Correspondence

differences such as clinical remit, *Alex J Mitchell, Amol Vaze, evident at all after one year.2 Later
therapeutic relationship, list size, and Sanjay Rao studies confirmed this.3
appointment duration. [email protected] It is also mistaken to believe that
The latter is picked up by Suresh Department of Liaison Psychiatry, Leicester General treatment guidelines accompanied
Pathak in his impressively large UK Hospital, Brandon Unit, Leicester LE5 4PW, UK by discussion groups will do much to
study against the 12-item General 1 Mitchell AJ, Zimmerman M, McGlinchey JB, improve the situation, as the negative
Young D, Chelminski I. Accuracy of specific
Health Questionnaire. We deliberately symptoms in the diagnosis of major depressive result of the Hampshire Depression
excluded disorders defined by use of disorder in psychiatric out-patients: data from Study showed.4 What is needed is
the MIDAS project. Psychol Med 2009;
the General Health Questionnaire 39: 1107–16.
video feedback of actual interviews
from our meta-analysis (preferring 2 Zimmerman M, Galione JN, Chelminski I, et al. between GPs and their patients, with
semi-structured interviews), but we A simpler definition of major depressive behavioural rehearsal of skills that
disorder. Psychol Med 2009; published online
have since analysed all such studies July 23. DOI:10.1017/S0033291709990572. assist in the detection of emotional
involving distress and mild depression 3 Bushnell J. Frequency of consultations and distress.5
and found that GPs have even more general practitioner recognition of There are important reasons why
psychological symptoms. Br J Gen Pract 2004;
difficulty identifying less severe 508: 838–42. GPs miss depression, and they involve
conditions. In collaboration with 4 Mitchell AJ. Reluctance to disclose difficult both GPs and their patients. Some
diagnoses: comparison of psychiatrists and
colleagues at Brown University, we oncologists. Support Care Cancer 2007;
GPs regard the detection of distress
have attempted to identify the most 15: 819–28. as a much less important secondary
influential symptoms contributing 5 Walters K, Buszewicz M, Weich S, King M. task to detection of physical disease,
Help-seeking preferences for psychological
to depression and to simplify current distress in primary care: effect of current and many patients are also mainly
criteria.1,2 mental state. Br J Gen Pract 2008; 58: 694–98. concerned with excluding a serious
Pathak’s comments also highlight physical disease as the cause of their
an additional, much overlooked symptoms. Missing depression often
factor underlying diagnostic errors, Are general practitioners involves tacit collusion between GPs
namely willingness to disclose and and patients, with the patient not
document. In some cases, clinicians
unable to diagnose wanting to admit their distress, and
identify patient-reported symptoms depression? preferring to deal with it themselves.
but do not believe psychological I declare that I have no conflicts of interest.
issues are clinically significant.3 In his provocative Comment (Aug 22,
Related to this finding, most p 589),1 Peter Tyrer cites our paper2 and
David Goldberg
[email protected]
clinical studies highlight the lowest argues that “benefit seemed to accrue
Institute of Psychiatry, King’s College London,
recognition rates in studies that from a fairly simple intervention:
London SE5 8AF, UK
use medical records (chart review). the administration of a short
1 Tyrer P. Are general practitioners really unable
Clinicians are reluctant to deliver questionnaire”. This is an important to diagnose depression? Lancet 2009; 374:
difficult diagnoses and hesitant to error: the questionnaire was merely 589–90.
record such diagnoses in records for a screening device used to identify 2 Johnstone A, Goldberg D. Psychiatric screening
in general practice: a controlled trial. Lancet
fear of error, introducing stigma, and patients with hidden psychiatric 1976; 1: 605–08.
upsetting the patient.4 disorders. Those identified went on to 3 Goldberg DP, Privett M, Ustun B, et al. The
effects of detection and treatment on the
Where there are inadequate treat- be randomly assigned intervention— outcome of major depression in primary care:
ment options, reluctance to disclose even if this was merely a discussion of a naturalistic study in 15 cities. Br J Gen Pract
is perhaps understandable, but, for the patient’s current life problems—or 1998; 48: 1840–44.
4 Thompson C, Kinmonth AL, Stevens L, et al.
depression, effective treatment is no intervention. What helped patients Effects of a clinical-practice guideline and
nearly always available, provided help was therefore the general practitioner practice-based education on detection and
outcome of depression in primary care:
is offered and the patient is willing (GP), not the questionnaire. Hampshire Depression Project randomised
to accept it. Yet research suggests Much has been learned since those controlled trial. Lancet 2000; 355: 185–91.
that only about half of those with early days: depressions detected 5 Goldberg D, Steele J, Smith C. Training family
doctors to recognise psychiatric illness with
distress will accept an initial offer by GPs are more severe than those increased accuracy. Lancet 1980; 2: 521–23.
of professional help.5 Hence one that they miss, and such patients
of the most important aspects of are likely to have a worse previous Author’s reply
depression management is to agree history and to have been seen many In my Comment I did state that, in
an intervention that is acceptable to times.3 In the original study, results the original study,1 benefit seemed
the patient. of detection of distress were most to accrue from the administration of
We declare that we have no conflicts of interest. marked between 3 and 6 months a short questionnaire, but I followed
after the consultation, and were not this by explaining that the better

1818 www.thelancet.com Vol 374 November 28, 2009