Should we screen for depression?BMJ 2006; 332 doi: https://doi.org/10.1136/bmj.332.7548.1027 (Published 27 April 2006) Cite this as: BMJ 2006;332:1027
- Simon Gilbody, senior lecturer in mental health services research1 (, )
- Trevor Sheldon, pro vice chancellor1,
- Simon Wessely, professor of epidemiological and liaison psychiatry2
- 1 Department of Health Sciences, University of York, York YO10 5DD
- 2 Academic Department of Psychological Medicine, Guy's, King's, and St Thomas's School of Medicine, Institute of Psychiatry, London SE5 8AF
- Correspondence to: S Gilbody
- Accepted 8 March 2006
The quality and outcomes framework will soon reward primary care doctors who screen for depression in England and Wales. This article scrutinises the rationale and evidence to support such screening
“All screening programmes do harm; some do good as well.”
Depression is common in primary care and hospital settings, but it is often not recognised by healthcare professionals.2 3 This has led to calls for screening programmes to aid detection and management.4 We use the criteria of the UK National Screening Committee to judge whether screening would do more good than harm.5 We drew on our experience in preparing a Cochrane review of the evidence for screening for depression.6
Depression screening as national health policy
In the United States screening for common mental health problems is thought to be effective and is a cornerstone of the agenda to improve mental health; population level screening programmes are supported by the drug industry.7 w1 w2 Similar national programmes have been advocated in Australia.w3 In England and Wales, screening has been supported more cautiously by the National Institute for Health and Clinical Excellence (NICE), which recommends that it should be offered to people at high risk of depression.w4 Screening may become health policy in England and Wales, since primary care doctors will be rewarded for “enhanced services for depression” within the quality and outcomes framework (QOF), which will include a screening programme.w5
In the past screening programmes have been implemented without due consideration of their effectiveness, their ethical and clinical implications, and their impact on finite healthcare resources.8 w6 Consequently, the National Screening Committee has been established in the United Kingdom; this committee works to specific criteria to help ensure that screening “does more good than harm.”5 These criteria pertain to the condition, the test, the treatment, and the screening programme.
The key National Screening Committee criteria are:
The condition should be an important health problem
The epidemiology and clinical course of the disease should be adequately understood.
Depression qualifies as a major public health problem, with an annual incidence of 8-12%.w7 Reductions in quality of life are comparable to those seen in major chronic physical diseases,w8 and the economic consequences of depression are profound; £8bn (€11.5bn; $14bn) each year in the UK and $83bn in the United States.w9 w10
The epidemiology, clinical course, and consultation patterns of people with depression are well understood.9 Surveys consistently show that a substantial proportion of patients with depression are missed by clinicians.10 w11 However, more recent longitudinal studies have shown that many of these patients are identified later during the course of consultations.11
Cross sectional surveys of depression tend to pick up depression and transient distress, found in response to psychosocial problems and life events.12 When these populations are followed up, a substantial proportion of people identified as a “case” by screening will have symptoms that resolve within two to four weeks. Thus for mild to moderate depression, evidence based treatment guidelines recommend an initial period of watchful waiting before active intervention.w4
Depression is thus a potential target for screening in terms of population morbidity. However, because of the transient nature of mood changes in many people, screening might detect large numbers of false positives. Because many “missed” cases are identified during later visits, screening may not be efficient.
The key National Screening Committee criteria are:
The screening test should be safe, simple, precise, and validated; a suitable cut-off value should be defined and agreed
The test should be acceptable to the population.
A variety of standardised questionnaires or brief consultation questions are available to detect depression, and these tests have reasonable psychometric properties (for questionnaires: median sensitivity 75%, median specificity 85%).13 14 However, the low prevalence of depression (< 10%) means that even sensitive and specific instruments will have low positive predictive value (< 50%).w12 Low positive predictive value will make the test less acceptable to clinicians, because patients will be followed up unnecessarily.w13
The assumption is that tests are acceptable to patients and clinicians, but this has not been well researched. Indirect evidence of poor acceptability comes from two sources. With respect to acceptability to patients, the uptake of screening tests for depression is generally low when they are offered in healthcare settings: 30-60% of patients in primary care decline to participate in clinic screening interviews offered by researchers or clinic nurses during routine attendance.15 w14 With respect to acceptability to healthcare professionals, questionnaires are rarely used once trials of practice based screening have ended.w14 The degree to which additional consultation questions are adopted or implemented in practice has not been evaluated and cannot be assumed on the basis of validation studies alone.14
Thus, the criteria of the National Screening Committee regarding test performance and acceptability are not clearly met.
The key National Screening Committee criteria are:
An effective treatment should be identified through the screening programme, with evidence that early treatment leads to better outcome
Clinical management of the condition and patient's outcomes should be optimised for all healthcare providers before the screening programme is offered.
The effectiveness of drugs and psychological interventions for depression is now established and forms the basis of evidence based guidelines.16 w4 However most guidelines focus on moderate to severe depression. In general, patients with undetected depression have milder forms of depression, which often resolve without intervention, than patients with identified depression.2 17w15 Psychological intervention and drugs are not as effective when mild depression persists compared with moderate depression.w4 The outcomes of patients with detected and undetected depression are similar when they are followed up over 6-12 months.w15-w17
The quality of care for patients with recognised depression falls short of evidence based guidelines.w18 Clinicians prescribe subtherapeutic doses and do not continue drugs for long enough to prevent relapse. Follow-up is poor: patients do not return for repeat prescriptions or for assessment of the response to treatment, and non-adherence with treatment is common.
Thus, the criteria of the National Screening Committee that benefit for patients as a consequence of screening and that optimised treatment should be in place before implementation of a screening programme are not met.
The screening programme
The key National Screening Committee criteria are:
High quality randomised controlled trials should provide evidence that the screening programme effectively reduces morbidity
The screening programme should be clinically, socially, and ethically acceptable to health professionals and the public
The benefit from screening should outweigh the physical and psychological harm
The cost of the screening programme should be economically balanced in relation to expenditure on medical care (value for money).
Most importantly, screening programmes should be shown to improve the detection, management, and outcomes of depression. Several randomised studies have failed to show that treatments are effective,18 w12 but industry sponsored observational studies have generally been more positive.19w14
Our Cochrane review (based on more than 6000 patients) concludes that routine feedback of the results of screening to clinicians results in a marginal increase in the rate of diagnosis of depression.6 However patients' outcomes are not improved at 6-12 months as a consequence of screening. These results need to be considered alongside the results of an earlier review conducted on behalf of the US Preventive Services Task Force, which was more supportive of screening programmes, especially within a comprehensive primary care programme for managing depression.7w19 One influential study in the task force report recruited patients via a practice screening programme and offered enhanced care, consisting of face to face education of patients, telephone support, management of drugs, psychotherapy, and structured follow-up.20 Clinicians were also offered guidelines, practice based education, and face to face support from specialists. Screening was only one element of this complex intervention, and positive outcomes cannot be assumed to be due to screening alone.w20 Thus, evidence is scant that screening alone results in improved care and outcomes; this is a key element of the National Screening Committee criteria.
UK guidelines on depression mention screening of high risk groups, such as patients with chronic physical illness, alcohol problems, or a history of depression.w4 Such a strategy seems appealing, but in our Cochrane review we found no studies that evaluated this strategy.6 Such a programme would be more complex than screening all attendees. High risk patients would have to be identified by people who deliver the screening (practice nurses, researchers, or receptionists). In the absence of randomised data to support this strategy, it is hard to support this approach.
A favourable benefit to harm ratio is another criterion of the National Screening Committee. The benefits that might be expected are minimal. The US Preventive Services Task Force found no empirical data on the harms of screening.7 Potential harms include the stigma associated with depression; the risk of labelling transient distress as illness; and probable discrimination by insurance companies. Identifying a large number of patients with undetected depression could divert resources from patients with greater need, who would benefit more. Screening also increases the length of time needed for consultation in primary care (average eight minutes for patients who have positive results on screening), when follow-up interviews and further diagnostic investigations are required.19w14
Screening programmes must be cost effective if they are to take priority over competing interventions. The Cochrane review found no randomised studies on cost effectiveness, and decision modelling has been used in the absence of prospective data.21 Several criteria would need to be met for screening to have a cost utility below $50 000 per quality adjusted life year. Administration, scoring, and feedback for screening instruments (printing, administrative staff time, and increased doctor time) would need to cost less than $3.00 (£1.80) per patient. The prevalence of depression would need to be more than 13% (higher than is usually seen in primary care). Screening would need to result in intervention in more than 80% of patients, and therapeutic benefit and remission would need to be seen in more than 85% of patients who screened positive. These criteria are unlikely to be achievable.
Screening alone cannot improve the management and outcome of depression, and the ratio of costs to benefits is unacceptable. This does not mean that screening has no part to play. Several studies have shown that integrated management programmes for depression (some of which incorporate screening) are more effective than usual care. Thus, the systems in place to manage depression in primary care and general hospitals are inadequate.w18 Collaborative care, case management, and stepped care are underpinned by randomised evidence and are promising candidates for integration into usual care.22 Screening patients who have concurrent physical illness, such as diabetes (as suggested by NICE guidelines), is an effective strategy, but only when used within a collaborative care system.23 However the individual contribution of screening as an active ingredient in individual packages of care is not clear. For many people, depression is a life-long and relapsing condition.w20 Population strategies aimed at reducing morbidity are more efficient when targeted at minimising the chronic effects of existing depression, rather than identifying more minor psychiatric morbidity.w21
Opportunistic screening and population level screening for depression do not fulfil the criteria of the National Screening Committee. However, the assumption has been made that screening for depression should be recommended, based on the prevalence of the disorder; the psychometric properties of screening tools; and the availability of effective interventions in the form of drugs.17 The criteria of the National Screening Committee provide an analytical framework that helps focus discussions on how to improve the inadequate management of depression. Screening for depression is an unhelpful diversion from more fundamental questions about the most efficient and effective way of organising and delivering care.24 w18 Screening should only be considered as part of a package of enhanced care. Without this, moves to implement screening will be associated with increased costs and no benefit.
Depression is common in patients in primary care and hospital settings but often is not recognised by healthcare professionals
Opportunistic screening and population level screening for depression have been supported in recent policy recommendations in the US and UK
The UK National Screening Committee has issued clear criteria, and all screening programmes should be judged against these criteria before implementation
The use of these criteria indicates that screening for depression is unlikely to be a clinically effective or cost effective way to improve the mental wellbeing of the population
Editorial by Scott and p 1030
Contributors and sources SG and TS have a long research interest in screening for mental health problems in primary care, and SW has an interest in the value of screening in general hospitals and in military populations. TS is interested in the evaluation of screening as health policy. This article arose from a shared interest and exchange of views regarding screening, in light of recent policy developments. SG and TS recently prepared a Cochrane review of screening for depression, which required extensive searches of a comprehensive range of databases. SG, TS, and SW formulated the ideas, and SG produced the first and subsequent drafts after contributions and comments from TS and SW. SG is guarantor.
Competing interests None declared.
Additional references w1-w22 are on bmj.com