4 Challenges in the diagnosis of depression and anxiety
People suffering from depression or anxiety often seek informal help at first – consulting friends, neighbours and family, and relevant websites and books. Only if the problem persists are they likely to seek professional help. Typically, the professional consulted will be the family doctor or GP.
People experiencing emotional distress may seek out a GP because they are experiencing physical symptoms such as back pain, heart palpitations, sleeping difficulties, tiredness, loss of appetite, etc. For such symptoms, GPs may need to exclude some conditions, such as hypothyroidism.
GPs often do make independent decisions about whether a patient is suffering from an emotional disorder or not, though if they feel uncertain, or if the disorder seems very serious, the patient may be referred to specialists in mental health diagnosis and care, such as psychiatrists.
Ideally, a GP would have the time and resources to carry out appropriate psychological and physiological tests, and to spend time assessing anyone who was suffering from an emotional disorder. However, as the average GP visit in the UK lasts only a few minutes, this is a counsel of perfection. Indeed a meta-analysis (see Box 5 for what a meta-analysis involves) of studies involving over 50 000 patients concluded that GPs do not recognise depression in a significant number of those who have it, and also frequently diagnose it in people who do not have it (Mitchell et al., 2009).
Box 5 Research methods: meta-analysis
A meta-analysis (‘meta’ means ‘high-level’ so meta-analysis means ‘high-level analysis’) considers the results of previous studies (published, and sometimes unpublished) on a specific topic to reach a more reliable overall conclusion. This is a very valuable process since it allows researchers to make sense of the often conflicting information that is presented by individual studies. In addition, meta-analyses can help in understanding precisely why individual studies show different or conflicting results. For instance, one study may show that Treatment X works, while another study may show that the same treatment doesn’t work. A meta-analysis might identify a variable (such as the ‘age of participants’) that explains the discrepancy: the first study may have been conducted with older people, and the second with younger people. This would suggest the possibility that Treatment X is effective with older patients but not with younger ones.
Meta-analyses need to be done carefully to try to make sure that the measures used in the different studies are comparable. For instance, if an emotional disorder such as major depression is assessed in different ways in some studies than in others, then this could confuse the results of the meta-analysis.
Mitchell et al. (2009) selected studies where GPs were making routine ‘unassisted’ diagnoses – based on their own judgment, ‘without specific help from severity scales, diagnostic instruments, education programmes, or other organisational approaches’ – that is, the way GPs normally make diagnoses.
The accuracy of the GPs’ diagnoses of depression had been assessed independently in each of the studies included in the meta-analysis, using DSM or ICD criteria for depression. Thus Mitchell and his colleagues had information about the extent to which GPs got the diagnosis of depression right or wrong.
They found that: ‘In general, a motivated GP in an urban setting (where the rate of prevalence of depression is 20%) would correctly diagnose 10 out of 20 cases, missing 10 true positives. The GP would correctly reassure 65 out of 80 non-depressed individuals, falsely diagnosing 15 people as depressed’.
Activity 6 Misdiagnosing depression
What is the percentage of true cases of depression misdiagnosed? Are these false negatives or false positives?
Answer
10 out of 20, that is 50%, of people are misdiagnosed as OK, even though they are depressed. These are false negatives.
What is the percentage of non-depressed people incorrectly diagnosed as depressed? Are these false negatives or false positives?
Answer
15 out of 80, that is 18.75%, of people who are not depressed are incorrectly diagnosed as depressed. These are false positives.
The number of people misdiagnosed as false negatives or false positives (from above, 15 + 10 = 25 out of 100, 1 in 4, or 25%) is therefore substantial. Moreover, as GPs prescribe antidepressant drug treatment and make referrals for counselling and therapy, this means some people who need treatment will not be offered it, while others may be prescribed treatment they do not need.
Where drug treatments are offered to false positives, this can be problematic, as drugs typically have side effects and can be difficult to come off. Fortunately the evidence also suggests that GPs are less likely to misdiagnose serious cases of depression (Mitchell et al., 2009).
The above should not be seen as a criticism of the diagnostic abilities of GPs. As Tyrer (2009) points out, the diagnosis of depression is fraught with difficulty even for experts, so it is not surprising that misdiagnosis, especially of milder and moderate cases of depression, occurs. The fact that depression is often mixed with anxiety, as we consider below, may make diagnosis even trickier.
Activity 7 Factors affecting diagnosis of emotional disorders in a primary care setting
Take a few minutes to think about, and make a list of, factors that might help a GP (working in the normal way, without assistance) to assess a patient more accurately for an emotional disorder such as depression. This activity should help you appreciate the factors at play in a primary care setting that may impact on diagnosis.
Discussion
You may have thought of some of the following factors or come up with others:
- The GP is able to spend enough time with patient to probe in a sensitive way and explore if there are any underlying issues (e.g. somatic or social) if the patient appears upset or worried.
- The GP is sensitive to emotional signals from the patient.
- The GP has specific mental health training.
- The GP knows of, and can apply, the biopsychosocial approach.
- The patient recognises, and is willing and able to speak about, his/her personal emotional distress.
- The GP is familiar with the patient (i.e. the patient is not a new, but is a regular visitor to the surgery) so is aware of what is normal or not for the patient.
- The GP has known the patient or patient’s family for a while and knows the patient’s medical issues.
- Having one of the patient’s friends or family members present – they may be able to provide another perspective on the patient’s condition.
Some people complaining of somatic symptoms may not be sensitive to, or may be unwilling to acknowledge, emotional suffering in themselves. Some may feel that somatic symptoms will be taken more seriously by the medical profession, or be more amenable to medical treatment. Hence they may be more likely to mention these to a doctor than feelings of anxiety and depression. Personal and cultural values may come into play here, too – for instance, the evidence suggests that there may be a gender difference, with men less likely to report emotional distress than women.