Introduction
Common mental disorders – typically defined as comprising depression, anxiety and somatoform disorders (e.g. World Health Organisation 2009) – are a principal cause of reduced functioning and quality of life in old age. As the population ages, the wellbeing of older adults becomes more important in fiscal as well as moral terms. Yet recent research indicates older adults continue to receive a different response from services than their younger counterparts.
Advocates of older adults attribute this to institutionalised ageism (Age Concern 2008). The quantitative data they cite, however, have limited explanatory power. The data can identify numerical disparities, for example that older adults are significantly less likely to be referred to mental health services (Kendrick et al. Reference Kendrick, Dowrick, McBride, Howe, Clarke, Maisey, Moore and Smith2009), but in the absence of a detailed account of referrers’ decision-making, explanations are speculative.
This paper reports the findings of a thematic analysis of data collected when a secondary care mental health for older adults service enquired of GPs – their principal referrer – about their assessment and treatment of common mental disorder in older adults, their expectations and experience of referral to specialist services, and their perception of unmet need in this client group. The study found that GPs regarded these issues, though complex, as ‘bread and butter general practice’ and looked to secondary care services for consultancy and training rather than ongoing care management. It is anticipated that the findings will be of interest to clinicians and service managers seeking to optimise relationships and referral practices between primary and secondary care services for older adults.
Common mental disorder in older adults
The United Kingdom (UK) Enquiry into Mental Health and Wellbeing in Later Life (Age Concern 2007) reported that 2 per cent of community-living older adults in the UK meet clinical criteria for major depression and 11 per cent meet criteria for minor depression. A further 10–12 per cent experience depressive symptoms of sufficient severity to necessitate intervention. Clinical criteria for anxiety disorders are met by 2 per cent of community-living older adults, while a further 10–24 per cent experience anxiety symptoms. As in other age groups, mental disorder is associated with negative biopsychosocial outcomes, including impaired functioning (e.g. Penninx et al. Reference Penninx, Guralnik, Ferrucci, Simonsick, Deeg and Wallace1998), increased mortality (e.g. Schulz, Drayer and Rollman Reference Schulz, Drayer and Rollman2002) and increased incidence of elder abuse (e.g. O'Keeffe et al. Reference O'Keeffe, Hills, Doyle, McCreadie, Scholes, Constantine, Tinker, Manthorpe, Biggs and Erens2007).
In England, the National Institute for Clinical Excellence has recommended that ‘the full range of psychological interventions should be made available to older adults’ (2004: 55), while the Scottish Executive has committed to ‘increase the availability of evidence-based psychological therapies for all age groups’ (2006: 4). Nonetheless the Healthcare Commission (2009) recognised age discrimination in mental health service provision in the majority of trusts.
Hypothesised barriers typically fall into three groups: patient barriers – unwillingness or inability to present; physician barriers – unwillingness or inability to diagnose and treat; and systemic barriers – inadequate provision of age-appropriate services (e.g. Age Concern 2008; Callahan, Dittus and Tierney Reference Callahan, Dittus and Tierney1996). The first is difficult to verify; those who do not present are by definition innumerable. Investigation of systemic barriers is hindered by local and national variation in services. Consequently, physician barriers are most commonly investigated.
GP role
In attitude surveys, GPs endorse the statement that detection of mental disorder in older adults is their responsibility (96.7%; Gallo, Ryan and Ford Reference Gallo, Ryan and Ford1999) and that they should actively seek symptoms (88.1%; Orrell et al. Reference Orrell, Scurfield, Cloke and Renshaw2000). Detection rates however are low. Mitchell, Vaze and Rao (Reference Mitchell, Vaze and Rao2009) conducted a meta-analysis of 41 studies comparing GP diagnosis of depression with diagnosis by clinical interview, eight were of older adults. Sensitivity and specificity were low – typically a GP with 100 clients of whom 20 were depressed would correctly identify only 10 of the 20 depressed patients, and correctly reassure only 65 of the 80 who were not – but were no worse for older adults than other age groups. Qualitative studies suggest that case-note review methods may underestimate awareness; GPs’ descriptions of consultations suggest that they are usually conscious of and ‘mentally note’ mental health issues in their patients, but may elect not to record or discuss such issues due to concerns about ‘labelling’ patients or medicalising social problems (e.g. Burroughs et al. Reference Burroughs, Lovell, Morley, Baldwin, Burns and Chew-Graham2006; Dew et al. Reference Dew, Dowell, McLeod, Collings and Bushnell2005).
Menchetti et al. (Reference Menchetti, Murri, Bertakis, Bortolottis and Berardi2009) found, perhaps unsurprisingly, that detection of depression in patients of all ages is over twice as likely where they present with a psychological complaint, but older adults tend not to do so. The Office of National Statistics (2003) found 90 per cent of older adults with a mental disorder had visited their GP in the preceding year with physical symptoms, but only 30 per cent with psychological symptoms. Again though the one-dimensional explanation that GPs are simply ‘missing’ symptoms of mental disorder is challenged by the qualitative literature, which suggests that GPs and older patients both recognise these, but prioritise physical health conditions in the setting of time- and resource-limited consultations (Drennan et al. Reference Drennan, Walters, Lenihan, Cohen, Myerson and Iliffe2007).
Common mental disorder in older adults is also considered to be under-treated (Chew-Graham, Baldwin and Burns Reference Chew-Graham, Baldwin and Burns2004). Callahan, Dittus and Tierney (Reference Callahan, Dittus and Tierney1996) found fewer than 60 per cent of patients rated as likely to be depressed had commenced any treatment plan within six months. Dearman et al. (Reference Dearman, Waheed, Nathoo and Baldwin2006) found 90 per cent of older adults diagnosed with depression were treated in primary care alone, one-third without medication. Inadequate training (e.g. only 70% of GPs felt their training was ‘adequate’; Orrell et al. Reference Orrell, Scurfield, Cloke and Renshaw2000) or lack of knowledge regarding treatment efficacy (e.g. 30.9% of non-psychiatrists believed treatments were ineffective for the elderly; Liu, Lu and Lee Reference Liu, Lu and Lee2008) are typically cited explanations. However, Alvidrez and Arean (Reference Alvidrez and Arean2002), surveying United States physicians’ knowledge about, experience of and attitudes to late-life depression, found physician attitudes were most predictive of willingness to refer for psychotherapy. Arean et al. (Reference Arean, Alvidrez, Feldman, Tong and Shermer2003) replicated these results with reference to anti-depressant prescription. Such findings support the suggestion that ‘ageist’ attitudes underlie neglect of this patient group.
The ageist attitude is typically summarised as ‘depression in the elderly is more fleeting or more “understandable” and therefore possibly not deserving of treatment’ (Schwenk Reference Schwenk2002: 158). Yet the statement is not wholly supported by empirical evidence. The former clause is; 77 per cent of family physicians endorsed the statement that ‘given the losses and chronic illnesses that older patients experience, depression is understandable’ (Gallo, Ryan and Ford Reference Gallo, Ryan and Ford1999: 251), while 58 per cent of non-psychiatric physicians attribute depression in primary care (at all ages) to recent misfortunes (Liu, Lu and Lee Reference Liu, Lu and Lee2008). Given that psycho-social risk factors for depression (social isolation, bereavement, poverty, etc.) are more prevalent in the elderly (e.g. Bruce Reference Bruce2002), this view is arguably valid. The latter clause though does not logically follow from the former, and is not empirically supported. Examinations of intention to treat, whether naturalistic (Callahan, Dittus and Tierney Reference Callahan, Dittus and Tierney1996) or vignette-based (Alvidrez and Arean Reference Alvidrez and Arean2002), show the majority of physicians plan some intervention, and it is premature to conclude that barriers to implementation are unfounded ageism.
Murray et al. (Reference Murray, Banerjee, Byng, Tylee, Bhugra and MacDonald2006) interviewed 30 primary care staff about their perceptions of depression in older people. Interviewees perceived depressed patients as burdened by gender and ethnic stereotypes, and unaware of the pathology and treatability of depression. Most interviewees took a psycho-social view of depression as ‘an understandable reaction to distressing circumstances’ (Murray et al. Reference Murray, Banerjee, Byng, Tylee, Bhugra and MacDonald2006: 1367), and were ambivalent about offering a medical solution to a social problem. Participants were not asked whether their views would differ were the patient group younger, but all were clear that depression should not be viewed as a normal aspect of ageing. Such findings suggest there is more to the ‘understandability phenomenon’ than the casual dismissal of distressed older adults by nihilistic GPs, yet the likelihood of self-selected participants and the possibility of socially desirable responding mean replication and elaboration is essential.
Secondary care role
Secondary care services, whether for adult mental health generally or specifically for older adults, can provide specialist assessment and treatment and facilitate access to home-based or day-centre support, among other services. The evidence suggests that contact with such services improves outcomes (Royal College of Psychiatrists 2006). However, older adults are much less likely than younger adults to be referred to such services (Age Concern 2007; Kendrick et al. Reference Kendrick, Dowrick, McBride, Howe, Clarke, Maisey, Moore and Smith2009) and development of specialist mental health services for older adults in the UK has not kept pace with those for younger adults (Healthcare Commission 2009).
Studies, whether qualitative or quantitative, have tended to focus on diagnosis and management within primary care, and not on the relationship and referral practices between primary and secondary care. Where this has been touched upon – for example, Burroughs et al. (2006: 374) interviewing primary care practitioners remarked on their perception that ‘the mental health service won't see anyone who hasn't got severe and enduring mental illness’ – it is not fully explored. This is felt by the authors to be a significant gap in the literature.
The research context
The study was undertaken as part of the first author's doctoral training in clinical psychology, during a six-month placement with NHS Borders’ Mental Health for Older Adults Team (MHOAT). MHOAT is a consultant-led, secondary care community mental health team, providing psychiatric and psychological assessment and treatment to adults over 70 in a predominantly rural area with a total population of 116,000. What is known about older adults with mental disorder in the NHS Borders area is consistent with the wider literature. Extrapolating from 2011 census data (Scotland's Census Results Online 2013) and using figures provided by Age Concern (2007), approximately 2,000 older adults in the Borders would meet clinical criteria for depression, and over 500 criteria for anxiety disorder at any point. GPs make around 400 referrals annually to MHOAT, accounting for approximately 80 per cent of total referrals to the team. Of these, fewer than 20 per cent are patients whose primary problem is defined by MHOAT as functional mental illness; the common mental disorders as described, plus schizophrenia and delusional disorders, and disorders of behaviour or personality. Approximately 80 per cent of referrals annually are for organic mental illness; the dementias, diseases of the nervous system and disorders due to substance abuse (unpublished data). Referrals of over-seventies to Psychological Services or the area's self-help service are negligible.
At the time the study was undertaken, there was a perception within the team that the Scottish Government's recent commitment to improving diagnosis and management of dementia (Scottish Executive 2006) was monopolising team resources, to the detriment of services provided to patients with common mental disorders. Concerns were raised that GPs were discouraged from referring such patients to MHOAT due to dissatisfaction with the quality of patient care.
Empirical data to support or refute this perception were incomplete. It was proposed that a qualitative exploration of GPs’ assessment and treatment of common mental disorders in this patient group, and their expectations and experience of referral to MHOAT, would be a useful means of identifying – and if necessary clarifying – referrer concerns, with a view to service adaptation.
Method
Recruitment and procedure
An outline of the proposed research and an invitation to participate in a practice-based small group interview to last one hour was issued to every GP in the NHS Borders South Team area via email. This was a convenience sample, and no inclusion or exclusion criteria were applied. Invitation materials summarised the data regarding referrals for common mental disorder to MHOAT, stated that MHOAT were interested in the factors influencing the decision to refer or not, and listed the proposed interview questions (see Figure 1). Written consent for audio recording and transcription of their interview was obtained from all participants. To maximise participant anonymity no demographic data were recorded.
Figure 1. Prompt questions used in group interview. Note: MHOAT: Mental Health for Older Adults Team.
Small group interviews took place in participating GPs’ practices, and were facilitated by the researcher and a Consultant Psychiatrist from MHOAT. Interviews were semi-structured; a series of ten ‘prompt’ questions (see Figure 1) were referred to, but participants were encouraged to be expansive on points arising. Nine GPs, three male and six female, participated in three small group interviews (N=5, N=2 and N=2), generating 151 minutes of digital audio recording. Interviews were transcribed according to guidelines from McLellan, MacQueen and Neidig (Reference McLellan, MacQueen and Neidig2003) and transcripts were subject to Thematic Analysis.
Ethical considerations
A research proposal was submitted to the Clinical Governance Facilitator of the NHS Trust, and to a representative of the University of Edinburgh, School of Health and Social Science's Research Ethics Committee. The project was deemed to be a service evaluation and no complex ethical issues were identified. However, the interviewers were known professionally by participants and were acting as representatives of a service provider. Therefore despite encouragement to be candid, it is possible that considerations such as a desire to preserve good working relationships consciously or unconsciously influenced participants’ responses. An opportunity to submit follow-up comments anonymously was presented but not taken by any participant.
Thematic Analysis
Thematic Analysis is a process of recognising and categorising patterns of meaning in a body of data. Described as a ‘foundational method’ (Braun and Clarke Reference Braun and Clarke2006: 78), Thematic Analysis adopts a systematic approach to the identification of patterns within the data, but strives to retain the context of each data unit, allowing hierarchical or other relationships between themes to be considered.
Unlike some qualitative methodologies, for example Interpretive Phenomenological Analysis (e.g. Smith and Osborn Reference Smith, Osborn and Smith2008), Thematic Analysis is not driven by a theoretical approach to data. As such it can be used flexibly. Users can consider data at the manifest level, where phrases are taken at face value and the code is a description of what is said, or at a latent level, where codes represent the researcher's interpretation of what is meant (Braun and Clarke Reference Braun and Clarke2006). The process of defining initial codes may be deductive; codes are derived from existing literature and test the data's concordance with prior findings or theories. Alternatively it may be inductive; original codes are suggested by review of the data itself. The latter approach is particularly useful in relatively new areas of research where appropriate theoretical frameworks are not yet established (Joffe and Yardley Reference Joffe, Yardley, Marks and Yardley2004). Thematic Analysis was selected for use in the present study for this reason.
Thematic Analysis was conducted according to guidelines laid out by Braun and Clarke (Reference Braun and Clarke2006). Analysis was carried out by the first author, with graduate-level experience of qualitative research methods, under the ongoing supervision of the third author. In the first phase, interviews were transcribed, and annotated where research-relevant comments were made. These annotations formed the basis of the second phase: coding. Coding was at the latent level, and inductive. Synonymous annotations were synthesised to produce 91 initial ‘codes’: labels summarising the content of a section of data, permitting organisation. Returning to the transcripts, every utterance was scrutinised for the presence of every code.
The third phase was to search for groups of codes which might coalesce into a theme. Groups were sought which maximised the extent to which all the codes within a theme reflect similar ideas while maintaining clear distinctions between themes. Some idiosyncratic codes were discarded from analysis at this point. Re-examination of transcripts, coding and themes occurred continuously in phases three and four. The fourth phase considered relationships between themes, using a thematic map to illustrate groupings of codes and connections. Multiple iterations were completed before the selection of the thematic map which best represented the data as interpreted by the researcher. Themes were then named and described. Participants were invited to review the analysis, no comments or corrections were received.
Results
Thematic analysis of transcripts generated seven themes, of which one had two subthemes. Themes are described below, and their relationships are illustrated in Figure 2.
Figure 2. Thematic map indicating relationships and included codes. Notes: GP: general practitioner. MHOAT: Mental Health for Older Adults Team.
Cohort effects
All but one respondent explicitly noted perceived differences between older and younger adults experiencing mental disorder, often in beliefs and attitudes. Older people were reported to attribute symptoms differently, and to have more rigid and strongly held beliefs about stigma, the desirability of coping unsupported, and the implications of failure to do so.
It's overcoming stigma and I think for an older generation of people sometimes that's harder than it is maybe for the younger generation who see that actually maybe depression, anxiety are no different to having asthma, epilepsy or diabetes. Whereas I think it's still seen sometimes as a sign of weakness or something if you're from an older generation. (GP9)
Elderly people are probably more accepting, more stoic. They say it's ‘cause I'm old and I'm lonely, rather than because I have a depressive illness. (GP8)
The doctor–patient relationship was described as different with older adults, again due to patient attitudes. GPs believed the quality of the relationship was more important to older patients, but that – perhaps as a result – their personal influence was greater.
Sometimes if someone's younger they kind of nip in, they don't mind which dentist or GP they see. Whereas historically the older generation would have THE family doctor, the same person, so could be a little wary about opening up on a first occasion. (GP9)
If I suggest it, they maybe will go along because I suggest it. Again that's the age and culture. (GP6)
Declining physical health was also felt to distinguish older adults, impacting by restricting contact, and by overshadowing mental health complaints. Most respondents noted the tendency of older adults to present with physical symptoms, and to be resistant to psychological explanations.
The more difficult group are probably people who are at home, or housebound for physical reasons. And it's probably easy for those people to not really be as aware of the possibility of depression. And we're focusing on their physical illnesses. (GP7)
[I saw] a very anxious and rather low chap presenting with a lot of physical symptoms. He wasn't absolutely convinced that it was because he was anxious and possibly depressed that the problem was arising. (GP8)
GP role
All respondents indicated that responsibility for detection and initiation of treatment of mental disorder lay with the GP. Four made explicit statements to that effect. While only four expressed confidence in managing mental disorder in the primary care setting, no GPs expressed apprehension or concerns about doing so.
Mood is something we see a lot of. You meet disturbance throughout all age groups and therefore you've got to have a degree of competence with it to survive within general practice. (GP6)
Depression is absolutely bread and butter general practice, and we jolly well ought to be quite good at it. It's a slightly different ballpark in the elderly for a number of reasons that we've covered. I think the elderly probably present in a slightly different way, but nonetheless I don't think it's beyond our capabilities. (GP8)
The doctor–patient relationship was described as central to the GPs’ role. Its development, often over many years, was reported to facilitate disclosure, to permit the GP a measure of directness in discussion, and to lend influence to their advice. The relationship itself was described as having therapeutic value.
That probably is why for some people with mild anxiety or depression it does remain in primary care, because the key to actually managing that without prescription of medication, is the therapeutic relationship between GP and patient, not a referral to a consultant who they've not met who they're not going to open up to. (GP9)
A clear conflict was identified between the development of such relationships, and pressures of work. The majority of respondents expressed concern about the impact of limited resources, specifically time, on quality of care.
Assessment
Respondents identified several factors taken into account when assessing older patients. Again the doctor–patient relationship was described as pivotal, with knowledge of patients’ circumstances and typical demeanour offering important cues for in-depth questioning.
We've also known the patients for many, many years. You can tend to pick out some subtle changes. (GP2)
If you've known someone for a long period of time, or you've got the beauty of referring back to their notes, you can see things coincide with, an anniversary of a bereavement or something even very historic. (GP9)
The process of becoming embedded in the local community and developing relationships meant others with personal or professional relationships with patients felt able to share concerns. All but two respondents commented on the importance of these accounts in prompting and enhancing assessment.
We get people coming in, relatives of older people saying so and so is not looking after themselves and they're not going out, that sort of thing. So there is to some extent a knowledge of the patient that you're dealing with. That's not a measurable thing, that's more a sort of hunch. (GP1)
Beyond this benefit of longer relationships with older patients, respondents indicated that their assessment was similar to that offered to younger patients.
I probably don't do it that much differently from a younger person, except more awareness that physical symptoms are more often in presentation in older people with depression, I keep that in mind. (GP7)
Routine assessment for mental disorder, particularly the administration of standardised measures such as Patient Health Questionnaire (PHQ-9) as part of general health screens, was described as being less useful than individualised, opportunistic assessment prompted by detailed knowledge. Nonetheless several respondents highlighted their utility in augmenting assessment.
In one instance in particular it was interesting to show the lady, who was over the seventy age group, how much she'd improved. (GP7)
Something like a PHQ-9 I find quite often you'll find that they suddenly, they'll fill up [become tearful] at one of the questions and you think ah hah, we've found something here. (GP8)
Overall the sense was that assessment of older patients, and younger, was a holistic process, integrating information from various sources, many of which were dependent on a confiding relationship.
Decision-making
As with assessment, clinical decision-making was described as multi-factorial and proceeding on a patient-by-patient basis. The principal decision discussed was whether or not to involve specialist services at an early point. Two frequently cited indicators of early involvement were severity – in particular presence of suicidal ideation, atypical presentation and differential diagnosis of organic brain disease. Comorbid physical health concerns, particularly concerns about concurrent medication, may also prompt referral. In the absence of these indicators, secondary care would typically only be considered where GP-administered treatments had proven ineffective.
If it was something very, very severe, or with suicidal intent or something, then I think absolutely it would be the first step. But my feeling would be in the first instance to utilise my relationship with the patient, and to try formulary advised medication. (GP9)
I wouldn't think of referring them to you without having tried two or three different antidepressants. Unless there was some, often I suppose dementia or memory problems may play a part and I'm not sure about the diagnosis. (GP7)
In addition to patient-specific factors, general comments were made about the suitability, acceptability and efficacy of treatments, particularly psychological/counselling approaches, for the older patient group. While some respondents judged that older clients were more likely to opt for a medical intervention, most drew no distinction between older and younger clients with regard to treatment preferences or response. The only systemic factor, cited by four respondents, was that of waiting times.
Some people think you're fobbing them off, if you don't give them a pill. (GP3)
We have got the option of offering psychological therapy. In my limited experience with the number of older people that I've seen with depression since that started, most of them are not interested in that. (GP7)
Intervention
Potential interventions identified fell into three categories: pharmacological, psychological or social. Typically GPs stressed that more than one approach might be considered or implemented. Where practice-specific treatment options were available – e.g. self-help coaches – these were highly valued.
The broad approaches would be pharmacology, psychology and as a last resort referral to the specialist services. But there's also the social side of things. (GP1)
There's a long wait for formal psychology, so lifestyle coach, social work, SSRIs [Selective Serotonin Reuptake Inhibitors]. Usually in conjunction. (GP3)
Importantly, respondents were clear that GP contact without prescription or referral was often a treatment in its own right. More than half of respondents made this point explicitly.
When we're looking at what the options are managing depression in older adults, well, actually for some people it really is that supportive talking therapy. Not necessarily needing to be provided by clinical psychology, but something that can be done through the nurturing relationship hopefully between doctor and patient. (GP9)
Often when I've done a consultation I think well what did I do in that, and really I didn't do anything apart from just listen for a few minutes and nod in the right places and agree with them on certain things. That's actually sometimes all that's required. (GP1)
Although the core themes of Assessment, Decision-making and Intervention are presented linearly in the thematic map, in fact they are recursive. Respondents described a process of continuously incorporating new information, such as treatment response, into assessment, and revising clinical judgements accordingly.
Role of secondary care
The central role of secondary care services in the context of mental disorder was depicted as supporting GPs in this ongoing process. Every respondent stated that they principally seek from secondary care consultation, advice and training, not direct input. This arrangement serves the dual purposes of enhancing GP competence and conserving secondary care resources.
You're the consultant in charge of a team and we consult you for advice and often the advice is what we're looking for. (GP2)
I would hope that MHOAT would continue to provide that sort of support and guidance to GPs because we can't keep up to date with all these changes [to mental health legislation]. (GP9)
Although in general GPs expressed satisfaction in the relationship between primary and secondary care, some unfulfilled expectations were identified. These accorded with the envisaged role of ‘consultant’; a speedy and clear response to enquiries, and – given that input was required only in extremis – a rapid response to referral. No respondent indicated that inadequate service by MHOAT impacted the decision to refer.
More than a health issue
An over-arching theme, repeatedly referred to by all participants, was mental disorder as more than a health issue. Rather it was seen as a problem whose origins, and solutions, lay in wider social functioning. The theme was divisible into two sub-themes: social problems and social solutions.
Social problems
Respondents identified social problems, general to all age groups but more prevalent in older adulthood, e.g. poverty, isolation and bereavement, as factors in mental disorder. One respondent commented on changing social structures and subsequently changing demands on older adults. Stigma, due to lack of public awareness, was viewed as adding to the burden of mental disorder.
Most of them are so isolated it's quite terrible actually. You know we all want to live in our own house, and then when actually everyone dies off around us, we're left in some pile, this hovel or whatever it is. And you can't get anywhere and things like that. I do to some extent feel they're abandoned. (GP6)
There was a lack of clarity among respondents as to whether a presentation of anxiety and depression in such circumstances warranted clinical diagnosis or medical intervention. While most endorsed the view that mental disorder in these cases was understandable, they distinguished between understandable and acceptable. The question was not whether mental disorder in older adults should be addressed, but whether medical services were the optimum means of doing so.
This group of adults that aren't getting treated, who say they're depressed but not getting treatment? Those are maybe just the sad lonely people who haven't got clinical depression but they need something more from society. (GP7)
Most of the time they are fed up, and they can't get out and you think ‘Gosh, if it was me in this situation’ and so do you … offer them antidepressants? (GP7)
Social solutions
Respondents highlighted several potential social solutions, for example provision of better day centre-type facilities, and greater utilisation of voluntary services such as Red Cross Befriending. There was recognition though, that such solutions might carry their own stigma, were limited in scope and availability, and were a poor substitute for the ‘organic’ social support of a well-integrated community network.
You know we're daft if we're not using people like Red Cross, BVCCF [Borders Voluntary Community Care Forum] and the neighbourhood link workers. I think we need to be smarter about involving resources that are already out there. (GP9)
Building a big community centre and saying you can go there, maybe it wouldn't solve anything. Maybe they'd just be old, empty buildings with the soup going cold. (GP6)
Some of that [wellbeing] is about just feeling part of your community, and being naturally supported by your community and the supports it can offer. Which might be as simple as someone popping in for a cup of tea. (GP9)
Raising public awareness of mental disorder in old age, both to challenge stigma and encourage help-seeking, was cited by six respondents as the social solution which might offer the greatest benefit and be most easily achieved. An age-appropriate format was a key concern.
[Information sharing] in an appropriate manner in settings where people socialise. Whether it be day centres, day hospitals, whether it be in coffee groups and lunch groups, whether it be a coffee morning in the Legion [a social club], whether it be in shops, in public libraries, on the bus, in adverts. Sexual health have done it, why can't the older adult team do it? (GP9)
Discussion
This qualitative study challenges the assumptions of existing quantitative literature in several ways. Firstly, respondents’ description of assessment as an ongoing process integrating detailed patient knowledge and accounts of significant others in the context of a long-established relationship, calls into question the use of any brief self-report measure as the ‘gold-standard’ by which detection rates are established. GPs have been charged with under-diagnosing mental disorder because their ‘hit-rate’ does not match that of questionnaires. Yet the picture painted in this analysis is not of inattention, but rather of doctors attuned to their patients in a manner which standardised assessment might struggle to replicate. While a questionnaire may define a patient as ‘depressed’ after one visit, a GP's assessment may still be in progress, or they may have the personal knowledge to correctly attribute symptoms to a response to loss. This reflects the existing literature suggesting that GPs may be mindful of, and even manage, emotional distress without explicitly diagnosing or treating (Dew et al. Reference Dew, Dowell, McLeod, Collings and Bushnell2005).
GPs have also been charged with failure to treat older adults due to belief that mental disorder in old age is understandable and therefore unworthy of intervention. This study suggests that apparent non-treatment is often a conscious decision to utilise the therapeutic nature of the doctor–patient relationship, according to patient preference. Again the picture is one of careful consideration not casual disregard. With reference to GPs’ beliefs, like the quantitative data this study confirms the first clause, but not the last. Respondents did attribute a great deal of mental disorder to financial, familial or other social factors, and did query whether either medication or psychological therapy was a legitimate response in this circumstance. They did not, however, equate understandability with acceptability. Their view was not that mental disorder in old age is not a legitimate problem, but that it is not entirely a medical problem.
Participants clearly located barriers at the patient level; older adults either do not present at all, or their psychological symptoms are masked by their physical ones. This may be viewed as self-exoneration on the part of GPs, yet respondents openly acknowledged concern that some patients slip the net, and did not take up the opportunity to attribute blame to secondary care services. They also noted the responsibility of the health service to raise awareness of the issue, ignorance being characterised as a principal barrier.
At a local level, concern that low referral rates reflected GP dissatisfaction with secondary care capacity or performance was lessened. GPs emphasised the value of specialist services as a source of advice and training, and gave a clear account of the circumstances under which they would refer. But they also described a sense of competence and confidence in managing the bulk of cases in the primary care setting. This finding contradicts the earlier suggestion that GPs feel they lack skills and training (Burroughs et al. Reference Burroughs, Lovell, Morley, Baldwin, Burns and Chew-Graham2006; Orrell et al. Reference Orrell, Scurfield, Cloke and Renshaw2000). This may be due to familiarity of the issues to the participant group; the Scottish Borders has a high average population age, and a tradition of long relationships with GPs, or may be due to national variation in training, or lack of detail and context in survey responses.
One striking feature of this analysis is its correspondence to that of Murray et al. (Reference Murray, Banerjee, Byng, Tylee, Bhugra and MacDonald2006). The views of an English, urban participant group – particularly with regard to the beliefs and attitudes of the patient group, the role of social factors in mental disorder, and the propriety of medicalising a perceived social problem – have been closely replicated in a rural Scottish setting. These themes were also observed in a GP interview study (with reference to patients of all ages) in New Zealand, a setting geographically – though arguably not culturally – distant from the UK (Dew et al. Reference Dew, Dowell, McLeod, Collings and Bushnell2005). The parallels lend weight to the validity of each of the studies.
The findings of this study will primarily be of interest in national settings where GPs, family physicians or similar serve as the first point of contact for patients with physical and mental health conditions, and as gatekeepers to secondary services. If the perceived needs of referrers generalise, organising secondary services to complement and support primary care capabilities rather than taking over patient care (in the case of common mental disorders) could optimise resource use. It would also be interesting to compare the perceived roles of specialist services in health-care systems where patients have direct access to these. The overarching theme, that of mental disorder as more than a medical problem, is applicable in any country or culture, and international variation – for example in the social status of older adults, urbanisation, or availability of health and social care – may point the way to social solutions.
The study has limitations. Joffe and Yardley (Reference Joffe, Yardley, Marks and Yardley2004) suggested no Thematic Analysis can be truly inductive, since the researcher has predetermined the areas of interest. In this case a relationship between the prompt questions devised by the researcher, and the core themes of assessment, decision-making and intervention can be plainly seen. However, coding and grouping of codes were carried out in a systematic fashion and generated results unanticipated by the researcher. Feedback was sought from participants to ratify the proposed thematic map.
The small sample size of this study was sufficient to allay the initiating concerns of the team regarding levels of service anticipated and experienced. However, there is the possibility that only those GPs favourably disposed towards MHOAT would be willing to participate in their research. To exclude a self-selection bias and to provide a comprehensive and academically rigorous analysis of GP views would require recruitment of a much larger sample.
Similarly, no personal or professional participant information was recorded, making it impossible to demonstrate sample representativeness, or to compare the views of, for example, the very experienced versus the recently qualified, or those with a special interest in mental health to those without. To some extent the consistency of views offered in interviews suggests this latter would be a fruitless exercise were it possible. However, quantitative studies (e.g. Alvidrez and Arean Reference Alvidrez and Arean2002; Arean et al. Reference Arean, Alvidrez, Feldman, Tong and Shermer2003) have found a relationship between demographic and training variables, and attitudes and actions, and it would be instructive to attempt to replicate these findings using a qualitative methodology. The participant numbers required to make such comparisons meaningful was beyond the scope of this project. However, the failure even to record this type of information precludes the possibility of incorporating these interviews into any future data set which might consider these comparisons, and is a significant failing.
This study has not sought, nor has it found, a definitive answer to how GPs manage or would like to manage mental disorder in older adults. At a wider level what it has done is highlight the incongruence between assumptions about GPs attitudes and intentions based on categorical data, and their own accounts. At a local level it has made explicit GPs’ perception of the needs of older patients with regard to mental health. This finding will help to focus scarce secondary care resources on areas of perceived need: consultation and training for primary care.
Acknowledgement
This paper was submitted in partial fulfilment of the degree of Doctorate in Clinical Psychology at the University of Edinburgh.
