Use of rating scales by consultant psychiatrists

The number and diversity of rating scales and tests of psychiatric status has increased dramatically in the last 20 to 30 years. Tests are in screening, e.g. for dementia the Mini Mental State Examination and the Abbreviated Mental Test; to aid clarification of the diagnosis in difficult cases, e.g. Walton-Black New Word Learning Test, to differentiate between depression and dementia; and in assessing severity of symptoms, e.g. Hamilton Depression Rating Scale.The number of rating scales for the assessment of the severity of depression has increased from the original Hamilton to at least nine specific instru ments currently available. In addition, there are at least six general instruments which include a depression scale, e.g. the depression scale of the Minnesota Multiphasic Personality Inventory. research work performed depressed cohorts, striking feature is the wide diversity in the use of rating scales. The lack of uniformity does not facilitate comparison between studies. point

The number and diversity of rating scales and tests of psychiatric status has increased dramatically in the last 20 to 30 years. Tests are in screening, e.g. for dementia the Mini Mental State Examination and the Abbreviated Mental Test; to aid clarification of the diagnosis in difficult cases, e.g. Walton-Black New Word Learning Test, to differentiate between depression and dementia; and in assessing severity of symptoms, e.g. Hamilton Depression Rating Scale.
The number of rating scales for the assessment of the severity of depression has increased from the original Hamilton to at least nine specific instru ments currently available. In addition, there are at least six general instruments which include a depression scale, e.g. the depression scale of the Minnesota Multiphasic Personality Inventory.
When one looks at clinical practice and research work performed with depressed cohorts, the striking feature is the wide diversity in the use of rating scales. The lack of uniformity does not facilitate comparison between studies. This point was addressed at the Dahlem Conference in 1982 (Kupfer & Rush, 1983). This conference was composed of a group of clinical investigators of depression. When considering the severity of illness the group rec ommended the use of at least one clinician rated method with proven reliability and validity. They considered that a self-report symptom scale may also be helpful but not essential. There was no consensus about which symptom severity rating scales should be used.
There is considerable criticism of the usefulness of rating scales, for example, that the information obtained is limited, yet unwarranted assumptions are made due to the apparent objectivity of the scales. Some psychologists regard extensive use of tests as a measure of lack of sophistication. However, estab lished research methodology would be impossible without the use of such instruments. Use of psycho logical tests and rating scales by consultant psychiatrists has not been explored to date.
The aim of this study was to determine the most frequently used rating scales by a sample of consult ant psychiatrists and to determine the characteristics of those using them.
Our hypothesis was that consultants in teaching centres are more likely to use rating scales.

The study
We used a two-stage postal survey. All consultants employed in the South West Thames region in sum mer 1989were sent a questionnaire. They were asked to rate on a I to 6 scale their familiarity and use of 32 common rating scales. They were also invited to name other rating scales and the extent to which they used them. Non rcsponders were sent a reminder. Characteristics of responders, such as subspecialty, age, qualifications, years in psychiatry, type of hospital, whether tutor or MRCPsych examiner, were elicited.

Findings
Ninety-nine consultants were surveyed and 73 ques tionnaires (74%) were returned. The most frequently used tests were:-Hamilton Depression Rating Scale (HAM-clinician rated depression scale). Beck Depression Inventory (BECK -self rated depression scale), Wechsler Adult Intelligence Scale (WAISgeneral intelligence test administered by psychol ogist), Ravens Coloured Progressive Matrices (RAV-general intelligence test administered by psychologist), Millhill Vocabulary Scale (MIL general intelligence test psychologist administered), Minnesota Multiphasic Personality Inventory (M MPI -self-administered questionnaire rating personality indices), Clifton Assessment Procedure for the Elderly (CAPE -nurse rated functional assessment), General Health Questionnaire (GHQ-self-admin istered psychological screening questionnaire), Eysenck Personality Questionnaire (EPQ-self administered questionnaire rating personality indices), Hospital Anxiety and Depression Scale (HAD self-administered anxiety and depression scale) (See Fig. 1).
Of the tests in the questionnaire, the least fre quently used were the: Sandoz Clinical Assessment -Geriatric (clinician rated mental state and functional assessment) used by 1  Four of these were found to be significant: Millhill -qualifications (P<0.02), HAM -examiner status (P < 0.001), BECK -type of hospital (P < 0.05) and Millhill -examiner status (P< 0.05).

Comment
A criticism of the study is that it was carried out in only one NHS region introducing a bias towards Rice and Donnelly certain scales because of consultants having similar opinions or being influenced by the same academic centre.
Another criticism is the selection of the 32 rating scales from current literature for the questionnaire. To find any that might have been over-looked we asked consultants to inform us of other tests they used. Nearly 50 were cited but none popular enough to be mentioned by more than two consultants.
In the analysis of the characteristics of those using the more common scales, an attempt was made to separate out consultants from more academic set tings, e.g., university hospital, tutors, examiners and those with additional qualifications. General adult psychiatrists were compared with the subspecialities, those over 39 years of age and under 20 years in psychiatry. Few of these factors were significant. Those variables that were associated may have been chance findings within the 35 analyses carried out. Our hypothesis that consultants in teaching centres are more likely to use rating scales was disproven. There was no marked difference between teaching and other centres. Nevertheless candidates being examined by South West Thames examiners would be advised to acquaint themselves with the HAM and Millhill.
The study shows that three scales are used much more commonly than others, the HAM, BECK and WAIS. This suggests that most consultants keep to a few well-known scales even though such a variety exists.