Depression should be managed like a chronic diseaseBMJ 2006; 332 doi: https://doi.org/10.1136/bmj.332.7548.985 (Published 27 April 2006) Cite this as: BMJ 2006;332:985
All rapid responses
Three related papers concerning Depression in the same issue of the
BMJ; Scott, 985-986, Gilbody et al 1027-1030 and Layard 1030-1033 (BMJ 29
April 2006) deserve comment.
We have argued that degradation of clinical use of the term
“depression”, that has facilitated excessive use of antidepressants has
also resulted in a loss of definition between “the illness that we call
depression”, and responses to adversity. The former undoubtedly benefits
from an illness model; the efficacies of drug and psychological treatments
clearly outweigh the costs of labelling and discrimination. That is not so
clearly the case for sub-threshold, sub-syndromal or, in ICD-10 terms,
mild to moderate “depression”. Gilbody and colleagues’ paper recognises
this, and contributes a timely warning against the risk of medicalising
emotional distress in response to the Quality and Outcomes Framework
(QOF). Applying a pathological medical model when natural adaptive
reactions to adversity are operating can be harmful.1
Jan Scott’s editorial deserves comment, as advice upon what primary
care should offer to individuals with depression. Here there is no
distinction between “the illness that we call depression” (which may
benefit from a chronic disease management paradigm) and emotional distress
(which might well be harmed by such an approach). Again, this is despite
the fact that the evidence base so thoroughly reviewed in developing NICE
guidelines for the management of depression does not support an illness
model for so-called mild to moderate, or sub-threshold depression.
It is timely to see Lord Layard’s views in print in the BMJ. The so-
called Layard hypothesis, that it would be economically sound to invest in
psychological therapies because the result would be improved quality of
life, increased revenue income and savings on incapacity benefit is
generating considerable interest and activity amongst psychotherapeutic
and mental health service provider communities. His position is that
cognitive behaviour therapy (CBT) can be effective across a wide spectrum
of anxiety and depressive conditions. We contend that CBT is another term
that has become degraded by over use. The evidence demonstrating efficacy
of CBT in the treatment of depression is limited to the treatment of
moderate to severe depression, and in the treatment of anxiety, to the
defined anxiety disorders; not to anxiety itself.
However the evidence for even this comes from comparative trials with
drug treatment rather than with “practical help and supportive care” which
we suggest is what is required when emotional distress is an adaptive
call. Furthermore such evidence does not necessarily translate directly
into support for the complex machinery of psychological treatment centres,
where effectiveness will depend on assessment, management, treatment,
education, communication and other ‘black box’ factors operating
throughout the pathway.
We have no wish to gainsay the realistic prospect of improvements in
public health and wellbeing that could arise from stronger investment in
psychological therapies. However, benefit could be restricted if the
adopted approaches do not explicitly recognise the potentially harmful
consequences of an illness model. Converging evidence points to the fact
that the active “ingredient” in psychological therapy is non-specific
factors mediated by the overall quality of the relationship between
therapist and client. This further emphasises the importance of
contributions from non-specialists. Both the NICE Depression Guidelines
and the NICE Anxiety Guidelines advocate a stepped care approach. This
amounts to providing therapeutic input of differing levels of intensity
and sophistication, determined by need and response. As a result a
distinction can be maintained between responses to adversity that take the
form of help-seeking from a healthcare setting, which might well be best
met by a short term, nonspecific practical or supportive intervention2 and
the illness that we call depression, which might well benefit from a
formally defined CBT, avoiding the socially harmful effects of identifying
the former as an instance of the latter.
1. Ørner RJ, Siriwardena AN, Dyas JV, Middleton HC, Shaw I, Woodward
L. (2004) The NICE guideline on depression. Research recommendations for
primary care. Primary Care Mental Health;2:137-9.
2. Ørner RJ, Siriwardena AN, Dyas JV. Anxiety and depression: a
model for assessment and therapy in primary care. (2004) Primary Care
Competing interests: No competing interests
I'd like to commend the whole thrust of the Rapid Response to the
Scott editorial posted above by Prof David Pilgrim.This is to query the
validity and explanatory power of narrowly biomedical thinking and
intervening in the domain of human pain and distress.
Do we take it from Scott that it is settled that "depression" is
always a timeless, free-standing, internally coherent, universally valid,
pathological entity with a life of its own 'out there'? Classification
systems like ICD or DSM (generally called 'international' but in fact
merely 'Western') provide definitions which seem to assume this- though
there are disclaimers in small print to the effect that diagnostic
categories are not facts of nature (as, say, a tree is)but cobbled-
together phenomena emerging as committee decisions. Indeed it was not
inevitable that depressed mood should be seen as the cardinal symptom, and
furnish the name of the whole syndrome.Other symptoms could have been
used:"sleep disorder syndrome" or "concentration and drive disorder
Scott recycles the statement the depression is under-recognised and
under-treated. What is the evidence for this? Some might point to a few
community surveys using quantitative instruments supposedly tapping
"depression". Such instruments, with their demand characteristics and
narrow focus on "symptoms",generate inflated prevalence estimates because
of their structural inability to assess the whole person-in-life context.
We do not have an epidemic of depression, we have an epidemic of anti-
depressant prescribing (up two and a half times in a decade) in an age of
the medicalisation and professionalisation of unhappiness and the problems
of living.Good news for the pharmaceutical industry.(1)
We require sociological, anthropological, philosophical and indeed
also political frameworks for the arena of human pain and distress to be
properly understood in all its nuances and ambiguities, shaped by context
and culture, and above all centred on meaning (no medical model captures
meaning). The depression-as-disease model does have some purchase (there
is a subset of seriously ill people) but as general formulation it says
less about the world than about the dominance of medicalised ways of
(1) Summerfield D. Depression:epidemic or pseudo-epidemic? Journal of
Royal Society of Medicine 2006; 99: 161-2.
Competing interests: No competing interests
The recent editorial and responses to it about depression assume that
the latter has scientific validity. The concept has legitimacy in the
medical profession and amongst lay people for now, but is it valid?
Depression has certainly entered the vernacular, in Eurocentric cultures,
in the past hundred years. Diagnostic checklists from the American
Psychiatric Association's Diagnostic and Statistical Manual (DSM) and the
World Health Organization's International Classification of Diseases (ICD)
increase the probability of reliability. However, the latter is a
necessary but not sufficient condition for validity. It is also not
suprising that doctors trained with the same checklist system will agree
with one another; but about what precisely?
Grounds for doubting the validity of the diagnosis are legion (1, 2).
They include its large overlap with other categories, including anxiety,
personality disorder and adjustment disorder, as well as an inconsistent
history in psychiatric nosology about unimodal versus bimodal case
distribution, and exogenous/reactive versus endogenous claims. The latter,
causation-implied, labels are supposed to be irrelevant under DSM, which,
since 1980, has asserted aetiological neutrality, but clinicians still use
them in practice (3). Also, what of cultures which have no word for
depression, are they ignorant and mistaken about the pandemic in their
midst (4)? To insist on a global pandemic position is a form of cultural
Moreover, the current emphasis on drug treatment assumes that
depression has a self-evident and well understood biological substrate,
when it does not. To argue that 'antidepressants' rectify both serotonin
depletion and feelings of misery is like arguing that a headache is due to
a lack of aspirin in the brain. The biochemical orthodoxy about medicinal
effectiveness is undermined by evidence of the strong placebo effect in
published studies (5).
Depression is not a free standing affliction 'out there', awaiting
discovery, but a socially constructed dustbin serving the interests of
psychiatry, the drug companies and lay people who, for a range of social
and biographical reasons, experience an intensification of misery and seek
help (or have it imposed upon them). Human suffering, which comes in all
shapes and sizes, is not inherently a disease but is simply about being
alive. It is a variegated existential state, which can be shaped,
triggered, endured, resolved, avoided and escaped from in various
Depression, like other functional psychiatric diagnoses, exemplifies
the 'epistemic fallacy'- that to name something is to make it real (7).
Confusing the map with the territory and individualising misery as a
medical condition in skin-encapsulated patients flow from a focus on this
historically-situated reification, which for now we call 'depression'.
There are three questions implied by the above doubts for for
psychiatric traditionalists, with their headlong rush into more diagnosis
and treatment. First, the pre-empirical scientific question: are we
framing the problem correctly? Second, the philosophical question: by
turning miserable people into patients, does this illuminate or obscure
the wide ranging existential dimensions to human suffering? Third, the
social policy question: can human suffering always be addressed
effectively by the technical fix of medical interventions?
(1)Pilgrim, D. and Bentall, R.P. The medicalisation of misery: a
critical realist analysis of the concept of depression. Journal of Mental
Health, 1998, 8, 3, 261—74.
(2)Dowrick, C. Beyond Depression: A New Approach to Understanding and
Management. Oxford: Oxford University Press. 2005
(3)McPherson, S. and Armstrong, D. Social determinants of diagnostic
labels in depression. Social Science and Medicine, 2006, 62, 1, 50-58.
(4) Wierzbicka A. Emotions across Languages and Cultures: Diversity
and Universals. Cambridge: Cambridge University Press. 1998
(5)Moncrieff, J. and Kirsch, I. Efficacy of antidepressants in
adults. British Medical Journal, 2005, 331, 155-157, 16th July.
(6)Brown, G. and Harris, T. Social Origins of Depression London:
(7)Bhaskar, R. Reclaiming Reality. London: Verso. 1991
Competing interests: No competing interests
Although few would disagree with the premise that depression is often recurrent and could therefore be considered a chronic (if episodic) disease, Scott’s editorial on “this much misunderstood disorder” in some places increases rather than reduces widespread ignorance.
Scott states that most depressive disorders and not mild. In fact in numerous epidemiological studies most
episodes are indeed mild. When studied in a large and generally unselected population there is a correlation
between the prevalence and symptoms. Specifically presentations with fewest symptoms are most common and
those with maximum symptoms (eg 9 of 9 DSMIV major depression symptoms) least common - in what is normally
described as a half-normal distribution. 1 The result is sub-syndromal mood disorders (increasing recognized as
important)2 are more common than mild (by ICD10) or minor depression by RDC criteria which are in turn at least
four times as common as moderate and severe depressive disorders (by either ICD10, RDC or DSMIV).3
Of even greater concern, is the repetition of the much misquoted citation that “15% of all patients with depression
will eventually commit suicide.” No one wishes to deny that suicide in the affective disorders is an extremely
important topic but lets not condemn many sufferers with such deterministic statements without first getting our
facts straight. Two much cited reviews of largely hospitalized depressed patients suggested that the proportionate
mortality (the percentage of the dead who died by suicide) was 15% but it these were later misquoted as referring
to case-fatality rate (the percentage of the original sample who died by suicide) in all patients with mood disorder.45 As most people suffering depression are not hospitalized it should be obvious that this figure cannot be applied to
“all patients” as Scott suggests. Fortunately, several groups have been prepared to study this area scientifically. In
an excellent meta-analysis, Bostwick et al (2000) found (case-fatality) the suicide rate to be 2% not 15% (2.0%
for outpatients diagnosed with an affective disorder) and only 6.0% for suicidal inpatients.6 Some will rightly
criticise a meta-analysis for lack of sample size in one single prospective study. However, more recently Høyer and
colleagues in Denmark followed up a high risk sample of 53,466 patients previously admitted for depression as
inpatients. Suicide was the cause of death in 3141 (6%) cases.7 Interestingly risk was higher in those with unipolar
depression than those with bipolar disorder.
I suggest it is time for BMJ editorials to undergo the same degree of scrutiny as major papers otherwise potentially important messages such as those carried here will be lost under a cloud of erranous scientific "facts"
1 Rucci P , Gherardi S, Tansella M, et al Subthreshold psychiatric disorders in primary care: prevalence and
associated characteristics. Journal of Affective Disorders 76 (2003) 171–181
2 Pincus HA, Davis WW, McQueen LE. 'Subthreshold' mental disorders. A review and synthesis of studies on minor
depression and other 'brand names'. Br J Psychiatry. 1999;174:288-296.
3 Barrett J, Barrett J, Oxman T, et al: The prevalence of psychiatric disorders in a primary care practice. Arch Gen
Psychiatry 1988; 45:1100–1106
4 Guze SB, Robins E: Suicide and primary affective disorders. Br J Psychiatry 1970; 117:437–438
5 Goodwin FK, Jamison KR: Suicide, in Manic-Depressive Illness. New York, Oxford University Press, 1990, pp 227–
6 Bostwick, J.M., Pankratz, V.S., 2000. Affective disorders and suicide risk: a reexamination. Am. J. Psychiatry 157
(12), 1925– 1932.
7 Høyer EHOlesena AV. Mortensen PB. Suicide risk in patients hospitalised because of an affective disorder: a
follow-up study, 1973–1993. Journal of Affective Disorders 78 (2004) 209–217
Competing interests: No competing interests
It is very disappointing that, once again, a BMJ editorial is
perpetuating the myth that 15% of people suffering from depression will
eventually commit suicide.1
A much-cited meta-analysis in 1970 found that 15% of people suffering
from depression committed suicide.2 This study focused primarily on people
hospitalised with severe depression, so it should never have been
generalised to the broader population of people with depression.
Several rigorous studies have subsequently been published refuting
the 15% claim. A meta-analysis by Bostwick & Pankratz found a
hierarchy of lifetime suicide prevalences: 8.6% in people ever
hospitalised for suicidality; 4% for affective disorder patients
hospitalised but not specifically for suicidality; 2.2% for mixed
Boardman and Healy analysed data from a database of suicide cases in
North Staffordshire, and used psychiatric prevalence rates form the US
National Comorbidity Survey to calculate lifetime suicide risk in people
with depression: 2.4% for any affective disorder; 1.1% for uncomplicated
cases with no mental health service contact.4
Blair-West et al. also found a much lower risk: 'The suicide risk in
major depression as it is currently defined diagnostically is of the order
of 3.4% rather than the previously accepted figure of 15%'.5
They noted: 'Because every major textbook quotes a suicide risk in
major depression of 15%, every good psychiatry trainee and, quite
reasonably therefore, any speaker who needs to emphasize the seriousness
of major depression as a public health concern, uses this figure too. What
is probably the most surprising is that a single paper, that by Guze and
Robins, could be so uncritically accepted and so widely promulgated'.
36 years later it is still being promulgated by the BMJ. Why?
1. Scott J. Depression should be managed like a chronic disease. BMJ.
2006 Apr 29;332(7548):985-6.
2. Guze SB, Robins E. Suicide and primary affective disorders. Br
Journal Psychiatry. 1970;117:437–8.
3. Bostwick JM, Pankratz VS. Affective disorders and suicide risk: a
reexamination. Am J Psychiatry. 2000 Dec;157(12):1925-32.
4. Boardman AP, Healy D. (2001, November). Modelling suicide risk in
affective disorders. Eur Psychiatry. 2001 Nov;16(7):400-5.
5. Blair-West GW, Mellsop GW. Major depression: does a gender-based
down-rating of suicide risk challenge its diagnostic validity? Aust N Z J
Psychiatry. 2001 Jun;35(3):322-8.
Competing interests: No competing interests