Journals Library

Primary care patients’ views and experiences of treatments for depression, and of those delivering their care: a secondary analysis of qualitative data

Project title

Primary care patients’ views and experiences of treatments for depression, and of those delivering their care: a secondary analysis of qualitative data

Project reference


Final report date

08 December 2015

Project start date

01 April 2014

Project end date

30 September 2015

Project duration

18 months

Project keywords

Depression; Primary Care; Qualitative Research; Secondary Analysis and Synthesis; Usual Care; Patient-Practitioner Relationships

Lead investigator(s)
  • Dr Katrina Turner, School of Social and Community Medicine, University of Bristol
NIHR School Collaborators
  • Dr David Kessler, School of Social and Community Medicine, University of Bristol
  • Professor Jenny Donovan, School of Social and Community Medicine, University of Bristol (was a co-applicant on the study but her time was not costed, as she works outside of academic primary care; provided methodological input and support)
  • Dr John Percival, School of Social and Community Medicine, University of Bristol (was the part-time Research Associate employed on the study, on a fixed term contract, to lead on the secondary analysis and synthesis of the data)

Project objectives

The aim of the study was to undertake a secondary analysis and synthesis of four qualitative data sets in order to bring together primary care patients’ views and experiences of treatments for depression, and of being treated. Specific study objectives were to:

  1. Assess what similarities and differences exist between patients’ views and experiences of different treatments;
  2. Investigate what is important to patients with depression in terms of delivery of care, characteristics of the practitioner and consulting style, and whether this changes depending on what care is being given;
  3. Establish whether patients receiving more than one treatment at the same time or sequentially, view them as having similar or different roles, and as supporting or conflicting with one another;
  4. Define what usual care entails for patients with depression, and whether this varies between different groups of patients with depression;
  5. Establish whether different patient groups have particular concerns and perceived treatment needs.

Changes to project objectives

The analysis conducted mainly focused on addressing objectives 2 and 4, as on analysing the data it became evident that it was best suited to addressing these two study objectives. However, when undertaking this analysis, consideration was given to whether, for example, different patient groups appeared to have particular concerns (objective 5), and this was commented upon when disseminating findings.

Brief summary


The data set

We conducted a secondary analysis and synthesis of data collected between 2006 and 2011 during four qualitative studies that were all nested within large, multi-centred, pragmatic depression trials funded by the NIHR HTA. Each of these qualitative studies had entailed conducting in-depth interviews with trials participants to explore their experiences of the trial and the treatments received. In total, the data set consisted of 165 interview transcripts.

The trials

Participants had been recruited to the trials through GP practices located in the UK. In each trial, randomisation had occurred at the level of the individual and participants had been allocated to one of two treatment arms. In three of the trials, participants were randomised to either an intervention arm or to ‘usual GP care’. Treatments assessed within the intervention arms were face-to-face cognitive behavioural therapy (CBT), online CBT, and facilitated physical activity. Usual GP care was defined as the care the individual would normally receive from his/her GP. In the remaining trial, participants were randomised to receive either listening visits (i.e. non-directive counselling) or an antidepressant. Treatments were delivered by various individuals - GPs, CBT therapists, Physical Activity Facilitators and Research Health Visitors - and aimed at different patient groups, i.e. patients with a new episode of depression, patients with treatment resistant depression, and women with postnatal depression.

Secondary analysis and synthesis of the data

The first stage of the analysis entailed two members of the research team (KT and JP) independently reading and re-reading a sample of 8 transcripts to gain an overall sense of the data and to consider how the data should be analysed. These transcripts were purposefully sampled from across the four studies to ensure maximum variation in terms of trial arm, participant gender and age. It was agreed that a thematic approach should be taken, as this would enable comparisons to be made across the four data sets. Having agreed this approach, KT and JP re-read and independently coded the sampled transcripts. They then met to compare and discuss their coding. This discussion led to the development of a coding frame. Having agreed this coding frame, transcripts were imported into NVivo and electronically coded. Data pertaining to specific codes were then retrieved and summarised in tables to enable central themes to be identified and comparisons to be made within and across the data sets. Further transcripts were sampled for coding, using the purposeful approach mentioned above, until data saturation was reached, i.e. no new themes emerged. During the course of the analysis, 68 transcripts were analysed in total.

Below quotes have been tagged with the participant’s trial and study number.


Objective 2: What patients value in terms of delivery of care and practitioner characteristics

The accounts of participants from all four studies indicated there are two complimentary sets of practitioner attributes that patients value when receiving care. The first set, helping provide the foundation for the relationship, is based on the practitioner’s bearing (characterised by approachability, empathy, supportiveness and active listening). The second set, helping sustain and develop the relationship, is based on the practitioner’s enabling role with the participant (exemplified by enhancing patient decision making and encouraging patient self-care).


Participants detailed how they valued practitioners who were accessible on an informal, ‘human’ level, and whose receptivity created space to admit and discuss emotional difficulties. Ways practitioners were described as conveying approachability included being friendly, not being condescending and sharing relevant personal information. Participants who had consulted practitioners who they viewed as unapproachable, commented upon the lack of ‘connection’ or ‘relationship’ with them.


Participants who felt their practitioner had understood them, felt more accepted and less stigmatised. They felt able to openly talk to the practitioner about how they were feeling and felt more confident in their own potential to change:

He [online CBT practitioner] knew how I was feeling... When you’re talking to someone [and] they know how you’re feeling, it’s easier to try and change the way you feel. Trial 3, participant 2

Practitioners who had not demonstrated empathy were described as not providing the level and type of support the individual felt they needed.


Participants reported a lack of confidence and self-esteem and, as a result, valued practitioners who gave them practical and emotional support. Such support could help participants cope better and encouraged them to comply with treatment.

Active listening

Across the four studies, participants mentioned that talking to a practitioner had made them feel less isolated and had given them an opportunity to vent rarely expressed feelings, which they found, in itself, emotionally beneficial. Talking was also described as helping patients gain a greater perspective and understanding of their situation which, in turn, had helped them to change their viewpoint.

The accounts suggested that patients felt listened to when the practitioner conveyed focussed attentiveness and ‘really listened’. It was evident this could be achieved within a short, 10 minute appointment and that, if active listening was not conveyed, the time available, however long, would be viewed as insufficient:

You had that hour but the whole time you could see the [CBT practitioner] looking at the clock…and like ten minutes before [the end] she’s going, ‘right I think we need to stop there now’, and it was like, why?, you know, I wasn’t asking for hours and hours. Trial 2, participant 3

The above four attributes were described as establishing the practitioner-patient relationship. Once established, this relationship could then be used as a platform on which participants could develop their own capabilities, particularly in relation to decision-making and self-care. For example, participants talked about how discussions with a practitioner increased their understanding of depression and their ability to evaluate treatment advice. Such discussions were also described as helping them set more realistic goals and to realise the importance of attending more readily to their own needs.

Objective 4: Defining usual care

Four key themes were identified within the participants’ accounts of usual care: reluctance to consult a GP about mental health issues, participants viewing GPs as not well placed to deal with depression, limited time and treatment options, and variation in the quality and amount of care given. These themes were evident across all four data sets.

Reluctance to consult a GP

Participants across all four studies mentioned they were hesitant to consult a GP about their mental health. Reasons given included previous unsatisfactory GP consultations, worrying that they would be viewed as ‘crazy’ or a ‘time waster’, feeling tired and lacking in confidence, feeling it was not appropriate for them to seek help because there was nothing physically wrong with them, and being concerned that they would need to initiate a conversation about their mental health. Some participants linked this concern to the fact that they had experienced little continuity in terms of which GP they had seen. This led to them feeling that they would ‘need to start again’ in terms of explaining how they felt, feeling they had no relationship with a specific practitioner and predicting that, even if they did talk to a GP, there was no guarantee they would see that particular GP again.

It’s a case of you see whoever’s there. “Well, can I make an appointment to see my doctor?”, “Well, no, because all the appointments are booked up”… you’ve got to ask on the day, and you see, the day when she’s there, by the time you get through on the phone or whatever, probably the appointments have gone. Trial 3, participant 10

Another reason participants gave for not seeking help was because they assumed their GP would prescribe them an antidepressant; a treatment which they did not want due to a fear of side-effects, being stigmatised by others and becoming medication dependent.

GPs as mental health practitioners

Some participants described GPs as generalists, who did not know about mental health issues, and a few participants stated they felt GPs viewed patients with mental health problems as an ‘inconvenience’. GPs were also viewed as being limited in what they could do because of the context in which they worked; participants felt they could only offer 10 minute appointments and prescribe an antidepressant. This meant some participants had low expectations about what care GPs could provide.

Interviewer: So if I was to ask you can GPs help with depression?

Participants: They (GPs), they alone, individuals can’t really, because, I mean they direct you… A GP directs, you know, to a specialist or what-have-you… if she was going to prescribe a course of drugs or something, pills, then that’s how she would help, isn’t it?... I mean, they don’t have time. You’ve got ten minutes… she has no resources that she’s holding back from me, you know, there’s nothing, nothing more she can offer really. Trial 3, participant 19

Limited time and treatment options

It was evident that participants who had consulted their GP, had mainly been offered 10 minute appointments and prescribed an antidepressant. Within the accounts there was little evidence that participants felt antidepressants had been prescribed because the GP thought they were the most appropriate treatment for them personally. Medication was described as the only way, or an easy way, for practitioners to treat depression.

Variation in the quality and amount of care given

Some participants thought very highly of their GP. They described their GP as being sympathetic, supportive and empathising with their situation. For these participants, their GP was someone who they could talk openly to without worrying that they would be judged or negatively viewed. These participants had continuity in terms of the GP they saw, and some of them detailed seeing their GP regularly, i.e. every 2 or 4 weeks. Interestingly, the accounts did not suggest that these participants had received a greater range of treatment options or longer appointment times, but rather their GP had interacted with them in a way that meant they had felt listened to and understood.

In contrast, other participants detailed how they felt their GP had dismissed their symptoms, patronised them or not listened. Participants found such interactions upsetting and had responded by registering with a different GP or not seeking help, even though they were aware they needed it.

It was just something about the doctor, I don’t know. She seemed very patronising… she made me feel little and stupid and thick so I thought “oh, I’m not going back to see you again”… I know I need to see her because I know in myself I’m not feeling right (women had previously attempted suicide). But then I don’t want to go… Trial 4, participant 15

Participants who did not experience continuity of care or regularly saw their GP also talked about not being followed-up. As these participants obtained their medication through repeat prescriptions, this meant their medication was not reviewed for several months. In some cases it also meant individuals who described themselves as not coping, were not in contact with a practitioner.


Primary care patients with depression value a practitioner who is approachable, supportive, empathic and actively listens. Such attributes help to establish a positive practitioner-patient relationship and, having been established, this relationship can support patients in developing their own understanding of depression, treatment choices and ways to manage their depression. It is evident that patients view a practitioner’s attitude and bearing as important, if not more important, that the actual treatment they prescribe and the time they give to the patient. It is also apparent that what patients value does not differ depending on the treatment they are receiving (e.g. an antidepressant or CBT), and therefore the practitioner they are consulting (e.g. a GP or a CBT therapist).

In terms of what usual care entails for primary care patients with depression, whilst the findings indicate there is variation in the quality and amount of care patients receive, patients largely perceive usual GP care as being limited to 10-minute appointments and being prescribed an antidepressant. The findings also show that individuals taking part in a trial who are randomised to usual care may be reluctant to consult a GP about mental health problems and, in some cases, may not seek help. This differs from the experiences of individuals randomised to an intervention arm, as being in this arm would automatically lead to opportunities for changes and increases in care. It is also evident that having initiated contact, patients may experience a lack of continuity of care and follow up, and consult a practitioner who they feel dismisses their symptoms; this is likely to be very different from the attention and consistency they might have received in a trial’s intervention arm. Such findings indicate that there are important differences in trial participants’ experiences of intervention and usual care arms, which relate not only to the treatment they receive but also to their experiences of accessing care and how care is delivered. Future researchers may want to try and address some of these differences, in order to more appropriately evaluate the effectiveness of a treatment.

Plain English summary

Depression is a common illness, treated mainly in primary care. Treatment options exist, for example, antidepressants or psychological interventions, such as cognitive behavioural therapy (CBT), but treatment adherence is often poor and some patients remain depressed despite receiving both medication and psychological intervention. Researchers have explored patients’ views and experiences of different treatment options, but little is known about what treatment patients with depression usually receive in general practice. It is also not known what practitioner characteristics patients value when receiving treatment, and whether these change depending on what care is being given. This study aimed to address both these gaps in knowledge.

The study entailed a secondary analysis and synthesis of data collected during four different interview studies, all of which had been nested within primary care depression trials. Treatments assessed within the trials included antidepressants, face-to-face CBT, online CBT, facilitated physical activity and listening visits (i.e. non-directive counselling). These treatments were delivered by various individuals - GPs, CBT therapists, Physical Activity Facilitators and Research Health Visitors - and aimed at different patient groups with depression (patients with a new episode of depression, patients with treatment resistant depression, and women with postnatal depression). From a total data set of 165 interview transcripts, we purposefully sampled and reanalysed 68 transcripts.

Findings indicated that patients with depression mainly view and experience usual care as being limited to 10-minute appointments and being prescribed an antidepressant. Findings also suggested that individuals who take part in clinical trials, who are randomised to receive usual care, may be reluctant to consult their GP about their mental health and, in some cases, may not seek help. It was also evident that if contact was made, patients may experience a lack of continuity of care and follow up. Such experiences are likely to be different from those experienced by individuals randomised to a trial’s intervention arm.

Patients’ accounts regarding what they value when receiving treatment, suggested they want to consult a practitioner who is approachable, empathic, supportive and actively listens, and that these attributes are valued irrespective of the type of practitioner they are consulting. The presence of these practitioner characteristics had a positive impact on patients’ engagement with treatment, and appeared to be more important to patients than the amount of time or specific treatment a practitioner was able to provide.


Published articles

  1. Percival, J., Donovan, J., Kessler, D. and Turner, K. (2016), ‘She believed in me’. What patients with depression value in their relationship with practitioners. A secondary analysis of multiple qualitative data sets. Health Expectations. doi: 10.1111/hex.12436.

Planned articles

The following two papers are currently being drafted and will be submitted to Trials:

  1. Turner K, Percival J, Donovan J, Kessler D. Defining usual care in depression trials to improve recruitment and the evaluation of complex interventions: a synthesis and secondary analysis of multiple qualitative data sets.
  2. Turner K, Percival J, Kessler D Donovan J. The value and challenges to synthesising multiple qualitative data sets: experiences from the STRIDE study.

Public involvement

As this study involved an analysis of existing data, our focus was very much driven by the material we had. However, two of the original studies had PPI members who commented on patient information sheets and letters to patients, and discussed with the qualitative researcher employed to do the interviews what language should be used and what areas should be covered during the interviews. In addition, we are currently talking with our PPI co-ordinator, in Bristol, about how best to present our findings to individuals who have been diagnosed with depression, so that we can hear their views on our findings and plans for dissemination.


The published paper will be of interest to health care practitioners working in the area of mental health, as it argues that what patients with depression value in their relationship with practitioners does not differ depending on the treatment they are receiving (e.g. an antidepressant or CBT) or the practitioner they are consulting, e.g. a GP, health visitor or CBT therapist.

The usual care paper to be submitted to Trials will highlight there are important differences in the experiences of individuals recruited to clinical trials, who are allocated to intervention and usual care arms. These differences relate not only to the treatment they receive but also to their experiences of accessing care, how care is delivered, and what treatment and practitioner expectations they have. The paper will be of interest to researchers involved with clinical trials.

The final paper listed above will be methodological in nature. It was presented as an oral presentation at the International Clinical Trials Methodology Conference, Glasgow, November 2015. The presentation was well received and we have been invited to submit it to the Qualitative Methods, Trials and Systematic Reviews series, which is a cross-journal series between Trials and Systematic Reviews. The paper will detail how we synthesized multiple qualitative data sets and highlight the challenges we faced when doing so. It will be of interest to other qualitative researchers who aim to use this methodology.

This project was funded by the National Institute for Health Research School for Primary Care Research (project number 207)

Department of Health Disclaimer

The views and opinions expressed therein are those of the authors and do not necessarily reflect those of the NIHR School for Primary Care Research, NIHR, NHS or the Department of Health.