What are dually diagnosed patients' problems represented to be in mental health? A WPR analysis of the multistability purpose of digital health records.

No previous studies have investigated how political measures, opinions and views of people with dual diagnoses, organisational requirements and professional values are purposefully communicated, mediated and/or integrated in digital records in mental health care. It remains unclear how health records function as both clinical vehicles for documentation, audit and quality assurance in patient care as well as political vehicles of power to articulate and reproduce idealised relations among actors and their roles in mental health. Informed by Bacchi's 'What's the problem represented to be' (WPR) approach, we consider how problems of dually diagnosed persons with co-occurring mental health conditions and drug use (DDPs) are represented and how the tools required to fix the patients' problems are communicated and legitimised in 10 digital health records consisting of 3830 pages of data. In the discussion, we distinguish their multiple purposes by discussing how their structure is shaped by broader discourses in health care. Then, we consider how the structure of records delineates professionals' autonomy, discretionary action and service provision in ways that limit possibilities for including mental health patients' experiences and exclude service users' voices from care planning.

INTRODUCTION

In mental health, persons dually diagnosed with co‐occurring mental health and drug use (DDPs) problems represent a dilemma‐ridden challenge for professionals in terms of providing coherent and effective treatment and help (Brekke, Lien, & Biong, 2018; Ness et al., 2014). Professionals’ discretionary actions and ways of providing mental health care to DDPs are known to be shaped by interactions between mismatching policies, organisational requirements such as patient consent and flow and incommensurable professional values (Bjerge et al., 2020). However, the dilemma‐ridden nature of this challenge also derives from different models of care, lack of specialised knowledge about drug treatment, as well as cultural and gendered orders of worth within the hierarchies of mental health (Johansen, 2018; Oute et al., 2018; Schepelern Johansen, 2005). Despite health records being a digital technology provided by political stakeholders that both shapes and reflects certain views about what matters in daily service provision (Moerenhout et al., 2020), existing research is less clear about the structuring and mediating effects of the different purposes of health records in mental health (Aaslestad, 2007; Berring et al., 2015; Hamilton & Manias, 2006). No previous studies have investigated how political measures, value‐laden opinions, views of DDPs, organisational requirements and professional values are purposefully communicated, mediated and/or integrated in digital records in mental health care. This article considers how the structure, communicative style and purposes of records shape service provision and what ramifications their constitution may have for professionals’ discretionary action and service users’ participation in mental health‐care planning.

BACKGROUND

Analysis of communication in/of health records is surely not new. Especially within nursing research, the constitution and written communication of/in health‐care records have been scrutinised (Ameel et al., 2020; Hamilton & Manias, 2006; Heartfield, 1996; Hyde et al., 2005; Moerenhout et al., 2020; Mohr, 1999; Zabielska, 2021). Most of this research primarily draws on a Foucauldian framework (Ameel et al., 2020; Hamilton & Manias, 2006; Heartfield, 1996; Hyde et al., 2005, 2006; Irving et al., 2006; Mohr, 1999), Habermas (Hyde et al., 2005) or linguistic theories (Berring et al., 2015; Hov et al., 2020). Recent research tends to draw more on post‐phenomenological theory and perspectives from science and technology studies (STS) to consider how digital records become actants that can bring into effect various issues such as concerns about professionalism or systemic rights violations following the technologies’ capacity to serve multiple purposes (de Boer, 2021; Moerenhout et al., 2020).

These studies have in common that they critically examine the developing discourses, communication and language used in constructing patients, patients’ needs and professional responses to them in records. An extensive Norwegian study illuminated the progression of four historical stages in mental health records between 1890 and 1990 and suggested that the genre and role of the narrator’s voice developed from considering ‘the patient as text’ towards presenting patients with a more humane understanding (Aaslestad, 2007). However, the existing research generally suggests that patients are primarily articulated within a biomedical discourse, even in patients whose illness(es) and suffering also relate to substance use, poverty and homelessness as well as to economic and legal factors (Hamilton & Manias, 2006; Heartfield, 1996; Hyde et al., 2005). Some studies specify how the day‐to‐day communication about patients’ and professionals’ roles are shaped by specific expectations of patients’ behaviour and intentions (Hamilton & Manias, 2006; Irving et al., 2006; Mohr, 1999). In this line of thought, studies also point to how patients’ voices are hardly represented in records and how this supports an authoritarian position of medical professionals (Aaslestad, 2007; Hyde et al., 2005, 2006; Zabielska, 2021). The existing body of research on the communicative style and rhetorical function of records favours analyses of their content rather than their structure. Thus, this research primarily shows how they are either considered as technologies to document and store data on incidents, symptoms and prescriptions of medication or technologies reflecting the guiding principles, discourses and defining powers of professionals in health care.

To grasp the variety of functions of records, Moerenhout et al. (2020) have considered their ‘multistability’ purpose: ‘Multistability means a technology can be put to multiple purposes or has multiple stabilities or variations. These different stabilities can be (but do not have to be) predetermined in the design’ (p. 230). They illustrate the idea of multistability by way of the famous image in which a rabbit and a duck can be seen. The observer can either see a rabbit or a duck in the same image, but both animals cannot be seen at the same time (Moerenhout et al., 2020). However, de Boer (2021) points to how this post‐phenomenological idea of the multistability purpose means that technologies can be used for a variety (‘multiple’) of purposes but that their purposes are not infinitely many (i.e. the ‘stability’). This means that these ideas can be applied to conceptualise how records can serve multiple purposes such as documentation, storage and interprofessional information handover, while also flagging or promoting the inherent role expectations of medical discourse and/or the values of recent years’ neoliberal emphasis on risk management and cost reduction in contemporary health care (Moerenhout et al., 2020).

However, the existing literature seems less clear in regard to how health records work as clinical tools to document, store and audit data as well as how they work as political vehicles of power to articulate and reproduce idealised relations among actors and their roles in mental health. At first glance, critical discourse analysis (Fairclough, 2020) or discourse theory (Laclau & Mouffe, 2001) seems like purposeful analytical choices to study the purposes of records. But apart from their dissimilar distinctions of discourse and society that do not address welfare policy, they would more exclusively be concerned with considering linguistic features, intertextuality and potentials for social change or examining hegemony and contestation, respectively. For the purposes of covering the above gap in the existing research, it is more suitable to use generic concepts from Bacchi’s (2009) analytical ‘What is the problem represented to be’ approach (WPR) as this enables us to flesh out the structure of the actual medical content, arguments and political reasoning informing representations of the patients’ problems in the health records.

WPR: ANALYTICAL FRAMEWORK

The WPR approach is a welfare‐sociological approach that presents a set of particularly useful ideas and questions to ask when considering communication in/of health records. WPR has previously been applied in policy analysis, both within alcohol and drug studies (Duke, 2020; Houborg et al., 2020; Lancaster et al., 2019) and in critical mental health (McPherson & Oute, 2021; Oute & Johansen, 2021). As opposed to critical discourse analysis and discourse theory, it explicitly draws on the Foucauldian concept of ‘problematization’ as a way to analyse taken‐for‐granted assumptions that constitute something/someone as an ‘object of thought’ (Bacchi & Goodwin, 2016; Houborg et al., 2020; Lancaster et al., 2019). Bacchi argues that policies do not simply mirror problems that already exist independently ‘out there’ as isolated and evidence‐based attempts to respond to them. Rather than predetermining the nature or content of political problematisations and a range of desired tools for intervention, Bacchi’s WPR approach provides generic concepts to analyse how both politicised ways of presenting problems and legitimising proposals for change are generic traits of welfare rationality in/of policy texts or in our case health records (Bacchi, 2009). This resembles Foucault’s works on ‘governmentality’, as Bacchi points to the importance of analysing how various ways of thinking about and exercising government always involve how welfare rationalities and historical knowledge practices make problems become ‘real’ (Bacchi & Bonham, 2014). This perspective implies how rationalities and knowledge render problematic subjectivities to become thinkable in ways that make them amenable to specific modes of governance directed towards transforming the problematic subject into its often implicit and idealised binary negation (Rose, 1999).

This constructionist lens is useful in elucidating what the generic ‘problem’ of mental health and drug use is represented to be and assessing what tools are presented as required means to fix patients’ problems in the data. However, such considerations also call for an investigation of how interventions are justified (Bacchi & Goodwin, 2016). This framework provides a lens to grasp how dominant ways to present certain target groups’ problems are used to legitimise the use of certain tools to remedy the problems (Bacchi, 2009). We use the WPR approach to analyse how generic problems are defined, what terminology is used and how it informs the interventions and remedies suggested in the expectation of certain outcomes in mental health records. As political constructions are promoted in/of digital technologies (Bacchi & Goodwin, 2016; Lancaster et al., 2019), the WPR approach also enables us to investigate how problem representations and legitimisations of certain interventions are co‐produced by the purposeful structure and communicative style in/of the records. That is how the purposes of digital records signal broader contemporary ways of formulating and justifying idealised schemata for representing, analysing and rectifying reality in ways that align with broader political and/or governmental rationalities (Bacchi & Goodwin, 2016; Rose, 1999). In this sense, Bacchi’s WPR approach also informs further discussion of how problem representations are permeated by pre‐existing ideologies and historical orders of the worth of mental health and beyond.

AIM

The aim of this article is twofold. Informed by the WPR approach, we consider how DDPs’ problems are represented and how the range of tools required to fix the patients’ problems are described and legitimised in/by digital health records. Then, we distinguish their multiple purposes by discussing how broader discourses shape the structure of records in health care. Finally, we discuss how the structure of the records delineates professionals’ autonomy, discretionary action and service provision in ways that limit possibilities for including DDPs’ experiences and exclude users’ voices from care planning.

METHODS

We collected health records at two treatment units providing in‐patient care for people with severe mental health and substance use disorders in the Capital area and a Southern part of Denmark. In collaboration with the ward management and staff, the records were purposefully collected by identifying admitted or formerly admitted patients who met the following inclusion criteria: Adult (+18) persons who had formal ICD10/11 diagnoses for a co‐occurring mental health condition (F20, F30, F60) and a drug misuse/addiction (F10‐F19) while also having been enrolled in employment services within the last 2 years.

RECRUITMENT

Based on these criteria, the staff identified relevant patients who were then asked if they were willing to give their written, informed consent to participate by letting their entire record be used as data in the study. They were also informed that declining consent would not affect their treatment in any way. After receiving consent from the patients, the staff ensured a digital copy of each of the patients’ digital records from the various versions of electronic patient record systems with different local interfaces entitled COSMIC, EPIC and ‘the health platform’.

SAMPLE

The sample included the complete psychiatric treatment records of eight male and two female patients. Other information about medical history (fractures, wounds, diabetes and cardiovascular disease) was excluded. The sample varied from 21 to 58 years of age with an average age of 40 years. In terms of ICD‐diagnoses (F), the sample records belonged to patients diagnosed with severe mental health conditions such as schizophrenia spectrum, affective and personality disorders (F20.0, 20.9, 21, 22.0, 60.3, 33.0 and 60.9) and addictions to different substances, predominantly opioids, cannabis, benzodiazepines and Ketamine (F11.22, 12.2, 14.2, 14.8, 19.1). The records did not report the patient’s total number of admissions.

DATA

The included records consist of a total of 3830 pages of data. The extracted data stem from notes from consultations and observations during in‐patient treatment at the above‐mentioned two units. The records are structured into different sections in which notes are designated to different types of medical professionals (doctors, psychiatrists, clinical psychologists, nurses and social workers), who describe the condition of the patient and the required treatments, a list of prescribed drugs and attachments that include information/data from other welfare systems (e.g. drug treatment, employment and housing services). Professionals’ consultation notes and observations are organised in different and again designated sections such as physical condition, diagnoses, addictions and social concerns that refer to a range of pre‐defined and quality‐assured problem areas that must be considered or ticked off in the record by the professional at each consultation or shift.

ETHICS

In Denmark, social studies are not formally regulated by host institutions, and the Danish National Committee on Health Research Ethics only assesses biomedical studies such as psychosurgery or clinical trials. Nevertheless, psychiatric frontlines and Danish Health authorities both informally and formally regulate the access to obtaining patients’ consent to allow the inclusion of their records in the study (Oute & Bjerge, 2019). Access to the records was thus thoroughly regulated, although approval to conduct this study from the host institution and the Danish National Committee on Health Research Ethics is not a legal requirement, as WPR studies are a form of inquiry rooted in the social sciences. In accordance with the Helsinki declaration and Danish Health Law, the researchers obtained written informed consent from the 10 patients included in this study. Each of the patients consented to researchers utilising their entire health record and all of its content for the purposes of this study. The researchers also abided by the Act of Processing Personal Data by getting approval from the Danish Data Protection Agency, journal 2015‐57‐0002; case number 2016‐051‐000001.

ANALYSIS

To consider how digital health records work as clinical–political vehicles in mental health care, the coding was informed by Bacchi’s WPR framework as outlined above. Bacchi’s notions of problematisation, legitimisation and tools for intervention guided how the first author deductively coded the content of the records within the generic, pre‐defined categories of (1) problem representations, (2) legitimisations and (3) tools for interventions. The coding worked up and specified the content of three analytical categories that we called (A) the representation of the complex patient, (B) legitimising health promotive, illness preventive and individualising interventions and (C) the representation of tools for intervention. NVivo 11/12 was used to code data and manage categories. A research assistant and the second author validated the coding and categorisation, which were adjusted until consensus was reached. The joint coding process revealed that the pre‐decided, generic categories accurately resembled the pre‐structured articulations of the patient subjects, the patients’ diagnoses, risk determinants, treatment goals and suggested interventions as predicted by Bacchi (2009). This insight led us to maintain the pre‐decided structure of categories and use this structure to present their specific articulations within these generic categories below.

THE REPRESENTATION OF THE COMPLEX PATIENT

The representation of the patient relies on the deficit model. This implies that the generic patient subject is presented as deficient throughout the records. Moreover, the bio‐psycho‐social model is used to articulate and distinguish how the patient’s deficiencies are constituted by three subtypes of risk determinants and treatment indications. In effect, their structure and pre‐decided categories represent the patient as the negation of an idealised notion of a responsible, mentally healthy, self‐managing and neither burdensome nor risky subject. However, more subtly, the patient subject is also presented as the implicit negation of an authoritarian and unwavering professional subject by way of a shadowy medical voice throughout the records (see Discussion section).

The deficit model signals a common feature of the representations of the patient’s addictions, substance use disorders and distress. This way of representing the patients’ negatives (deficits) implicitly signals a potential to turn them into positives (Harper & Speed, 2014; Keane, 2017). As the patient is continuously articulated using the abbreviation ‘PT’ across the records, this communicative style depicts the patient as an addicted and ambiguous object who lacks insight and illness recognition, self‐control and responsibility (cf. addiction), motivation to comply with treatment and the ability to regulate emotions and cope with distress. In effect, the patient such as the 27‐year‐old male described below is presented as a complex and wavering subject with the potential to change:

Things went pear‐shaped last Thursday. He had an appointment with a friend. She called him up when he was on the bus, and she asked him if it is ok to bring two others along. He chose to say it is ok. When he arrived, they had already started to misuse. Alcohol, cocaine and joints were on the table. He got overwhelmed by craving, and he sensed that he was really not capable of resisting the urge. He misused two lines of cocaine and took four hits off the joint. He instantly felt bad and paranoid (under surveillance) and got tired.

The excerpt depicts how a DDP is presented as poorly emotionally regulated, not adequately motivated and addicted. This way of forecasting the patient’s poor emotion regulation and addictive behaviour rhetorically informs the presentation of the patient’s urge and ambiguity about complying with the medical and cognitive–behavioural treatments. Resembling the articulation of all cases in this study, these signs were presented as symptoms of an underlying mental illness. The constitutive symptoms of patients’ conditions were presented using generic descriptions of delusional and black/white thinking, high risk of physical illness, poor emotion regulation, sleep deprivation, high levels of aggression, violence, and unpredictability, poor illness recognition, inadequate coping strategies and lack of treatment compliance:

His symptoms are: Black/white thinking—lack of impulse control—difficulties about letting go—relational issues. He pushes people away with his aggressive behaviour—contentious in cases of impulsive frustration—unstable affect (particularly anger)—lack of endurance—unstable and moody—intense and unstable relations.

As with this 38‐year‐old male patient, the DDP is generally depicted ambiguously, highlighting the belief that patients can become self‐managing via motivation and participation in treatment. This potential is also articulated by hinting that patients feel shameful due to a recognition of their inappropriate behaviour, for example, by stating that ‘afterwards he [patient] thinks that it is really fucked up’. In the same way, this potential was also suggested by outlining how effective treatment had been:

In many ways, PT has recovered during the admission; he has become mentally and physically rehabilitated; he has recovered entirely from his cannabis addiction. Furthermore, he has quit benzodiazepines, and he is for the time being, considered well‐medicated. The patient displays no paranoid delusions in the ward where he feels well and does not think that he is being surveilled by anybody. However, he maintains that he is being surveilled and harassed when he is in his flat, and the patient firmly believes that these [experiences of being surveilled etc.] are real and has no recognition of it being a sign of illness. The patient therefore considers that he could just as well stop taking the antipsychotic medication.

The representation of the patient subject, such as the case of this 58‐year‐old male, can be seen as hopeful. However, the potential of turning the patient’s negatives into positives is presented as being determined by three physical, behavioural and social risk factors that refer back to the bio‐psycho‐social model that has been widely adopted in western psychiatry (Read et al., 2009).

Risk determinants and treatment indications

The articulation of these determining elements emerged in several ways across the records, for example, as descriptions of how the bio‐psycho‐social model is used in assessments of patients:

We have reviewed the bio‐psycho‐social model, and she [the 35‐year‐old female patient] stated that she grew up with two parents who were addicts that did not validate her.

The bio‐psycho‐social determinants are described as physical health concerns, formal and informal presentations of diagnoses and addictions as well as depictions of the patient’s social background. The latter includes assessments of family history and patients’ exposure to burdensome social circumstances such as poverty, educational and vocational status and sexual and relational trauma during childhood or early adulthood.

Physical health determinants

In segregated sections of the records, the patients’ physical health determinants are portrayed via a pre‐structured focus on the so‐called KRAM factors that include diet (K), smoking (R), alcohol (A) and exercise (M) (Eriksen, 2016). This preventive focus on physical risk factors emerged as presentations of how patients are submitted to different measures such as pulse, blood pressure (BP), electrocardiograms and screening for obstructive pulmonary disease, weight gain and obesity (BMI). These determining risk factors for the patient’s ill health were presented via articulations of screening and preventive measures for migraine, healthy diet, increased exercise and decreased alcohol intake, while each of these pre‐defined physical measures was also presented as subject to change via specific intervention.

Diagnoses and addictions

The patients’ characteristics were articulated by way of short depictions of the patients in conjunction with both formal diagnoses and informal labels. In some cases, the narrator’s subtle voice in the records articulated the formally used ICD‐10 diagnoses such as ‘DF20.0 Schizophrenia, DF11.22 opioid‐addiction syndrome, DF12.2 cannabis addiction syndrome’ as being based on semi‐structured assessments such as the SCID‐II. However, diagnoses were presented without referral to any structured assessment in most cases.

More often, informal terminology was thus used to present the patients. A common trait of these articulations often encompassed the use of value‐laden depictions of the patient as being non‐compliant or poorly motivated. For example, it is explained that ‘Abilify [psychotropic medication] is tried but, for the time being, compliance is poor’, or elsewhere it is stated that ‘pt expressed deficient motivation’. Furthermore, patients and their problems were generically labelled as addicts and/or addicted in the records, for example, ‘[32‐year‐old male] known with drug addictions and ADHD. Previous admissions with suicidal behaviour and ideation, admitted now. Be aware of veneficium [poisoning]’. This jeopardising way of representing patients emerged in many forms, and the data held several conclusions such as the case of a

[32‐year‐old male] who in the record is described with addictive behaviour and suicidal ideation. In addition, a history of violence and mood swings is described. Hence, the patient must be characterized as dangerous.

Anamnesis and social history

The patients’ characteristics were also presented via depictions of the medical and family history, which were often presented as the root social causes of the patient’s ill health. This portrayal focussed on determining risk factors and behavioural issues, mental illness and addictions in patients’ proximate relations or past:

PT [42‐year‐old male] finishes school but starts to smoke cannabis at the age of fifteen. At first for pleasure with friends. PT starts as a carpenter apprentice but drops out due to increasing daily misuse of mixed substances such as Ketamine and later cocaine and heroin.

The representations of the social or historical constitution of the subject’s problems were connected to depictions of turmoil, social isolation, exclusion, strained and abusive family relations and previous addiction among the close family relations: ‘The patient had a chaotic upbringing with an addicted father using alcohol and illicit drugs’. The depiction of the social background of the patients’ problems was often underscored by representations of the patients’ traumatic childhood experiences. This emerged via descriptions of how the patient ‘had a hard childhood with an addicted mother and a father who exercised incest on her. She was also raped by friends of her mother’. In doing so, the subject’s mental health problems and addictions were displayed as an effect of family disruption, abusive relations, social marginalisation, homelessness and early life trauma.

As the above bio‐psycho‐social elements are embedded in the structure of the records, they guide the collection of clinical data about the patient. Furthermore, they also determine how professionals’ generic portrayal of the patients shapes and legitimises the proposition and use of certain interventions that, to some extent, align with the patients’ generic problems.

LEGITIMISING HEALTH PROMOTIVE, ILLNESS PREVENTIVE AND INDIVIDUALISING INTERVENTIONS

The above portrayal of the complex patient informs risk‐oriented and managerial justifications of the patients’ needs for intervention that cover short‐ and long‐term goals for patients’ physical health, behaviour, illness prevention and individualised treatment. How the patient’s needs for intervention are legitimised emerged via representations of three generic types of goals. In turn, physical, behavioural and social goals are presented as both short‐ and long‐term.

Physical goals

The goals are also presented via a pre‐structured, political focus on the illness—preventive KRAM—factors that claim to promote health by diminishing the individual’s risk of ill health by improving diet, decreasing smoking, diminishing alcohol consumption and encouraging physical exercise (Eriksen, 2016). These objectives are presented via descriptions of a range of measures; for example, the patient has to ‘lose weight’, ‘comply with the procedure of monitoring blood pressure’, map out the patient’s diet and enhance knowledge for the patient to choose a new and healthier lifestyle.

PT [a 45‐year‐old female] followed the Wednesday procedure and has been weighed. Abdominal width is measured, and BP + BMI (Body Mass Index) has increased to 35,8, which sets off a diet plan. PT wishes to lose weight but has no idea about healthy and unhealthy diet and the amount of nutrition the body needs.

Additionally, health‐promoting objectives regarding exercise were in some cases simply presented as ‘exercise more’ but were often articulated in a slightly more detailed form, for example, ‘within four weeks […] the physiotherapist supports me in being physically active’. Taken together, the political imperative to be physically healthy implicitly justifies these measures for the individual to minimise health risks and diminish the burden of cost to society (Eriksen, 2016).

Behavioural goals

The goals pertaining to patients’ diagnoses were articulated by presenting how remission (withdrawal of symptoms) of the psychiatric disorder was monitored and ensured. Abstinence was formulated as ‘quitting drug misuse’, and sobriety was presented as

Treatment objective: to control, treat and monitor the mental illness and to support the patient in managing everyday life requirements in a more independent way. Moreover, to support PT in achieving a better life quality so that she keeps clean.

Such goals were combined with long‐term personal goals to minimise distress, enhance illness management and increase long‐term motivation and treatment adherence:

PT’s [21‐year‐old male] objectives (for treatment): 1. End cannabis misuse and use alternative strategies in cases of urge to misuse. 2. Phase out benzodiazepine treatment.

Short‐term goals were set to end misuse; for example, by describing that: ‘Within six weeks, I [21‐year‐old male] will be capable of regulating my feelings by using emotion regulation skills, [from DBT] so that I will feel better.’ The short‐ and long‐term objectives refer to the generic patient’s transformative potential to achieve remission, become abstinent and enhance cognitive–behavioural skills to control the urge to use drugs and recover. While the patient’s behavioural deficits legitimise the use of medical treatments and CBT/DBT (cognitive–behavioural treatment/Dialectical Behaviour Therapy) interventions, these goals legitimised interventions by promising a reduction of harms for the individual. Such goals would also minimise costs related to social benefits, sickness leave and exclusion from the labour market (Bjerge et al., 2020).

Social goals pertaining to medical history and social background

Social objectives pertained to diminishing determining risk factors and improving mental health. As opposed to the former physical and behavioural goals, social objectives were often communicated using what can be seen as a more humane voice (Aaslestad, 2007). While ‘sympathy’ was only explicated in a few cases, the ‘humane’ representations of goals relating to medical history and social background emerged via a more personalised voice in the records:

She [a 35‐year‐old female] is heterosexual. She has been exposed to many sexual assaults from early childhood, but also later in life. Currently, she feels that it is difficult to reject men who ask her to have sex, which is something that she works with in DBT training.

Representations of sexual trauma and strained family relations in early life justified goals to promote health by focussing on aiding the patient in developing a new sense of self, managing difficult emotions and developing strategies to maintain ‘healthy’ (sexual or non‐sexual) relations via DBT training. Such short‐term goals were justified by a promise of diminishing exposure to and improving personal management of abusive relationships and childhood trauma, thereby promoting mental health and preventing relapse.

Generic long‐term objectives were connected to continuing outpatient treatment and improving occupational, financial and housing situation in everyday life outside psychiatry:

A new place to stay. Required work to arrive at the objectives: to be active in treatment, e.g., participating in groups and social life on the ward; finding other housing opportunities.

Echoing recent developments in Nordic mental health policy (Oute & Johansen, 2021), the representation of the goals to fix the patient’s lack of housing, financial security or employment suggests that the responsibility of fixing the patients’ social problems is indeed supported in mental health care. However, these are mainly individualised and considered the responsibility of municipal and civil actors.

THE REPRESENTATION OF TOOLS FOR INTERVENTION

This final theme describes how interventions such as medical treatment, CBT, DBT and the recovery star are presented as tools to individualise and remedy patients’ problems in the records. The suggested range of tools is presented as meaningful and legitimate ways to intervene with a particular view to remedying the physical, behavioural and social risk determinants that concur with the generic representation of the subject’s bio‐psycho‐social problems.

Short‐term goal interventions

The tools suggested to arrive at the physical goals cover monitoring tools such as BP, abdominal width, weight and BMI measures. Structuring or planning interventions such as diet plans and planning of physical exercise and physical education (led by dietitian, physiotherapy or nurse, respectively) were commonly presented as ways to prevent future illness. Finally, medication was commonly used to remedy elevated BP and heart or pulmonary diseases.

Interventions referring to the behavioural goals cover various medical treatments to ensure remission and prevent relapse. These medical tools were described in the medication chart in the records, and they encompass prescriptions of various psychotropic medications. Antipsychotic drugs such as Olanzapine, Aripiprazole, Clozapine and various benzodiazepines such as Nitrazepam and Diazepam and finally antidepressants (particularly SSRIs) and mood stabilisers such as Sertraline, Fluoxetine, Lithium and Carbamazepine were prescribed. The drugs were recommended to be taken orally and/or as injections to remedy, but not necessarily cure, schizophrenia and affective disorders. The drugs were prescribed to be taken one at a time, but often several of these drugs were prescribed in combination with each other to remedy the patient’s complex health problems. To manage the subject’s addictions, substitute treatment interventions such as Methadone (liquid and pills) and Buprenorphine were prescribed.

Aligning with patients’ personal goals, interventions such as NADA treatment (needle treatment for abstinence symptoms) were at times suggested. Tools from the cognitive–behavioural toolbox were often presented by describing that ‘we decided that he prepares a list of pros and cons on doing cocaine’. These tools were presented as ways to enhance the patient’s illness and addiction management, increase long‐term motivation, compliance, treatment adherence and prevent relapse. Cognitive restructuring (CR), motivational techniques (e.g. pro and con lists), chain analyses, cognitive enhancement, mindfulness training (e.g. meditation) and DBT such as emotion regulation techniques were presented: ‘I must participate actively in DBT and do homework. Contact person supports me in DBT. Doctor: Weekly consultation and medication control’.

Finally, managerial techniques were articulated as ways to arrive at short‐term behavioural goals. In particular, patient involvement was highlighted to ensure participation in treatment by articulating that the ‘patient consents to the treatment plan’. Consent was thus presented as a measure to increase motivation to adhere to treatment during admission while also enhancing the patients’ potential to become self‐managing. In particular, two interventions were presented as tools to foster the patient’s self‐management. First, a locally developed version of the recovery star that was informed by the CHIME framework was used to ensure patients’ self‐management (Boholdt‐Petersen et al., 2019; Leamy et al., 2011). The patient and professional collaborated on rating each item from 0 to 10 within the following parameters: Management of mental and physical health, self‐care, life mastery, social networks, relations, substance abuse, responsibilities, identity, sense of self, trust and hope. Second, structured ‘psychoeducation’ for both patients and relatives was presented as a means to enhance self‐management to optimise treatment adherence and illness prevention. This aligns with a responsibilisation agenda from the Nordic region and the UK, known as a general mental health policy to increase patients’ self‐management and adherence by individualising patients’, relatives’ and civil actors’ treatment responsibilities (McPherson & Oute, 2021).

Long‐term goal interventions

The patient’s consent was also presented as a key measure to enable the involvement of municipal, regional or civil actors who could potentially manage and achieve the patient’s long‐term goals after admission. The patient’s consent to in‐house collaboration and referral to a range of external actors were presented as a prerequisite to go beyond short‐term interventions pertaining to diminishing health risks and relational and social problems factors over time.

Long‐term interventions to remedy social health determinants included collaboration with, but almost exclusively referral to, other welfare actors from outside mental health care. Such actors were presented as tools to ensure a safe and foreseeable transition from in‐patient to outpatient or community care and to ensure contact with probation services. It could also be to maintain long‐term medical and addiction treatment, to enable the patient’s participation in occupational services or competitive employment, to improve the patients’ often poor financial situation and locate a temporary or permanent place to stay for the patient and to enable management of educational, occupational and family problems. However, often without considering their specific purposes, the records presented such actors from outside mental health as tools to address the root social causes of the patients’ ill health. Actors such as outpatient professionals, mobile treatment units, peer support workers, municipal mentors, lawyers, housing services and homelessness workers, child service professionals, outpatient treatment staff, GPs, drug treatment professionals, probation officers, Clubhouse (Fountain House) members and/or relatives such as parents, siblings or friends were presented.

DISCUSSION

In following Bacchi’s constructivist approach to analysing how articulatory practices in/of records constitute politicised subjectivities (Bacchi, 2009), we shall discuss how broader discourses in mental health shape the political target subjects before considering the multistability purpose of the records and its implications.

The social constitution of records

Although our heterogeneous sample records belonging to DDPs are derived from Danish in‐patient treatment, our analysis echoes Harper and Speed’s (2014) acknowledged analysis of the persistence of the deficit model in mental health. This suggests how the generic construction of the patient relies on an underlying aetiology of mental health problems that asserts that mental patients are constituted by a range of deficits that ‘are predicated on existing negatives (deficits) being reframed as potential positives’ (p. 47). As the bio‐psycho‐social problem areas are embedded in the very communicative structure of the records, the presentation of these determining risk factors forecasts that the patient is a costly, physically risky, mentally ill, poorly regulated, addictive and socially marginalised subject. However, the records do not articulate a traditional pessimistic notion of a chronic and addicted mental health patient (Davidson et al., 2009). Rather, the records reflect a bio‐political notion of the target subject that promotes that the patient’s bio‐psycho‐social negatives can be transformed into a (Rose, 2019): (1) less costly, healthy and non‐risky; (2) self‐managing, compliant, responsible, insightful and self‐controlling (cf., addiction as loss of self‐control) and (3) socially included, employed and autonomously living subject who is not dependent on welfare services.

However, the former binary formulation goes hand in hand with and is shaped by the articulations of the professional that are communicated subtly in the records via the use of a passive voice. The constant use of passive communication—that is, ‘PT is described’, ‘is tired’ or ‘is known as’—articulates the professional as the patient subject’s field of discursivity (discursive outside). The representation of the looming presence of the professional in the records reflects a subject that constantly and objectively monitors the patient, insightfully supports appropriate forms of coping and has a superior capacity to ensure the patient’s participation and treatment compliance. This passive representation suggests that the generic professional aligns with the position of the Aristotelian notion of ‘an unmoved mover’ (Jørgensen, 2005). Although this communicative style might not represent an intentional choice of frontline workers, the representation of the shadowy voice brings into effect the professional as a distant and unquestionable god‐like figure. This communicative style thus functions as a way to uphold superior authority to make true judgements about and on behalf of the patient, for the benefit of the patient and for lessening the patients’ burden on society.

Multistability purpose

The authoritarian voice goes hand in hand with the pre‐structured political ideals for minimising health risks and optimising self‐management. It does so, not only by remaining passive and unquestionable, but by communicating in ways that align with the inherent ideology of recent years’ austerity politics. However, user experiences have largely been excluded in policy and system development or considered tokenistic at the individual level (Andersen & Pors, 2019; Beck, 1997; McPherson & Beresford, 2019; Rose, 2019). Thus, it reiterates previous research to find that users’ perspectives were structurally and rhetorically excluded from the records (Rose, 2019; Zabielska, 2021).

However, while the communicative style of health records might be constituted by quasi‐objectivising medical discourse, the digital representation of the patients in our sample is simultaneously shaped by a political risk‐management model. This insight can be seen as a novel contribution to the existing body of research. Hence, our analysis questions the reliability of previous analyses. It does so by showing how their structure (that shapes professionals’ practices) is politicised rather than considering records as tools that reflect medical professionalism and action (Aaslestad, 2007; Berring et al., 2015; Hamilton & Manias, 2006; Hyde et al., 2005, 2006; Moerenhout et al., 2020). This illustrates the previous distinction about whether records metaphorically represent a rabbit/duck and indicates how previous studies have focussed on either the rabbit or the duck (Kerkin et al., 2018; Moerenhout et al., 2020). The issue of reliability might be an epistemological concern. By virtue of the constructivist WPR approach, our analysis is more exclusively concerned with the political nature of the structure and representations in/of the records. Thus, it considers to a lesser extent how records function as clinical vehicles to document data and share observations over time, and audit and quality assurance documentation on symptoms or prescription of medication. This disconnection between professional content and politicised structure and communication resembles Rose’s research of the interstices between ‘many psychiatries’ (Rose, 2019). In contrast to Rose’s suggestion that mental health is medically dominated, our analysis of the bio‐political purposes of the records suggests otherwise. This raises the wide‐ranging question if communication in/of educational materials or grant proposals in psychiatry would be under similar political influence.

Ramifications for professionals’ discretionary action and provision of care

While adding to our theoretical understanding of the multiple purposes of records, our analysis of their politicised structure and multistability purpose also speaks to a hotly debated issue. As mentioned, Danish Health authorities implemented different versions of digital health record systems such as COSMIC, EPIC and the health platform in the Danish regions that oversee health care (Bjerge et al., 2020). During the era of austerity politics, these systems have been implemented with a view to assure quality and optimise service provision (Bentzon & Rosenberg, 2021). In response, medical actors have criticised the records for being ‘a digital destruction’ by arguing that the politically structured records erode the foundation for medical professionalism by destroying medical professionals’ possibilities and legitimate rights to make clinical decisions autonomously (Bentzon & Rosenberg, 2021). This signals a rabbit/duck situation as the clinical actors advocate for a need to alter the politicised records in favour of clinical–professional purposes. However, the case of the implementation of digital records in Danish health care echoes Kleinman’s 25‐year‐old prediction, in which he (Kleinman, 1995) argued that political requirements to pre‐structure care provision on evidence‐based risk categories, such as the pre‐structured categories, would erode health professionals’ autonomy and discretionary action and the political recognition of academic professionalism in medicine. This questions whether or how health records will play a pivotal role in forcing professionals to risk losing political recognition by reclaiming their professional–academic care values or formally align them with political ideals about risk management to gain recognition in years to come.

This dilemma is facilitated by the structure of the records, but it also displays a larger but well‐known conflict in mental health policy and care. This conflict is found in the UN and WHO policies of providing recovery‐oriented and community health care. Similarly, the clinical recognition of patients’ explanatory models and experiences of their distress and illness have been seen as key features of what Kleinman denoted as academic psychiatry (Kleinman, 1991; United Nations, 2017). However, the risk‐oriented and managerial policy that informs the structure of the records and legitimises the daily evidence‐based treatments formally antagonises the relevance of patients’ experiences of distress and their explanations about what is needed to remedy their suffering. The politicised structure of the digital records thus undermines the formal possibilities of professionals to autonomously make decisions to provide care that pays attention to and includes patients’ experiences in accordance with academic psychiatry. Hence, our analysis suggests that this well‐known and overarching dilemma is literally inscribed into the very structure of what is formally considered contemporary mental health care. Perhaps, more importantly, it also draws attention to the fact that digital records indeed delineate medical professionalism and autonomy. This explains to some degree why service provision to mental patients (with and without addictions) represents a dilemma‐ridden challenge for professionals that goes far beyond professionals’ lack of knowledge about treatment, formal training and good intentions (Brekke, Lien, Nysveen, & Biong, 2018; Ness et al., 2014). As discretionary action is always characterised by a process of considering various care values and making incommensurable policies work at street level, the antagonism calls into question how professionals make these incommensurable requirements meet in mental health care. Thus, the effects of the politically pre‐determined focus on determining risk factors, individualising illness management and the rhetorical and structural exclusion of patient experience in the records might impose a moral appeal for professionals to tinker actively with service construction. If so, it questions how professionals might deliberately ignore the political requirements of the records and systematically fail to document how they ‘go an extra mile’ to collaborate with users in the shadows of mental health in order to provide meaningful and adequate help to service users (Nygaard‐Christensen et al., 2018; Oute & Rudge, 2019).

Ramifications for patients’ rights to be heard

In recent years, the WHO and the UN have called for a paradigmatic revolution in mental health in the wake of decades of the unethical persistence of over‐medicalisation, use of coercion, systematic violations of patients’ rights to be heard and dehumanising practices (United Nations, 2017). Similarly, mental health policies have made requirements about PPI (Patient and Public Involvement) and aligning services with a recovery orientation across the Nordic region, the UK and most western industrialised countries (Harper & Speed, 2014; McPherson & Oute, 2021). These policies emphasise that professionals should elicit and honour patients’ experiences within the context of supportive relationships, community inclusion, meaningful employment, citizenship and material surroundings that promote personal recovery in their life outside psychiatry (Davidson et al., 2008; Davidson et al., 2009). This points to a paradox because the structure of the records entirely rules out or at least downplays the clinical recognition of patients’ voices by pre‐excluding their real‐life experiences. However, the antagonising political emphasis on recovery is therefore not only a paradox; the conceptual presentation of recovery in the records is deeply flawed. The politicised clinical use of recovery is incoherent as recovery is presented as an in‐house intervention to individualise treatment responsibility and illness prevention while at the same time presenting patients’ social health concerns as the responsibility of external actors outside of mental health and systematically silencing patients. This suggests that the politicised nature of health records and its promotion of patient involvement and recovery as an individualising treatment intervention maintain the existing power structures of mental health rather than genuinely attempting to humanise mental health‐care practices in accordance with patients’ rights and the UN’s (United Nations, 2017) recommendations. This insight highlights the urgency of recent developments of care plans from the UK such as Refocus and the Power Threat Meaning framework that attempt to take their points of departure in patients’ experiences of distress, although it remains unclear what the implementation of such plans might look like across Nordic welfare contexts (Bird et al., 2014; Johnstone & Boyle, 2018).

CONCLUSION

This analysis of health records suggests that the records predominantly work as political vehicles rather than clinical tools. The politicised structure of the records and their pre‐defined medical content articulate a generic, idealised subject of a complex patient that calls for medical, cognitive–behavioural and individualising measures to minimise health risks, ensure self‐management of health and addiction and refer the responsibility for social concerns to external actors outside mental health. This illuminates how political ideology shapes professionals’ practices. It illuminates how a bio‐political structure of records (Rose, 2019) delineates professionals’ possibilities to make autonomous decisions structurally and how it structurally and rhetorically excludes users’ perspectives and goes against current policies of recovery‐oriented care. This calls for future studies that critically examine how professionals actively deal with the political delineation of professional discretion and exclusion of patient perspectives on a day‐to‐day basis in mental health care. Further, it calls for studies of how developments of new forms of care plans that are inclusive and sensitive to patients’ experience could be implemented in mental health care in Nordic welfare systems.

AUTHOR CONTRIBUTIONS

Jeppe Oute: Conceptualisation (lead); data curation (lead); formal analysis (lead); funding acquisition (supporting); investigation (lead); project administration (equal); resources (lead); software (lead); validation (lead); writing—original draft (lead); writing—review & editing (lead). Bagga Bjerge: Conceptualisation (equal); formal analysis (supporting); funding acquisition (lead); Methodology (supporting); project administration (lead); resources (supporting); software (supporting); writing—original draft (supporting); writing —review & editing (supporting). Larry Davidson: Conceptualisation (supporting); formal analysis (supporting); investigation (supporting); resources (supporting); supervision (supporting); writing—original draft (supporting); writing—review & editing (supporting).