A transcendental phenomenological approach was adopted to identify the essence of the lived experience (Moustakas, 1994). Essences are made up of objects (e.g., physical objects, ideas) and their subjective characteristics (e.g., feelings, thoughts, memories and emotions). The essence of a phenomenon describes both the noema, ‘what is experienced’, and the noesis, ‘how it is experienced’. The noema refers to the objective, physical aspects of the experience, for example, what medications are used, or what happens physically when medications are used (or not). The noesis refers to the internal context of experiencing something, for example, feeling scared to take medications or using medications as part of a routine. To uncover the essence of a phenomenon, researchers must pursue experiential data of those who have lived experience of the phenomenon itself that can tell us what happens and how it happens (Moustakas, 1994).
Phenomenologists ‘bracket’ their presuppositions in an effort to reduce bias (Moustakas, 1994). I did this in preparation for this investigation of medication use in older people with cancer. As a clinical pharmacist, I expected people with cancer to be reluctant to use medications due to anticipated negative side effects. As a healthcare professional, I practice evidence-based medicine (i.e.,
Experiences of medication use for cancer 73 using research from clinical trials, case reports and regulatory bodies to construct my knowledge and practice). I believed that chemotherapy, hormonal and adjuvant anticancer treatments were particularly toxic, more than medications for other diseases, acting both as a poison and a cure (Martin, 2006). I believed that these side effects, in addition to the burden of using medications for other co-morbidities, contribute to poor medication use for older people with cancer. I believed my role was to encourage patients to tolerate these side effects and to use treatments as they were prescribed to achieve optimal outcomes of their cancer treatment. In other words, my primary role was to instruct, guide and encourage patients to use pharmaceuticals to treat their cancer - so as to minimise the risk of death from the disease. Reflecting on these previously held beliefs raised questions about patients’ experiences of using medications for cancer and how these pharmaceuticals and the side effects they create fit into patients’ everyday lives.
A convenience sample of participants was recruited from four general practices in North East England, United Kingdom, between April and June 2014. General practices were selected to include areas of low (n = 1), intermediate (n = 2) and high levels of deprivation (n = 1) using the Index of Multiple Deprivation (McLennan et al., 2011 ), to include a diversity of socioeconomic experiences. To support recruitment, general practices displayed posters that read ‘Have you had cancer? We are trying to find out about patient experiences of treatment. Ask your GP for more information’. General practices screened expressions of interest and forwarded contact details of people who met the inclusion criteria: adults over 55 years old with a previous diagnosis of any type of cancer, who had past or current experience of receiving treatment for cancer, who were conversant in English and had capacity to consent to take part in research. A limitation of this recruitment strategy is it may have inadvertently excluded people who had experiences of declining, discontinuing or not using cancer treatments following their cancer diagnosis.
Participants were given an information sheet that explained the project was being completed by a doctoral student and included my post-nominals (MRPharmS) that enabled participants to identify that I was a pharmacist. 1 then made contact to arrange the interview. To prevent my identity as a pharmacist influencing data collection, prior to each interview participants were informed that, although 1 had pharmacy training, I could not offer any clinical advice during the interview, would not share anything they disclosed with their usual care team and wanted to know both about using and not using medications for cancer as prescribed (see Rathbone and Jamie, 2016, for more information about the ethico-legal tension this raised). In taking this approach, 1 hoped to motivate participants to provide honest and open accounts of their experiences, rather than sharing what they expected a pharmacist would want to hear, such as positive stories of medication use as prescribed (Brenner & DeLamater, 2016). Originally the aim was to recruit 15 participants based on theoretical data saturation (Crotty, 1998; Moustakas, 1994); however, once data collection had started, it became apparent that much smaller numbers, approximately five participants, were adequate to reach saturation. A small number of additional interviews, inthis case three, were conducted to confirm saturation had been reached (Crotty, 1998). See Table 6.1 for demographic data.
Data collection included a single in-depth, semistructured interview with each participant. Participants were interviewed at home for between 68 and 115 minutes, with a mean interview length of 90 minutes. Data were collected between May 2014 and June 2015 using a topic guide, which included three open-ended questions:
- (i) ‘what was it like being prescribed medicines?’
- (ii) ‘what are your experiences of taking your medicines as prescribed?’
- (iii) ‘what are your experiences of not taking your medicines as prescribed?’
These questions were open and, initially, did not specifically focus on cancer, as 1 wanted the participants to frame their experiences in their own way. Further probing relevant to their cancer diagnosis included determining the medications used for cancer and how each participant felt about using them. Interviews were audio recorded and additional notes, as relevant, were made regarding emphasis, gestures and how participants interacted with their medications during the interview.
Interviews were transcribed verbatim and anonymised by the interviewer (APR). Transcripts were quality checked by a postgraduate research associate in the same research group. Data were thematically analysed via manual coding followed by computer coding using NVivo. Coding was conducted inductively and iteratively by a single author and verified by supervisors and postgraduate research associates during data discussion group meetings between interviews and following the completion of all interviews. Codes were generated, deconstructed and merged until dominant themes were clarified. This approach was guided by the works of Crotty (1998), Moustakas (1994) and Halldorsdottir (2000).
The study received approval from institutional and national ethics committees.