Mini-Mental State findings from an age 60+ random community sample (N = 1681) indicate that score is related to education, age and race (but not sex) and to functional status, but not to selected aspects of physical or mental health. Adjustment for demographic characteristics, particularly education, is recommended lest cognitive impairment be overestimated.

Depressive thoughts were commonly reported by a sample of patients attending a south London general practitioner. Their occurrence corresponded more closely with a psychiatrist's than the general practitioner's diagnosis of depression; a ‘scale’ composed of questions about depressive thinking discriminated well between depressed and non-depressed men, but was a much poorer discriminator in the women.

This paper describes the development of a self-rating scale to measure adjustment to cancer. The Mental Adjustment to Cancer (MAC) Scale has been completed by 235 patients with various types and stages of cancer. It is easily understood by, and acceptable to, patients. Data relating to reliability and internal consistency indicate that the MAC scale is a valuable research tool for clinical studies of patients' psychological adjustment to cancer.

This paper examines the power of the sporadic v. familial method as applied to schizophrenia and major depression. The model used assumes aetiological heterogeneity with a subpopulation of cases due to a ‘major’ environmental event and the remainder resulting from a generalized single major locus. The findings suggest that, for sample sizes to which it is commonly applied, the sporadic v. familial classification has low power to detect aetiological heterogeneity. When applied to nuclear families, substantial power requires at a minimum 100–150 proband families. If the proportion of environmental cases in the population is low, or the ‘test’ for environmental aetiology in probands does not have high sensitivity and specificity, the required sample sizes are considerably larger. Adding monozygotic twins increases the power of the method, but including second-degree relatives does not. The optimal approach to the sporadic v. familial method will differ as a function of the frequency of the disorder and the relative effort and expense of examining probands versus family members. Other methods should be considered for discriminating genetic and environmental forms of illness.

The impact of the transition to parenthood for dual-earner couples was examined in a longitudinal study. Data were collected by questionnaire from couples expecting their first child, during pregnancy and after maternity leave, and from childless couples at two separate time points. The impact of the transition was minimal overall, although women experienced more pressure than men and there were some cases of extreme stress. New single-earner parents reported a greater decline in life satisfaction than dual-earners.

The factor structure of the 60-item version of the General Health Questionnaire was explored, using data collected in a community study in a rural area of northern Spain. Six principal components, similar to those previously reported with this instrument, were found to provide a good description of the data structure.

The 30-item and 12-item versions of the GHQ were then disembedded from the parent version, and further principal components analyses carried out. Again, the results were similar to previous studies: in each of the three versions analysed here, the two most important components represented a disturbance of mood (‘general dysphoria’)– including aspects of anxiety, depression and irritability– and a disturbance of social performance (‘social function/optimism’).

The principal component structure of the GHQ-60 was then utilized to calculate factor scores, and these were compared with PSE ratings using Relative Operating Characteristic (ROC) analysis. While four of the six factors discriminated well (area under the ROC curve 0–75 or more) between PSE ‘cases’ and ‘non-cases’, only one, depressive thoughts, was a good discriminator between depressed and non-depressed PSE ‘cases’.

The value of a new method of scoring the General Health Questionnaire (CGHQ) has been assessed in a representative sample of 6317 British community residents aged 18 +. The CGHQ, which takes account of the chronicity of psychiatric symptoms, is found to be superior to the conventional scoring method in yielding a wider range of scores, a more normal distribution and in its close association to a well validated measure of neurotic illness.

On the other hand, the conventional GHQ and the CGHQ score appear equally sensitive to a number of socio-demographic variables, including social class, marital status and employment. An important finding is that the two methods identify different individuals as cases. Compared with the GHQ, the CGHQ identifies a higher proportion of the middle-aged. The implications of these findings with respect to the optimal method of scoring the GHQ are discussed.

Less educated elderly people are commonly found to perform more poorly on the Mini-Mental State Examination (MMSE). This educational level difference has been attributed by some research workers to test bias. To assess whether the MMSE is biased against the poorly educated, its validity was assessed separately in the more- and less-educated members of a community sample. No evidence was found to indicate that the test is a biased measure of cognitive impairment.

A structured respondent-based version of the SADS-L interview was compared with a semi-structured investigator-based version of the same interview in terms of efficiency in assessing lifetime psychopathology. There was substantial agreement across most Research Diagnostic Criteria (RDC) categories, but there was less agreement on the rating of symptoms. The implications of these similarities and differences between the techniques are discussed.

In a defined population of 112000, an attempt was made to discover every patient with multiple sclerosis. Using strict diagnostic criteria, 91 patients were identified. A search of psychiatric records for the same population revealed that 15 (16%) of these patients were referred to psychiatrists between the onset of their symptoms and the diagnosis of multiple sclerosis. Ten (11 %) were seen with symptoms attributable to multiple sclerosis. These symptoms were recognized as neurological in only two cases, while the other eight patients received a variety of psychiatric diagnoses.

Possible reasons for diagnostic error included the subjective nature of many early symptoms, histrionic behaviour, and psychiatric disturbance which drew attention away from physical symptoms. The results underline the caution needed when patients with physical symptoms are referred for psychiatric assessment.

Three studies are reported using the Geriatric Mental State in one of its community forms on a total of 647 subjects aged over 65 and living in their own homes. The concordance between the computer diagnosis AGECAT and psychiatrists' diagnosis is at least as good in these replication samples as in the original studies (Copeland et al. 1986). AGECAT has been shown to be useful for epidemiological surveys and as a diagnostic guide for non-medical interviewers.

Visual Analogue Scales (VAS) provide a simple technique for measuring subjective experience. They have been established as valid and reliable in a range of clinical and research applications, although there is also evidence of increased error and decreased sensitivity when used some subject groups. Decisions concerned with the choice of scoring interval, experimental design, and statistical analysis for VAS have in some instances been based on convention, assumption and convenience, highlighting the need for more comprehensive assessment of individual scales if this versatile and sensitive measurement technique is to be used to full advantage.

Alternative procedures for coding nonresponse on the Mini-Mental State Examination (as error or as correct) results in different classification of 13% of a random sample of 1931 subjects aged 60 and over. Comparison of responders' and nonresponders' ability to perform activities of daily living, and examination of the relative difficulty level of omitted items indicates that, in epidemiological surveys, scoring nonresponse as error is more likely to be correct.

This paper discusses the relevance of assessing the nature of intimate relationships and reports on the development of such an instrument. The Intimate Bond Measure (IBM) is a self-report measure assessing two key underlying dimensions, care and control. Its properties are assessed in separate studies, establishing its high test-retest reliability, the homogeneous nature of the isolated dimensions, its insensitivity to broad socio-demographic influences and its minimal sensitivity to depressed mood state. Support for its validity, in terms of both perceived and actual characteristics of care and control, is demonstrated. It provides a simple and efficient measure of central constructs underlying intimate relationships, and is of potential use in studies attempting to assess the relevance of intimate relationships to the onset and course of psychiatric disorders.

In a study of pre-morbid personality in 56 head-injured subjects, subjects' self-ratings of pre-morbid personality were compared with informants' ratings of the subjects' pre-morbid personality on two personality questionnaires (the Eysenck Personality Inventory (EPI) and the Marke-Nyman Temperament Scale (MNTS)). Correlations between self-ratings and informantratings were positive and significant for all three MNTS and for EPI Extraversion and Lie scales, but not for EPI Neuroticism, where the lack of subject-informant correlation was attributed to contamination of the self-rating of the trait measure by current abnormalities of mental state. Further analyses supported previous evidence that the MNTS ‘Validity’ scale may predict the development of psychiatric symptoms.

This paper describes the development, validation and use of a computerized assessment for minor psychiatric disorder based on the Clinical Interview Schedule (CIS; Goldberg et al. 1970). There was good agreement between the computerized assessment and the CIS administered by psychiatrists, both in assessing overall severity and in defining ‘cases’ of psychiatric disorder. Individual symptoms elicited by the computer and the CIS were compared, and the levels of agreement found were similar to those from inter-observer studies of standardized interviews. Subjects from a variety of non-psychiatric settings regarded the assessment as acceptable, accurate and easy to use. It is concluded that this computerized assessment of neurotic symptoms is valid and reliable. It eliminates observer bias, it is an efficient use of research resources and it may have clinical applications in primary care.

One-hundred-and-seventy-nine women attending their first antenatal clinic at two general practices in South London were asked to complete the 30-item General Health Questionnaire (GHQ-30) as part of a larger prospective longitudinal study into childbirth-related affective disorders. The women were later interviewed at home and a standardized psychiatric interview administered. Relative operating characteristic (ROC) curves were used to determine the optimum cut-off point on the GHQ-30. Thirty-five per cent of the women were high scorers on the GHQ-30 (≥6) and 29% were found to be ‘cases’ at interview (CIS Overall Severity Rating ≥2).

Evidence is summarized suggesting that existing self-report questionnaires for the assessment of Type A behaviour pattern suffer from major methodological shortcomings. Type A individuals may be unwilling or unable to accurately self-report. The female spouse of the male patient can potentially provide a more objective, accurate ‘trait’ description less contaminated by ‘state’ variables such as diagnosis of illness or medication. The development of a spouse-report questionnaire is described and pilot data from 41 couples are presented, including comparison of the self-report and spouse-report responses on 46 questionnaire items and five scales. Unexpectedly high levels of agreement were found between self and spouse reports. Possible explanations of such agreement are critically examined.

The Psychological Stress Evaluator (PSEV) is a machine for detecting signs of stress in the voice. It was tested using tape recordings of negotiators during a prolonged terrorist siege in which a number of death threats were issued. The study aimed to test whether the PSEV could distinguish between the traces of terrorist voices before the death threats from voice traces at other points. A scoring system using the proportion of voice oscillations falling above defined amplitude levels was used. PSEV scores did rise significantly more just before the actual shooting of a hostage. Unfortunately, the apparatus proved to be slow and cumbersome in practice and of limited, or no, use during the course of incidents similar to the one analysed.