When Treatment Pressures Become Coercive: A Context-Sensitive Model of Informal Coercion in Mental Healthcare

Abstract

Treatment pressures are communicative strategies that mental health professionals use to influence the decision-making of mental health service users and improve their adherence to recommended treatment. Szmukler and Appelbaum describe a spectrum of treatment pressures, which encompasses persuasion, interpersonal leverage, offers and threats, arguing that only a particular type of threat amounts to informal coercion. We contend that this account of informal coercion is insufficiently sensitive to context and fails to recognize the fundamental power imbalance in mental healthcare. Based on a set of counterexamples, we argue that what makes a proposal coercive is not whether service users will actually be made worse off if they reject the proposal, but rather whether they have the justified belief that this is the case. Whether this belief is justified depends on the presence of certain contextual factors, such as strong dependency on professionals and the salient possibility of formal coercion.

Keywords:

• Informal coercion
• informed consent
• leverage
• psychiatry
• treatment pressure
• voluntariness

INTRODUCTION

In their everyday practice, mental health professionals use various communicative strategies to influence the decision-making of mental health service users and improve their adherence to recommended treatment and social rules. These communicative strategies are discussed in the literature as “informal coercion,” “treatment pressures,” and “psychological pressures” (Hotzy and Jaeger 2016; Klingemann et al. 2022; Potthoff et al. 2022; Yeeles 2016). In what follows, we will refer to these communicative strategies as “treatment pressures” and use the term “informal coercion” only to refer to coercive use of treatment pressures.

Treatment pressures should be distinguished from interventions that are commonly referred to as “formal coercion,” such as involuntary hospital admission, involuntary treatment, seclusion, and physical and mechanical restraint. These interventions are termed “formal” coercion because they are typically regulated by law (Wasserman et al. 2020), whereas treatment pressures are typically unregulated (Yeeles 2016). Conceptually, the key difference between the two types of intervention is that formal coercion is carried out against the will of service users, while treatment pressures are used precisely to influence the behavior of service users via their will (Potthoff et al. 2022).

Empirical evidence indicates that the use of treatment pressures is highly prevalent in mental healthcare services (Hotzy and Jaeger 2016; Yeeles 2016). The most influential model of treatment pressures in mental healthcare was developed by Szmukler and Appelbaum (2008). This model incorporates aspects of treatment pressures identified in the earlier literature, such as the differentiation between positive pressures (e.g., persuasion and inducement) and negative pressures (e.g., threats) (Bindman et al. 2005; Iversen et al. 2002; Lidz et al. 1995, 1998) and the idea that influence and informal coercion exist on a continuum (Curtis and Diamond 1997; Lovell 1996; Olsen 2003; Sorgaard 2007).

According to Szmukler and Appelbaum’s model, treatment pressures comprise persuasion, interpersonal leverage, inducements, and threats.¹ Persuasion involves rational argumentation; interpersonal leverage involves announcing conditional changes in emotional attitudes within interpersonal relationships; inducements involve making conditional proposals to make people better off if they accept the proposal; and threats involve making conditional proposals to make people worse off if they refuse the proposal (Szmukler and Appelbaum 2008; see also Szmukler 2018). The various types of treatment pressures can be arranged hierarchically to reflect the level of pressure, ranging from persuasion at the bottom end to threats at the top end. Table 1 provides an overview of Szmukler and Appelbaum’s model and illustrates each type of treatment pressure with an example drawn from mental healthcare (see also Gather et al. 2019; Hempeler et al. 2022).

Table 1. Szmukler and Appelbaum’s (2008) model of treatment pressures.

Type	Means	Example
Persuasion	Rational argumentation	“I propose you take this tranquilizer. Last time, it really helped you to calm down.”
Interpersonal leverage	Announcing conditional changes in emotional attitudes within interpersonal relationships	“We have always worked together so well. I would be very disappointed if you were to quit the group therapy sessions.”
Inducements	Making conditional proposals to make people better off if they accept the proposal	“If you come here every month to receive your depot injection, we will help you find a job that suits you.”
Threats	Making conditional proposals to make people worse off if they turn down the proposal	“Please take your antipsychotic medication, or else we will cancel your daughter’s visit tonight.”

While pressure and coercion are often used synonymously in the debate, Szmukler and Appelbaum explicitly refrain from using the term “coercion” in relation to the full spectrum of treatment pressures because they think that the term “is best applied to specific types of pressure” (2008, 234). Referring to Wertheimer’s (1987) moral baseline theory of coercion, Szmukler and Appelbaum state that “contemporary ideas about “coercion” suggest the use of this term should be reserved for particular types of threat” (2008, 243) and argue that the use of treatment pressures like persuasion, interpersonal leverage and inducements is not coercive.

Wertheimer distinguishes between an “analytical” and a “justification” question regarding informal coercion: the former asks “what constitutes or counts as [informal] coercion,” whereas the latter asks “what justifies the use of [informal] coercion” (1993, 239).² In this article, we set out to answer the analytical question. We thus ask: What is informal coercion? Or put in a different way, when is the use of treatment pressures coercive? Answering these analytical questions amounts to spelling out the truth conditions for claims about informal coercion (Wertheimer 1993).

The conceptual analysis of informal coercion that we propose in this article is informed by qualitative research on service users’ perspectives on treatment pressures (Allison and Flemming 2019; Canvin et al. 2013; Norvoll and Pedersen 2016; Potthoff et al. 2022; Verbeke et al. 2019). Our account challenges Szmukler and Appelbaum’s account of informal coercion and the moral baseline theories that form its basis. In this article, we show that due to a fundamental power imbalance between professionals and service users in mental healthcare, contextual factors and service users’ interpretation of communicative interactions play a key role in determining whether the use of treatment pressures is coercive. We argue that the use of treatment pressures like persuasion, interpersonal leverage, and inducements can be coercive and, accordingly, that service users can be coerced into accepting treatment even if professionals do not issue any threat. While our analysis is focused on the use of treatment pressures and informal coercion in inpatient settings, we believe that the results of our analysis also apply in outpatient and community mental healthcare settings and other settings characterized by a fundamental power imbalance. Context-specific factors must be considered in the application of our findings in other contexts.

The structure of this article is as follows. We first expound Wertheimer’s (1987, 1993) theory of informal coercion. Subsequently, we draw on Rhodes’ (2000a, 2000b) subjective account of coercion to demonstrate the relevance of contextual factors and the recipient's beliefs in determining whether a proposal is coercive in contexts that involve a fundamental power imbalance. After demonstrating that a fundamental power imbalance exists between professionals and service users in mental healthcare, we develop two case vignettes to demonstrate that Szmukler and Appelbaum’s account of informal coercion is incomplete. We propose an alternative analysis of the case vignettes which shows that contextual factors and the recipient's beliefs can render the use of treatment pressures coercive even if no explicit threat is issued. We close by replying to five potential objections to our context-sensitive account of informal coercion and deriving clinical recommendations from our account.

WERTHEIMER’S MORAL BASELINE THEORY OF COERCION

Szmukler and Appelbaum (2008) base their distinction between coercive and non-coercive treatment pressures on Wertheimer’s (1987) philosophical theory of coercion and his (1993) application of the theory to the context of mental healthcare. Wertheimer’s theory is inspired by Nozick’s (1969) previous work on coercion, in which Nozick analyzed for the first time the truth conditions for claims like ‘Person A coerces person B to do φ’. Nozick’s work sparked such a lively debate that subsequent theorists explicitly adopted or implicitly presupposed his theoretical framework (Anderson 2023).

Since on Szmukler and Appelbaum’s model only threats are potentially coercive, understanding the distinction between threats and offers is crucial. According to Wertheimer (1987, 1993), person A issues a threat to person B if and only if A proposes that B do φ and A indicates that she will make B worse off than in the moral baseline position if B does not accept the proposal. By contrast, A makes an offer to person B if and only if A proposes that B do φ and A indicates that she will make B no worse off than in the moral baseline position if B does not accept the proposal. The moral baseline is set by reference to the way that A morally ought to treat B.³

By way of illustration, suppose that a professional tells a service user, “Please take your medication, or else you cannot leave the hospital.” Let us for the sake of the argument assume that legal and moral rights coincide. If the service user was admitted to the hospital voluntarily and has a moral right to leave the hospital, the professional indicates that she will make the service user worse off than in the moral baseline position if she rejects her proposal. Her proposal is thus a threat. If, on the other hand, the service user was involuntarily admitted to the hospital by court order for a period of six weeks and was eventually granted conditional leave subject to treatment adherence, the proposal constitutes an offer instead. This is because the service user will not be made worse off than in the moral baseline position if she rejects the proposal; rather, she gains the prospect of being made better off by being able to leave the hospital if she accepts it.

Wertheimer contends that not all threats are coercive (1987, 1993). A threat is coercive only if the threat makes it the case that B has no reasonable alternative but to accept the proposal. “Your money or your life” is the paradigmatic example. By contrast, a threat to make a person only marginally worse off than in the moral baseline position does not make turning down the proposal an unreasonable alternative. Take, for example, a cashier who hands over the entire cash register because someone threatens to take one of her lunch sandwiches if she does not hand over the money. In this case, the cashier has not been coerced into giving away the store’s returns of the day because despite the threat she still had a reasonable alternative besides handing over the money. This type of threat is thus not coercive—though of course still morally problematic.

RHODES’ MOBSTER EXAMPLE

Rhodes (2000a, 2000b) criticizes moral baseline theories of coercion such as Wertheimer’s (1987, 1993) by developing an example that underlines the importance of contextual factors and the recipient's beliefs for determining whether proposals are coercive. This is his example of the notorious mobster (Rhodes 2000b, 47–48). The example is as follows: a notorious local mobster approaches a grocer and proposes to buy a basket of apples below the market price. While the mobster has no intention of threatening or harming the grocer, the grocer submits to selling the apples at a cheaper price for fear of repercussions (Rhodes 2000b).

Wertheimer’s (1987, 1993) account of coercion would yield that the mobster’s proposal is not coercive. The reason for this is that the mobster does not indicate that he will make the grocer worse off than in the moral baseline position if he does not accept the proposal. Still, most of us will feel that the grocer is coerced into selling his apples below the market price. This can be explained by reference to contextual factors and the beliefs that the grocer has in view of these factors. If just any regular customer had made this proposal, the grocer would likely have perceived the proposal as an attempt to bargain. Given the local mobster’s reputation, however, it seems reasonable for the grocer to believe that repercussions will be unavoidable if he rejects the proposal and that he has no reasonable alternative but to sell his apples below the market price. Accordingly, Rhodes’ example demonstrates that the beliefs that the recipient has in light of the proposal context play a crucial role in determining whether a proposal is coercive.

Moral baseline accounts define coercion in terms of whether A indicates that she will make B worse off than in the moral baseline position if B does not accept the proposal. We will use the adjective “recipient-focused” to refer to moral baseline accounts that grant a central role to B’s beliefs about whether A indicates that she will make B worse off than in the moral baseline position if B does not accept the proposal. While Szmukler and Appelbaum acknowledge that recipient-focused accounts of informal coercion can be valuable “for the clinical context,” they ultimately reject them in favor of Wertheimer’s account, explaining that the latter “comports better with the frameworks we will propose for justifying coercive interventions” (2008, 237).

THE POWER IMBALANCE IN MENTAL HEALTHCARE

The central importance of context in evaluating treatment pressures and informal coercion in mental healthcare is a recurring finding in both observational studies (Lidz et al. 1993; Sjöström 2006, 2016) and qualitative studies with service users (Allison and Flemming 2019; Canvin et al. 2013; Norvoll and Pedersen 2016; Potthoff et al. 2022; Verbeke et al. 2019). For example, an interview study with service users carried out by Potthoff et al. (2022) revealed that service users see contextual factors such as the style of communication, the quality of the therapeutic relationship, and the spatial surroundings of communication as essential to evaluate the coerciveness of an interaction.

The interactions between mental health professionals and service users are shaped by a fundamental power imbalance (Lidz et al. 1993; Rugkåsa et al. 2014; Schramme 2017; Sjöström 2006, 2016). Based on the available empirical research on treatment pressures and informal coercion in mental healthcare, we have identified two general types of contextual factors in which this power imbalance finds expression, and which structurally change the nature of communicative interactions between professionals and service users. These are (1) the dependency of service users on professionals and (2) the possibility of the use of formal coercion.

Let us turn to dependency first. This type of power imbalance is most evident in the case of service users who have been admitted to a mental health hospital involuntarily on a court order. Because these service users cannot decide to leave the ward at their own discretion, they strongly depend on the professionals working on the ward for receiving adequate care. Similar considerations may hold for service users who are voluntary inpatients. Empirical studies indicate that many service users assess their own legal admission status incorrectly (Iversen et al. 2002; Katsakou et al. 2011; O'Donoghue et al. 2014), with one UK study among 100 service users recently admitted to a mental health hospital in South London finding that two-thirds of service users who were admitted voluntarily were uncertain about whether they were free to leave the hospital if they wanted to (Bindman et al. 2005). Service users who are voluntary inpatients may also feel that they lack alternatives because they depend fully on the treatment team for receiving care. The following statement of a service user illustrates that service users who are voluntary inpatients may choose to comply with recommended treatment primarily because they are concerned that otherwise they will lose access to the treatment they need: “And if I didn’t take my pills, then they would discharge me. And by then I had gotten so sick [that I wouldn’t have been able to stay at home], so I was forced [to take my pills]” (Norvoll and Pedersen 2016, 209).

Service users’ dependency on professionals is not limited to the provision of medical care but extends to domains such as support for securing housing, work, social benefits, and legal advice due to the gatekeeping function professionals oftentimes have (Grimen 2009; Maung 2019). Although these psychosocial domains should be part of standard care, access is often made contingent on treatment compliance (Monahan et al. 2005; Robbins et al. 2006).

The possibility of the use of formal coercion increases the power imbalance between professionals and service users considerably (Lidz et al. 1993; Rugkåsa et al. 2014; Sjöström 2006, 2016). For service users, noncompliance can result in being subjected to involuntary hospital admission, involuntary treatment, seclusion, or physical or mechanical restraint. The possibility of the use of formal coercion changes the nature of communicative interactions between professionals and service users (Allison and Flemming 2019; Sjöström 2006, 2016) and gives it “an underlying coercive dimension” (Rugkåsa et al. 2014, 891). The following remark made by a service user when reflecting on communication about medication with the community mental health team underlines this: “If I stopped my medication, I’d be sectioned [i.e., involuntarily admitted to hospital] in 24 hours, I guarantee” (Gault 2009, 510). Another vivid illustration is the following report from a service user who was an inpatient: “And I eh refused to take [tablets] at first … and then they came round with their heavies to try and inject me and then I stuck out my hand and took [the tablets]” (McGuinness, Dowling, and Trimble 2013, 730).

The possibility of the use of formal coercion does not have to be invoked verbally or intentionally for it to become a salient possibility for service users. Empirical research indicates that the physical and organizational structure of the ward can also make the use of formal coercion a salient possibility (Larsen and Terkelsen 2014; Potthoff et al. 2022; Sjöström 2016, 2006). A pertinent example is the following quote from a service user who felt pressured during admission to the hospital:

There was a table with chairs, but also a patient bench, and these belts were attached to this patient bench and they are used when in doubt … but that was in there by default, and that irritated me because I thought, why is this patient bench there now, or why is it there with these devices? (Potthoff et al. 2022, 9–10).

CLINICAL COUNTEREXAMPLES TO SZMUKLER AND APPELBAUM’S ACCOUNT

In what follows, we will illustrate the relevance of service users’ dependency on professionals and the possibility of the use of formal coercion in determining whether proposals made by professionals involve informal coercion. We will use the philosophical method of cases (Machery 2017). According to this method, a philosophical account of the truth conditions of a concept is prima facie undermined by prompting everyday judgments about particular hypothetical cases which contradict this account. Our procedure is as follows: We will present two case vignettes, one involving the possibility of the use of formal coercion and one involving dependency. Each case vignette will take two different courses to prompt everyday judgments showing that the use of treatment pressures can involve informal coercion even if it does not involve threats.

We developed the case vignettes based on a synthesis of findings from our interview study with service users (Potthoff et al. 2022) and actual cases encountered in clinical practice by research team members with clinical expertise and experience. We discussed the interview findings and clinical experiences in the research team, distilled key aspects from them, and constructed the case vignettes using these key aspects. Team members with clinical experience reviewed the constructed case vignettes to ensure that they were realistic.

The Possibility of the Use of Formal Coercion

Let us start by considering the following case vignette:

Alex was diagnosed with schizophrenia 5 years ago and has been involuntarily admitted to a mental health hospital several times. Recently, the voices in his head have started coming back and he has increasingly been feeling persecuted by a group of mysterious people wandering around the neighborhood. After Alex had locked himself up in his room, his mother brought him to the mental health hospital. Alex agreed to speak to a psychiatrist but does not want to stay in the hospital as an inpatient.

We can imagine several ways for the narrative to continue. Consider a first scenario:

Scenario 1: Upon arrival at the mental health hospital, a security guard approaches Alex and brings him to the admission room. The first thing Alex notices upon entering the room is the bench used for mechanically restraining people located in the corner of the room. It reminds him of a particularly distressful situation during his last hospitalization when six people forced him on the bench and strapped him up. He still doesn’t understand why they did that. He felt intimidated and helpless and never wants it to happen again. As Alex takes his seat in front of the psychiatrist’s desk, he spots the security guard waiting outside the room on a chair next to the door.

After a short wait, the psychiatrist enters the room. At the end of their conversation, Alex tells her that he wants to go home. The psychiatrist is aware that there are no grounds for obtaining a commitment order for Alex and therefore tries to persuade him into staying. She says: “From what I’ve heard, I don’t think you’re doing well at the moment. I know you think you can handle it by yourself, but I don’t believe you will get better just like that. I really think you should stay here for a while.”

Alex looks at the bench and then at the security guard sitting just outside the door. He fears that he will be prevented from leaving by means of physical force if he doesn’t agree to stay, so he responds: “Okay… I’ll stay.”

Now consider an alternative scenario and sequence of events:

Scenario 2: Upon arrival at the mental health hospital, a friendly nurse greets Alex and asks him how he’s doing. Alex is too troubled to reply but the nurse is patient and calms him down. She explains that the psychiatrist wants to see him and talk to him about how he’s feeling at the moment. The nurse offers to accompany him to the admission room and wait there with him until the psychiatrist arrives. As Alex takes his seat in front of the psychiatrist’s desk, it strikes him that the room looks just like an ordinary doctor’s room.

After a short wait, the psychiatrist enters the room. At the end of their conversation, Alex tells her that he wants to go home. The psychiatrist knows that there are no grounds for obtaining a commitment order for Alex and therefore tries to persuade him into staying. She says: “From what I’ve heard, I don’t think you’re doing well at the moment. I know you think you can handle it by yourself, but I don’t believe you will get better just like that. I really think you should stay here for a while.”

As Alex thinks about it, he realizes that the psychiatrist might have a point. He thinks about it again and then responds: “Okay… I’ll stay.”

Szmukler and Appelbaum’s account (2008) of informal coercion yields that neither of the scenarios involves informal coercion. To see why, note that the psychiatrist’s statements are exactly the same in the two scenarios: she issues no threat and merely tries to persuade Alex to stay at the hospital by means of rational argumentation. Since Szmukler and Appelbaum’s account focuses on the explicit content of the proposal, it yields that the level of pressure exerted on Alex is the same in the two scenarios and must be located at the lower end of the spectrum of treatment pressures.

Most people will find this result counterintuitive. While Szmukler and Appelbaum’s account correctly yields that no informal coercion is exerted in the second scenario, we assume that most people will feel that the first scenario involves informal coercion. Moreover, even if some people may have doubts about whether that is the case, it cannot reasonably be denied that the level of pressure faced by Alex varies significantly across the two scenarios. Szmukler and Appelbaum’s account cannot accommodate this variation.

To explain why the level of pressure varies across the scenarios, let us first turn to the first scenario and start by defining the moral baseline. To prevent implicit disagreements about the location of the moral baseline from influencing people’s intuitions about informal coercion, we will assume, for the sake of the argument, that Alex’s legal and moral rights coincide. It is stipulated in the description of the case vignette that there is no legal basis for involuntary commitment. On the assumption, it thus follows that it is morally impermissible to keep Alex at the hospital against his will: Alex has the moral right to choose whether he wants to stay or go home. Now consider Alex’s beliefs. Alex believes that he will be prevented from leaving the hospital by means of physical force if he rejects the psychiatrist’s proposal. Considering the assumed moral baseline, Alex thus believes that he will be made worse off than in the moral baseline position if he rejects the proposal.

The psychiatrist knows that she cannot obtain a commitment order for Alex. Accordingly, it is plausible to assume that she does not intend to keep Alex at the hospital by means of physical force if he were to make a move to leave. The psychiatrist, in other words, does not intend to make Alex worse off than in the moral baseline position if he rejects her proposal. This means that Alex’s belief is false.

Even though Alex’s belief is false, there are several contextual factors present in the first scenario that seem to render his belief justified. The security guard sitting right next to the door conveys the impression that Alex will be prevented from leaving the hospital by means of physical force and the bench with the straps in the admission room suggests an even more drastic deprivation of liberty. Both contextual factors make the possibility of the use of formal coercion salient and justify Alex’s belief that the psychiatrist will ensure that he will be prevented from leaving the hospital by means of physical force. Given this belief, we can furthermore understand why Alex sees accepting the psychiatrist’s proposal as his only viable option.

Let us compare this with the second scenario. In this scenario, neither the relevant belief nor the relevant contextual factors are present: Alex does not believe that he will be prevented from leaving the hospital by means of physical force if he does not accept the psychiatrist’s proposal; and there are no contextual factors that make the possibility of the use of formal coercion salient and which could hence justify such a belief.

The analysis of the two contrasting scenarios thus reveals that whether a recipient of a proposal has certain beliefs and whether these beliefs are justified plays a key role in determining whether the proposal is coercive. Before we formulate our analysis in a more formal way, let us first consider a case vignette involving dependency.

Dependency

Consider the following case vignette:

Mika has bipolar disorder and has been staying on a psychiatric ward for several weeks now. Because of her illness, Mika is not able to live by herself anymore. She has explained to her treating psychiatrist that she wants to move to an assisted living facility after her stay at the hospital. The treating psychiatrist and a social worker are currently organizing such a place for her. Mika knows that the psychiatrist must submit a report about her medical condition and treatment compliance to the assisted living facility, based on which it will be decided whether she is granted a place.

Mika didn’t sleep well last night and feels tired. When the nurse comes by to remind her of the group therapy session in the afternoon, Mika tells him that she doesn’t want to participate today and would rather rest.

After their conversation, Mika lies down on her bed and thinks of the report to be submitted to the assisted living facility. She worries about what to do if she cannot move there because Corey, a fellow service user from her ward who was treated by the same psychiatrist, just recently got rejected. Mika wonders if the psychiatrist denied Corey support because he was sometimes reluctant to participate in treatment. After all, she thinks, psychiatrists have a great deal of discretion in writing reports.

Again, we can imagine multiple ways in which the narrative might continue. Consider a first scenario:

Scenario 1: Just as Mika lies there wondering about this, her treating psychiatrist comes to check in on her on his daily round. After a brief exchange about the status of her application to the assisted living facility, he says: “Mika, nurse Leo told me that you don’t want to go to the therapy session today. I was so impressed with how well you’ve done so far. Maybe you’ll reconsider going. I would be very disappointed if you didn’t attend today’s session.” Mika is worried that the psychiatrist will portray her in a bad light in the report to the assisted living facility if she doesn’t attend today’s therapy session. Her mind wanders off to her current home situation. She finally responds: “Yes, you’re right, I’ll go.”

Now consider an alternative scenario and sequence of events:

Scenario 2: Just as Mika lies there wondering about this, her assigned peer support worker comes to check in on her. He says: “Mika, nurse Leo told me that you don’t want to go to the therapy session today. I was so impressed with how well you’ve done so far. Maybe you’ll reconsider going. I would be very disappointed if you didn’t attend today’s session.” Mika thinks about how she doesn’t want her peer worker to be disappointed in her. She pulls herself together and responds: “Yes, you’re right, I’ll go.”

Szmukler and Appelbaum’s (2008) account of informal coercion again yields that neither of the scenarios involves informal coercion. To see why, note that neither the psychiatrist nor the peer support worker indicates that they will make Mika worse off than in the moral baseline position if she rejects the proposal. If, following Szmukler and Appelbaum’s account, we focus on the explicit statements of the persons who make the proposal, it is clear that the psychiatrist and the peer support worker merely indicate that they would be disappointed if Mika were to skip the therapy session. Although their disappointment would make Mika worse off in some sense, people have no moral right that others not be disappointed in them. Consequently, Mika would not be made worse off than in the moral baseline position if she rejects the proposal. That means that on Szmukler and Appelbaum’s account, the psychiatrist and peer support worker merely use interpersonal leverage and do not issue any threat to motivate Mika to attend the therapy session. Accordingly, Szmukler and Appelbaum’s account yields that the level of pressure exerted on Alex is the same in the two scenarios and should be located within the lower range of the spectrum of treatment pressures.

Again, most people will feel that this result is counterintuitive. While Szmukler and Appelbaum’s account correctly yields that no informal coercion is exerted in the second scenario, we assume that most people will feel that the first scenario involves informal coercion. And again, even if some people may have doubts about whether that is the case, the level of pressure faced by Mika clearly varies across the two scenarios, and this is something that Szmukler and Appelbaum’s account cannot accommodate.

To explain why the level of pressure varies across the scenarios, note that Mika believes that her psychiatrist will submit a negative report about her to the assisted living facility if she rejects his proposal that she attend the therapy session. Let us assume that supporting Mika in finding a safe and stable living situation is part of the psychiatrist’s professional duties. It is hence plausible to assume that Mika has a moral right that her treating psychiatrist support her in finding a safe and stable living situation and not write a negative report about her on account of one missed therapy session. Accordingly, Mika will probably believe that she will be made worse off than in the moral baseline position if she skips the therapy session.

We can stipulate that the psychiatrist does not intend to submit a negative report to the assisted living facility if Mika rejects his proposal. This implies that Mika’s belief is false. Note, however, that in the first scenario there are several factors that seem to justify Mika’s belief. The first is that the person who uses interpersonal leverage to motivate Mika to go to therapy is also responsible for writing the report that is necessary for her application to the assisted living facility. The second is that psychiatrists have quite some discretion in writing reports, combined with the suggestion that the psychiatrist has denied another service user support in finding housing for the wrong reasons. Taken together, these factors seem to justify Mika’s belief that she will be made worse off than in the moral baseline position if she rejects her psychiatrist’s proposal. Furthermore, since Mika depends upon a positive report to get access to adequate housing and cannot imagine living independently anymore, it is understandable that she sees herself as having no reasonable alternative but to go to therapy.

Let us compare this with the second scenario. In this scenario, Mika does not believe that she will be made worse off than in the moral baseline position if she decides to skip the therapy session. Such a belief would also not be justified in the context because the peer worker who announces that he will be disappointed if Mika were to skip the therapy session and hence uses interpersonal leverage to motivate Mika to go to therapy is not responsible for writing the report to be submitted to the assisted living facility. Furthermore, the statement made by the peer support worker leaves Mika with two alternatives, neither of which seems unreasonable: she can either go to the group therapy session or accept that she will disappoint the peer support worker.

A CONTEXT-SENSITIVE ACCOUNT OF INFORMAL COERCION IN MENTAL HEALTHCARE

In analyzing the case vignettes, we showed that the possibility of the use of formal coercion and service users’ dependency on professionals are relevant contextual factors in determining whether the use of treatment pressures is coercive. We demonstrated that if these contextual factors are present, the use of treatment pressures can become coercive even if it does not involve an explicit threat.

Based on our analysis of the case vignettes and incorporating insights from Wertheimer (1987, 1993) and Rhodes (2000a, 2000b), we propose the following account of informal coercion in mental healthcare:

A coerces B into φ-ing if, and only if,

1. A proposes that B do φ;

2. B believes that P, where P stands for the proposition that A will make B worse off than in the moral baseline position if B does not do φ;

3. given the circumstances in which A makes the proposal, B is justified in believing that P; and

4. in light of her belief that P, it is understandable that B believes that there is no reasonable alternative but to φ.

Conditions 2 and 3 mark the departure of our account of informal coercion from that of Szmukler and Appelbaum (2008) and Wertheimer (1987, 1993). Where the latter accounts focus on whether A indicates that she will make B worse off than in the moral baseline position if B does not do φ, our account focuses on whether B justifiably believes that this is the case. Our account is thus recipient-focused, though it is not subjectivist. A subjectivist recipient-focused account would include our condition 2 but omit our condition 3. That is, on such an account, it would be relevant only whether B believes that A indicates that she will make B worse off than in the moral baseline position if B does not do φ. In our view, by contrast, it is relevant not only whether B believes that this is the case (condition 2), but also whether B’s belief is justified given the circumstances of the proposal (condition 3). We have shown that, in contrast to Szmukler and Appelbaum’s account, our analysis of informal coercion can explain why our intuitions vary across the contrastive scenarios of the two case vignettes.

Recall, however, that while many people will think that Alex and Mika have been coerced into accepting treatment in the first scenario of the vignettes, others may have doubts about whether this is the case. This poses no challenge to our account because our analysis of informal coercion can explain this disagreement, while the other accounts cannot.

To see why Szmukler and Appelbaum’s account cannot explain this disagreement, recall that in both vignettes it is stipulated that the professionals do not indicate that they will make Alex and Mika worse off than in the moral baseline position if they reject the proposal. If only threats can amount to informal coercion, as Szmukler and Appelbaum hold, there can thus be no disagreement over whether Alex and Mika have been coerced: they clearly have not.

Similar considerations hold for a subjectivist account that includes our condition 2 but omits our condition 3. Recall that in the first scenario of each case vignette, Alex and Mika believe that the psychiatrist will make them worse off than in the moral baseline position if they reject the proposal. If informal coercion merely depends on whether the recipient has this belief, there can again be no disagreement over whether Alex and Mika have been coerced: they clearly have.

In contrast to the accounts just mentioned, our analysis of informal coercion can explain the disagreement over whether Alex and Mika were coerced into accepting treatment in the first scenario of each case vignette. It reveals that the disagreement is over whether the beliefs of Alex and Mika (i.e., their belief that they will be made worse off than in the moral baseline position if they reject the psychiatrist’s proposal) are justified. People who think that the beliefs of Alex and Mika in the first scenarios of the case vignettes are justified will affirm that the first scenarios involve informal coercion; and people who think that these beliefs are unjustified will deny this. The fact that people’s judgment about whether a particular situation involves informal coercion varies with their judgment about whether the relevant belief is justified in the context only confirms that our conceptual analysis of informal coercion is correct.

Although questions about whether the relevant belief is justified depend on contextual factors and are hence necessarily interpretive, we hold that there are right answers to these questions. We rely here on insights from philosophical hermeneutics. Gadamer (1975), among others, has shown that meaning and understanding are essentially contextual and interpretative. At the same time, he has argued extensively and convincingly that interpretation is not merely subjective and that interpretations can be better or worse, true or false, depending on the extent to which they cohere with features of the context. Accordingly, although there may certainly be disagreements over whether a given belief is justified, these disagreements can in principle be resolved in a conversation in which the parties mutually test the extent to which their interpretations cohere with features of the context.

OBJECTIONS AND REPLIES

Our account of informal coercion in mental healthcare could give rise to potential objections. A first potential objection is that we make the question of whether service users are coerced into accepting a professional’s proposal a merely subjective question. This is incorrect. Although service users’ beliefs play a pivotal role in our account of informal coercion, we specify as a necessary condition not only that a service user believes that she will be made worse off than in the moral baseline position if she rejects the proposal, but also that this belief is justified in the context. As we have described above, the question whether the belief is justified can be settled by reference to contextual factors and is thus not merely subjective. By way of illustration, suppose that in the first case vignette Alex had the delusional belief that if he were to reject the proposal to stay at the hospital, the psychiatrist would give mysterious people walking around in his neighborhood the sign to attack him upon his return home. Assume, furthermore, that in light of this delusional belief, Alex sees no alternative but to stay in the hospital. While a purely subjectivist recipient-focused account would yield that Alex was coerced into staying at the hospital in this scenario, according to our account this scenario does not involve informal coercion because Alex’s belief is unjustified.

A second objection that might be raised is that the inclusion of contextual factors and beliefs in the analysis renders informal coercion vague and difficult to assess in clinical practice. Recognizing this, we think this is simply a fact of the matter: the reality of informal coercion in mental healthcare just is very complex. In this, it is no different from other types of communicative interaction. It is widely accepted in the philosophy of language that the meaning of an utterance is largely determined by its context (Wittgenstein 2009). Accordingly, we would object in return that traditional moral baseline accounts of informal coercion oversimplify the phenomenon. That said, the vagueness of the phenomenon should not be overestimated. We have identified two contextual factors that play a key role in determining whether a communicative interaction involves informal coercion. Even if these are probably not exhaustive, our analysis provides substantial guidance to professionals for communication with service users.

A third possible objection is that our account renders all use of treatment pressure in mental healthcare coercive because there is always some level of dependency and some possibility of formal coercion in this context. True as the latter claim may be, it does not imply that this dependency is always so strong or the possibility of the use of formal coercion so salient that it would justify a service user’s belief that rejecting a proposal will leave her worse off than in the moral baseline position. Our analysis of the case vignettes precisely illustrates that our account can discriminate between situations in which the use of treatment pressure is coercive and situations in which it is not.

A fourth possible objection is that our account renders the scope of professionals’ responsibility too wide by allowing for the possibility of unintentional informal coercion. We acknowledge that our account allows for the possibility of unintentional informal coercion and that it broadens professionals’ responsibility in comparison to ordinary accounts. Yet, we deny that our account renders the scope of responsibility too wide. To be able to see why, note that the implied scope is not as broad as it may seem at first sight. The reason is that on many theories of responsibility, one typically is morally responsible for performing a morally impermissible action only if one performed that action intentionally, and one typically has a valid excuse if this condition of intentionality is not fulfilled (Scholten 2020; Wallace 1994). It can thus be true that a professional unintentionally (and impermissibly) exerted informal coercion and yet false that she is morally responsible for doing so. That said, it becomes harder for professionals to sustain the excuse of ignorance once they have been made aware of the possibility of unintentional informal coercion and the contextual factors that contribute to it. Once aware of this, professionals may be held morally responsible for exerting informal coercion unknowingly on account of culpable ignorance (see Rudy-Hiller 2022). We take this to be a plausible implication.

A fifth and final possible objection is that our account implausibly allows for unintentional informal coercion. In support of this objection, one could cite the intuitive idea that there can be no coercee without a coercer. Yet note that our account is not in conflict with this idea: we do not claim that there can be a coercee without a coercer, but only that there can be a coercee without someone being aware of being a coercer. Moreover, we consider it a strength of our account that it allows for unintentional informal coercion because this aligns well with the empirical evidence. Empirical studies found that service users report a higher prevalence of informal coercion than professionals. For example, two empirical studies found that in more than half of the cases in which service users reported informal coercion, nurses asserted that they merely used persuasion (Eriksson and Westrin 1995; Westrin and Nilstun 2000). Valenti and Giacco (2022) furthermore showed that even among professionals there is disagreement about the coerciveness of treatment pressures. Other empirical studies indicated that professionals tended to underestimate the level of treatment pressure and that the degree of underestimation increased with the level of pressure (Elmer et al. 2018; Jaeger et al. 2014; Schori et al. 2018). These findings suggest that professionals sometimes exert informal coercion unknowingly, and our account of informal coercion provides an explanation of these findings.

Relatedly, we consider it a strength of our account that it takes the perspectives of service users seriously and grants their experiences and beliefs a pivotal role in the assessment of informal coercion, though without turning the assessment of informal coercion into a merely subjective issue. The inclusion of these experiences and beliefs in our conceptual analysis of informal coercion supports service user testimonies that would be discarded on traditional moral baseline accounts of informal coercion.

CONCLUSIONS AND CLINICAL RECOMMENDATIONS

Based on a set of clinical counterexamples, we have argued that Szmukler and Appelbaum’s (2008) account of informal coercion in mental healthcare is insufficiently sensitive to context. We proposed an alternative account of informal coercion in mental healthcare and showed that it can explain our intuitions about a set of case vignettes. Our proposed recipient-focused account challenges the view that only threats can be coercive and allows for the possibility that professionals exert informal coercion unknowingly.

The following clinical recommendations can be derived from our account:

• Professionals should reflect critically on their communication with service users and consider the context of communicative interaction (e.g., who communicates with a service user when, where, and how).

• Hospital staff and professionals should work toward reducing the power imbalance in mental healthcare by removing relevant contextual factors (e.g., by adopting a respectful attitude toward service users and changing ward architecture).

• Professionals should be transparent in their communication and make explicit that service users can turn down treatment offers without facing disadvantages beyond those possibly entailed by the treatment rejection itself.

• Professionals should take seriously the testimonies of service users who report having been coerced into accepting treatment also in cases where no explicit threat was issued.

AUTHOR CONTRIBUTIONS

All authors contributed to the design of the work and participated in team sessions in which each step of the argument was discussed until consensus was reached. CH and MS wrote the various drafts of the manuscript and EB, SP, and JG revised them for important intellectual content. All authors agree to the article’s arguments and conclusions and gave their approval for the final version to be published.

ACKNOWLEDGMENTS

We would like to thank George Szmukler, Thomas Schramme, Johannes Müller-Salo, our colleagues in Bochum and audiences in Oxford and Varese for their valuable comments on earlier versions of the manuscript.

DISCLOSURE STATEMENT

The authors report there are no competing interests to declare.

Additional information

Funding

This research is part of the project SALUS (2018-2024) and is supported by the German Federal Ministry of Education and Research (grant number 01GP1792).

Notes

1 Other models are available in the literature which overlap with Szmukler and Appelbaum’s (2008) model to varying degrees. For example, Olsen (2003) distinguishes between advice, social pressure, manipulation of social forces, and manipulation of resources; Blumenthal-Barby et al. (2013) distinguish between direct recommendation, appeal to patient’s values and goals, intentional framing of information or options, offering concrete incentives, leveling concrete threats, and deception; Canvin et al. (2013) and Valenti and Giacco (2022) differentiate between leveraged and non-leveraged pressures; and Pelto-Piri et al. (2019) suggest expanding Szmukler and Appelbaum’s (2008) model with three additional types of influence, namely cheating, using a disciplinary style, and referring to rules and routines. We will focus on Szmukler and Appelbaum’s (2008) taxonomy of informal coercion because it has been the most influential and will stick strictly to the terms as we define and illustrate them here.

2 Authors in the philosophical debate refer to what we call “informal coercion” by means of the term “coercion”. To avoid potential confusion of informal and formal coercion among readers familiar with the mental health ethics literature, we use the term “informal coercion” in this article, except in sections 2 and 3, where we engage exclusively with the philosophical literature.

3 Answering the question of where to set the moral baseline is beyond the scope of this article. We follow Wertheimer (1987) in understanding the moral baseline as being determined by the rights of a proposal’s recipient and remain neutral regarding the content of these rights. Dunn et al. (2012) propose to determine the moral baseline based on the mental health professionals’ duties of care.