Identifying psychiatric illness among general medical patients.

The recognition of psychiatric illness by physicians and surgeons comes about either by the doctor noticing some cue that suggests psychological disorder or because the patient's complaints cannot be accounted for by known organic disease. In either case the doctor must be able to follow up with appropriate questions which will enable him to make a confident diagnosis of a psychiatric disorder. That sounds straightforward, but it is not. Over half of all medical patients with psychiatric illnesses diagnosable according to research criteria will not have their illnesses detected by the medical staff looking after them'-4-for five main reasons. Firstly, many such patients do not provide any cues, either non-verbal or verbal, that suggest a psychological disorder, though they will readily describe their symptoms if they are asked directly. Secondly, patients often mention depression or anxiety at the beginning of the interview together with their presenting somatic symptoms, yet only the latter are picked up and discussed further. Thirdly, medical histories are often taken in conditions of little privacy: patients may admit problems in a side room that were denied on the open ward.' Fourthly, a known organic cause for a patient's symptoms does not exclude a psychiatric disorder; indeed, such disorders will often be found to be responsible for exacerbating the patient's complaint. Finally, and rather sadly, even when they suspect a psychiatric disorder many clinicians are not confident of their ability to make psychiatric assessments. One approach to the problem is for the clinician to make use of the bewildering variety of pencil and paper tests that may be given to patients in order to detect those who are likely to have a psychiatric disorder. Most such illnesses turn out to be anxiety states and depressive illnesses, and the screening tests are mainly concerned with the symptoms of these. None ofthem make diagnoses, but several will provide a profile of scores in addition to a general score indicating the probability that a disorder is present. The account which follows is mainly concerned with those scales which have been well validated in general hospital settings. Since symptoms related to anxiety are correlated with symptoms related to depression many clinicians prefer a single scale that gives an indication of anxiety-depression.