“Voluntarily Admitted Against My Will”: Patient Perspectives on Effects of, and Alternatives to, Coercion in Psychiatric Care for Self-Injury

Abstract

Introduction:

Various forms of coercion are used in the psychiatric care of patients with self-injurious behaviors, but there is little research on how these are perceived by the patients. The aim of this study was to investigate how 6 patients, who had received care for self-injurious behavior, perceived coercion and how they think coercion could be avoided.

Methods:

This study employed a qualitative design with 6 semistructured interviews and interpretative phenomenological analysis.

Results:

Three main themes were identified: keep voluntary care voluntary, apportioning control and responsibility, and dialogue and participation. Constant supervision was described as the most destructive form of coercion. To enable self-responsibility, a reduction of control and supervision was advocated. Calls were made for a treatment based on the assumption that there is a desire, on behalf of the patients, to get better.

Conclusions:

The use of coercion in the psychiatric care of patients with self-injurious behavior can be reduced by increasing predictability, by listening to the patient with genuine interest, and by involving the patient in decisions regarding their treatment.

Introduction

Self-injury, defined as deliberate and direct bodily harm without suicidal intent (1), has become an increasing concern in psychiatry. Common examples are cutting, carving, banging, and hitting oneself. The most commonly reported reason to self-injure is to regulate negative affect (2), but other reported reasons are self-punishment, interpersonal influence, and stopping dissociation (3). The prevalence of self-injury has been reported as 40% to 80% in young psychiatric patients and 20% in adult patients in Western countries (4). The use of coercion is widespread in the treatment of patients with severe self-injurious behaviors, based on what is called a paternalistic perspective within medical ethics (meaning that professional staff should make decisions in the patient’s best interest) (5).

Two alternatives to paternalism in the normative medical ethics literature are autonomy and reciprocity (6). Autonomy means that well-informed patients have the right to make decisions regarding their care, even if this would go against their best interests seen from a professional perspective. Reciprocity means that patients are expected to participate as full partners in the delivery of care. Although paternalism, autonomy, and reciprocity may appear to be 3 incompatible perspectives, it has been suggested that they represent 3 complementary contributions to the ethics of psychiatry(6).

Interventions aiming to bring reluctant patients to accept treatment do not necessarily involve explicitly compulsory measures. Szmukler and Appelbaum (7) describe a hierarchy of treatment pressures, ranging from various means of informal coercive practices, such as persuasion, interpersonal leverage, inducements (offers), and threats, ending with formal compulsory treatment. Research indicates that informal coercion occurs frequently and tends to be intertwined with the relationship to health-care professionals, who may apply coercion, either intentionally or unintentionally, as part of the interaction with patients (8).

Research on coercion has increasingly focused on the patients’ subjective experience of coercive methods (8 –10). Perceived coercion includes a sense of lack of influence, control, freedom, and choice (11) and does not necessarily correlate with factual coercion (12,13). The degree of perceived coercion has been found to correlate with the patient’s ratings of the therapeutic relationship (12,14).

Listening to people’s experiences is essential in developing a person-centered care where alternatives to coercive measures are identified, tested, and evaluated. In their systematic review of coercion in psychiatry, Newton-Howes and Mullen (15) identified 5 qualitative studies of perceived coercion. These studies reported a number of negative aspects of perceived coercion—a common theme was the experience of “being dehumanized through a loss of normal human interaction and isolation” (p. 469). Similar results are also described in a later interview study (16).

Although perceived coercion in voluntary patients seems to be generally associated with negative experiences, involuntary treatment, as such, is not always negatively perceived (17,18). For example, when Katsakou et al (17) explored involuntary patients’ views retrospectively by means of in-depth interviews, they identified 3 groups of patients with different views on their involuntary hospitalization: those who believed that it was right, those who thought it was wrong, and those with ambivalent views. It has also been found that, although dissatisfaction with treatment among involuntary inpatients could be predicted by measures of perceived coercion, it could not be predicted by the number of coercive measures (restraint, forced medication, seclusion, etc) as documented in the medical records (18).

One reason that coercive measures are not necessarily perceived as negative is that they need not be associated with disrespect and not being listened to. Previous research indicates that when involuntary patients feel that they are given a chance to participate in decisions regarding their care and health professionals are genuinely interested in their well-being, they find it easier to accept compulsory treatment (19,20).

There is little research on perceived coercion in patients with self-injurious behavior. Taylor et al (21) carried out a systematic review of 31 studies of attitudes toward clinical services following self-harm among not only patients but also patients’ friends and relatives. The results showed that participants with negative experiences reported a perceived lack of patient involvement in management decisions, inappropriate staff behavior, and lack of staff knowledge, whereas participants with positive experiences reported greater participation in care decisions and perceptions of staff as sympathetic.

Most relevant for the present research topic is a study (22) where 19 self-injuring individuals who had been treated in psychiatric inpatient care were asked to write about their experiences and their thoughts about alternatives to coercive treatment methods. Content analysis of their reports, which varied in length from a couple of paragraphs to several pages, led the authors to identify 3 categories of answers: (1) a wish for understanding instead of neglect, (2) a wish for mutual relation instead of distrust, and (3) a wish for professionalism instead of a counterproductive care.

Although these results give some hints about how patients with self-injurious behavior would like to be treated, they provide little information about how they perceive the various forms of coercion they have been subjected to. In-depth interviews have the potential of producing detailed data on such experiences. The purpose of the present study was to explore how 6 patients who had been in inpatient care for self-injurious behavior perceived coercion as part of their treatment and how they thought coercion could be avoided.

The study was carried out in Sweden, where compulsory psychiatric care is regulated by The Compulsory Psychiatric Care Act (LPT) and the Forensic Psychiatric Care Act (LRV; 23). Coercive measures, such as forced medication, external body searches, restraint, seclusion, and reduced access to electronic communication, are regulated in these laws. In Sweden, some psychiatric patients with serious self-injurious behaviors have been admitted to high-security forensic psychiatry wards, despite not having been prosecuted, a widely criticized solution.

Method

This study employed a qualitative design using semistructured interviews with 6 participants.

Participants

The participants were 6 women, with an average age of 36.5 years, who had engaged in self-injury and had been subjected to coercive measures during treatment. All participants had been in care for periods ranging from 4 to 12 years, during which they had periodically been in round-the-clock care against their will. This included care given under formally voluntary terms. Aside from self-injury behaviors, they had received several other diagnoses including borderline personality disorder, obsessive-compulsive personality disorder, obsessive-compulsive disorder, anorexia nervosa, post-traumatic stress disorder, major depression, bipolar disorder, and autism. The coercive measures they had been subjected to were forced medication, restraint by belt, external body searches, and seclusion.

Recruitment of participants

Information about the study was spread via social media, the organization Self-Harm and Eating Disorders Organization, and the national anti-stigma campaign Hjärnkoll. The information letter asked “Have you ever been treated for self-injurious behavior against your will?” and asked if they would like to be interviewed about their experiences. A total of 16 individuals expressed their interest in participating. Of these, 10 passed 3 basic criteria: (1) having received compulsory care at least once in connection with deliberate self-harm, (2) having experienced some kind of coercion as part of psychiatric care at least once to prevent self-harm, and (3) not being subject to compulsory psychiatric care at present.

A short telephone interview was carried out with these 10 individuals to establish that they had received treatment for problems corresponding to the proposed diagnosis nonsuicidal self-injury in Diagnostic and Statistical Manual of Mental Disorders, 5th Edition (24) and that they had been subject to legally regulated coercion in connection with their treatment. The interviewer established this on the basis of the examples provided by the respondent. For logistic reasons, only 6 of the 10 women who took part in the brief telephone interview were able to go through with the in-depth interview.

Procedure

Semistructured interview

Data were gathered through semistructured interviews, carried out by the 2 first authors (3 interviews each). The interviews lasted between 1 hour 15 minutes and 1 hour 45 minutes and followed an interview guide with 9 basic questions. Two of the questions were about the coercive measures they had experienced under voluntary and involuntary care, respectively. The other questions were open questions that were followed up by asking about personal experiences and reflections. The respondents were not given any definition of coercion or self-injury, neither in the initial information about the study nor during the interviews, in order to allow room for their own experiences and understanding of coercion.

Ethics

The participants were informed that they had the right to end the interview and their participation at any time, without having to state a reason. It was also made clear to them that they could withhold any information they were not comfortable sharing. An institutional review board at Lund University gave positive feedback on this study concerning research ethics.

Data Analysis

The interviews were tape recorded and transcribed and analyzed using interpretative phenomenological analysis. Interpretative phenomenological analysis is a qualitative method that includes the following steps: (1) reviewing the material from each interview, with an aim to enter the participant’s world; (2) commencing an analytical dialogue with this material, making notes; (3) starting to formulate themes on the basis of these notes; (4) identifying connections between the different themes and creating clusters of themes that point to the aspects that are most interesting and relevant to the purpose of the study; (5) repeating these steps with each interview, approaching it on its own merits; and (6) seeking connections between interviews, searching for the themes that are most central and contribute the most insight (25). Both interviewers participated in the analysis of all interviews.

Results

The analysis led to the formulation of 3 overarching themes, which are illustrated below: (1) keeping voluntary care voluntary, (2) the apportioning of control and responsibility, and (3) dialogue and participation.

Keeping Voluntary Care Voluntary

The respondents’ accounts show that even interventions given under judicially voluntary terms can be experienced as coercion. The respondents described various kinds of threats as recurring elements, even under voluntary care; for example, threats of changing voluntary care into compulsory care, transfer to another ward or unit, reduced benefits, or being undressed by force.

Especially with the medication, or, if I was admitted voluntarily and had to take [medication], then “you know if you don’t take this it will become compulsory care,” so there was always an underlying threat. I call it being “voluntarily admitted against my will”…They would say that ‘‘if you don’t stop harming yourself, or if you carry on like this, you’ll be transferred to forensic psychiatry, and you know it!” It was an underlying threat, “carry on like this, and that’s where you’ll end up!” (Participant 3)

The threats were often considered worse than the formally coercive interventions. One participant expressed that formally coercive interventions, such as forced medication, were easier to handle, as they were of a passing and limited nature, while interventions involving informal coercion could continue for days, sometimes weeks.

Apportioning Control and Responsibility

The type of control described as the most negative was constant supervision. This element, which is not legally classified as a coercive intervention, was perceived by the majority of the respondents to be the most invasive form of coercion they had been subjected to. Constant supervision was described as a destructive intervention with far-reaching negative consequences.

…because it lasted like for a long period. You didn’t have constant supervision for an hour, it went on for days. But an injection, you got it like, it was very quick…And they were over. But the constant supervision lasted for days, weeks sometimes, like months. (Participant 1)

The respondents reckoned that constant supervision contributed to their existence being robbed of its meaning, and they described how it became almost impossible to maintain a view of oneself as competent and capable of taking responsibility, when, sometimes for several weeks, they would not be left alone even during visits to the toilet. To encourage and enable trust and self-responsibility, the respondents reckoned that elements of control and supervision need to be reduced.

Another example is that when staff focused on removing opportunities for self-injury, it increased the patients’ fixation on injuring themselves in ever more creative ways. One respondent reasoned about the confiscation of tools:

I found new ways of self-injuring, and they just took away more and more. And then I had to get even more creative…It became some sort of game, where they tried to protect me, and I just tried to find new ways, it didn’t lead anywhere. (Participant 2)

Zero tolerance of self-injury was interpreted as aiming for total control of the patient, which in the short term could prevent self-injury, but which in the long term was a trigger to further more serious self-injurious behavior.

Instead of controlling and disrupting a behavior through zero tolerance, the respondents were looking for acceptance of relapses and a constructive treatment with a common goal of finding functional strategies. The respondents stressed the importance of understanding that self-injury fulfills a function and conveyed that you can’t take away one strategy without offering another.

Dialogue and Participation

All of the respondents stressed the importance of understanding that it is ill health and disability that are behind self-injury, not wickedness. They suggested that treatment based on the assumption that people who injure themselves at the same time want to feel better could reduce the use of coercion. By listening to the needs and wishes expressed by the patients, it should be possible to create agreements around the layout and operation of the treatment which would benefit trust and thereby positive outcomes:

You could turn it around and say “How do you want us to help you during this time so you don’t feel you have to self-injure,” or something like “what do YOU need?” (Participant 4)

Respondents advised that during interactions they should be listened to by health-care professionals who had a genuine interest of finding out what their problems were about. The importance of understanding that self-injury can fulfill several functions was brought up, as was the need to be curious about the individual and the specific situation from the start of the treatment process.

Respondents also said that it is as important to inform the patient as it is to listen to the patient. Without any insight into how, when, and why things are done, the care becomes incomprehensible and unpredictable:

If they consider that I don’t understand why, they could try to explain why. And if I don’t understand today, then they can try to explain tomorrow. Sooner or later I’ll get it! (Participant 5)

Some of the respondents still had some level of contact with psychiatry, but they all tended to avoid treatment, in one way or other. They stated that, as a result of their experiences, they had come to avoid contact with around-the-clock psychiatric care. Some went as far as seeking to avoid all forms of health care. They described how they initially had faith and had sought help, but that they, due to negative experiences and broken trust, had come to dismiss treatment.

Discussion

The present results illustrate (1) that informal coercion, in the form of threats and constant supervision, can be perceived as worse than formal coercion among at least some patients with self-injurious behaviors and (2) that patients prefer a treatment situation characterized by dialogue and participation in treatment planning. The latter is in accordance with previous findings both among patients with self-injurious behavior (21,22) and among other patients who have been subject to coercion (19). However, the former finding has to our knowledge not been reported previously.

Why are threats and constant supervision sometimes perceived as worse than formal coercion? One reason stated by some respondents is that formally coercive interventions are of a passing and limited nature, whereas informal coercion can continue for days, sometimes weeks. Another possible reason is that informal coercion, such as the use of threats, is more difficult to combine with a genuine interest and respect for the patient and may therefore be perceived as extra humiliating. Also, the respondents’ description of constant supervision as destructive for their experiences of meaning, competence, and responsibility echo previous findings about “being dehumanized” (14).

Previous studies point in a similar direction. For example, in a Norwegian study (26), perceived coercion was studied among 223 patients who were admitted to acute wards. The results showed that the patient’s formal legal status was a poor predictor of perceived coercion; the only significant predictors were the use of negative pressures (threats and force) and process exclusion (not having their viewpoints taken into account).

The respondents in the present study conveyed that experiences of coercion had made them feel violated long after cessation of care and had led them to avoid contact with the health-care systems. This confirms earlier research findings that patients who have been subject to coercion, even where they have in hindsight understood the reasoning behind the coercion, continue to feel violated and dissatisfied (27).

The respondents also described how a controlling environment with “zero tolerance” for self-injury can actually increase their fixation with self-injury, so that they harm themselves more frequently and more severely. The respondents’ feedback highlights a need for active measures to counteract the fixation on self-injury, a fixation that risks drawing staff and patients into destructive power struggles. There is strong support in research that nonsuicidal self-injury serves the function of reducing negative emotions (3), and it is important that treatment does not bring about further difficult emotions and traumatic experiences. It has previously been shown that certain interventions can trigger the inclination to self-injury in patients with self-injury issues (18).

The majority of the respondents describe constant supervision as a measure that psychiatry ought to avoid, as it has contributed to a tangible sense of personal loss of control and self-esteem. The respondents’ descriptions of constant supervision, which is not a formal measure of coercion, highlight the importance of not limiting investigations to formal measures of coercion (9,19,28). This is in accordance with the analysis by Favazza (29), that supervision, especially in the form of “constant supervision,” encourages regression and risks exacerbating destructive behavior.

The results suggest that real participation and communication should be prioritized, because comprehensibility and predictability increase the likelihood of cooperation, which in turn can decrease the risk for the development of situations that require coercive interventions. A difficult but important balancing act seems to be required, to inform patients about requirements and potential consequences of digression without using threats of coercion. A key to avoid coercion may lie in providing information about the background and purpose of interventions. If paternalistic reasoning has a role to play in the treatment of patients with self-injurious behavior, it is probably essential that it is explicitly balanced by an emphasis on the patient’s autonomy and reciprocity—seen as 3 complementary ethical perspectives in psychiatric care (6).

Limitations

The small sample and the phenomenological method of analysis with its focus on idiographic data forbid us from drawing any general conclusions about the population of individuals who engage in self-injurious behaviors. Moreover, the recruitment process may have favored a selective sample, as information was distributed through social media, a patient organization, and an anti-stigma campaign, channels which might favor a certain perspective. A main strength of qualitative method, however, is that it is sensitive to the respondents’ subjective experiences in a way that is difficult to achieve with quantitative studies that use standardized questionnaires with predefined response alternatives. For example, the methodological decision not to use a preset definition of coercion in the interviews paved the way for the finding that “constant supervision” was perceived as a most offensive measure.

Conclusion

Informal measures such as threats and constant supervision can sometimes be experienced as more violating than formal coercion. Although compulsory treatment is sometimes needed, there is much that can be done in terms of reducing informal coercion in the psychiatric care of patients with self-injurious behavior. This includes the sharing of information about treatment to increase predictability, listening to the patient with a genuine interest in their experiences, and involving the patient in treatment decisions.

Author Biographies

Ellen Gerle, MA, MSc, has a degree in philosophy from Södertörn University, and in psychology from Lund University, Sweden. She is a licensed psychologist who works at a psychiatric unit for victims of war and torture in Malmö, Sweden. She is also involved in research on gambling.

Anne Fischer, MS, has her degree in psychology from Lund University, Sweden. She is a licensed psychologist working at the Compulsory School Administration in Malmö, Sweden.

Lars-Gunnar Lundh, PhD, is professor emeritus at Lund University, Sweden, and a licensed psychologist and psychotherapist. He is involved in research in a number of different areas, includingpsychotherapy, mindfulness, non-suicidal self-injury, insomnia, anxiety disorders, and alexithymia.