Introduction
Depression in young people is common (Cooper and Goodyer, Reference Cooper and Goodyer1993) and, while much of it appears to be self-limiting, there is also evidence of high levels of persistence and recurrence in a substantial minority (Lewinsohn et al., Reference Lewinsohn, Clarke, Seeley and Rhode1994). Depression in this age group is frequently associated with functional impairment, and can be severe and prolonged, often not resolving over eighteen months (Goodman et al., Reference Goodman, Ford and Meltzer2002). There are associations between depression in the teens and affective disorder (Pine et al., Reference Pine, Cohen and Cohen1999; Harrington, Reference Harrington2001) and impaired psychosocial functioning in early adulthood (Rao et al., Reference Rao, Hammen and Daley1999; Asarnow et al., Reference Asarnow, Jaycox, Duan, LaBorde, Rea and Tang2005). Even young people with low levels of depressive symptoms and associated impairment – the so-called ‘subsyndromal depression’ – have been shown to have impaired outcomes (Angold et al., Reference Angold, Costello, Farmer, Burns and Erkanli1999). There is some evidence that intervention can reduce psychological morbidity (Walker et al., Reference Walker, Townsend, Oakley, Donovan, Smith, Hurst, Bell and Marshall2002) and it is therefore possible that early intervention might alter the experience of mental ill-health in later life (Jacobson et al., Reference Jacobson, Churchill, Donovan, Garralda and Fay2002).
The NICE guidelines on depression in children and young people (NICE, 2005) argue for the need for enhanced detection and risk profiling in community settings. While there has been considerable emphasis on primary care training to enhance identification and management of adult psychiatric disorder, little work has addressed psychopathology in children and young people and there is not yet direct evidence on how the NICE recommendations could be best achieved.
General practitioners (GPs) may be seen as being best placed to undertake detection and risk profiling tasks. The majority of registered young people consult their GP each year. Most attend with physical symptoms; behavioural or emotional complaints account for only 2% of presentations (Kramer and Garralda, Reference Kramer and Garralda1998) although the prevalence of depressive disorders among young attenders appears to be higher. The only study carried out in the UK using research criteria for diagnosis of depression indicated up to 20% prevalence among young people consulting their GP (Kramer and Garralda, Reference Kramer and Garralda1998). Rates in urban and suburban populations, using questionnaires rather than diagnostic interviews, were similar (Yates et al., Reference Yates, Kramer and Garralda2004).
There are a number of reasons why GPs might not readily adopt this new role. Although GPs believe that depressive presentations among teenagers are becoming increasingly common (Vandana and Ambelas, Reference Vandana and Ambelas2004) and do identify most of those with severe psychological morbidity, they also fail to identify the majority of those with depressive disorders (Angold et al., Reference Angold, Costello, Farmer, Burns and Erkanli1999). There is some evidence that GPs spend less time in consultation with young people (Jacobson et al., Reference Jacobson, Churchill, Donovan, Garralda and Fay2002) and many practitioners feel that teenagers are hard to communicate with and worry about over-medicalizing their lives (MacFarlane and McPherson, Reference MacFarlane and McPherson1995; Iliffe et al., Reference Iliffe, Gledhill, da Cunha, Kramer and Garralda2004). Many young people with psychological disorders do not take up psychological therapies in any sustained way (Westman and Garralda, Reference Westman and Garralda1996), although such therapies appear to be effective in reducing symptoms and case prevalence in those who do engage (Harrington, Reference Harrington1998).
However, educational interventions can produce sustained positive changes in consultations (Sanci et al., Reference Sanci, Coffey, Patton and Bowes2005), and training GP registrars in how to identify and respond to mental health problems in young people increases identification of psychiatric disorders among general practice attenders (Bernard et al., Reference Bernard, Garralda, Hughes and Tylee1999). We have shown that a therapeutic intervention significantly increases case identification among GPs in one practice, where it was used regularly within routine consultations, albeit with considerable variation between doctors (Gledhill et al., Reference Gledhill, Kramer, Iliffe and Garralda2003).
Although there is a case for educational interventions that enhance diagnostic and therapeutic skills in GPs managing depression in young people, it is unclear how such interventions should be constructed and used. There are many studies on interventions aiming to improve the primary care identification and management of depressive and other psychiatric disorders in adults (Cooper, Reference Cooper2003), but surprisingly few focussed on teenage depression The studies in adults have had mixed or negative results, perhaps because of reliance on anti-depressant medication at the expense of active patient involvement (Rost et al., Reference Rost, Nutting, Smith, Elliott and Dickinson2002), or on the use of psychological interventions that were too complex for use in primary care (King et al., Reference King, Davidson, Taylor, Haines, Sharp and Turner2002). Although training programmes in the detection and management of depression may improve the skills of GPs, it does not necessarily translate into better outcomes for patients, or benefits for the health service (Gask et al., Reference Gask, Dowrick, Dixon, Sutton, Perry, Torgerson and Usherwood2004). However, multifaceted programmes that integrate improvements in detection, treatment and follow-up have been shown to be effective in depressed adults (Katon et al., Reference Katon, Von, Lin, Simon, Walker and Unutzer1999; Pignone et al., Reference Pignone, Gaynes, Rushton, Burchell, Orleans and Mulrow2002; Gilbody et al., Reference Gilbody, Whitty, Grimshaw and Thomas2003; WHO, 2004).
More attention to the process of the doctor–patient consultation has also been advocated as a way of unravelling the problems in enhancing skills in primary care(Cooper, Reference Cooper2003). This paper describes GPs’ perceptions of the opportunities and difficulties of working with teenagers, and recognizing and responding to depression in this age group.
Methods
Data was collected during interviews with GPs in a single, university-linked inner London group practice with a registered population of 14 000 patients. The practice was involved in a developmental project on recognition of and response to depression in young people, and the interviews were carried out after an educational programme had been used in the practice, in order to gain feedback on the usefulness and applicability of the programme. The educational programme promoted the use of screening questions for psychological distress, followed by a brief assessment for symptoms of depression, and then an intervention using cognitive behaviour therapy and interpersonal therapy principles. GPs were trained in the intervention and asked to use it with all patients between thirteen and sixteen years of age who presented within a given six-month time period. A full account of this has been published elsewhere (Gledhill et al., Reference Gledhill, Kramer, Iliffe and Garralda2003). Face-to-face interviews were selected as the optimal method for eliciting insider perspectives and frameworks of meaning (Britten, Reference Britten1995), and were carried out by one experienced researcher with clinical experience in child and adolescent psychiatry (JG). All GPs who had been involved in testing the intervention participated in the interviews, which took place within the practice, and the same interview schedule was used for all. A semi-structured interview schedule was developed following a review of the literature on obstacles to recognition of depression in young people, and is shown in Box 1. Interviews lasted between 40 and 60 min, were tape-recorded and the tapes were transcribed verbatim, for analysis. Computer software was not used for this analysis.
Box 1 Themes and topics in the semi-structured interview schedule
Reflecting on the intervention that the GPs had tested what was their overall impression of the intervention’s usefulness? What were the most difficult and the most useful parts?
What were their views about conceptualizing certain adolescent difficulties as ‘psychiatric disorder’ and was it realistic to encourage GPs to distinguish between emotional upset and depression in teenagers?
Had the GPs views about teenagers and mental health in young people changed as a result of the study?
Had using the intervention increased workload or created any problem for everyday work? If so, what problems?
In what situations did it seem inappropriate to use the intervention? When did the GP not use the screening questions?
What effect did a parent’s presence have on the GPs use of the intervention?
What issues arise when psychological disorder is ‘unmasked’?
What did the teenagers seen by the GP think about the intervention, when it was used?
Did anything cause the GP discomfort in these consultations?
What was the value of the training course and materials, and how could they be improved? Were skills acquired that might be used in the future? If so, describe.
A grounded hermeneutic approach was used to understand the data obtained (Addison, Reference Addison1992). Analysis was carried out by close examination of the transcribed texts and identification of the meanings embedded within them, all transcripts being read and key items of content identified by all seven authors independently, prior to group discussion. Individual items of content were combined into themes, about which working hypotheses or interpretation could be generated. An analytic induction technique (King et al., Reference King, Davidson, Taylor, Haines, Sharp and Turner2002), was used with initial hypotheses being checked against empirical data, particularly for falsifying evidence. These interpretations were revised and reapplied to the data by the researchers individually and in group discussion, until agreement was reached about the meanings and relationships of interpretations and the researchers were satisfied that negative or deviant cases had been identified and incorporated into the analysis (Murphy et al., Reference Murphy, Dingwall, Greatbach, Parker and Watson1998). Through a process of progressive focussing (Rost et al., Reference Rost, Nutting, Smith, Elliott and Dickinson2002) on the research question the research team identified the over-arching themes arising from the interviews, and these are presented in this paper.
Findings on the implications for the acquisition of new skills have been published elsewhere (Iliffe et al., Reference Iliffe, Gledhill, da Cunha, Kramer and Garralda2004). This paper focuses on practitioners’ understanding of depression in young people.
Results
Nine GPs participated in the study – six women and three men.
Two over-arching themes arose from the transcripts. The first was the way GPs understood the use of primary care by teenagers, and the second was how GPs construct depression in teenagers as a problem on the boundary between pathology and normality. A number of sub-themes emerged within each over-arching theme, and these are marked in italics below.
Teenagers as service users
Teenagers are thought about as different from other patients. They are perceived as being relatively infrequent attenders, with a tendency to default from follow-up (see Box 2). This perceived low profile makes the introduction of a medical agenda during consultations feel risky for the practitioner.
Box 2 Teenagers as service users
‘…adolescents as group don’t present as much to general practice’.
‘Teenagers don’t come back to see me when I say so, that′s the thing’.
‘…it might put adolescents off coming in, if they hardly ever come and then they get confronted with that’.
‘As a group they tend to present in an emergency setting which is the least time we can offer them’.
‘The adolescents they kind of breeze in and breeze out and breeze in to somebody else another time’.
‘20% of adolescents came booked and 80% of the ones I saw came in an emergency surgery … .
‘Sometimes it was difficult if there was a parent there …… a couple of others where I felt if I’d been, if the adolescent had been on their own with me, then it would have been more comfortable ….. but it had been their decision for the parent to stay’.
‘… there were a couple of situations where a parent was present and said ‘Well I don’t, you know, he hasn’t got any problems, have you?’ …… and then they said ‘I’d like to stay’.
‘there was one where I screened and didn’t make an intervention… it was partly time, the consultation had gone on a long time and his mother was there and it was complicated, it was this complicated family …and erm, I thought I’d done enough for one day really’.
‘difficult if was a parent there, if the adolescent had been on their own with me then it would have been more comfortable’.
They use emergency clinics or appointments, and do not book appointments ahead. Part of the difference between teenagers and adults lies in their perceived impulsivity or inconsistency. There was no explanation or further discussion of this belief, which is seen as a fact of life, just how teenagers are (see Box 2).
The presence of parents could be problematic. The fact that teenagers were not necessarily separable from their parents could make consultations difficult, particularly if the objective was to identify depression (see Box 2).
Teenagers are complex. This means that practitioners have to make sometimes complex judgements about how far to proceed with questions about a teenager’s psychological state, which introduced another element of risk, since misjudgement could affect future consultations. Teenagers could be difficult to understand, and hard to communicate with, adding to the sense of risk. Lack of experience in consulting with teenagers created further concern, but there was also a sense that trust could be built between the doctor and their young patient (see Box 3).
Box 3 Teenagers as service users
‘A young 16 year old came twice about his spots and then a third time about his spots, and I gave all the cues of … you know, giving him the opportunity to talk about anything else … I think I gave him enough cues but actually being able to find some way of actually being able to hear, allow him to, you know, perhaps he needed to check me out 3 times, that’s why he kept coming back but you know to be able to discuss that’.
‘Sometimes it’s appropriate to force the pathway open … and sometimes it doesn’t feel appropriate, and I can’t tell you how I know when it is and when it isn’t too easily but there are situations when you know that you really need to go through this whether they like it or not. And there are situations where I feel perhaps that isn’t the right thing to do, and that I’ve got to use time to sort it, then I can. ‘This is how far you go this time’, and the next time you parcel up in another bit of information gathering, and the third time you come back to some other point. And that’s, I mean that’s the typical working style in general practice, you don’t do anything in a block necessarily’.
‘…it felt inappropriate with adolescents who consulted with a pretty straightforward problem who you… if you saw somebody, who you hadn’t attended for 3 years but was 16 and had sprained their ankle on Saturday playing football and needed a sick note, it just didn’t feel appropriate to be going in (to a discussion about their mood)’.
‘Other times adolescents deny their symptoms when actually they are depressed’.
‘Important to know that you are using the right language for them’.
‘Talking to teenagers is quite difficult, cause they are so monosyllabic… and to have sort of useful phrases to use, e.g. with suicidal ideas’.
‘It is important to know how they view their symptoms, they don’t understand their symptoms in the same way as an adult does’.
‘I think adolescent medicine is a kind of neglected side of your career, I don’t feel that you get much training and suddenly you’ve faced with adolescents…’.
‘I think they like having someone who listens to them, I think that’s rewarding for them’.
‘……it’s the whole business of trying to make sure that your consultations are perceived as helpful by the patient, and at times it’s very clear that people do want an opportunity to talk about, something other than the presenting problem but equally it’s clear that at times people just think you’re being invasive’.
Lack of time militates against thinking about depression in teenagers. Unravelling complex psychological problems requires time as well as judgement, and this time may not be easy to find (see Box 4).
Box 4 Teenagers as service users
‘May be I’m oversensitive about the time in psychiatry cases but I think it should take longer to answer, to sort of go through a psychiatric history in general, than if you unearth a particular problem for say an organic symptomatology’.
‘… hardly anybody sees me for 10 minutes in an emergency surgery’.
‘I think there is a dilemma that you do need to spend a lot of time working out whether it is a psychiatric disorder or not. Whether or not you need to label them and only label them if it’s a necessary intervention that is going to help them’.
‘Well it’s completely exhausted end-of-week burn-out feeling where you think ‘Oh hell, I’ll do this next time’ …it’s lack of time, pressure, … lack of energy, you know reserves running flat really. Cutting corners … minimising your expenditure so that you can focus on what you really are absolutely certain you must do’.
‘I think I found it very difficult to do the extra work in an emergency surgery and it would put me off unless somebody had some florid psychiatric symptoms to try and opportunistically pick up depression in an otherwise seemingly happy adolescent who’s come with an unrelated problem’.
‘I mean if they were booked cases, you know, it wasn’t so worrying, you know, kind of able to sort of spend… quite relaxed about it, say you know, let’s have whatever is coming. Or if it’s the emergency clinic, I think ‘I hope they say ‘No’ (to the initial screening question for depression).
‘I hoped that they don’t give positive answers so that then I wouldn’t have to spend more time with them’.
‘Not enough time, not only to identify the problem but also to make intervention at the same time’.
‘We need longer time to go through a psychiatric history in general’.
Teenagers and depression
Although some of the GPs seemed to have no great difficulty in recognizing depression in young people, we identified three other types of response. The first was surprise, the second was normalization of depressed mood and the third was a challenge to the scientific credentials of depression as a diagnostic category in young people.
Surprise. There was a sense that young people were almost invisible, and surprise – once they had become visible – how much they could be ‘troubled’ (see Box 5).
Box 5 Teenagers and depression
SURPRISE
‘…I hadn’t thought about teenagers specifically’ (when discussing depression as a problem of adolescence).
‘Before the study I hadn’t realised that teenagers could be as troubled as they were, and I have suddenly realises that actually a lot of them haven’t had the advantages that we did, and therefore you know, like the parenting for example, and so that was very useful to realise, to take a step back and look at you know what their experiences are compared to luckier people basically’.
NORMALIZING
‘I find it difficult sometimes that if a young person comes in and is a bit miserable, that it’s not, you know, that it could actually be a clinical depression rather than actually, you know, to do with, separating out normal adolescent upbringing and depression is difficult’.
‘I’m not sure that the way forward is through detecting things that are going to get better on their own (…) I’m slightly reluctant to medicalise normal experience (…) the normal ups and downs of adolescence’.
‘…I think adolescence is a particular time of flux and angst, where kids go through all sorts of changes, and what is normal for an adolescent I can’t say that I know what the sort of normal range of feelings are (…) they go through their own existential crisis (…) the time when somebody leaves school, makes decisions about what they’re going to do for the rest of their life, stops being a child and so dependent (…) it’s a time of great change’.
‘In most children it’s self-limiting anyway that, intervention or no intervention, they don’t attend, they don’t take their medication, there not a lot of point and it’s a waste of time’.
‘I think that some normal symptoms could have labelled somebody as being depressed when they weren’t’.
‘Sometimes is difficult to make a diagnosis because adolescents can be very negative in the way they talk, they moods are very white and black…and if they come to see you in a black mood then probably it’s not a depression’.
Normalization: mood changes are part of normal teenage development. Depression is seen almost exclusively as a normal part of adolescence, rather than a disorder. There is, therefore, a risk of over-diagnosis and potentially harmful labelling (see Box 5).
Challenging the psychiatric model. There were some statements that challenged the relevance of psychiatric thinking in teenagers, counterposed to an acceptance by some practitioners that depression in young people not only existed but also was manageable within general practice (see Box 6).
Box 6 Teenagers and depression
CHALLENGING THE DIAGNOSIS
‘Depression isn’t necessarily based on any pathology that we can see under a microscope- somebody decided it was this and that; these feelings for this amount of time and that’s how we diagnose it’.
‘One of the problems with psychiatric disorders is that they are produced by psychiatrists who, by definition only see a sub sample of the population of disorders and therefore the disorder’s definition can be very skewed’.
‘how you make a research tool into something that could be adapted into a clinical context is the big issue … From our point of view clinically what we have to do is maintain a dialogue and let everything fit into the dialogue appropriately and therefore the adaptation (of the research tool) to the dialogue is the critical thing’.
‘Well, I’d want to know that it was worth it, if you told me that we were missing one in two depressions… that adolescents under the age of 16 were committing suicide at an exponential rate, then I would see that there was a point in it. If you tell me that in most…. it’s self limiting anyway, that erm intervention or no intervention, they don’t attend, they don’t take their medication, there’s not a lot of point and it’s a waste of time’.
RECOGNIZING DEPRESSION
‘I don’t think I have a problem about conceptualizing that age group as having a depressive disorder’.
‘I haven’t changed the idea of adolescents being depressed, but it has changed the idea that I can do something, the understanding of how they view their symptoms, to link their symptoms into other things, they feel very isolated’.
‘I didn’t feel that I had unearthed problems or made problems that didn’t exist or medicalised a kind of normal adolescent psyche’.
‘I am glad to uncover depression in adolescents, I think they like having someone who listens to them, I think it is rewarding for them’.
‘Before the study I hadn’t realized that teenagers could be as troubled as they were, and I have suddenly realized that actually a lot of them haven’t had the advantages that we did, and therefore you know, like the parenting for example’.
‘Before the study I just though, you know ‘it is just a row’, and I didn’t give the weight that I might have in certain cases’.
Discussion
What this study shows
GPs in this study thought about teenagers as a separate and difficult patient category, and characterized them as: 1) intermittently and impulsively using GP services, but not engaging, 2) being complex in their presentation of depressed mood and thereby time-consuming, 3) ‘going through a phase’ that does not necessarily have consequences, 4) being embedded in families, which makes addressing their individual needs more complicated. These characterizations did not necessarily prevent the GPs from recognizing and responding to depression in young people, but they did seem to complicate their responses.
How this relates to the literature
We see three issues that emerge from our findings as paramount: the nature of teenagers as a group different from other patients, normalization of teenage depression and the challenge to the relevance of a psychiatric diagnosis.
Teenagers make different use of services from other patients. Attendance rates by teenagers are relatively high, at two to three consultations per year (Department of Health, 1992) but teenagers get shorter consultations than older adults (Jacobson et al., Reference Jacobson, Wilkinson and Owen1994). Even when GPs perceive that young people have psychological problems they do not always explore these and a specific management or follow-up plan is put into place in only a minority of cases (Martinez et al., Reference Martinez, Reynolds and Howe2006).
Although there is some (dated) evidence that many young people believe that their GP is unsympathetic, or feel uncomfortable in consultations with their GP (Balding, Reference Balding1996), more recent research suggests that teenagers are generally satisfied with the care that they receive in general practice (Jacobson et al., Reference Jacobson, Mellanby and Donnovan2000). Similarly, the attitudes of teenagers may be less of a barrier to general practice care than their perceived status as a special, problematic group might suggest (Churchill et al., Reference Churchill, Allen, Denman, Williams, Fielding and von Fragstein2000).
Mood changes are part of normal teenage development. GPs do recognize and respond to severe depression in young people, as they do in adults (Hyde et al., Reference Hyde, Evans, Sharp, Croudace, Harrsion, Lewis and Araya2005), so the normalization of mood changes is not absolute. Between 75% and 80% of young people do not experience distress of any depth or significance during their teens (Meltzer et al., Reference Meltzer2000), and the belief that young people are miserable, hyperemotional and at war with their parents and the world does not correspond with the epidemiology of psychological distress in teenage populations (Arnett, Reference Arnett2007). The emphasis given by these GPs to the normality of mood changes in young people is congruent in one sense with what is known about the low prevalence of psychological disturbance in this age group, and the resilience and optimism of young people.
What is incongruent is the perception that ‘adolescence is a time of flux and angst’, because this could lead to significant psychological distress being mis-categorized as normal for the age. Depression in the teenage years can be persistent and associated with recurrent depression. The reluctance to view low mood and associated symptoms as depression appears to reflect a risk assessment that gives greater weight to the hazards of labelling unhappiness as depression, than to the identification of depression and intervention to minimize current and future impairment.
‘Normalization’ of depression in teenagers may flow from taking the psychosocial view of problems that is the hallmark of general practice. We know that GPs in socio-economically deprived areas may see depression as a normal response to disadvantage and adverse life events, and may be reluctant to respond medically to such patients because of their inability to influence the underlying disadvantage (Chew-Graham et al., Reference Chew-Graham, Mullin, May, Hedley and Cole2002). A comparable process may occur with depressed teenagers, who may be seen in the family context rather than as individuals, particularly if their psychological distress appears linked to seemingly intractable family circumstances.
Challenging psychiatric diagnoses. There was a tendency to challenge the validity of a psychiatric diagnosis in teenagers and to minimize the importance of depressed mood. This appears to be different from GPs’ responses to depression in adults, where organizational issues, referral options, therapies and stigma influence clinical thinking, but not rejection of depression as a valid category (Railton et al., Reference Railton, Mowatt and Bain2000).
The time factor stressed by some of the GPs in this study is subtly different to that identified with depressed adults. GPs acknowledge that dealing with depression in adults typically requires longer consultations than other conditions, especially for first consultations, but nevertheless can be accommodated within routine care (Pollock and Grime, Reference Pollock and Grime2003).
However, there is a major difference regarding depression in young people since drug treatment is the last therapeutic option for mild and moderate depression in young people, used only when cognitive behaviour therapy has been implemented without response. GPs may feel confident enough to use medication in adults but not in teenagers, especially with the recent controversy about the use of selective serotonin reuptake inhibitors in this age group (Cotgrove and Timimi, Reference Cotgrove and Timimi2007), and this could be an obstacle to recognition and response. There is also evidence that GPs do not feel confident using structured psychological approaches to depression in adults, which can be perceived as too rigid for the problems encountered in general practice (Pierce and Gunn, Reference Pierce and Gunn2007).
Limitations of the study
The GPs in this study were interviewed at the end of a project in which they had received training in the identification of depression in young people, and so may have had both a special interest in depression in young people and also a different perception of the problem because of the training. We did not attempt to assess attitudes in any depth before starting the training, and in retrospect feel that this could have been useful. The GPs worked in the particular environment of an inner London practice, with caseloads and methods of working that might not be the same in other settings and populations. Therefore, their views may not be generalizable to other GPs. However, ‘generalizability’ is not a meaningful idea in qualitative research, since the relevance and applicability of the findings is determined by the reader (not the researcher), who reaches a conclusion about how well the themes identified in this group correspond to his/her experience. All qualitative analysis is a process of reduction and it is recognized that this can compromise the totality of the qualitative data (Burnard, Reference Burnard1998). We attempted to offset this bias by having the data analysed by three (is this three or seven) individuals from different disciplines, with an iterative approach to compare themes with text.
Implications for research and practice
GPs’ knowledge and skills appear to influence the recognition and appropriate management of depression, and GPs with a declared interest in mental health or who have had mental health training are more likely to see more patients with depression and more likely to provide appropriate mental health assessments and treatments. Conversely, GPs who report greater professional unease in managing depressed patients and a stronger belief in the inevitability of the course of depression are less likely to follow up depressed patients (Richards et al., Reference Richards, Ryan, McCabe, Groom and Hickie2004). If our findings are typical, professional development designed to equip GPs with the skills and confidence to recognize and respond to depression in young people may need to address two perceptions, the ‘normal’ (as opposed to frequently self-limiting but sometimes disabling) nature of depressed mood in this age group, and the perceived under-utilization of general practice by teenagers. Such training may also need to address the complexities of young people’s positions within their families, and to be realistic about what therapies can be delivered in general practice. These hypotheses should be tested more widely in different settings.
Acknowledgements
We thank the general practitioners who took part in this study.
Conflicts of interest: none declared.
Ethics approval: this study was approved by St Mary’s local research ethics committee.
