ACT as the least restrictive solution. The Norwegian ACT evaluation did not show a reduction in the number of admissions, but there was a significant reduction in compulsory admissions and total inpatient days. Patients with co-occurring substance abuse had significantly fewer involuntary inpatient days, despite severe problematic substance use at 2 years follow-up [42]. Although the ACT model has been criticized for being paternalistic and coercive, studies show that patients are generally more engaged and satisfied with ACT than traditional community-based services [25]. The Norwegian ACT evaluation also revealed a high user satisfaction; patients subject to CTOs were more satisfied with ACT than voluntarily enrolled patients [43], and they also reported the highest degree of recovery [44].

The participants emphasized that many patients that were referred to them from hospital wards and high-security facilities were already subject to CTOs and that the CTO decision-making in the ACT team typically involved multiple interacting and complex risk factors. Similar to what is reported in other studies, our participants emphasized that a multidisciplinary team approach and close monitoring provided more comprehensive understanding of risk [45]. In addition, the team’s position enabled the early detection of warning signs, quick responses and the stepping up of the intensity of interventions during crises all of which allowed for more focused preventative efforts.

The use of CTOs was sometimes presented as a long-term safety measure to prevent harm that could result from patients’ symptom severity, co-occurring substance abuse, frequent readmissions and a history of victimization, deliberate self-harm or violence. However, there was substantial variation in interpretation of the dangerousness criterion, and the two RCs that were not part of the treatment team expressed more doubts about the accurate prediction of risk. This “risk as difficult to predict” finding has been reported in other studies. Feiring and Ugstad [20] found that many RCs were reluctant to assess whether a patient was at risk of harming others or societal consequences of untreated mental disorders. Assessing patients’ potential dangerousness is also challenging for GPs [9]. In Norway, the MHA has been founded on a strong treatment philosophy [46], and clinicians may find it less stigmatizing to refer to patients’ treatment needs than to dangerousness.

Our study participants emphasized the importance of the context in which psychiatric assessments were made. While ACT allowed for comprehensive assessments, reflective team discussions and multiple interventions, the functional split between inpatient and outpatient care was seen as a challenge. Although the ACT teams were mainly involved in admissions and discharge, disagreements between the ACT psychiatrist and the responsible ward clinician and uncoordinated changes to treatment plans were typical concerns. An ongoing debate in other European countries concerns which of the organizational models in mental health care, i.e. continuity of care or a functional split of responsibility between inpatient and outpatient care, is most effective [47–49]. Lindgren et al. [50] found that continuity across treatment settings was associated with better long-term outcomes. Despite lack of solid evidence of whether specialization or continuity of care is more effective (length of stay) [48], studies show that patients and clinicians prefer continuity across inpatient and outpatient settings [48, 49, 51, 52].

Many RCs described lack of inter-agency collaboration regarding a small subgroup of patients with co-occurring substance abuse with higher risk of victimization and violence during acute phases of illness as a main concern. Bed pressure and lack of treatment resources was seen to place clinicians in a position where they have to balance discretionary standards for civil commitment and ambiguous benchmarks for what constitute good practice and acceptable risks. Based on a recent amendment in the Mental Health Care Act, from September 2017, involuntary admissions and CTOs can only be used if the person lacks decision-making capacity, unless the person is deemed to constitute a risk to his/her own life or the safety of others. The amendment also involves a narrowing of the dangerousness criterion, which no longer includes the risk of harm to the person’s own health. Although capacity is a cornerstone in autonomous decision making, it is well known that psychiatric patients’ capacity to make informed decisions can fluctuate [53]. Many RCs in our study stated that they often applied the treatment criterion to justify the need for CTOs. The RCs will now be required to use a more structured approach to consider patients’ capacity to make treatment decisions, as of yet we have no knowledge of whether the recent amendment influences RCs’ decisions within an ACT setting.

The strengths of this study is that it draws on a rich dataset, and that we know that participants work in ACT teams with moderate to high fidelity to the model. One limitation is the small number of participants. Although the aim of qualitative studies is not to generalize the results, we cannot exclude the possibility that other RCs would have provided a more nuanced and deeper understanding of the RCs’ reasoning and CTO decision making. Since a majority of the participants were psychiatrists, the inclusion of other participants may have enabled a focus on possible variations between professionals in attitudes and decision-making practices. It would have been interesting to complement this research with experiences of CTO decision making in an ACT setting from the perspectives of family members and hospital clinicians.

The RCs found that many of the patients enrolled in ACT are hard to reach and difficult to treat. The ACT model’s holistic and long-term treatment perspective was described as a more attuned way of working with patients with complex needs than traditional outpatient services. The participants emphasized that CTOs were mainly founded on patients’ clinical needs, and also that establishing stability and safety for patients enrolled in ACT is often a lengthy process. The main finding, ‘feeling more confident and secure through shared responsibility’ illustrates that the focus of the ACT model on frequent patient contact, shared caseload and comprehensive service provision was considered as a major improvement on traditional community services in that it could facilitate more nuanced assessments and reduce coercion. However, in some cases the RCs described CTOs combined with ACT as a long-term safety measure to prevent harm. Many RCs described a small subgroup of patients with co-occurring substance abuse who had frequent encounters with the police, where bed pressure and lack of inter-agency collaboration was presented as a main challenge.