Evaluation of depression treatment effectiveness may be less straightforward compared with other medical conditions. Multiple definitions are used, for example remission, response or recovery from depression, and inconsistencies within definitions exist.Reference Prien, Carpenter and Kupfer1–Reference Jacobson and Truax3
Remission is widely recognised as the most favourable outcome of treatment for depression and primarily relies on changes in the amount and severity of depressive symptoms.Reference Keller4,Reference Zimmerman, McGlinchey, Posternak, Friedman, Attiullah and Boerescu5 In clinical trials, depressive symptoms are measured by clinician-rated or patient-rated depression rating scales, which is a pragmatic way to determine treatment effects. Although the use of symptom rating scales is important for objectively measuring treatment outcomes, this might not be optimal in the case of depression. First, clinical trials often use treatment outcomes defined by researchers and the clinical community,Reference Fava6,Reference Sartorius7 instead of reflecting patients’ values regarding relevant treatment outcomes.Reference Bridges and Jones8 Second, depression is a mental disorder with a high probability of recurrence and chronicity.Reference Torpey and Klein9 In specialist mental healthcare, around 85% of patients with major depressive disorder experience a recurrence within 15 years.Reference Hardeveld, Spijker, De Graaf, Nolen and Beekman10 In those patients, remission might not be a realistic treatment outcome but would rather be an exception. Indeed, many remitted patients continue to experience ongoing deficits in functioning or quality of life.Reference IsHak, James, Mirocha, Youssef, Tobia and Pi11,Reference Verduijn, Verhoeven, Milaneschi, Schoevers, van Hemert and Beekman12
Symptom-based scales will likely remain the standard for treatment outcome assessment in randomised controlled trials and clinical practice. However, it may be worthwhile to develop measures that assesses relevant domains other than symptom resolution.Reference McGlinchey, Zimmerman and Atkins13 Incorporating patients’ perspectives and attitudes towards health and illness, and taking a more patient-centred approach in the assessment of treatment outcomes has gained greater interest recently.
Relevant outcomes of treatment
Patients, their spouses and the community demand treatment that produces relevant outcomes, i.e. significant return on investment.Reference Awad14,Reference Trujols, Portella, Iraurgi, Campins, Sinol and de Los Cobos15 Patients that actively engage in treatment decisions have higher satisfaction scores and better clinical outcomes.Reference Deen, Fortney and Pyne16,Reference Clever, Ford, Rubenstein, Rost, Meredith and Sherbourne17 Assessment of treatment effects may be biased if it is based on treatment outcomes that have limited relevance to patients and clinicians in the consulting room, having a negative impact on policy decision-making.
Little evidence is available about what patients in specialist mental healthcare who have experienced depression would define as relevant outcomes of treatment. A study performed in primary care demonstrated that patients value a broad range of indicators of recovery from depression, for instance, managing the depression, functioning as before and enjoying activities as before the depression.Reference Johnson, Gunn and Kokanovic18 In a quantitative study, patients perceived symptom resolution as only one important factor. The presence of features of positive health (for example optimism), a return to one's usual level of functioning, and feeling like your usual self, were, according to patients, better indicators of remission from depression.Reference Zimmerman, McGlinchey, Posternak, Friedman, Attiullah and Boerescu5 Finally, two studies investigating relevant treatment goals showed that improvements in functioning in social and occupational domains were also considered important.Reference Battle, Uebelacker, Friedman, Cardemil, Beevers and Miller19,Reference Uebelacker, Battle, Friedman, Cardemil, Beevers and Miller20
To be able to develop treatments that are in line with patients’ needs and to provide effective specialist depression care, it is crucial to know what patients value as relevant outcomes of depression treatment in specialist depression care. In addition, it is important to know what clinicians perceive as relevant outcomes of treatment for their patients. Discrepant views may well bring out diverging expectations, leading to disappointment and treatment failure, unnecessary prolongation of treatment or treatment discontinuation. To our knowledge, clinicians’ perspectives on relevant outcomes of depression treatment have not been explicitly studied before. The aims of the study are (a) to investigate patient-relevant outcomes and clinician-relevant outcomes of depression treatment, and (b) to investigate whether any discrepancies exist between patients’ and clinicians’ views regarding these relevant outcomes.
Method
Data for this qualitative study were collected by means of focus group interviews to stimulate exchange of views and experiences between patients. Due to their time constraints, we conducted semi-structured interviews with clinicians.
Ethical aspects
In line with the Dutch Medical Research involving Human Subjects Act, the Medical Ethics Review Board (METc) of the University Medical Center Groningen exempted this research from full review. A waiver from the METc was obtained because there was not an infringement of the physical and/or psychological integrity of the participants. Prior to the start of the focus groups, written informed consent was obtained from all participants. All participants agreed on audiotaping the interviews and usage for scientific research after anonymisation.
Selection of participants
Both patients and clinicians in the specialist mental healthcare setting were selected by purposive sampling. Patients were eligible for participation in the study if they (a) have (or had) a depressive disorder as main diagnosis; (b) had experience with depression treatment, and (c) were willing to share their views and experiences on relevant treatment outcomes for depression treatment in a focus group.
Clinicians were eligible for participation in the study if they (a) treat(ed) patients with depressive disorders, and (b) were willing to share their views on relevant treatment outcomes for depression treatment in a semi-structured interview.
Recruitment of participants took place in the mental healthcare organisations connected to the Rob Giel Research center (RGOc; a collaboration of six regional mental healthcare providers in the Northern-Netherlands). Patient Councils of these mental healthcare organisations distributed flyers in waiting rooms and day care areas. The call was also posted on the RGOc website (www.rgoc.nl) and distributed via the RGOc newsletter. Clinicians were recruited via the RGOc network, RGOc website and RGOc newsletter. Interested participants could sign-up by sending an email or by telephone. One of the researchers (K.K.) approached all participants by telephone for an eligibility check and asked them via snowball sampling for the recruitment of other potential participants.
In total 18 individuals were eligible: 12 patients and 7 clinicians. One eligible patient dropped out of the study as he forgot to attend the focus group meeting.
Data collection
We conducted four focus group interviews in three focus groups between November 2016 and June 2017. During the first interview, which was a pilot, we were unable to discuss all the topics of the interview guide. Therefore, these participants were interviewed a second time to discuss the undiscussed topics of the interview guide and to verify the results of the first interview.
The focus group interviews with patients were moderated by F.J. (PhD, psychologist and epidemiologist) and K.K. (MSc, health scientist). Both researchers are women and have previous experience and training in conducting and executing qualitative research. The researchers did not know the participants in advance nor did they have a therapeutic relationship with the patients. Therefore, a short introduction round took place prior to the start of the focus group interviews. The researchers went openly into the focus group interviews.
The interview guide for the focus group interview was pilot tested with the first focus group after which we made small adaptations. Topics in the interview guide included: personal experiences with treatment for depression and treatment goals, used outcomes during treatment, views on the definition of remission/treatment success for treatment evaluation.
The focus group interviews with patients took place in a meeting room in a mental healthcare organisation that was closest to where the patients lived. Besides participants and researchers, no one else was present during the focus group interviews. The number of participants in the focus groups varied between three and five.
Seven semi-structured interviews with clinicians were conducted by K.K. in May and June 2017. The interview guide for the semi-structured interviews included the same topics as the interview guide used for patients. The semi-structured interviews with clinicians took place in the consulting room of the practising clinician.
During the (focus group) interviews, we made field notes and analysed those as preparation for the next interview. The duration of the focus group interviews and semi-structured interviews ranged between 73 and 86 min and between 40 and 85 min, respectively. We continued with data collection until data saturation was reached.
Data analysis
All focus group interviews and semi-structured interviews were audiotaped and transcribed verbatim. We made a summary of all interviews and returned the summaries to the participants for a member check. We used the phenomenology approach to examine participants’ subjective experiences on a particular phenomenon. Data were coded using thematic content analysis. Two interviews were open coded by two researchers (K.K. and F.J.) with paper and pencil for comparison. After comparison and agreement on the codes, all transcripts were coded using the software package ATLAS-ti version 8.0.40.0 (ATLAS.ti Scientific Software Development GmbH) to ensure systematic coding of the data. Subsequently, main themes were derived from the data and identified from the codes. Finally, we integrated main themes and research data to compare patients’ perspectives and clinicians’ perspectives using the one-sheet-of-paper approach.Reference Ziebland and McPherson21 Quotations from the participants were translated from Dutch by a native English speaker and are presented to illustrate the themes and findings. See supplementary material (available online at https://doi.org/10.1192/bjo.2020.27) for the consolidated criteria for reporting qualitative research checklist that was used to report this research.
To ensure validity of the study data, peer debriefing took place on a frequent basis with a third researcher, M.A.A. (PhD, social scientist). From the start of the study (prior to data collection), during data collection, coding of the data and reporting of the findings, this unbiased researcher provided feedback to ensure credibility.
Results
Table 1 presents the characteristics of the study sample. Out of 11 patients, 8 patients were women. The patients ranged in age from 22 to 69 years. Nine patients were currently in treatment for their depression. Two patients were currently not in treatment and were in remission and/or had finished treatment at the time of the focus group interview. Most patients experienced multiple depressive episodes.
Characteristics of interviewed patients and clinicians
Six clinicians were women and one was a man. They ranged in age from 32 to 64 years. The clinicians worked in five different mental healthcare organisations. We interviewed four psychiatrists, one specialised psychiatric nurse and two psychologists.
Patients’ perspectives on relevant treatment outcomes in depression treatment
Main themes that emerged from the analysis of the data, relating to relevant treatment outcomes from the patient's perspective, are described below and examples of quotes for each theme are illustrated in Table 2.
Quotes for each theme from the patient's perspective
Social functioning and interpersonal relationships
The majority of patients mentioned goals related to social functioning (defined as an individual's ability to perform and fulfil normal social rolesReference Hirschfeld, Montgomery, Keller, Kasper, Schatzberg and Moller22) and interpersonal relationships as important goals of depression treatment. Normalisation of social functioning was considered important (Table 2, quote 1). It included getting out of bed, continuing normal daily activities and functioning as before the depression. One patient stated that it was acceptable to use antidepressant medication, if necessary, for obtaining normalisation of social functioning (Table 2, quote 2). Patients saw undertaking activities again with friends and family as a good indicator of social functioning.
However, patients who had experienced multiple depressive episodes or patients who were diagnosed with chronic depression had a different view on functioning. They stressed that they needed to find new ways of functioning they would consider as satisfactory given circumstances, even though it would not quite be in the same way as before, as illustrated in the next quote:
‘You can also find other ways, can't you? Functioning very differently from how you used to, and yet, um, find satisfaction. That you've found a new mode, let's say. So you could still have symptoms, but you have improved as it were.’ (Participant 13, woman, age 41)
Over one-third of the patients mentioned that their depression severity and stage of life determined their treatment goals related to functioning: young patients strived for returning to a full life, such as being part of the labour force, starting a family, having life goals like any other person, whereas middle-aged patients adjusted their goals and ambitions. They focused especially on personal relationships and functioning well within their family again. Being ‘free of depressive symptoms’ was not mentioned by any patient as a relevant treatment goal when patients experienced recurrent depression or became chronically depressed.
Prevention of future recurrences
More than half of the patients mentioned that they perceived long-term outcomes of depression treatment as very meaningful. They considered it important to learn how to integrate techniques on how to cope with depression in daily life and to obtain skills for the prevention or early signalling of future depressive episodes (Table 2, quote 3 and 4).
Acceptance of illness and managing the depression
As a result of patients experiencing several depressive episodes and remaining vulnerable, it appeared that accepting that depression is part of a patient's life was a way forward. One could still have a good life, and managing the depression became a goal in half of the interviewed patients as illustrated in the following quote and in Table 2 (quote 5 and 6).
‘…At some point you start to adjust your expectations and at some point you also realize okay well, in three years’ time I'd like to be here and here, what do I have to do to achieve that? And that's what I've more or less achieved and now I also think, okay and I'll never be entirely rid of it and I'll continue to have that vulnerability.’ (Participant 2, woman, age 22)
Achievement of personal goals and societal expectations
The majority of patients had personal goals that were set during their depression treatment, for example improving self-esteem, absence of suicidal thoughts, not being a burden to others, or structure in daily routine. One-third of the patients mentioned that their own expectations and goals changed after experiencing several depressive episodes, realising that full recovery is not attainable.
Some patients experienced societal pressure during treatment. They felt that their treatment goals were not always in line with those set by the treating clinician, mental healthcare organisation or society, as illustrated by quote 7 and 8 in Table 2. Finally, some patients got the impression that clinicians focused too much on having no residual symptoms as the clinical end-point, and that clinicians relied too much on depression severity scales, without looking at the individual patient (Table 2, quote 9).
Clinicians’ perspectives on relevant treatment outcomes in depression treatment
Main themes identified from the clinician's perspective are described below and illustrated in quotes for each theme in Table 3.
Quotes for each themes from the clinician's perspective
Symptom reduction/clinical improvement
All clinicians mentioned symptom reduction or clinical improvement as an important treatment outcome in practice. They focus specifically on the main symptoms of depression: low mood, loss of pleasure in activities and concentration. In addition, the majority of clinicians also use depression severity rating scales to evaluate symptom reduction or clinical improvement (Table 3, quote 1 and 2).
Social functioning and interpersonal relationships
All clinicians mentioned that patients should be able to take up former social and personal roles, as illustrated in the following example and quote 3 in Table 3.
‘That it gradually improves in terms of mood, that the level of activity improves, do they go back to school or not, do they start a course, do they leave the house, sometimes it's …, it depends how serious it is, to also look at that, not only to have the symptoms gone but also that they become active again and, say, resume their normal development somehow and keep going.’ (Participant 10, woman, age 59)
Participating at work was not seen as a necessary goal of treatment from the clinician's perspective (Table 3, quote 4).
Patient satisfaction and quality of life
The majority of clinicians noted that quality of life and patient satisfaction with treatment or satisfaction with functioning were also relevant when determining whether depression treatment was successful (Table 3, quote 5, 6 and 7). More than half of the clinicians mentioned that they ask whether their patients are satisfied with the results of treatment.
Achievement of predetermined personal goals
Most clinicians mentioned that achieving goals defined at the start of the treatment was a relevant indicator of treatment success (Table 3, quote 8). According to clinicians, these predetermined goals mainly relate to personal goals and may differ between patients, but should preferable be agreed upon by patient and clinician. One clinician mentioned that it would be helpful to base treatment success on achievement of predetermined treatment goals, as it is hard to tell whether a patient is in full remission (Table 3, quote 9).
Similarities and discrepancies between patients’ and clinicians’ perspectives
In Table 4 relevant outcomes according to both perspectives are summarised. Restoring social functioning and interpersonal relations, regaining quality of life, and achieving personal goals were mentioned most often as important indicators of treatment success by both clinicians and patients. In the last column of the table, we have added outcomes that are frequently used in randomised controlled trials for comparison with the outcomes in our study.Reference Prien, Carpenter and Kupfer1,Reference Nierenberg and DeCecco2 Apparently, in trials reduction of symptoms, i.e. (time to) recovery, remission, response, or a functional/administrative outcome are considered relevant outcomes.
Discrepancies and similarities from patients’ perspective, clinicians’ perspective and outcomes used in randomised controlled studies
✓, present.
a. ‘Functioning as before’.
b. Results depression rating scales + clinical judgement.
Importantly, evident discrepancies between clinicians’ judgements and patients’ perspectives also emerged. Clinicians emphasised symptom reduction as a relevant treatment outcome, whereas for patients this was not a prominent theme. Although normalisation of social functioning and interpersonal relationships may go hand in hand with symptom reduction, some patients argued clinicians focused too much on being free of symptoms, whereas patients would rather learn how to manage their depression. Next, only patients and not clinicians made a clear distinction between treatment outcomes in a first-episode of depression versus recurrent/chronic depression. Notably, clinicians mentioned patient satisfaction with treatment as a relevant outcome in the treatment of depression.
Discussion
Main findings and comparison with findings from other studies
This qualitative study provides new insights into what patients and clinicians perceive as relevant outcomes in the treatment of depression in the specialist care setting. From a patient's perspective, the main themes identified were social functioning and interpersonal relationships, prevention of future recurrences, managing depression and achieving personal goals. Clinicians on the other hand were mainly focused on symptom reduction and clinical improvement, social functioning and interpersonal relationships, patient satisfaction and achievement of predetermined personal goals. Patients and clinicians agreed on the majority of issues, but differences were also found.
Importantly, usual end-points defined in trials, i.e. remission and response, only to a limited extent appear to reflect what matters in clinical practice. This may indicate that the efficacy observed in trials may not reflect actual preferences, which in turn results in disappointment with treatment effects. The focus group interviews undeniably demonstrated that patients perceived other outcomes relevant in addition. Especially after experiencing several depressive episodes, patients adapt their treatment goals. Our results indicate that we may need to use different end-points when treating a first episode versus recurrent or chronic depression, so that treatment goals and success can reflect all stages of the disease.Reference Porter23 To inform more extensively on clinically relevant results, Rush et al proposed including a second end-point, such as daily functioning, in clinical trials in difficult-to-treat depression.Reference Rush, Thase and Dube24
As in other chronic diseases, patients have reported still experiencing a good quality of life despite the fact that they have not fully recovered or achieved remission, the so called ‘disability paradox’.Reference Albrecht and Devlieger25 Clinicians on the other hand, emphasised symptom reduction, treatment satisfaction and achievement of predetermined treatment goals, of which the latter are not generally taken into account in clinical trials either.
There is an ongoing debate whether the World Health Organization definition of health, described as ‘a state of complete physical, mental and social well-being and not merely the absence of disease or infirmity’26 still fits in a world where chronic diseases are highly prevalent. Newly proposed definitions suggest including the ability of people to develop strategies to cope with the disease while maintaining their perceived quality of life.Reference Huber, Knottnerus, Green, van der Horst, Jadad and Kromhout27 Learning to cope with depression and learning to function satisfactorily are in line with the proposed definitions of health, especially in chronic conditions.
By using qualitative methods, patients and clinicians can elaborate freely on what they consider important without being guided by suggestions from researchers, pre-defined questionnaires or clinical expectations. Our findings are in accordance with findings from the few previous studies that were performed on defining recovery from depression, especially regarding functioning in several domains and returning to one's usual level of functioning.Reference Zimmerman, McGlinchey, Posternak, Friedman, Attiullah and Boerescu5,Reference Johnson, Gunn and Kokanovic18–Reference Uebelacker, Battle, Friedman, Cardemil, Beevers and Miller20,Reference Obbarius, van Maasakkers, Baer, Clark, Crocker and de Beurs28 In addition, we noted that patients gradually modify their treatment goals after several reoccurrences, which, as far as we know, has not been described before. Interestingly, relevant outcomes related to different domains of functioning (such as physical, social) are also found to be important in other disorders (for example heart failure, diabetes, psychosis).Reference Kraai, Vermeulen, Hillege and Jaarsma29–Reference Morrow, Haidet, Skinner and Naik31
Limitations
In this study, a few factors should be taken into consideration. First, most of the interviewed patients had either chronical depression or had gone through several depressive episodes, which was a representation of the specialist mental healthcare treatment setting. In the Netherlands, specialist mental healthcare providers deliver in-patient, out-patient and community care treatment to patients with severe mental health problems. In contrast, patients with mild-to-moderate non-complicated psychiatric disorders receive treatment from a general practitioner with additional support from a specialised mental healthcare nurse. Alternatively, they may be referred to generalist mental healthcare. Patients with a single episode only were less represented. However, the patients in our study that experienced multiple episodes of depression mentioned the same relevant treatment outcomes for a first episode as patients who experienced a single episode only. Our study sample is likely to be representative for patients in our catchment area. The ratio of men to women diagnosed with depression is 2:3.
Second, most participants were of White ethnicity. Culture and ethnicity might have influenced treatment goals deemed relevant in depression care. Third, within the varied group of clinicians not all professions were equally well represented. However, we verified our results by grouping the type of clinicians and it appeared the results were very consistent between the professions. Therefore, we think that our sample of clinicians has provided representative themes, and data saturation was reached. Finally, a focus group interview with both patients and clinicians might have resulted in more in-depth discussions and interaction between both views. On the other hand, this could also have introduced social desirability bias in patients.
Implications
The results of this study are relevant for both research and clinical practice. Traditional end-points used in clinical trials only partly reflect factors that are deemed relevant to patients and clinicians alike. Several outcomes considered relevant by patients and clinicians, such as functioning and quality of life, would be relatively easy to incorporate in outcome assessment for depression, as validated questionnaires are available. In addition, by developing or adding questionnaires related to the achievement of personal goals and patient's treatment satisfaction, this field would move even further towards value-based care. The current study therefore provides relevant input for the further development of patient-reported outcome measures. The development of such self-reported instruments necessitates direct input from patients, which is still broadly disregarded in conventional research.Reference Trujols, Portella, Iraurgi, Campins, Sinol and de Los Cobos15,Reference Neale and Strang32
Likewise, goal-setting appears to be very important for treatment evaluation from both the patients’ and clinicians’ perspectives, but is barely used in randomised controlled trials to evaluate treatment. To ensure that patients and clinicians speak the same language, they should be more explicit in the common goals they have in mind for treatment. How patients and clinicians can improve common goal-setting warrants further investigation.
To conclude, in treatment-outcomes research, the focus should go beyond symptom resolution. Both patients and clinicians value other outcomes, especially in recurrent or chronic depression. When treatment efficacy and effectiveness are measured in terms of value-based end-points, treatment outcomes are more meaningful, and improvement in treatment success rates becomes achievable.
Supplementary material
Supplementary material is available online at https://doi.org/10.1192/bjo.2020.27.
Data availability
As the participant consent for the collection of data did not explicitly or implicitly include details of sharing their anonymised data, we are legally and ethically not allowed to upload the data for sharing. Data are from the IMPROVE study, whose researcher may be contacted at k.kan@umcg.nl.
Author contributions
Conception and design, or analysis and interpretation of data: conception and design: K.K., F.J. and M.A.A.; analysis: K.K., F.J. and M.A.A.; and interpretation of data: all authors. Drafting the article or revising it critically for important intellectual content: drafting: K.K., F.J. and M.A.A.; revising: all authors. Final approval of the version to be published: all authors.
Funding
This work was supported by an unrestricted grant from Stichting De Friesland, Leeuwarden, the Netherlands (grant number DS29). The mission of Stichting De Friesland is to invest its capital in initiatives and developments aiming to improve healthcare and public health in the broadest sense. Stichting De Friesland had no role in the study design; collection, analysis, and interpretation of the data; in report writing; nor in the decision to submit the article for publication.
Declaration of interest
None.
ICMJE forms are in the supplementary material, available online at https://doi.org/10.1192/bjo.2020.27.
Open access
eLetters
No eLetters have been published for this article.