Discursive construction of the patient in online clinical cancer pathways information

Aim: To explore how the patient is constructed and socially positioned in discourses of web-based pathways information available to people with cancer in Norway. Design: Mixed qualitative and quantitative design, using Corpus-Assisted Critical Discourse Analysis. Methods: The study, conducted in January 2020, examined the language of one general and six specific web-based cancer pathways information brochures. The approach combines analysis of word frequencies and concordance lines using corpus analysis software to identify the ‘linguistic fingerprint’ or ‘aboutness’ of the text prior to further qualitative critical discourse analysis. Results:


| INTRODUC TI ON
The new cancer strategy in Norway, Living with Cancer (Ministry of Health & Care Services, 2018), is a continuation and update of the Government's earlier cancer strategy and sets out its ambition for a service where patients participate more, have an influence on their own treatment and get help to manage their illness and everyday life.
This strategy focuses on the question of what is important for the patient and their needs.
They were implemented in Norway in 2014 and last updated in (Health Directorate, 2014. The goal is to contribute to a quick investigation and start of treatment without unnecessary waiting time (Helse.norge.no, 2020;Jensen et al., 2015). These pathways are supposed to achieve the best outcomes for people with cancer.
Although Scandinavian countries share a health care model with the primary goal of equal access to welfare, changes to governance systems such as New Public Management (NPM) are placing ever-increasing demands on efficiency and productivity, raising questions and ethical dilemmas about patient social positions and agency. This kind of governance appears to undermine an emphasis in public documents about transferring more power to individual patients that has developed since the adoption of the Universal Declaration of Human Rights by the United Nations in 1948. However, paternalistic ideology, where patients are assumed to comply and play minimal and passive roles, has dominated health care (Beauchamp & Childress, 2009;Collins, Britten, Ruusuvuori, & Thompson, 2007).
Indeed, Dahlborg Lyckhage, Pennbrant, and Boman (2017), Lassen, Ottesen, and Strunck (2018) and Aasen and Dahl (2019) identify ideological struggles in legal discourse across Scandinavian countries as national governance systems continue to exert hegemonic power by strongly influencing patient choice and autonomy.
This mixed messaging about patient participation and autonomy in the face of NPM is evident in Norway. On the one hand, recent public documents underline the rights of patients to participate, influence and make personal choices in health care (Ministry of Health & Care Services, 1999a, as set down in Norwegian law (Ministry of Health & Care Services, 1999a, 1999b (Aasen & Dahl, 2019). It is a discourse that validates the power of medicine, the healthcare system and health personnel over the needs, rights, participation and autonomy of the patient (Aasen & Dahl, 2019). This study analyses the discourses in the Norwegian cancer pathways and investigates how the patients are constructed.

| Background
The World Health Organization has promoted a person-centred approach, with a global goal of humanizing health care by ensuring that it is rooted in universal principles of human rights: The overall vision for people-centered health care is one in which individuals, families and communities are served by and are able to participate in trusted health systems that respond to their needs in humane and holistic ways…  Aasen, Kvangarsnes, Wold, and Heggen (2012) described the concept of patient participation as involving the relationship between patients and the health-care team, who engage in a process of power exchange. Participation 'does not necessarily require shared decision-making but rather a dialogue with shared information and knowledge and mutual engagement in intellectual and/or physical activities influenced by the context'.
Person-centred practice is now almost in the nursing and healthcare discourse and as a global movement . The person-centred nursing framework of McCance (2010, 2018) has influenced practice and policy development. As Richards, Coulter, and Wicks (2015) (p. 3) argue, the challenge remains one of overcoming 'system' inertia and paternalism and that a significant part of this change is the need to shift the discourse away from person-centred 'care' and to a unified discourse of person-centred 'cultures'.
The implementation of care pathways has been concerned with improving the healthcare system, reducing unnecessary practice variation, enhancing coordination and continuity and achieving better clinical outcomes (Faber, Grande, Wollersheim, Hermens, & Wlwyn, 2014). The length of hospital stay has decreased and, as a result, the costs have reduced, with fewer in-hospital complications observed (Rotter et al., 2010). However, this standardization of organizational procedures and evidenced-based care also has an evident drawback. Care pathways have mainly been developed by healthcare specialists and have been disease-based (Røstad, Garåsen, Steinsbekk, Sletvold, & Grimsmo, 2013).
Studies connected to cancer pathways point to the importance of patient participation and communication. Active patient participation in formulating the treatment plan alongside clinician's support and encouragement were fields that had a positive health outcome to patients (Ishikawa & Yano, 2008;Street, Makoul, Arora, & Epstein, 2009). Over recent decades we have perhaps assumed that there has been a shift from paternalism to patient participation. We may consider that patient autonomy is respected and information and decisions are shared between physician and patient (Ishikawa & Yano, 2008). However, a systematic review of end-of-life care pathways for improving outcomes in caring for the dying (Chan, Webster, & Bowers, 2016), maintained that there are substantial concerns about safety and quality of care associated with the pathway implementation and limited available evidence concerning the clinical, physical, psychological or emotional effectiveness of end-of-life care pathways.
There are now 28 pathways for cancer in Norway, providing patients with greater predictability for assessment and an active role in timely treatment, not least through online resources. However, there has been no research into whether this online material supports or advances participation.

| Aim
The aim of this study is to explore how the patient is constructed and socially positioned in the discourses of web-based pathways available to people with cancer. The following research question was addressed: How are the patients constructed in the cancer pathways in Norway?

| Design
A mixed qualitative and quantitative design was employed, using Corpus-Assisted Critical Discourse Analysis. This study uses a Corpus-Assisted Critical Discourse Analysis to examine the language of web-based cancer care pathways information in Norway. This is a 'tried and tested' mixed methods approach (Crawford, Gilbert, Gilbert, Gale, & Harvey, 2013) which has become established over the last 15 years (see Adolphs, Brown, Carter, Crawford, & Sahota, 2004;Atanasova, Koteyko, Brown, & Crawford, 2017;Crawford, Brown, & Harvey, 2014). Typically, the approach combines analysis of word frequencies and concordance lines using corpus analysis software to identify the 'linguistic fingerprint' or 'aboutness' of the text (Scott, 2006) prior to further qualitative discourse analysis.
In this study, we use AntConc 3.2.1. software (Anthony, 2010) to capture the 'aboutness' of the language used in the care pathways documents to underpin subsequent qualitative discourse analytic work (Louw, 1993;Sinclair, 1991). This subsequent phase will comprise a constructionist discourse analysis (Tuominen, Talja, & Savolainen, 2002), which draws on the tradition of Critical Discourse Analysis (Baker, 2012;Fairclough, 2003;Lin, 2014). We can think of the quantitative component as a 'baseline' to support the claims of the subsequent analysis (Crawford et al., 2013) whereby we examine the language as a social and political practice that advances a particular version or construction of the reality of care pathways.
A discourse represents a particular part of the world from a specific perspective (Fairclough, 2003). Discourses can possess different types of order; notably one aspect of this order is dominance. Critical discourse analysis seeks to uncover how power and dominance are manifested through language (Fairclough, 2001). Fairclough (2001) suggests that ideology, hegemony and power are three central concepts in discourses. He defines ideology as a set of beliefs and attitudes shared by members of a group. The most powerful or hegemonic discourse in any field emerges from the social struggle for dominance as ideological meanings are established. Fairclough (1992) also made a distinction between power in a discourse and power behind a discourse. The power behind the discourse is the hidden effect of discourse, an effect of power. It influences a person's identity, relationships, knowledge and beliefs (Fairclough, 1992(Fairclough, , 2001(Fairclough, , 2003.

| Sample and data collection
The study examines web-based cancer care pathways information We created a small corpus of searchable file of these six TXT formatted texts called the Online Cancer Care Pathways Corpus (OCCPC) and subjected it to quantitative and qualitative analysis.

| Ethical considerations
No formal ethical scrutiny was required or undertaken.

| Data analysis
We used Corpus Analysis first to establish the quantified 'linguistic fingerprint' of the OCCPC (9,464 words) set out above in terms of raw frequencies and collocations and then Critical Discourse Analysis to describe, interpret and explain the data (Fairclough, 1992(Fairclough, , 2001(Fairclough, , 2003 in a three-step process:

| Description of the text
The transcribed interviews were subject to linguistic interpretation (Fairclough, 2003;Vagle, 2005) by asking the following: Who wrote the text and for whom? What authority lies with the patient? Are patients marked as agentive or passive subjects? What kind of evaluative language is used? Are personal pronouns or names used? How are modal auxiliary verbs (will, can, should, must) used?

| Interpretation of the discursive practice
The themes were interpreted in terms of the relationship between the processes of production of the text and the distribution and the process of interpretation of the text (Fairclough, 1992(Fairclough, , 2001. Then interdiscursivity was analysed by examining the particular mix of genres and styles of discourses (Fairclough, 2003), i.e. how different discourses were expressed in the text.

| Explanation of the social practice
Finally, the discursive practice was explained as a part of the contexts of social practice guided by the concept of ideology, hegemony and power in Fairclough's theory of discourse. The discursive practice in the combined corpus of texts is presented in the 'Results' section and is explained as a social practice in the 'Discussion' section.

| Rigour
The trustworthiness of the data was secured by describing the documents and interpretation process thoroughly (Lincoln & Guba, 1985).
The critical discourse analysis is just one of many readings. It is a cultural production where we bring our preunderstanding of the field into the analysis (Rogers & Schaenen, 2014). As registered nurses the first and third author brought a preunderstanding of the field to the analysed documents, which can be both a strength and a limitation; a strength by knowing the field and limitation by the possibility of being blind to non-nursing aspects in the texts (Dahl, 2017). The fact that the second author does not have an affiliation to the Norwegian health system balanced the analysis and results. Sensitive to the position and resources applied in analysis (Fairclough, 2001), the online cancer pathways documents were read and compared by the researchers in an ongoing process and the analysis was supported by quotations. Using both quantitative data (computational analysis, frequency counts) of several documents and qualitative text analysis, provides a reliable mean of identifying patterns in the documents (Crawford et al., 2014).

| RE SULTS
This study analysed cancer pathways web-based information to patients in Norway, including the general information and six of 28 specific pathways. All the 28 brochures of pathways started in the same way, beginning in general terms about procedures and tests and then assessment phases 1, 2 and 3 with comments and timescales: You have been referred for assessment by the specialist health service because there is a suspicion of a cancer diagnosis. Specialized procedures and tests will determine whether or not you have cancer. If it is shown that you do have cancer, your treatment will be planned in consultation with you.
The corpus of combined texts about pathways for cancer care, or OCCPC, as described above, constructs patients in particular ways in relation to their authority through the use of evaluative words, personal pronouns (instead of names), modal auxiliary verbs (will, can, should, must) and active or passive words. In the first instance, we examined the frequency of words used in the OCCPC (Tables 1 and 2) to achieve a broad 'linguistic signature' for the combined text to assist a critical discourse analysis.
The patients are named as patient/s or referred to the personal pronoun forms you and your. The pronoun you is one of the most frequent words in the OCCPC and reveals a marked passivity in the construction of the patient in the text. All the 28 brochures have the same text. For example, the word you is collocated in phrases that remove agency, as in 'you will be given', 'you will be told', 'you have been referred for assessment', 'give you more information'. In effect, the patients get or receive information. The sentence: 'Treatment will be planned in consultation with you' is foregrounded at the start of each 28-pathway document, however, the word dialogue is only

| The medical and interprofessional expert discourse
The word 'we' is used only five times in the OCCPC and typically describes the health personnel and the cancer organization: We work often interprofessionally. The personal pronoun 'we' does not include the patient. The words cancer, treatment, assessment, procedure, tests, shall and will had a high frequency. As made apparent in the following examples, patients are given or told what to do. They are compelled to act rather than invited to participate. Patients must do certain things or be subject to what the service provider/s has to or have to do: Information will be given about possible treatments and what will happen next. You will be told about the procedures and tests that are relevant for you and when they will be carried out… You must do preparation before the test … You must come back to move the drain. Sometimes we have to start a treatment -Often we have to consider how long and how much. That you have another illness that has to be investigated.
The only ground for patient agency was the collocation physical activity, mentioned 16 times in the OCCPC. For example, we The overall construction of patients as passive as opposed to agentive can be seen clearly in the wording of the following example from the OCCPC:

Procedures and tests for suspected cancer When
you attend your first appointment at the hospital, you will be told about the procedures and tests that are relevant for you and when they will be carried out.
Various types of procedures and tests are normally carried out. During the assessment, your suspected cancer will either be confirmed, and you will be given

115/114
Will/ be (shall in Norwegian) medical questions will be answered by a doctor (all brochures) information also will be given (IS-0502E) the cancer pathway will be concluded (IS-0467E) any decision concerning your treatment will be planned in consultation with you (all brochures) you will be told about the procedures and tests (all brochures) the cancer pathway coordinator will book the necessary appointments for your procedure (IS-0502E, IS-0467E, IS-0465E) your doctor will tell you what to expect (IS-0465E) what sort of treatment you will need? (IS-0502E) the treatment should be considered the patient's wishes and individual situation (general information) sometimes wish the patient to defer the assessment and treatment (general information) 4/6/1 Participation/communication/dialogue the goal is that the patient and relatives experience good information, involvement, participation and dialogue with the health service. (general information) participation and communication give possibility for preparer the pathway (general information) 5 Nursing -Only in the general information in the same paragraph.
In cancer pathways, nursing relates to the disease and the treatment of the disease. (general information) a diagnosis, or it will be disproven. You may also be referred for assessment for an illness other than cancer.
If the assessment indicates that you have cancer, you will be given information on possible treatments and what will happen next.
In summary, the medical and interprofessional discourse of the OCCPC constructs the patient as a person who must listen to and do as health professionals decide. The patient is presented as a passive individual without perceptions and knowledge about their own situation and thereby lack power.

| Pathways discourse
The Norwegian national cancer pathways guidance at Helsenorge.no insists that pathways should ideally provide predictability and security for the patients and their relatives: The aim is to speed up patient assessments and start of treatment and minimize waiting times, to make the assessment period as predictable and stress-free as possible for you.
Perhaps unsurprisingly given the focus, the word pathways have a high frequency (147) in the text and the concept is defined in the text as follows: The pathways define the number of calendar days that each part of the assessment should take. These are called pathway timescales.
That said, this pathways discourse underlines patient passivity. It is the hospital's business to ensure the patient completes the pathway. It is as if the patient is placed on a particular conveyer belt: The hospital will do its best to ensure that patients with suspected cancer complete their cancer pathway within the time framework described in the pathway concerned. This distancing effect, constructing the patient as a passive recipient in a process not their own, is evident in the patient role reduced to asking questions: If you have any questions, you can contact the cancer pathway contact. This constructed passivity is further evident in discourse of the patient as subject to program codes, as follows: The pathways time will be monitored by defined codes at different measurement points. Pathways prosess start, assessment start/ attendance at the spesialist health services, investigation end/decision and treatment start. See the code guides.
The patient is effectively constructed as a de-personalized disease on care pathways. The patient is granted no agency in timely movement along the pathways. This is the preserve of health professionals Here, the OCCPC constructs the patient in a very limited way as a person who might have something to say in response to the enquiry of a health professional. In other words, the scope for participation remains in the gift of experts. The patients are constructed as an active person in the first paragraph of the information, in each of the six documents enfolded into the OCCPC. And then being marked as a passive person without perceptions and knowledge about their own situation, who has to listen to and do as health professionals decide. And finally reduced to a disease and a code that generates standardized pathway procedures and timescales (Figure 1).

| D ISCUSS I ON
This study analysed the pathways information to patients, establishing discourses that construct the patient differently (Figure 1), represent different ideologies (Fairclough, 1992) or sustain different political perspectives. We found that the medical and interprofessional expert discourse and the pathway discourse dominated in the care pathways texts, supporting the powerful position of health personnel. Røstad et al. (2013)  Norway, this is also dominated by the discourse of expertise (Aasen & Dahl, 2019). Analysis of other similar legislations in Denmark (Lassen et al., 2018) and Sweden (Dahlborg Lyckhage et al., 2017) show that the power still belongs to the healthcare system and promotes an ideology of paternalism.
The new cancer strategy in Norway (2018) (Helse.norge.no, 2020) which is intended to reduce unnecessary practice variation, improve coordination and continuity (Faber et al., 2014) and follow governance systems (NPM) in their focus on efficiency and productivity. However, the costs have been reduced by implementing standardized pathways (Rotter et al., 2010). In the cancer pathways, NPM is visible in the purpose of the pathways: 'to speed up patient assessments and start treatment and minimize waiting time, in order to make the assessment period as predictable and stress-free as possible for the patient' (Greve, http

| Standardized pathway timescales versus person-centred care
The patients get to know their pathways when they have received a diagnosis and then they are allocated a code. This code effectively means that the patients have been standardized. All patients with the same code get the same timescale. In effect, coding deems patients as similar rather than unique individuals. Richards et al. (2015) (p. 3) argue that person-centredness can only happen if there is a person-centred culture and discourse. The cancer pathways did not evoke a person-centred culture. The notions of participation and dialogical caring practices were mostly absent.
The cancer pathways in the Norwegian information brochures start with a short paragraph, constructing the patient as an active person in interaction with the health personnel, but this is the only time patient participation discourse occurred in these texts.
Participation 'does not necessarily require shared decision-making but rather a dialogue with shared information and knowledge, dialogue and power exchange (Aasen et al., 2012

| Limitations
The data we analysed are limited to a Norwegian context and relatively small number of texts. A limited number of documents can F I G U R E 1 The process of discourses and constructing the patient in cancer pathways [Colour figure can be viewed at wileyonlinelibrary. com] The participation discourse: The patient is constructed as an active person in interaction with health personnel Medical and interprofessional expert discourse: The patient is constructed as a passive person without knowledge about his/her own situation, who has to listen to the health professional Pathwaysdiscourse: The patient is constructed as a disease/diagnosis and a code which generates standarizied pathway timescales however be valuable in corpus-assisted discourse analysis (Atkins & Harvey, 2010). The results can be deemed transferable to similar contexts of care pathways for cancer and possibly for other diagnoses in other countries.

| CON CLUS ION
This study offers insight into the construction of patients in online clinical pathways documentation for cancer treatment, with a goal to contribute equal and quick investigation and start of treatment without unnecessary waiting time (Helse.norge.no, 2020). The analysis of the information on cancer care pathways in the OCCPC and its constituent texts revealed how governance systems such as New Public Management demands on efficiency and productivity influence and shape the discourse of cancer pathways. The World Health Organization has promoted a person-centred approach where participation and a partnership of equals are key. Neither a person-centred approach nor caring dialogue was present in the specific pathways.
The patients are first constructed as participating and active only in a very brief participation discourse in the context of a dominant expert and pathway discourse which reduces patients to 'a disease' and 'a code'. The dominant ideology of the pathways was found to be paternalistic whereby patients were constructed as passive persons who get standardized treatment. This can result in a negative impact on the patient`s health outcomes in that way that, when the patients have received a code and their program for treatment is decided, the patients' needs and wishes are no longer prioritized as important.
The individual patient joins a coded, standardized group and the care pathways texts encourage health practitioners to overlook their patients' unique and particular identities and lived experience.

CO N FLI C T O F I NTE R E S T
No conflict of interest has been declared by the authors.

PE E R R E V I E W
The peer review history for this article is available at https://publo ns.com/publo n/10.1111/jan.14513.