Routinely administered questionnaires for depression and anxiety: systematic review
BMJ 2001; 322 doi: https://doi.org/10.1136/bmj.322.7283.406 (Published 17 February 2001) Cite this as: BMJ 2001;322:406
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.
To the Editors:
Gilbody et al recently published a systematic review of the effect in
primary care settings of routinely administered questionnaires on the
recognition, treatment, and outcome of psychiatric disorders, particularly
depression. 1 They reviewed randomized trials published through the year
2000 and concluded that the routine administration of such questionnaires
is a "costly exercise" that "has not been shown to influence clinicians'
behavior."
On behalf of the US Preventive Services Task Force, we recently
performed a broader systematic review of the effectiveness of routine
depression screening. In addition to those reported in Gilbody et al, we
identified six randomized trials of screening that examined recognition,
treatment, or clinical outcomes. 2-7 Each study used a validated screening
instrument and gave feedback to providers of the screening results; some
also confirmed results from the screening instrument with a criterion
standard or gave systematic support to providers and patients to improve
the quality of care after recognition of the diagnosis/condition. Gilbody
et al cited the Wells et al study in their discussion but did not include
it in their analysis; other studies were not addressed in the report.
Because several of these studies had positive impact on at least one major
outcome, we are concerned that failing to include them may have affected
the conclusions of the review. Gilbody et al included 1 study of
depression screening that we did not identify.8
Even with the addition of the extra studies, we agree with Gilbody et
al that interpreting the results of these screening trials is difficult
because of the heterogeneity of outcome measures and times at which
outcomes were assessed. In addition, because the minimal difference in
outcome rates that is considered clinically important has not been
defined, the question of whether existing trials have enough power to
exclude an important effect remains unclear.
Gilbody et al surmise that because the post-test probability of major
depression is only 50% after a positive screen, physicians may tire of
sorting out actual cases from those "who would not benefit clinically."
We argue, by contrast, that a 35%-50% probability of major depression is
quite high enough to justify the next step - i.e., a diagnostic interview
lasting 10 minutes. Failure to act on the results of a positive screen may
be related to lack of familiarity with diagnostic interviewing, lack of
skill in treating depression, or failure to appreciate the morbidity
associated with depressive disorders. Systematic efforts to provide
additional support for treatment and follow-up, as employed in the Wells
et al, 6 may improve the likelihood of improved outcomes.
Michael Pignone, MD, MPH
Bradley N. Gaynes, MD, MPH
Tracy Orleans, PhD
Kathleen N. Lohr, PhD
University of North Carolina - Research Triangle Institute
Evidence -based Practice Center
Reference List
1. Gilbody SM, House AO, Sheldon TA. Routinely administered
questionnaires for depression and anxiety: systematic review. BMJ 2001
322:406-9.
2. Zung WW, King RE. Identification and treatment of masked
depression in a general medical practice. Journal of Clinical Psychiatry
1983; 44:365-8.
3. Callahan CM, Hendrie HC, Dittus RS, Brater DC, Hui SL, Tierney
WM. Improving treatment of late life depression in primary care: a
randomized clinical trial. Journal of the American Geriatrics Society
1994; 42:839-46.
4. Reifler DR, Kessler HS, Bernhard EJ, Leon AC, Martin GJ. Impact
of screening for mental health concerns on health service utilization and
functional status in primary care patients. Archives of Internal Medicine
1996; 156:2593-9.
5. Williams JWJ, Mulrow CD, Kroenke K et al. Case-finding for
depression in primary care: a randomized trial. American Journal of
Medicine 1999; 106:36-43.
6. Wells KB, Sherbourne C, Schoenbaum M et al. Impact of
disseminating quality improvement programs for depression in managed
primary care: a randomized controlled trial. JAMA 2000; 283:212-20
7. Whooley MA, Stone B, Soghikian K. Randomized trial of case-
finding for depression in elderly primary care patients. J Gen Intern Med
2000;15:293-300.
8. German PS, Shapiro S, Skinner EA et al. Detection and management
of mental health problems of older patients by primary care providers.
JAMA 1987;257:489-93.
Competing interests: No competing interests
Editor - Gilbody et al’s(1) systematic review of psychiatric screening
questionnaires suggests that they are of little benefit in improving
psychosocial outcomes of patients with psychiatric disorders managed in
non-psychiatric settings. This is particularly the case if there is no
established system for administering, scoring and responding to ‘high
risk’ results. Based on our experience of using a screening instrument in
the context of a transplant clinic, we would argue that screening
questionnaires serve a useful purpose if targeted at appropriate patient
groups and if supported by a responsive psychiatric service.
Addenbrooke’s Hospital is a centre for liver transplantation and the
weekly post- transplant clinic has regular input from a liaison
psychiatrist. Compliance with medication is frequently impaired in
patients with mental illness and poor compliance in patients following
liver transplantation may have serious consequences for their graft
survival. For this reason, the identification and treatment of
psychiatric morbidity in patients attending the clinic is considered an
important part of their overall management.
As initial rates of psychiatric referral were lower than the
predicted 19% (based on prevalence studies of psychiatric morbidity
following liver transplanatation(2), it was decided to introduce the
Hospital Anxiety and Depression Scale(HADS)(3). This screening
questionnaire is specifically designed for patients with medical disorders
and takes into account the overlap in symptoms of psychiatric and physical
disorders. Patients attending the clinic were routinely screened with the
questionnaire and the psychiatrist assessed all those scoring above the
recognised threshold of 11 in the fields of anxiety or depression.
This intervention led to a significant increase in the detection and
subsequent treatment of anxiety and depressive disorders. Prior to the
introduction of the screen, 7 of 258 patients attending the clinic were
referred to and treated by the psychiatrist. Following the introduction of
the HADS, 41 of the first 164 clinic patients to complete the
questionnaire scored above the critical threshold. Subsequent assessment
by the clinic psychiatrist found that 33 of these patients had a
clinically significant psychiatric disorder warranting treatment (all met
ICD-10 diagnostic criteria for a psychiatric disorder). This shows a
statistically significant increase in cases identified (chi squared
=35.39, df=1, p<_0.001 and="and" a="a" much="much" lower="lower" rate="rate" of="of" false="false" positives="positives" than="than" referred="referred" to="to" by="by" gilbody="gilbody" et="et" al.="al." _20="_20" vs.="vs." _50.="_50." the="the" increase="increase" in="in" detection="detection" was="was" accompanied="accompanied" an="an" psychosocial="psychosocial" psychopharmacological="psychopharmacological" interventions.="interventions." p="p"/> In conclusion, the use of a screening questionnaire in a liver
transplant clinic led to improved patient care and a more efficient use of
an existing psychiatric resource. We feel that the resources required are
justified in medical settings such as these where failure to address
psychopathology may lead to a serious deterioration in physical health.
Scott Eaton
Specialist Registrar in Psychiatry
Cathy Walsh
Consultant Psychiatrist
email: smeaton@globalnet.co.uk
Liaison Psychiatry Service ,
S3 Outpatients,
Box 175,
Addenbrooke's Hospital,
Cambridge
CB2 2QQ
References:
1. Gilbody SM, House AO, Sheldon TA. Routinely administered
questionnaires for depression and anxiety: systematic review. BMJ
2001:322; 406-9
2. Commander M, Neuberger J, Dean C. Psychiatric and social consequences
of liver transplantation. Transplantation 1992:53; 1038-40
3. Zigmond AS, Snaith RP. The Hospital Anxiety and Depression Scale.
Acta psychiatr. Scand 1983:67; 361-370
Competing interests: No competing interests
Editor - We concur with Gilbody et al. 1 that further research is
needed to justify regular use of anxiety and depression questionnaires in
the routine clinical setting. However, we do not agree with the overall
negative message that has been put across. There are two important points
to consider: (a) the appropriateness of the questionnaires used; and (b)
the appropriateness of the clinical setting in which they are used.
The questionnaires used in the studies addressed by Gilbody et al.
may be appropriate for screening the general population, but may perform
suboptimally in the medical setting. High scores on some depression
questionnaires may be from the inclusion of items related to the patients'
physical illness, such as fatigue and sleep disturbance, as discovered in
investigations of rheumatoid arthritis (RA) patients.2 There are, however,
more suitable questionnaires for use in medical settings.
The Hospital Anxiety and Depression scale 3 (HAD) is a questionnaire
specifically designed for use with the physically ill by avoiding such
symptoms. Its application in RA patients 4 has shown that patients with
probable depression have worse functional ability. However, our analyses
of data from 125 RA outpatients indicate that even in the HAD there may be
one item that is answered in terms of retardation due to physical
disability rather than reflecting depression: 'I feel as if I am slowed
down' was answered as 'Nearly all the time' by 29 cases (23%), 'Very
often' by 44 (35%), and 'Sometimes' by 52 (42%); nobody responded 'Not at
all'. Paired comparisons of the original depression score and an adjusted
score (not including this item and thus scaled by 7/6) revealed a
significant difference (respective means = 6.60 and 5.15; t(124) = 21.96,
P<_0.001. similarly="similarly" probable="probable" cases="cases" of="of" depression="depression" with="with" scores="scores" _10="_10" or="or" greater3="greater3" decreased="decreased" from="from" _28="_28" _22="_22" to="to" _15="_15" _12="_12" when="when" using="using" the="the" adjusted="adjusted" score="score" suggesting="suggesting" improved="improved" specificity.="specificity." pearsons="pearsons" correlations="correlations" between="between" functional="functional" disability="disability" as="as" assessed="assessed" by="by" health="health" assessment="assessment" questionnaire="questionnaire" _5="_5" and="and" original="original" showed="showed" that="that" both="both" were="were" moderately="moderately" related="related" respective="respective" rs="0.33," p0.001="p0.001" _0.28="_0.28" p0.01="p0.01" indicating="indicating" there="there" is="is" still="still" a="a" link="link" uncontaminated="uncontaminated" physical="physical" symptoms="symptoms" disability.="disability." p="p"/> We agree that the primary care and general medical hospital
outpatients are possibly not the most appropriate settings for exploratory
investigation of psychological well-being, using existing instruments.
However, secondary care clinics for patients with chronic illness may be a
more relevant target for depression screening, as depression in such
patient groups is much more prevalent (for example, in RA 4). During our
ongoing studies of psychosocial aspects of RA, we have identified several
patients with high HAD depression scores. Feedback of this to the managing
physician has helped identify and treat the problem in the routine
clinical setting, as under current constraints, patients may rarely be
given the time to discuss their psychological well-being in detail with
their health professionals.
When familiar with the items, appropriate health professionals (for
example, the physicians themselves or specialist nurses) can score a paper
version of the HAD within a minute and address possible cases there and
then at minimal cost. Sensible administration and data management of
appropriate depression questionnaires in the correct setting, amongst high
risk patient groups could reduce the major health care and societal costs
related to continued depression by swiftly providing help for those at
need.
Gareth J. Treharne
health psychology postgraduate research student
School of Psychology, University of Birmingham, Birmingham B15 2TT
gjt884@bham.ac.uk
Antonia C. Lyons
lecturer in health psychology
School of Psychology, University of Birmingham, Birmingham B15 2TT
a.c.lyons@bham.ac.uk
George D. Kitas
consultant rheumatologist
Department of Rheumatology, University of Birmingham, Birmingham B15 2TT
g.d.kitas@bham.ac.uk
1. Gilbody SM, House AO, Sheldon TA. Routinely administered
questionnaires for depression and anxiety. BMJ 2001;322:406-409. (17
February.)
2. Callahan LF, Kaplan MR, Pincus T. The Beck depression inventory,
center for epidemiological studies depression scale (CES-D), and general
well-being schedule depression subscale in rheumatoid arthritis. Arthritis
Care Res 1991;4:3-11.
3. Zigmond AS, Snaith RP. The hospital anxiety and depression scale.
Acta Psychiatr Scand 1983;67:361-70.
4. Murphy H, Dickens C, Creed F, Bernstein R. Depression, illness
perception and coping in rheumatoid arthritis. J Psychosom Res 1999;46:155
-64.
5. Fries JF, Spitz PW, Young DY. The dimensions of health outcomes:
The health assessment questionnaire, disability and pain scales. J
Rheumatol 1982;9:789-93.
Competing interests: No competing interests
This meta-analysis is very useful, as is the previous response by
Coyne et al. However, the postive predictive predictive values mentioned
by Coyne, need to be taken in the context of the base line rate of
depression in the population studied. (This is, of course, Bayes's theorem:
the paper demonstrates this effect of prediction in a low vs. high resk
group in fig. 2).
The more accurate the questionnaire the better: adapting and
automating standard questionnaires such as the CIDI may prove in the long
term to be a better strategy.
We now need to consider what are the barriers to clinicians using the
information from the questionnaire, including the method (or person who
administers it) and barrier to using the results.
Competing interests: No competing interests
We applaud the recent meta-analysis 1 challenging the
conventional wisdom that routine screening is an efficient
means of improving the outcome of depression in general
medical settings. We also agree that the allocation of
resources that would be required to make such screening
feasible is not warranted. It appears, however, that the
authors could have made an even stronger case for their
conclusions.
Specifically, they may have overestimated the
efficiency of screening and underestimated pitfalls
inherent in its implementation. As we have previously noted,
2 the positive predictive value of an elevated score on a
screening instrument is typically less than 35%, rather the
50% they assumed in analysis. Furthermore, the considerable
professional resources consumed in resolving false-positive
cases decrease the resources available for efforts to
improve the outcome of already detected patients. The widely
touted PRIME-MD study demonstrated the latter problem. More
than 80% of the 1000 patients screened positive for a mental
disorder, and most of these cases proved to be
false-positives. Clinicians averaged an additional 8 minutes
in their follow-up interviews with these patients, an
increase of over 50% in the length of a typical primary care
visit. This effort, however, ultimately yielded only 16 new
prescriptions for antidepressant medication. 3 If
implemented without significant (and costly) support, this
program would drastically reduce the time and resources
available for follow-up with previously identified patients.
Another study recently demonstrated how difficulties in
obtaining patient acceptance and in integrating follow-up
interviews into the competing demands of biomedical care can
substantially reduce the yield of new patients in need of
treatment.4
Finally, there is the relatively recent
phenomenon, at least in North America, of the prevalence of
antidepressant prescription equaling or exceeding the
prevalence of depression among general medical patients.5
While some of this undoubtedly reflects increased detection
of depressed patients and alternative uses for
anti-depressant medication, much of the increase represents
inappropriate and inadequate treatment for depression. This
is especially important given repeated demonstrations that
increased detection does not translate into improved
outcomes in non-specialty settings. This situation is likely
to be aggravated through the implementation of routine
screening. Without efforts aimed at improving the outcome of
already detected depression in routine general medical care
and at reducing the inappropriate prescription of
antidepressants, routine screening remains part of the
problem, rather than the solution to untreated or
inadequately treated depression outside of specialty mental
health settings.
1. Gilbody, S.M., House, A.O., & Sheldon, T.A. Routinely
administered questionnaires for depression and anxiety:
Systematic review. British Medical Journal 2001;
322:406-09.
2. Coyne, J.C., Thompson, R., Palmer, S.C., Kagee, A., &
Maunsell, E. Should we screen for depression? Caveats and
pitfalls. Applied and Preventative Psychology 2000; 9:
101-21.
3. Williams, J.B.W., Spitzer, R.L., Linzer, M., Kroenke, K.,
Hahn, S.R., DeGruy,F.V., & Lazev, A. Gender differences in
depression in primary care. American Journal of Obstetrics
and Gynecology 1995; 173:654-59.
4. Payne, D., Hofman, R., Theodoulou, M., Dosik, M., Massie,
M. Screening for anxiety and depression in women with breast
cancer. Psychosomatics 1999; 40: 64-9.
5. Mamdani MM, Parikh SV, Austin PC, & Upshur, R.E. Use of
antidepressants among elderly subjects: Trends and
contribut
Competing interests: No competing interests
Brief questions better than screening questionnaires for depression
In response to the study by Gilbody et al (BMJ 2001;322:406-9) it
has been shown elsewhere that screening has not improved outcomes of
patients with depression. 1 Since that report there is one study that
asked patients a single question "have you felt depressed or sad much of
the time in the past year." 2 The screened group were more likely to
recover (48% vs 27% p<0.05)2 presumably due to the fact that those
with major depression were more likely to be recognised (45% vs 24%) and
treated (55% vs 28% p<_0.02. when="when" compared="compared" with="with" the="the" gold="gold" standard="standard" diagnostic="diagnostic" interview="interview" schedule="schedule" single="single" screening="screening" question="question" had="had" sensitivity="sensitivity" of="of" _85="_85" those="those" depression="depression" and="and" a="a" specificity="specificity" _66.="_66." p="p"/>Another screening tool of two questions has been developed in a written
form as opposed to a verbal question.3 These two questions are "during the
past month have you often been bothered by feeling down, depressed or
hopeless" and "during the past month have you often been bothered by
little interest or pleasure in doing things" and a yes to either of these
has a sensitivity and specificity of 0.96 and 0.66 for depression in
patients in whom substance abuse is excluded. Written questionnaires are
too time consuming for primary care physicians. The two question screening
question is easy to remember and has a positive predictive value of 33%
for a prevalence of depression of 15%. Any patient with a positive
response to either of these questions can be asked a few further DSM IV
questions to rule in or rule out depression. This is a much more practical
solution and we are hoping to test these two questions in a clustered
randomised controlled trial in general practice.
1. US Preventive Services Taskforce. Guide to clinical preventive services
second edition Report of the US preventive services task force
(screening). International medical publishing, Alexandria Virginia 1996.
2. Williams JW, Mulrow CD, Kroenke K, etal. Case-finding for depression in
primary care: a randomized controlled trial. Am J Med 1999;106:36-43.
3. Whooley MA, Avins AL, Miranda J, Browner WS. Case finding instruments
for depression two questions as good as many. J Gen Intern Med 1997;12:439
-45.
Competing interests: No competing interests