Psykologtidsskriftet

Original ArticlePeer reviewed
Open access https://doi.org/10.52734/KYTL2024

Diverging interests at work: a text analysis of mental health records

Hilde Flata1, Hanne Weie Oddli1 & Petter Aaslestad2

  • 1 Department of Psychology, University of Oslo

  • 2 Department of Language and Literature, NTNU – Norwegian University of Science and Technology

    • h.k.flata@psykologi.uio.no

Published: 24.04.2025

Background: Patients in mental health care can read their medical records in an online portal. Research based on interviews and surveys shows that this opportunity presents new challenges for healthcare personnel when writing medical records. This study examined what characterised the genre in the early phase of outpatient treatment. We aimed to formulate concepts that may raise awareness of the medical record genre and promote reflection on different ways of keeping medical records. Method: An interdisciplinary research group analysed text from six medical records from the first three months of treatment in outpatient mental health care. Results: The analysis resulted in four main categories: 1) professional authority in the text, 2) patient perspective in the text, 3) standardised text and 4) incoherent text. Subcategories nuanced the findings. Implications: We used the findings to consider how medical record writing can enhance the therapeutic relationship: healthcare personnel will be able to develop their repertoire in the direction of change-oriented records by expressing professional authority from more function- and participant-oriented perspectives, using experiential language and individualising standardised text to reduce text fragmentation. In this way, medical records may balance patient participation with a distinctly situation-based professional assessment.

Keywords: electronic health record, mental health, medical record genre, text analysis, open notes

Patients in mental health care can read their medical records in an online portal. Consideration therefore needs to be given to the different ways medical records are used: for communicating clinical assessments to colleagues, documenting health care for regulatory authorities, and potentially serving as a means of communication with patients during treatment (Flata, 2023).

Medical record access can be developed into a relational tool that enhances the treatment collaboration (Blease et al., 2020). For this to happen, clinical notes must be written in a way that does not compromise the collaboration. This requires greater awareness of the linguistic repertoire dictated by the medical record genre. However, the genre-based perspective has been largely absent in research. Medical records have generally been understood within a basic communication model, purely as a medium for conveying information (Berge, 1997). The characteristics of the current medical record genre and clinical notes in mental health care are an under-researched topic.

Research on digital access to medical records for patients with mental disorders is largely based on surveys and interviews with patients and clinicians. A literature review article found that patients generally felt they had more control over their treatment and greater trust in their clinician when their medical record was written using respectful language, contained no misinformation or unfamiliar diagnoses, and aligned with their own experience of the consultation (Schwarz et al., 2021). Many clinicians reported an increased workload and were ambivalent about whether the change was beneficial for treatment (Schwarz et al., 2021). In a Norwegian survey from the mental health care sector, several clinicians indicated that they refrained from documenting observations and assessments that might provoke the patient (Fagerlund et al., 2021).

An analysis of the genre can provide insights into aspects of medical records that have not come to light in interviews and surveys. The field could benefit from drawing on perspectives from humanistic research, which recognises that texts inevitably convey meanings beyond neutral information. Literary scholar Petter Aaslestad employed a narratological reading strategy to demonstrate how patients are made audible and visible in mental health records over the span of a century (Aaslestad, 2007). Two later Norwegian studies, based on close readings of child welfare records and medical certificates respectively, demonstrated how underlying contradictions and value-laden elements were expressed in the texts (Engebretsen, 2007; Aarseth, 2019).

In our study, the overarching goal was to raise awareness of the medical record as a genre. We examined its characteristics in the initial phase of outpatient mental health care. By analysing the structure and linguistic devices of medical records through the lens of literary analysis (Lothe, 2007a), we developed a conceptual framework for our findings. This framework was used to reflect on how medical record-keeping can balance the various considerations it must accommodate. Our reading aimed to highlight defining features of the genre. We analysed which elements appeared to be consistently regulated by genre norms and which deviated from conventional expectations of form and language.

Method

Design

Our analysis of patient medical records is the first part of a larger project, Digital Access to Mental Health Records, at the University of Oslo. The project examines the significance of patient medical records in the collaborative therapeutic relationship. In addition to the analysis of patient records, the project includes interviews with patients and clinicians about their experiences with reading and writing medical records respectively.

Our study design is thus a genre analysis, in which we examine the medical record entries with an emphasis on structure while remaining open to different interpretations of the text. The aim was to read the text as it was presented through what was expressed linguistically (Gadamer, 2012). In the study, we use the term writer rather than clinician to emphasise that it is the written word being analysed, not the writer’s intentions behind it.

Material

The text material consisted of medical records from the first three months of outpatient mental health care for six patients in 2022, and was limited to documents that the patient could read on helsenorge.no. Each record was made up of 5 to 17 documents, primarily outpatient notes. Four medical records had only one writer, one had two writers and the other had three writers, including a letter from another outpatient facility.

Except for one medical record, which was written after a change of clinician, all included a ‘Note after the first consultation’. The notes had sub-headings that serve as a template in the electronic medical record. The template is the employer’s guide for what needs to be documented, as governed by the regulation on medical records.

Recruitment

The medical records were collected from six patient-clinician dyads at two mental health outpatient clinics at a hospital in Oslo. The clinicians did not participate in the recruitment of patient participants.

Five clinicians participated: three men and two women. Three were psychologists with less than three years of experience, one was a clinical psychologist and one was a specialised nurse with therapy training. The latter two clinician participants had more than 15 years of clinical experience.

Six patients participated: three women and three men. The patient sample was not restricted based on referral reason or diagnostic category, but lack of capacity to consent was an exclusion criterion. Clinician participants’ patients were sent a digital invitation to take part in the project along with the regular invitation to the first consultation. The patient participants were aged between 23 and 45 years. Their treatment history varied; some had been treated multiple times, while for others, this was their first encounter with mental health care.

Theoretical foundation

Patient records belong to a medical text genre. They are an established text format that shapes the structure and interpretation of the texts. This framework exists at the intersection of legislative expectations, assessment and treatment protocols, professional standards and the clinician-patient interaction. In our study, a pragmatic genre perspective served as a common theoretical basis for the methodology. Genre is understood here as a contextual concept that highlights the expectations for language and text within the setting in which the text is used (Ledin & Berge, 2001).

One source of the pragmatic genre perspective is the dialogical view of language proposed by Russian linguist and literary scholar Mikhail Bakhtin (1895–1975). Through his analysis of Dostoevsky’s novels, he established a metaphorical use of the concept of polyphony. He demonstrated how the speech patterns and perspectives of different social strata resonate as voices in the author’s language, without being subordinated to the narrator’s voice (Claudi, 2013). Bakhtin (2005) views language as a form of social interaction, where speech always echoes other speech and is directed at someone. Closely related to Bakhtin’s theories, the French literary scholar Julia Kristeva introduced the concept of intertextuality in 1967. She writes: ‘Every text takes shape as a mosaic of citations, every text is the absorption and transformation of other texts’ (Kristeva, 1967, cited in Refsum, 2007, p. 100). Here she is highlighting how every linguistic utterance inevitably enters into a dialogue with an infinite number of other texts and text types. The relationships between them are a prerequisite for all language use (Refsum, 2007). The understanding of the text in our close reading is based on the concept of intertextuality. Our use of the term sender function refers to the idea that the sender in the text cannot be reduced to the writer, because the text carries intertextual references.

Analysis: close reading

We analysed the medical records using interdisciplinary close reading, based on the close reading of New Criticism (Lothe, 2007b, p. 155) and its subsequent development in structuralism. The theoretical foundation described above served as a common methodological basis for our dialogue.

By forming an interdisciplinary research group, we leveraged our diverse interpretative perspectives, grounded in the humanistic tradition of textual analysis and the qualitative research traditions in the social and health sciences. HF’s methodological grounding as a historian of ideas is in theories of text in context (Krefting, 2017). She was particularly interested in the role that genre context played in how the text could be shaped by the writer. As a psychotherapy researcher, HWO has a methodological grounding in qualitative approaches. She drew on the constructivist theoretical foundation of grounded theory (Charmaz, 2006) and reflexive thematic analysis (Braun & Clarke, 2019). As a literature scholar, PA has extensive experience in exploring narrative structures in various types of texts. For example, he has analysed text structures in patient medical records in several published works. His perspective is shaped by the legacy of structuralism, where his focus is on the potential significance of text structures. In this context, the question ‘how the text means’ is prioritised over ‘what the text means’.

Analysis process

All the authors read the entire material in several rounds. The plan was to start with a close reading and then hold group meetings to explore the differences and similarities in our interpretations. HF and PA exchanged views in a note-taking system within the material. HWO provided textual input and metacommentary during the group meetings.

We started by reading the medical records one at a time. They seemed somewhat fragmented, with copious notes on formal procedures and symptoms. After initial discussions, we identified three main questions we wanted to pursue further: How was the sender function established? What characterised the coherence and reasoning? In what ways was the text descriptively symptom-oriented or more change-oriented?

For each record, HF wrote an analysis note on how the patient and the health care were portrayed. As part of the hermeneutic process (see, for example, Gadamer, 2012), the analysis note was discussed in the group meetings, and new insights formed part of the pre-understanding for the next reading.

We further investigated how protocols, discharge summaries, the patient’s account and the writer’s clinical assessments could be analysed as voices resonating within the text. We were interested in how the writer seemed to transform the patient’s narrative into the genre’s more concise format, and how a diagnostic or treatment strategy was established in the text.

Finally, HF structured the material according to main categories that highlighted similarities, with subcategories showing nuances. Through the group meetings, we reached agreement on the categories presented in the results section of the article.

Reflexivity

HF has worked as a clinical psychologist for the past three decades and is also a qualified historian of ideas. She was interested in the linguistic repertoires the writer used to fulfil the various functions of the medical record, and the significance of the language for patients’ reading experience and the establishment of collaboration. HWO is a professor of clinical psychology and has broad-based experience with qualitative analysis methods in psychotherapy research. Her perspective highlighted how the medical record texts contributed to both understanding the problem and framing the treatment plan. PA is a professor of literary science and has experience in research on patient records using narratology as a reading method (Aaslestad, 2007). His perspective clarified how the text creates meaning based on its various structural levels. The range of scientific expertise captured, for example, how mental illness and change processes were portrayed in the medical records, while the analysis of the sender function and reasoning within the text stemmed from an interest in its structure.

HF made audio recordings of the group meetings and wrote summaries, which increased our awareness of how our different pre-understandings influenced our work. This is in line with a dialogical view of reflexivity as an ongoing individual and collective effort to clarify how various perspectives come into play when exploring phenomena (Stige et al., 2009).

None of the research group had explicit patient experience. We therefore sought input from professionals at the Change Factory, who are user representatives with patient experience. Findings related to the patient perspective were discussed with them.

Additionally, findings concerning standardised text and the coherence within medical records were discussed with a psychiatrist with many years of oversight experience with the Norwegian Board of Health Supervision. We sought her perspective on the results because it was important for the study that our understanding of the medical records sufficiently accorded with regulatory requirements.

Ethics

The research project was approved by the Regional Committee for Medical and Health Research Ethics (REK ref. 184 084). Inclusion was based on written consent.

Results

The analysis resulted in four main textual categories, which were present to varying degrees in all the medical records: 1) professional authority in the text, 2) patient perspective in the text, 3) standardised text and 4) incoherent text.

The term professional authority refers to clinical assessments of health status, care needs and interventions. The patient perspective concerns the patient’s perceptions and understandings of their experiences, problems and possible solutions. Standardised text was particularly found in risk assessments, assessment reports and under current status, where the writer is expected to convey their observations of the patient. Incoherent text refers to text with a fragmented structure or lack of internal logical consistency.

We present the categories with excerpts from the medical records, identified by letters (A, B, C, D, E, F). The text in parentheses following A, B, etc., refers to the subheading of the section of the medical record that the quote is taken from.

Professional authority in the text

Patient records are written by healthcare personnel treating the patient. In this sense, all text in medical records is formulated by a writer with professional authority. However, the way this professional authority was expressed linguistically varied within the structured format of the records.

In some records, the writer would draw on previous clinicians’ assessments, sometimes without citing the source. In such instances, the assessment was deemed to have unknown professional authority. A predominantly descriptive, symptom-oriented perspective was evident throughout. A more function-oriented perspective – where the writer also suggested ways to understand connections – was less common. When the patient was portrayed as an active participant in the process, this could be interpreted as an indication of participant-oriented professional authority.

Unknown professional authority

In some records, professional authority was not solely vested in the clinician. Referrals, discharge summaries and previous records could serve as explicit sources or be referenced without citation, as in the following example: ‘Long history of complex issues with follow-up from various age-appropriate services, both inpatient and outpatient. In summary: various diagnoses, such as schizotypal disorder, anxiety, substance use, depression, PTSD [post-traumatic stress disorder], bipolar disorder, borderline personality disorder and eating disorder, have circulated.’ (A, Current issue)

The record did not indicate here who made these assessments or how the previous assessments were thought to relate to current treatment needs. A lack of source citation was also observed in other records where the patient had a prior treatment history.

Descriptive, symptom-oriented professional authority

The patient’s symptom profile was described in all the notes from the first conversation, as in this excerpt: ‘Young man referred for depressive symptoms and suspected PTSD. [...] Upon questioning, the patient denied clear symptoms of reliving childhood experiences and a recent threatening situation. [...] Differential diagnosis suggests that anxiety disorder, adjustment disorder and depression are possible explanations for the clinical picture [...].’ (B, Assessment)

The writer’s formulation here clearly shows that the initial goal was to clarify the patient’s differential diagnosis.

Function-oriented professional authority

When the text included assessments that showed the writer’s understanding of the relationship between the patient’s symptoms and experiences, and possible mechanisms of change that guided the treatment, we referred to this as function-oriented professional authority. Function-oriented quality was rarely identified in the notes from the first conversation. The following description was an exception:

The patient is a young man [...]. In conversation, does not appear to have obvious signs of clinical depression, but there is also a clear tendency to hide how he is really feeling from others. Struggles with a deep despair that does not have a clear connection to current life circumstances. Signs of overadjustment in relationships and a lot of shame-related issues are evident. (C, Assessment)

The writer outlined possible connections here in the expression of suffering – despair in a patient who hides how he is feeling – which created a basis for establishing a plan for change.

In a clinical note from another treatment pathway, function-oriented professional authority was demonstrated through an educational approach, focusing on teaching the patient how to take better care of himself: ‘Creates a list on a whiteboard of signs that he is not doing well, and tools he can use. […] Tools: set social boundaries, do not make too many arrangements [...].’ (B, Current)

Participant-oriented professional authority

Participant-oriented professional authority seemed to be focused on the patient’s involvement in the assessment and treatment process. In the following excerpt, the writer notes that the patient perceived the situation differently than the writer: ‘It appeared to the undersigned that the patient was not entirely in agreement with the suggestion for follow-up until starting DBT [dialectical behaviour therapy], but that the patient possibly accepted it out of fear of being terminated. […] it may be necessary to address this in the next session.’ (A, Assessment)

The writer acknowledged that the patient may have withheld their opinion, and that she concealed her disagreement because she was afraid of being ‘terminated’. When read from a participant-oriented perspective, it was implicit that the writer expressed concern for the patient. He sensed that the success of their ongoing dialogue depended on the patient’s fear of being left too much to her own devices being addressed in their next conversation.

Patient perspective in the text

All the medical records contained text that seemed to express the patient’s perspective. This text was consistently written in the third person, with sentences often having an implied subject (‘Slept poorly’). It was generally possible to understand from the context when the patient seemed to be the source of the content.

One point in the template clearly invited the writer to render the patient’s perspective: ‘The patient’s goals and expectations for assessment/treatment/motivation.’ The writer is expected to give an account of what the patient wanted from the health service. However, the concept of motivation is ambiguous: is it the patient’s own assessment of their motivation or the clinician’s assessment of the patient’s motivation? The term allows for both possibilities. In our material, only one writer assessed the patient’s motivation under this heading: ‘Shows interest in treatment’. (C, Patient’s goals and expectations)

The communication of the patient’s perspective as symptom-oriented descriptions dominated the material. Depictions of the patient as an active participant were more seldom. We also identified a somewhat different textual aspect, where the portrayal of the patient’s perspective was influenced by emotions or values. In such cases, the portrayal appeared to be emotionally charged.

Symptom-oriented descriptions

The descriptions of the patient’s health status and situation tended to be noted as the patient’s symptoms. These descriptions were often in the form of a list, as in this example:

He reports relevant symptoms in the form of low mood, tearfulness, irritability and frequent mood swings, feelings of emptiness, apathy and a sense of being paralysed. Has difficulty initiating activities but maintains normal hygiene and activity as long as he has external structures that motivate him. (D, Current issue)

Here, the writer uses nominalised concepts, such as tearfulness and irritability. The symptoms mentioned are known in clinical contexts as criteria for depressive disorder. However, the last sentence could also be interpreted as contributing to the differential diagnosis of possible ADHD (attention deficit hyperactivity disorder), which was relevant for this patient.

Descriptions of the patient’s symptoms sometimes helped convey the depth of their suffering, as illustrated in this note from the first consultation: ‘The patient says he has suffered from depressive issues for as long as he can remember [...]. He describes a permanent feeling of ‘despair throughout his body’ and a strong sense of hopelessness.’ (C, Current issue)

Here, the writer has conveyed the patient’s experience by, for example, quoting the patient’s own words. The quote was marked with quotation marks, as required by the genre conventions. How a quote is perceived depends on the context in which it is placed. In this excerpt, the writer included the quote in a context that highlighted the intensity of the patient’s suffering.

The patient as an active participant

The medical record’s subheading about the patient’s goals was a clear invitation for the patient to give their input as an active participant. However, the active participant aspect could also come to the fore under ‘Current issue’:

The patient expresses a desire for assessment for ADHD/ADD after a previous private psychologist had assessed that his depressive symptoms were likely triggered/ongoing because of underlying attention deficits. He states that since then, he has read up on ADHD and that he recognises himself in several of the descriptions: [...] (D, Current issue)

The writer highlighted the patient’s own knowledge-seeking efforts. The patient’s self-knowledge, supported by the previous clinician’s professional authority, was used as the basis for planning the assessment.

Value-laden patient perspective

Disagreements or misunderstandings that arose between the patient and the clinician could be subtle or more obvious in the text. The excerpt below is an example of a text that openly conveyed that the patient brought up an error in their medical record:

The patient points out that he has read the entry in his medical record from the last appointment and reacted to something that the undersigned wrote in the internal referral note. He refers to the sentence about no change in sleep pattern and points out that it is incorrect. [...] He says that he read his medical records in the evening and felt frustrated and misunderstood when he read them. He goes on to describe thoughts of giving up on DPS (the outpatient clinic), feeling rejected and that there is no point continuing. The patient points out that he now understands that there is a point, but that on his way to the appointment today, he felt sceptical towards the undersigned (B, Current issue)

The patient’s perspective seemed to be honestly conveyed by the writer here. The writer’s presence in the text was minimal. However, the writer used the expression ‘the patient points out’ three times in this section. To ‘point out’ means to bring attention to something, often as a criticism, as in this case. The repeated use of the term carries a connotation that the clinician perceived the patient as challenging in the conversation.

A somewhat different aspect of value-laden language was found in another medical record: ‘Unwanted intimate relationship with clinician during adolescence allegedly has affected her’. (A, Background) The entry was so brief that the patient’s narrative was barely visible. However, it was reasonable to interpret ‘Unwanted’ and ‘affected her’ as expressions of the patient’s perspective. In contrast, the term ‘allegedly’ departed from the patient’s perspective. Today, the common meaning of ‘allegedly’ is that something is claimed (‘Alleged’, 2025). Use of this term therefore called into question the veracity of the statements. However, the writer may have used the term in a neutral sense to indicate that this was information that the writer was not in a position to evaluate.

Standardised text

Medical record entries are consistently structured according to standardised genre conventions. The greatest extent of standardised text was found in risk assessments, diagnostic reports and under current status. Findings from examinations, i.e. the patient’s responses to questions, were usually conveyed in a more individualised form, where the patient’s specific experiences and situation were described, often from the patient’s perspective, while the conclusions were consistently standardised formulations.

We identified two subcategories: individualised standardisation and standardised diagnostic assessment.

Individualised standardisation

Assessments and evaluations of suicide risk are anchored in national clinical protocols. Based on these, employers have drawn up guidelines for clinicians to follow in consultations and in patient record-keeping. This work has thus been standardised. However, the question of whether preventive measures are needed for the patient must be answered based on specific individual circumstances.

The following excerpt demonstrated individualised expression that was also in line with the standardisation:

The patient denies currently having suicidal ideation and specific plans. Expresses that she had suicidal thoughts the evening she destroyed her flat, felt disbelieved and a strong sense of injustice over what she had been through. [...]. The risk is assessed as present, but currently low. Sufficiently managed through further outpatient follow-up. (E, Suicide risk)

The excerpt painted an individualised picture of the patient in a high-risk situation. The fact that the episode was told from the patient’s perspective gave the assessment an individualised grounding. The excerpt also had a technical tone, with a list of risk factors and protective factors as the basis for the conclusion. The conclusion consisted of standardised formulations, which was consistent across all the risk assessments in our material.

Current status often consisted of phrases used in most of the medical records (such as ‘Oriented to time, place and situation’). However, standardised expressions could also be used in an individualised way, where they were regarded as relevant to the issue the patient was seeking help for.

Standardised diagnostic assessment

With the exception of two medical records indicating that the patient’s need for health care had already been sufficiently assessed, all records contained a standardised diagnostic assessment using the MINI (Mini-International Neuropsychiatric Interview) questionnaire, which at the time was part of the basic assessment.

The diagnostic reports often included brief formulations with codes that were incomprehensible without referencing the MINI questionnaire. Some reports had no conclusion, which was instead given under ‘Assessment’ in the dynamic medical record, as in this case: ‘No current or previous depressive episodes identified according to ICD-10 criteria. Anxiety symptoms suggestive of agoraphobia and social phobia observed, history of greater functional decline than currently. Not sufficient to make a diagnosis at present. But the symptoms described by the patient meet the criteria for PTSD [...].’ (E, Assessment)

Following further diagnostic clarification, the conclusion helped guide the treatment towards post-traumatic stress disorder.

Incoherent text

All the medical records occasionally showed a lack of internal consistency. We identified two subcategories. Although the letter genre was underrepresented in our material, one letter was in a subcategory of its own: inconsistent standardisation, due to its self-contradictory formulation. Most of the records contained fragmented text where the writer used less relevant standardised forms, presumably to meet the documentation requirement.

Inconsistent standardisation

Letters are a text genre with specific expectations regarding form, i.e. they follow a standardised format. For example, it is expected that the letter is addressed to a recipient and that the sender is clearly indicated. Our material included three letters, one of which was a rejection letter to a patient. As expected in the letter genre, the letter was directly addressed to the recipient: ‘[N.N] has referred you to us’, followed by ‘[we have] concluded that you are not entitled to an assessment/treatment in the specialist health service’. (A, Referral rejection). Other than the name of the hospital, there was no clear indication of who sent the letter.

The personal introduction then changed, and the recipient of the letter was no longer addressed in the second person. Both under the heading ‘Reason for referral’ and in the explanation for the rejection, the recipient was referred to in the third person. The letter was thus inconsistent in its approach to addressing the recipient.

After the explanation for the rejection, the letter again shifted to a standardised form of address: ‘We recommend that you contact your general practitioner (GP) or the health provider who referred you for further follow-up. You can ask your GP if there are grounds for a new assessment at another treatment facility.’ (A, Referral rejection). However, the recommendation was not tailored to the patient’s situation, as she had been referred from another outpatient clinic at the same hospital. It was unclear whether the patient was still entitled to receive treatment at the outpatient clinic that had referred her.

Standardised forms to meet the documentation requirement

At times, it seemed like the primary purpose of the medical record entries was to meet the documentation requirement. The following risk assessment was set out in a note from a writer who met the patient for the first time after a change of clinician: ‘Based on available information, the suicide risk is currently considered statistically elevated due to mental illness, but nevertheless low and without the need for specific interventions.’ (F, Assessment). Suicide risk did not appear to have been addressed or considered a relevant aspect of the patient’s experience of suffering, but the note documented that the new clinician had assessed the suicide risk and any potential need for interventions.

Such standardised formulations could give the medical record a fragmented feel, where the connection between the patient’s suffering and the efforts to address the situation was lost in formalities. Similarly, risk assessments for violence were often expressed in variants such as ‘No history of serious violence or threats. Not considered high risk’ or ‘Not observed’, contributing to the fragmented nature of the medical records. There seemed to be a disconnect between these formulations and the rest of the description regarding the patient’s experience of suffering and the efforts to address the situation.

Discussion

Patient access to medical records via an online portal presents a challenge for healthcare providers in terms of writing clinical notes without compromising the treatment collaboration. This is particularly the case when sensitive topics, such as mental illness, are to be described. The medical record genre is an inherited text format that guides how clinical notes are formulated, and the expectations for how they should be written can create challenges in relation to how patients will perceive the content. Until now, there has been no research on the structure of medical records as a genre in the context of patient access.

Based on our genre analysis, we developed a conceptual framework for genre characteristics, reflecting how the texts are formulated according to the genre’s conventions. We found that the medical records were characterised by four main textual categories: professional authority, the patient perspective, standardisation and incoherent text. Subcategories revealed the nuances within these. The categories thus highlighted the linguistic repertoires that the genre allowed for. We use this conceptual framework to explore how genre-appropriate ways of writing clinical notes can develop the genre, enhancing its potential for medical record access to serve as a relational tool in treatment (Blease et al., 2020).

Linguistic repertoires

The medical record entries were characterised by extensive symptom descriptions, which has the effect of objectifying the patient, who is viewed from an external perspective. This is particularly apparent when the symptom is described as a nominalised concept, such as tearfulness, rather than using more experiential language, like stating that the patient cries easily. This can partly be explained by the medical record format, which demands brevity, and by the structure, which encourages a linguistic repertoire influenced by the medical field’s tradition of objectifying descriptions (Ekeland, 2021). Clinical guidelines also typically note that differential diagnostic clarifications of symptoms are the first step towards appropriate treatment.

A key function of patient medical records is to describe the basis for diagnosis. However, there can be an imbalance in the dominance of descriptive symptom orientation. Both the clinician and the patient can lose sight of how the patient’s issues are connected to the broader context of their life. Following our analysis, we established the concepts of function-oriented and participant-oriented professional authority. These terms refer to linguistic repertoires for coherent understanding of change and for interaction with the patient. Some medical records also consisted of linguistic repertoires that describe the patient as an active participant. These are genre-appropriate alternatives to objectifying language that can reduce the risk of medical record access reinforcing the patient’s passive self-understanding.

Passivity and lack of coherence are also risks when writers refer to assessments from other sources without citing them (unknown professional authority). The lack of coherence and reliability undermines the transparency on which patient access to medical records is based. Our results do not offer solutions to the real dilemmas that arise when patients have access to their medical records. One of the main dilemmas is how to formulate assessments that could adversely affect treatment if shared with the patient too soon.

Clarification of context

Our study showed variations in writers’ approach to the medical record genre, including the extent to which they used standardised phrases. We do not consider the genre’s standardisations inappropriate per se; they save time and facilitate communication among professionals, but without clear contextualisation, they risk becoming empty phrases. By individualising the content, the assessments can be more specifically rooted in the experiences and suffering of the individual patient. Several writers used an experiential linguistic repertoire and individualised standardised text. Together with the use of function- and participant-oriented professional authority, this made the medical record more coherent and change-oriented. We found that this approach strengthens the possibility for the writer to establish an integrated understanding of how a particular patient’s problems can be addressed within the unique context of the individual therapeutic relationship (Berg, 2020). In our view, this would increase the potential for patient access to medical records to function as a relational tool (Blease et al., 2020).

In a conversation about patient perspective in the text, user representatives stressed that the findings should be considered in light of the self-understanding that patients formed through reading the content of their medical record. They described how being objectified, without any context to give meaning to the portrayal, reinforced a sense of powerlessness in their self-understanding. The user representatives were therefore interested in how the medical record conveyed whether the patient was listened to and involved (The Change Factory, 03.11.23). This input aligns well with findings from process research in psychotherapy. It is well-established that therapists who empower patients to take control of their lives (agency) and foster a shared understanding of goals and tasks in therapy build an alliance with the patient that enhances the likelihood of a successful treatment outcome (Horvath et al., 2011; Oddli et al., 2012). However, it cannot be assumed that all patients would prefer a more individualised medical record or one with less of a focus on symptoms. This could be further explored by talking to patients about their experiences with reading their medical records.

Strengths and limitations

A prerequisite for the credibility of the results is that our text material adequately reflects the medical record genre. One of the strengths of the study is that most of the material was written by clinicians who wanted to participate and who adequately adhered to the genre format. This provided us with a rich dataset.

Genre is primarily about form. For a genre analysis, it is therefore secondary that the content of the medical records varies by type of service, geography and writer. When reading medical records with a focus on form rather than content, the medical record genre appears to be relatively stable across different contexts (Aaslestad, 2007).

We have not explored whether differences in treatment cultures or the professional backgrounds of the writers may have contributed to systematic variations in the medical record texts. However, to increase the relevance of the findings for writers in comparable clinician roles, we wanted to include clinicians from different professions. Nevertheless, the sample was dominated by psychologists. One of the clinician participants is a registered nurse, but there were no doctors or participants with a background in social work. The study may therefore have missed nuances in the material that differences in clinical note-writing approaches across professions might represent (Malterud, 2017). However, as argued above, such nuances are of secondary importance for the genre study.

We sought to make the analysis process transparent by showing what the different perspectives contributed to the analysis. This strengthened the methodological integrity of the study (Levitt et al., 2017). The study’s transferability lies in the relative stability of the medical record genre and the concepts and categories we established from our results. These can be used to reflect on how medical records can meet statutory requirements, be professionally appropriate and simultaneously enhance the potential to strengthen the collaboration with the patient. Our findings are likely to have the greatest transferability to various service levels in mental health care, substance use treatment, general practitioners and emergency care. These are situations where reading their medical records can impact a patient’s self-understanding and understanding of change.

Future research

Medical records’ significance for collaboration in treatment can be further explored by incorporating medical record texts into interviews with clinicians and patients about their experiences with writing and reading clinical notes during the treatment process. Such interview analyses could provide insight into what aspects of the text can promote or hinder the treatment collaboration. Various aspects of the medical record genre could also be examined using data technology to analyse large text corpora. Our results could be discussed in relation to findings from such approaches.

Conclusion

The medical records were characterised by four main textual categories: professional authority, patient perspective, standardised text and incoherent text. The results from the study can be used to reflect on the linguistic repertoires enabled by the medical record genre when certain considerations need to be made in the writing of clinical notes. Individualising standardised text and using experiential language will help to contextualise the record entries. By being more mindful of not letting symptom descriptions dominate the text and focusing on function- and participant-oriented professional authority, the writer can develop a repertoire of more change-oriented notes with better internal coherence. Further investigation is needed to determine if this would increase the likelihood of patients feeling heard by a clinician with transferable clinical expertise when reading their medical records.

References

  1. Alleged. (2025). In Oxford English Dictionary. Retrieved February 19, 2025, from https://www.oed.com/search/dictionary/?scope=Entries&q=alleged

  2. Bakhtin, M. M. (2005). Spørsmålet om talegenrane (R.T. Slaattelid, Trans.). Ariadne. (Original work published 1929)

  3. Berg, H. (2020). Evidens og etikk: Hva er problemet med evidensbasert praksis i psykologi? Fagbokforlaget.

  4. Berge, K. L. (1997). Communication. In P.V. Lamarque (Red.), Concise Encyclopedia of Philosophy of Language (pp. 95–102). Pergamon Press.

  5. Blease, C. R., O’Neill, S., Walker, J., Hägglund, M., & Torous, J. (2020). Sharing notes with mental health patients: Balancing risks with respect. Lancet Psychiatry, 7(11), 924–925. https://doi.org/10.1016/S2215-0366(20)30032-8

  6. Braun, V., & Clarke, V. (2019). Reflecting on reflexive thematic analysis. Qualitative Research in Sport, Exercise and Health, 11(4), 589–597. https://doi.org/10.1080/2159676X.2019.1628806

  7. Charmaz, K. (2006). Constructing grounded theory. A practical guide through qualitative analysis. SAGE.

  8. Claudi, M. B. (2013). Litteraturteori. Fagbokforlaget.

  9. Ekeland, T.-J. (2021). Psykisk helsevern – en kunnskapsplattform. Nasjonalt senter for erfaringskompetanse innen psykisk helse.

  10. Engebretsen, E. (2007). Hva sa klienten? Retorikken i barnevernets journaler. Cappelen akademisk forlag.

  11. Fagerlund, A. J., Kristiansen, E., Johansen, M. A., Haumann, K., & Zanaboni, P. (2021). Elektronisk innsyn i journal for pasienter i psykisk helsevern: Helsepersonells erfaringer. Tidsskrift for Norsk psykologforening, 58(5), 380–389. https://psykologtidsskriftet.no/vitenskapelig-artikkel/2021/05/elektronisk-innsyn-i-journal-pasienter-i-psykisk-helsevern

  12. Flata, H. (2023). Når journalen blir klinikk: Digitalt journalinnsyn. Michael, 20(31), 184–191. https://www.michaeljournal.no/article/2023/02/Når%20journalen%20blir%20klinikk%20digitalt%20journalinnsyn

  13. Gadamer, H.-G. (2012). Sannhet og metode: Grunntrekk i en filosofisk hermeneutikk. (L. Holm-Larsen, Trans.). Pax forlag. (Original work published 1960)

  14. Horvath, A., Del Re, A. C., Flückiger, C., & Symonds, D. (2011). Alliance in individual psychotherapy. Psychotherapy, 48(1), 9–16. https://doi.org/10.1037/a0022186

  15. Krefting, E., Schaanning, E., & Aasgaard, R. (2017). Grep om fortiden: Perspektiver og metoder i idéhistorie. Cappelen Damm Akademisk.

  16. Ledin, P., & Berge, K. L. (2001). Perspektiv på genre. Rhetorica Scandinavica, 18(6), 4–16.

  17. Levitt, H. M., Motulsky, S. L., Wertz, F. J., Morrow, S. L., & Ponterotto, J. G. (2017). Recommendations for designing and reviewing qualitative research in psychology: Promoting methodological integrity. Qualitative Psychology, 4(1), 2–22. https://doi.org/10.1037/qup0000082

  18. Lothe, J. (2007a). Analyse. In J. Lothe, C. Refsum & U. Solberg (Eds.), Litteraturvitenskapelig leksikon (2nd ed., pp. 8–9). Kunnskapsforlaget.

  19. Lothe, J. (2007b). Nykritikken. In J. Lothe, C. Refsum & U. Solberg (eds.), Litteraturvitenskapelig leksikon (2nd ed., pp. 155–156). Kunnskapsforlaget.

  20. Malterud, K. (2017). Kvalitative forskningsmetoder for medisin og helsefag (4th ed.). Universitetsforlaget.

  21. Oddli, H. W., & Rønnestad, M. H. (2012). How experienced therapists introduce the technical aspects in the initial alliance formation: Powerful decision makers supporting clients’ agency. Psychotherapy Research, 22(2), 176–193. https://doi.org/10.1080/10503307.2011.633280

  22. Refsum, C. (2007). Intertekstualitet. In J. Lothe, C. Refsum & U. Solberg (Eds.), Litteraturvitenskapelig leksikon (2nd ed., pp. 100–101). Kunnskapsforlaget.

  23. Schwarz, J., Bärkås, A., Blease, C., Collins, L., Hägglund, M., Markham, S., & Hochwarter, S. (2021). Sharing clinical notes and electronic health records with people affected by mental health conditions: Scoping review. JMIR Mental Health, 8(12), e34 170. https://doi.org/10.2196/34170

  24. Stige, B., Malterud, K., & Midtgarden T. (2009). Toward an agenda for evaluation of qualitative research. Qualitative Health Research, 19(10), 1504–1516. https://doi.org/10.1177/1049732309348

  25. Aarseth, G. (2019). The language of work disability: A study of medical certificates written by Norwegian general practitioners [Doctoral thesis]. Det medisinske fakultet, Universitetet i Oslo. https://www.duo.uio.no/bitstream/handle/10852/68312/3/phd-Aarseth-2019.pdf

  26. Aaslestad, P. (2007). Pasienten som tekst: Fortellerrollen i psykiatriske journaler. Gaustad 1890–1990 (2nd ed.). Universitetsforlaget