This study forms part of a larger PhD on the building of trust in clinical trials, where an ethnographic approach, informed by Heideggerian practice theory was utilised. More details on this can be found in Reynolds [37]. Ethnographic observations over a 17-month period were conducted, supplemented with interviews, at a UK clinical trial centre based in a diverse region of London. According to the 2021 census, the area in which The Centre was located had self-reported ethnicity of: 61.3% white, 14.1% Asian, and 10.5% black [38]. The staff and participants observed reflected this local diversity being from a range of nationalities and ethnicities. This centre was chosen because of its experience and success in running clinical trials. For example, it was the highest recruiting centre for a large, transnational, long-term, hypertension study. As a result, the local area has seen rapid decline in cardiovascular event rates, and an improvement of around 300% above the national average. The Centre ran trials on many areas of cardiovascular disease, as well as diabetes. For this study, three trials which investigated treatment for hypertension were observed. The Centre, trials, staff, and trial participants have all been anonymised as a condition of this study’s ethical approval. All the names of people and trials that follow are pseudonyms.
The trials observed
All three trials observed were on different ways to manage hypertension. A summary of the trials can be seen below and in Table 1. These trials tested different interventions: combinations of already licensed drugs, the taking of an organic substance in drinking form, and an implanted device into a major artery. The trials were also a mix of commercially and non-commercially funded. This range was chosen in order to gain an understanding of trial practices across a range of trials within the same centre.
Snowdon
The Snowdon trial aimed to determine if a person’s response to antihypertensive drugs differs by “self-defined” ethnicity, with the ultimate goal of being able to deliver personalised treatment for high blood pressure. Participants would take combinations of already licensed hypertension medications. The Centre planned to recruit high double figures of participants for this trial. A non-departmental government body and a charitable organisation jointly funded the trial. The trial was being undertaken at over 10 sites in the UK.
Ben Nevis
Ben Nevis was a commercial study sponsored by a private biotechnology company. It involved implanting a device into a major artery and delivering it intravascularly via a delivery catheter. The primary objective of the study was to evaluate the safety and effectiveness of the device in reducing blood pressure in participants with refractory hypertension. Up to 200 participants were being recruited at 50 different sites worldwide with single-figure recruitment targeted at The Centre. It was a sham-controlled trial.
Scafell Pike
Scafell Pike aimed to determine whether a natural, organic substance in juice form reduced over thickening of the heart muscle (left ventricular hypertrophy, LVH) and stiffness of the arteries when given to hypertensive participants. Overall, this trial aimed to recruit triple digit figures of participants at The Centre. A UK university and a UK hospital charity sponsored the trial. The study was double-blind, randomised, and placebo controlled.
All of the trials provided travel expenses. Ben Nevis and Scafell Pike also paid participants a nominal amount for their time (£150). All of the trials hoped to find ways to manage participants’ hypertension by optimising and/or reducing patients’ medication. When speaking to participants about motivation, reducing medication was frequently mentioned.
Data collection
Observations occurred across the whole cycle of participant involvement with The Centre, including pre-screening, regular trial consultations, and post-trial visits, as well as observations of the running of The Centre from reception. All the observations were conducted by the first author, DJR. The majority of participants on the Snowdon trial were scheduled for Monday, Wednesday, and Friday mornings. For observations of this trial, DJR gained permission from the staff to arrive opportunistically and without prior scheduling for observations. Participants on the Scafell Pike and Ben Nevis trials were booked in at more varied times. For these observations, DJR would talk to the PIs at the start of the week about when visits were scheduled and specifically come to The Centre for them. All staff were consented to be observed by DJR at the start of the study. For trial participant consent, when participants arrived staff would ask them if they would mind having an ethnographer in the room during their visit. If they did not, DJR would enter, introduce himself and the study, and take written consent. DJR would then sit on a chair in the corner of the room. DJR took handwritten notes in a notebook during the consultations, before typing up fuller reflections at the end of the day, after observations had taken place. DJR’s placement and notetaking style were designed to be minimally disruptive to the running of the trials. Halfway through the time at The Centre, one of the clinical trial practitioners described DJR as “a piece of the furniture”. Reflecting on this, DJR believes that people got used to his presence over time and took it to mean that he was now immersed in The Centre (despite being overt).
During the observations, ethnographic interviews were also used when appropriate. These are informal, spontaneous interviews which take place during observations [39]. They allow the researcher to gain an understanding of what is going on during observation, from the point of view of those being observed. These took the form of clarifications and questions about what had been observed. Typically, on each trial visit there would be time when the staff would leave the room, for example, to collect study medication, or a piece of equipment needed for the next part of the consultation. DJR would often use this time to conduct the ethnographic interviews of the participants. These were informal and not recorded by Dictaphone. DJR would write down the responses in his notebook. When ethnographic interviews of staff occurred, they would happen after the participants had left at the end of the trial visit.
Healthcare professionals involved with the trials and some clinical trial participants were interviewed. On the Snowdon trial, two of the three clinical trial practitioners who worked on the trial were interviewed. By the time the interviews began the third was on maternity leave so not interviewed. The PI had been scheduled to be interviewed, but due to the COVID-19 pandemic, and the refocusing of staff towards this it was not deemed appropriate to add an additional burden to them in the form of an interview. For this same reason, neither of the two staff (one PI and one clinical trial practitioner) on the Ben Nevis trial were interviewed. The PI of Scafell Pike was interviewed as they were the person in charge of recruitment and the day-to-day running of the trial. This interview took place pre-pandemic lockdown. Participants on the trials were selected for interview based on their finishing of the trials at the planned time of the interviews commencing. Due to the COVID-19 pandemic, not as many interviews were conducted as had originally been planned with participants as it was decided not to invite participants for interview after the lockdown had been initiated, as all were classed as “vulnerable” and we did not want to add further demands to their lives at that stressful time. However, due to the ethnographic observation and ethnographic interview data previously gathered, we believe we had reached information power and a lower number of interviews did not detract from the work. Here, we understand information power [40]. At the end of the data collection period, it was believed that the additional data being collected was not revealing any meaningful new insights but rather repeating what had been seen previously. A summary of data collection can be seen in Table 2.
Data analysis
As DJR, a non-clinical social scientist, was eventually regarded by staff as “part of the furniture”, he recognised that this close rapport inevitably shaped what was noticed and how it was interpreted. To guard against over-identification, he kept reflexive fieldnotes and critically discussed emergent themes with co-authors, using his outsider disciplinary perspective to question taken-for-granted assumptions. The ethnography yielded 136 pages of fieldnotes along with 86 pages of transcripts from the interviews. These were thematically coded using NVIVO 12. Specifically, the data was analysed using thematic analysis, a qualitative technique for identifying and analysing themes was employed [41, 42]. Rather than counting words or phrases, this method allows for an in-depth analysis of the different areas identified in the data. The typing up of fieldnotes and the transcription of the interviews formed the first part of the analysis as it helped garner familiarisation with the data, and spot patterns and themes that had not noticed before. This took place as fieldwork was being conducted as DJR would type the handwritten fieldnotes onto a computer shortly after taking them, and interviews were transcribed within a week of their taking place. When formal data analysis started towards the end of the fieldwork, data was uploaded NVIVO 12. The data was read and reread, where initial codes were generated. These were then reviewed so that a cohesive story was being told. It must be noted that due to the large amount of ethnographic data that the story presented here is only one that could have been told, with others represented in other outputs. The data analysis was an iterative and reflexive process, with DJR repeatedly going back and rethinking and reworking codes after discussions with MP and GM. Codes were reduced or expanded and earlier coding revisited often as new insights were discussed and uncovered. Writing of findings is often seen as a stage that takes place after the analysis has been finished. However, the writing of findings is in itself a key part of analysis as it involves constantly returning to the data, reviewing the code, with the making new of connections.