Management of depression in UK general practice in relation to scores on depression severity questionnaires: analysis of medical record data
BMJ 2009; 338 doi: https://doi.org/10.1136/bmj.b750 (Published 19 March 2009) Cite this as: BMJ 2009;338:b750
All rapid responses
Rapid responses are electronic comments to the editor. They enable our users to debate issues raised in articles published on bmj.com. A rapid response is first posted online. If you need the URL (web address) of an individual response, simply click on the response headline and copy the URL from the browser window. A proportion of responses will, after editing, be published online and in the print journal as letters, which are indexed in PubMed. Rapid responses are not indexed in PubMed and they are not journal articles. The BMJ reserves the right to remove responses which are being wilfully misrepresented as published articles or when it is brought to our attention that a response spreads misinformation.
From March 2022, the word limit for rapid responses will be 600 words not including references and author details. We will no longer post responses that exceed this limit.
The word limit for letters selected from posted responses remains 300 words.
We thank Cameron and colleagues for their thoughtful response. We
don’t agree however that we’re proposing modifying scales ‘on the hoof’ in
suggesting that practitioners might adopt slightly different threshold
scores for considering intervention. We suggest this not on the basis of
our study alone but in light of the results of validation studies against
more extensive ‘gold standard’ diagnostic assessments, along the lines of
the suggestion made by Cameron and colleagues in fact. It is these studies
which have suggested that the validity of the measures in terms of
identifying major depressive disorder could be improved by using a more
conservative cut-off score of 12 rather than 10 on the PHQ-9[1,2], and a
less conservative cut-off of 10 rather than 11 on the HAD-D.[1]
However, notwithstanding that point, it is important to emphasise
that the validity of a symptom questionnaire score in identifying
significant depression is always going to be questionable in any
individual person. Validity is expressed in terms of the ability of an
instrument to identify cases at the group level, and there will always be
false positives and false negatives. This has two implications for
clinical practice.
Firstly, when assessing patients, practitioners should not be guided
entirely by the symptom score alone, but should consider the severity of
symptoms in terms of significant effects on work and daily activities,
their duration, and whether there is a past history of more severe
symptoms. The need to take these factors into account is stressed in the
quality and outcomes framework (QOF) guidance for the GP contract[3] and
it seems likely that the GPs in our study were doing just that, given our
findings.
Secondly, what’s more important in any individual patient is the
trajectory of scores rather than the absolute value at any given point in
time. We thank Panch for making that point above. This is the basis of a
new indicator in the QOF for 2009/10, DEP3, which awards practices
additional points for carrying out a follow-up questionnaire measure of
severity recorded 5-12 weeks (inclusive) after the initial recording of
the assessment of severity.[3] The PHQ-9 has been shown to be a responsive
and reliable measure for gauging response to treatment in individual
patient care,[4] and there is on-line guidance available on using it to
assess response to treatment.[5]
[1] Lowe B, Spitzer RL, Grafe K, Kroenke K, Quenter A, Zipfel S et
al. Comparative validity of three screening questionnaires for DSM-IV
depressive disorders and physicians' diagnoses. J Affective Disorders
2004; 78:131-140.
[2] Gilbody SM, Richards D, Barkham M. Diagnosing depression in
primary care using self-completed instruments: UK validation of PHQ-9 and
CORE-OM. Br J General Practice 2007; 57:650-652.
[3] BMA, NHS Employers. Quality and Outcomes Framework guidance for
GMS contract 2009/10. Delivering investment in general practice. London:
2009.
[4] Lowe B, Unutzer J, Callahan CM, Perkins AJ, Kroenke K.
Monitoring depression treatment outcomes with the Patient Health
Questionnaire-9. Medical Care 2004; 42:1194-1201.
[5] The MacArthur Initiative on Depression in Primary Care. Using
the PHQ-9 to assess patient response to treatment.
www.depression-
primarycare.org/clinicians/toolkits/materials/forms/phq9/treatment_response.
Competing interests:
I am the lead author on the paper to which this rapid response is being made. I was a member of the mental health expert reference group for the quality and outcomes framework.the
Competing interests: No competing interests
Dear Editor- It is disappointing to read the results of Kendrick’s
study on the management of depression particularly in relation to older
adults and to realise that little has changed since MacDonald published
his paper in this journal over 23 years ago (1). In 1986 he found that
although depression was being recognised in older people they were less
likely to be prescribed antidepressant medication or referred to
specialist mental health care. Despite the medical advances of two
decades, the developments of new antidepressant medications and increasing
availability of psychological interventions it would appear that older
adults are still underdiagnosed and undertreated.
Depression is the most common mental health disorder in older adults.
One in four older people living in the community have symptoms of
depression that are severe enough to warrant intervention (2). Older
people themselves are often slow to recognise and seek help for
depression in the mistaken belief that this is a normal part of ageing and
that treatment is not needed, a view unfortunately shared by some health
care professionals. Although depression in older adults may be more
prolonged than in younger people it responds equally well to therapeutic
intervention (3).Untreated, it may cause needless suffering for many
individuals and their families.
It is sobering to remember that depression is the leading risk factor
for suicide(4)and that older adults have a suicide risk of almost twice
that of the general population. Studies such as this, highlighting the
inequalities which exist in the treatment of depression, serve as a timely
reminder to constantly and vigorously challenge attitudes that create
barriers to the equitable treatment of older people.
1. MacDonald AJ. Do general practitioners ‘miss’ depression in
elderly patients? BMJ 1996; 282: 1365-6
2. Age Concern Policy Publication. 13th August 2008.
3. G. Alexopoulos. Depression in the elderly. The Lancet; 365: 1961-1970
4. Fiske A et al. Depression in older adults. Annual Review Clinical
Psychology. 2009; 5: 363-89
Dr Conor Maguire, Dr T Coughlan, Dr R Collins, Prof D O’Neill, Age
Related Health Care, Adelaide and Meath, National Children Hospital,
Tallaght, Dublin 24, Ireland.
Competing interests:
None declared
Competing interests: No competing interests
A comparison of the questionnaire scores and their relationship with
the rate of referral to the psychiatric services and antidepressant
prescription was conducted. At face value it appears that PHQ9 responders
may have falsely alerted the clinicians to act in a more aggressive
manner.
It may be a false assumption. It is probably likely that the
incidence is in fact higher among those who responded to the PHQ9. The
comparison at baseline shows a higher preponderance in chronic physical
illnesses among this group. The percentage of chronic physical illness PHQ
9 group is in fact 50% higher than in the HAD group. This alone could be
the reason for the disparity in severity assessment performed in this
investigations.
The suitability of questionnaire debate is not new and may probably
never be settled as more new questionnaires are produced for depression
assessment. The ideal use of the questionnaire, however, is in the
measurement of progress. For assessment of improvement the use of any of
the three validated questionnaires is acceptable.
Competing interests:
None declared
Competing interests: No competing interests
This and its associated paper discussed the importance of depression
severity questionnaires to provide guidance to general practitioners with
the aim of improving patient management. Although amongst other things,
it was noted that GPs take account of age and physical illness, but there
was no comment about why such factors are important.
What is important is that both increasing age and physical illnesses
such as cardiovascular and cerebrovascular disorders and type 1 diabetes,
for example, share the common features of having impaired blood flow
because of increased blood viscosity and reduced red cell deformability.
Such changes are well documented in the medical literature.
So it is not surprising that depressive illness has been shown to be
associated with reduced rates of cerebral blood flow. Perhaps the most
informative study was that of Bench et al (1) who reported, "Thus,
recovery from depression is associated with increases in regional cerebral
blood flow in the same area in which focal decreases in regional cerebral
blood flow are described in the depressed state, in comparison with normal
subjects." Therefore it is not surprising that during the systemic
impairment of blood flow associated with heart disease, stroke and
diabetes, that depression should develop. More recently, Navarro et al
(2) reported that the reduced rate of blood flow which had been observed
in the left anterior frontal region of the brain in elderly subjects
suffering from depression, returned to normal during remission.
As low intensity physical activity has been shown to lower blood
viscosity, it could be anticipated that physical activity would be helpful
in depression and this has been shown in several studies. For example,
in a randomised, controlled study of exercise on depressive symptoms,
Mather et al (3) stated, "Although anti-depressants may facilitate a more
rapid therapeutic response than exercise, after 16 weeks of treatment
exercise was equally effective in reducing depression among patients with
major depressive disorders." It is relevant also that among the various
risk factors, at least three of those factors (stressful events,
inactivity and cigarette smoking) are also associated with increased blood
viscosity.
Therefore, if impaired cerebral blood flow is a primary factor in
depressive illness, the most important question is, "Can the blood
viscosity problem be treated ?" If it is legitimate to extrapolate from
the observations in other studies, the answer would be in the affirmative.
Kromhout et al (4) reported in 1985 that in a 20-year follow-up study it
was found that a daily intake of 35 grams of oily fish was associated with
a 50% decrease in coronary heart disease. In 1986, Kamada et al (5)
reported the use of spin label technology to show that the poor
deformability of diabetic red cells was improved by sardine oil as a
dietary supplement which increased the fluidity of the lipid bilayer of
the red cell membrane and improved cell deformability. Those findings
suggest that in adequate doses, fish oil rich in omega-3 fatty acids could
be helpful in conditions with impaired blood flow, including depressive
illness.
Although there have been investigations into the effectiveness of
omega-3 fatty acids in depression the results have been variable and no
study was located which studied the effects of the oil on blood flow. A
2004 report from Finland stated that a low intake of fish was associated
with depression in women but not in men. A study from Ireland of the
association of depressed mood and fish intake in 10,602 men, reported
that, "These findings suggest that depressed mood is associated with fish
intake both directly and indirectly, etc." It should be noted that in the
Northern Hemisphere the common food fish are oily fish, but in the
Southern Hemisphere there are no common food fish which are good sources
of omega-3 fatty acids. Two epub 2009 papers also comment on the benefits
of a diet rich in oily fish. The first paper concluded,
"Our findings suggest that dietary intakes of fish and long-chain omega-3
fatty acids may be inversely asssociated with chronic depressive symptoms
in women." The abstract of the second paper commenced, "The
unsatisfactory results of the monoamine-based anti-depressant therapy and
the high occurrence of somatic symptoms and physical illness in patients
with depression imply that the serotonin hypothesis is insufficient to
approach the aetiology of depression."
It seems reasonable to conclude from the foregoing that depression is
the consequence of inadequate rates of regional cerebral blood flow. For
that reason alone, GPs should consider the potential benefits of
suggesting a daily dietary supplement of 6 grams daily of fish oil to any
patient presenting with the symptom of depression. We have found that 4
grams daily of fish oil was inadequate to maintain the resulting benefits.
References.
1. Bench CJ, Frackowiak RS, Dolan RJ. Changes in regional cerebral blood
flow on recovery from depression. Psychol Med 1995; 25: 247-61.
2. Navarro V, Gasto C, Lomena F, et al. Normalisation of frontal
cerebral perfusion in remitted elderly major depression: a 12-month follow
-up SPECT study. Neuroimage 2002; 16: 781-7.
3. Mather AS, Rodriguez C, Guthrie ME, et al. Effects of exercise on
depressive symptoms in older adults with poorly responsive depressive
disorder: randomised, controlled trial. Br J Psychiatry 2002; 180: 411-5.
4. Kromhout D, Bosschieter IH, Coulander C, et al. The inverse relation
between fish consumption and 20 year mortality from coronary heart
disease. N Engl J Med 1985; 312: 1205-9.
5. Kamada T, Yamashita T, Baba Y, et al. Dietary sardine oil increased
erythrocyte fluidity in diabetic patients. Diabetes 1986; 35: 604-11.
Competing interests:
None declared
Competing interests: No competing interests
actually, on a point of pedantry, the phrase was "not all that
wheezes is Asthma "
hence
'Not all that weeps is depression..'
But of course I do agree , the order of events, and their relation
one to another, makes all the difference of meaning in the world. Sadness
and Grief are not the same as Depression in need of drug treatment, but
often do share the same pervasive space in human understanding.
Competing interests:
None declared
Competing interests: No competing interests
Doctors listening to the chests of patients with left ventricular
failure will remember that 'All that wheezes is not asthma'. Similarly
when low mood stikes our patients we need to remember that 'All that weeps
is not depression'. Ajustment disorders, bereavement, dysthymia and
borderline personality traits can all present in low mood, and are not
depression, but are easily mislabeled.
The HADS BDI and PHQ9 are all markers of severity of depression, and for
PHQ9 and HADs secreening tools to judge the probability that depression is
present. They are not diagnostic tools.
In particular the fragile labile mood and need for instant relief because
of low frustration tolerance that feature in borderline personality traits
can give very odd results with theses tools. Often the score is heavily
shifted to the severe end.
In the study no attempt seems to have been made to establish the actual
diagnosis for the patients given the tools, other than that they were
given them.
We need to use diagnostic labels carefully, and certainly avoid the easy
trap of low mood is depression needs counselling.
Competing interests:
None declared
Competing interests: No competing interests
The general practitioners who did not prescribe antidepressants to
21% of patients in response to these two simple self assessments are to be
applauded for their circumspection. Those that failed to refer the 77% to
specialized services are not. As noted in this journal, in February 2007
Kirsch and colleagues concluded, "there seems little evidence to support
the prescription of antidepressant medication to any but the most severely
depressed patients." Your editorial giving the chemicals the benefit of
the doubt (as opposed to a demand for scientific rigor)also did not give a
rousing endorsement of a chemical only approach. (March 4, 2008)
The value of supportive therapy has been well documented. The
failure to refer more patients to supportive services needs to be
explained.
Competing interests:
None declared
Competing interests: No competing interests
Kendrick and colleagues confirm our finding that the PHQ-9 and HADS
depression subscale (HAD-D) classify patients' depression severity
differently (1). They found that practices using the PHQ-9 recorded larger
proportions of patients with moderate to severe depression than those
using the HAD-D (2). In light of this, it is perhaps fortunate that the
proportion of patients receiving antidepressant prescriptions and
referrals to specialist services were similar for both groups: after all,
it is entirely appropriate that GPs should treat cautiously the results of
assessments over which questions of validity remain (1), however much
patients like them (3).
The authors propose that a cut-off greater than or equal to 12 in the
PHQ-9 and greater than or equal to 9 in HAD-D might be "more valid". In
their study, this would have led to a similar proportion of patients being
classified as moderately to severely depressed. However, observing
similar proportions of cases does not mean that, had the same patients
been assessed with both measures, then the same individuals would fit
within the same cut-offs for both measures. To test this notion, we re-
analysed our sample of patients referred to primary care mental health
workers who had completed both PHQ-9 and HAD-D (1) using the cut-offs
suggested by Kendrick and colleagues. Cases for PHQ-9 and HAD-D remained
significantly different, with PHQ-9 classifying more patients as cases
(McNemar test chi square=6.509, p=0.011).
Clearly, altering the threshold along the lines proposed by Kendrick
and colleagues does not lead to a concordance between the PHQ-9 and HAD-D
in our sample. Rather than modifying the scales "on the hoof", surely it
would be more appropriate to compare the psychometric properties of the
HAD-D and PHQ-9 with the Hamilton Rating Scale for Depression (4). After
all, most of what is known about treatment effectiveness for depression in
relation to severity is based on this last assessment (5). Should the PHQ-
9 and HAD-D fall at this hurdle, then the QOF depression measures will
simply have to be reconsidered if we wish to keep faith with the evidence
base. For now, GPs should continue to exercise circumspection when
interpreting depression severity measure scores - and their clinical
judgement when treating people with depression.
References
1. Cameron IM, Crawford JR, Lawton K, Reid IC. Psychometric
comparison of the PHQ-9 and HADS for measuring depression severity in
primary care. Br J Gen Pract 2008;58:32-6.
2. Kendrick T, Dowrick C, McBride A, Howe A, Clarke P, Maisey S, et
al. Management of depression in UK general practice in relation to scores
on depression severity questionnaires: analysis of medical record data.
BMJ 2009;338: b750.
3. Dowrick C, Leydon GM, McBride A, Howe A, Burgess H, Clarke P, et
al. Patients' and doctors' views on depression severity questionnaires
incentivised in UK quality and outcomes framework: qualitative study. BMJ
2009;338: b663.
4. Cameron IM, Crawford JR, Lawton K, Sharma S, DuToit S, Hay S, et
al. Assessing the validity of the PHQ-9, HADS, BDI-II and QIDS-SR16 in
measuring the severity of depression in a UK sample of primary care
patients with a diagnosis of depression: study protocol. Primary Care and
Community Psychiatry 2008;13(2):67-71.
5. NHS National Institute for Clinical Excellence. Depression:
Management of depression in primary and secondary care 2004. Report No 23.
Competing interests:
None declared
Competing interests: No competing interests
Depression severity measurement in primary care
Kendrick & Dowrick justify their suggestion that the cut offs for
depression severity measures should be altered by appealing to studies
which have assessed the scales’ validity as case finders for major
depression 1, 2. We believe that making a change on this basis would be
premature. It is important to be clear about which aspect of the scales’
validity is under scrutiny. They might be used in three rather different
ways: to identify cases; to determine severity as a guide to treatment
selection; and to measure outcome. The weight of the validation lies with
identification1, 2, and we have no quarrel with that (though one study is
very small (n=96) and the other is conducted on a non-UK population).
But it is as a guide to selection of treatment that the PHQ-9 scale
limitations in particular are most important. QOF guidance states that
“assessment of severity is essential to decide on appropriate
interventions and improve the quality of care”3, and goes on to link this
statement with NICE recommendations. However, reading beyond the “Summary
of Recommendations” in the NICE guidance, we learn: “a key issue is
whether severity of illness can guide the use of antidepressant
medication. Unfortunately there is little data to help with this point.”4.
Importantly, such data that do exist are certainly not based on the PHQ-9:
most research has utilised the Hamilton Depression Rating Scale. Indeed,
Kendrick & Dowrick use exactly this outcome measure in their THREAD
study5. If the PHQ-9 fails to map onto the measures used to derive the
evidence base, then it cannot be used sensibly to make treatment decisions
which are framed by that evidence base.
Kendrick and colleagues emphasise that practitioners are taking into
account other factors in their treatment decisions in their linked papers
6, 7. One could just as easily conclude that practitioners are basing
their decisions entirely on other factors. Perhaps this is just as well -
in our view, the scales are not measuring the same constructs. Snaith
warned in 1993 that “unwary researchers will assume that all depression
scales assess more or less the same state”8. Clinicians should continue
to be wary when interpreting the scores of the HADS-D or the PHQ-9 -
regardless of the cut-off applied.
References
1. Gilbody S, Richards D, Barkham M. Diagnosing depression in primary
care using self-completed instruments: UK validation of PHQ-9 and CORE-OM.
Br J Gen Pract 2007;57:650-2.
2. Lowe B, Spitzer RL, Grafe K, Kroenke K, Quenter A, Zipfel S, et
al. Comparative validity of three screening questionnaires for DSM-IV
depressive disorders and physicians' diagnoses. J Affect Disord
2004;78:131-40.
3. NHS Employers and the General Practitioners' Committee. Quality
and Outcome Frameworks: guidance for GMS contract 2009/10. 2009.
4. NHS National Institute for Clinical Excellence. Depression:
Management of depression in primary and secondary care; 2004. Report
No.23.
5. Chatwin J, Kendrick T, THREAD Study G. Protocol for the THREAD
(THREshold for AntiDepressants) study: a randomised controlled trial to
determine the clinical and cost-effectiveness of antidepressants plus
supportive care, versus supportive care alone, for mild to moderate
depression in UK general practice. BMC Family Practice 2007;8:2.
6. Kendrick T, Dowrick C, McBride A, Howe A, Clarke P, Maisey S, et
al. Management of depression in UK general practice in relation to scores
on depression severity questionnaires: analysis of medical record data.
BMJ 2009;338.
7. Dowrick C, Leydon GM, McBride A, Howe A, Burgess H, Clarke P, et
al. Patients' and doctors' views on depression severity questionnaires
incentivised in UK quality and outcomes framework: qualitative study. BMJ
2009;338.
8. Snaith P. What Do Depression Rating Scales Measure?. British
Journal of Psychiatry 1993;163:293-8.
Competing interests:
None declared
Competing interests: No competing interests