Screening for depression in primary care is under increasing scrutiny. The first of the ‘key priorities for implementation’ in the NICE guidelines for depression reads as follows:
‘Screening should be undertaken in primary care and general hospital settings for depression in high-risk groups — for example, those with a past history of depression, significant physical illnesses causing disability, or other mental health problems such as dementia.’1
There are two problems with this recommendation. The first is that it is not entirely accurate to refer to this process as ‘screening’. Depression does not meet the widely accepted criteria for a disease that is appropriate for screening, in that:
‘Treatment given before symptoms develop should be more beneficial in terms of reducing morbidity or mortality than that given after they develop’.2
There is no preclinical phase in depression as far as we know. We can only justify screening if we can argue that the ‘undetected’ form of the disease is prevalent, and that its detection and treatment in primary care attenders reduces morbidity or mortality.
For this reason it would have been more accurate if the NICE guidelines1 had referred to their recommendations as ‘case-finding’ in high-risk groups, rather than ‘screening’. In a study from New Zealand published in this Journal,3 The MaGPIe Research Group found that screening all attenders in a primary care setting was far less effective than targeting patients who were at higher risk of depression. In their sample, one additional patient with DSM-IV disorder would be identified for every 20 patients screened: GPs were effective at identifying mental health problems in the patients they knew. However, if GPs screened patients they were seeing for the first time, one additional patient with DSM-IV disorder would be identified for every five patients screened.
The second problem is that there is good evidence from a systematic review of randomised controlled trials that the routine administration and feedback of psychiatric screening questionnaires has little effect on the psychosocial outcomes of those with psychiatric disorder managed in non-psychiatric settings.4
The merits of screening for depression in primary care have been researched and debated in the US for some time. While the US Preventive Services Task Force concluded: ‘The benefits of screening are likely to outweigh any potential harms’, it also acknowledged that:
‘Larger benefits have been observed in studies in which the communication of screening results is coordinated with effective follow-up and treatment’.5
The larger benefits were associated with models of effective follow-up and treatment, such as those described by Wells et al.6 In addition to feedback of the results of screening, the model included the provision of educational materials, assistance in treatment initiation and maintenance and access to nurse-led medication follow-up, or to cognitive behavioural therapy. Weingarten et al found that complex interventions of this sort were associated with improvements in ‘disease control’ in a number of disease management programmes, including depression.7 Such interventions go far beyond ‘screening’.
It is difficult to argue the case for screening in view of the lack of evidence for its effectiveness unless it is linked to complex interventions. The strategies recommended by NICE and The MaGPIe Research Group are likely to give a larger yield of undetected cases.1,3 However, the benefits of targeted case finding in high-risk groups have not been proven in terms of improved outcome for patients.
In our efforts to increase the rate of diagnosis of depression in primary care we may be forgetting the views of the patients. Historically many of the studies that described relatively low rates of detection of depression in primary care were cross-sectional in design.8 The argument has been that many patients present to their GPs with somatic symptoms of depression and that the underlying psychological disorder is ‘missed’. In longitudinal studies GPs identify a much higher proportion of depressed patients.9 This fits with the findings of The MaGPIe Research Group, that in routine practice:
‘GPs are effective at identifying mental health problems in patients they know.’3
This tells us what many GPs know already: that the diagnosis of a mental health disorder is a process that often takes more than one consultation and evolves in a context of trust. Both patient and doctor may need to be sure that the somatic symptoms of depression are exactly that, and not the symptoms of an underlying physical illness. What is often referred to as somatisation is not necessarily a fixed attitude, but can be an understandable expression of anxiety, or indeed, of the patient's priorities at that time.
Any drive to increase the rate of detection needs to be seen in a wider context. Between 1991 and 2002, prescriptions per head for all antidepressants in the UK increased 2.8-fold and the total cost (adjusted for inflation) increased by £310 million.10 There are several possible reasons for this, including the drive by the Royal Colleges of General Practitioners and Psychiatrists in the ‘Defeat Depression Campaign’ to identify and treat more cases of depression; an increase in the number of indications for antidepressants; the unprecedented publicity given to the serotonin-selective reuptake inhibitors, and the promotion of these drugs by the pharmaceutical companies; and perhaps, a greater openness about depression and an accompanying willingness to seek help. However, the fact that help has often come in the form of antidepressants may not be a response to the patients' agenda: an opinion poll among lay people in 1996 found that 85% believed counselling to be effective but were against antidepressants, and that 78% of those questioned regarded antidepressants as addictive.11 GPs may feel they have little else to offer their unhappy patients.
All the evidence suggests that we do not need to identify more cases of depression in primary care, but rather, increase the effectiveness of our management of those that have been identified. Kendrick et al found that although GPs prescribed antidepressants on the perceived severity of the depression, their ratings did not agree well with a validated screening instrument, and their assessment of patients' attitudes to treatment were only moderately related to patients' self-reports.12 In other words, we may not be delivering antidepressants to those who are most likely to benefit from them, and our assessment of our patients' attitudes to treatment are not as accurate or sensitive as we would wish. This means that we need to look more closely at the diagnostic criteria that GPs use to inform their management decisions, at who is being prescribed antidepressants, and what is happening to them. The multifaceted interventions described by Weingarten et al7 include provider and patient education and feedback, and structured follow-up. This model of chronic disease management for asthma and diabetes is now a part of primary care. We cannot afford to ignore the evidence that this approach may be at least as effective in depression.
- © British Journal of General Practice, 2005.