The medical schools of origin of 531 psychiatrists who had qualified in the U.K. in the period 1961–1970 were ascertained. Psychiatrists were for the purpose of the study operationally defined as those who had taken the Royal College of Psychiatrists' Membership examinations since its inception. There were large discrepancies, as much as five-fold, between different medical schools, and no ready explanation is forthcoming. There were also discrepancies in the proportion of women graduates opting for psychiatry.

One-hundred husbands, diagnosed as suffering from alcoholism, and their wives, were followed up twelve months after initial consultation and assessment. Follow-up information was complete in 89 cases. On the basis of both husband and wife accounts of the husband's drinking behaviour during the follow-up period, and their assessment of the drinking problem at twelve-months follow-up, 28 were classified as having a ‘good’ outcome and 29 as having a ‘bad’ outcome. In the remaining 32 cases outcome was considered ‘equivocal‘.

A composite measure of marital cohesion was predictive of twelve-month outcome classification, cohesive marriages being significantly more likely to have a good outcome. The measure of marital cohesion was based upon husband and wife reports of mutual affection and of husband involvement in family tasks, favourable spouse perceptions and meta-perceptions, and optimism about the future of the marriage. Composite measures of dominance balance within the marriage were not predictive of outcome.

Husband's job status, husband's self-esteem, and wife's reported hardship were not independent of marital cohesion, and were themselves predictive of twelve-months outcome. When these variables were partially controlled it was found that marital cohesion remained predictive for husbands with relatively low status jobs and husbands with relatively low levels of self-esteem. It is an over-simplification to state that either the marriage, the spouse, or the severity of the patient's condition is alone the cause of variation in outcome.

It is possible to integrate these findings with those of other studies on the influence of family variables on the outcome of conditions other than alcoholism. Together these studies suggest a general hypothesis linking a breakdown in the cohesiveness, or mutual rewardingness, of family relationships and unfavourable outcomes following treatment or consultation for psychological disorder.

Seventy-three male alcoholics permitted information to be obtained from official sources about time recorded as lost from work in receipt of sickness or unemployment benefits and about their weekly state insurance contributions. The average yearly time loss was 121.7 working days per person, comprising an average yearly loss through sickness of 86.1 and through unemployment of 35.6 working days respectively. By contrast the recorded national sickness loss for men in a comparable twelve months period averaged 15.9 working days per person.

Thirteen alcoholics showed over five years, prolonged deficiency in work attendance. State benefits to the subjects, over twelve months during the early 1970s, totalled £18,434.80. Diagnoses on their medical certificates underestimated incapacity from alcoholism.

In a double blind trial, 50 male chronic alcoholic patients were treated with either fenfluramine in a dose of 60 mg or 120 mg daily, or with identically prepared placebo tablets. Patients were interviewed on admission to the trial and then at four-weekly intervals for a period of one year and blood levels of delta-aminolaevulinic acid dehydratase (ALAD) and gamma-glutamyl transpeptidase (γGT) and fenfluramine were determined. The efficacy of fenfluramine at the two dose levels was compared with placebo on the basis of the number of lapses indicated by the clinical history and also by alterations in the biochemical indices.

Twenty-seven patients completed the period of observation, there being 9 in each of the three groups. Those receiving 120 mg fenfluramine daily showed significantly fewer lapses than either of the other two groups (p < 0.01) on biochemical but not on clinical criteria. Overall assessment revealed that 3 of the 9 patients receiving the high dose of fenfluramine had a good result during the period of the trial, but there were none in the 60 mg group or in those receiving placebo. More extensive trials of fenfluramine in the treatment of chronic alcoholism are indicated.

A rational approach to the use of monoamine oxidase inhibitors (MAOIs) is outlined. Patients suitable for treatment cannot be classified adequately using conventional diagnostic labels. They include those with primary symptoms of hypochondriasis, agoraphobia and social phobias, irritability, somatic anxiety and anergia; those with primary depressed mood, guilt, ideas of reference and personality disorders seldom respond. There is great variation in the interval between the first administration of these drugs and clinical response, and this may account for the inconsistencies in published trials. The type of drug and its dose may affect rate of response, as may biochemical factors, including acetylator and monoamine oxidase status. To obtain maximum benefit, a course of therapy with MAOIs should last for several months.

Two-hundred-and-seventy-nine patients (103 men, 176 women) were followed-up 1–2 years after an act of non-fatal deliberate self-harm. Of 155 patients offered a psychiatric out-patient appointment at the time, only 68 completed the treatment. A further act of deliberate self-harm was committed by 26 men and 41 women within twelve months. The factors most highly associated with repetition were previous psychiatric treatment, a previous act of deliberate self-harm, and a criminal record. These factors held good for a separate series of patients. Significantly more repeaters received prolonged psychiatric care after the initial episode of deliberate self-harm. The implications of these findings for the clinical management of such patients are discussed.

Temporal stability of symptom patterns was examined in 33 depressives studied initially at the height of illness and again some months later in a relapse following earlier recovery. Relapse symptoms were significantly less severe at the time of rating, and initial and relapse severity were not significantly related. Nevertheless there were significant correlations between initial and relapse scores on two factors orthogonal to severity. One of these reflected symptom patterns important to the endogenous-neurotic distinction, the other a contrast between depression and anxiety. These findings indicate some consistency in symptom patterns in successive episodes of depression, and thereby support the value of clinical classifications depending on symptom patterns.

The uses of chlormethiazole (Heminevrin) are briefly mentioned. Differing views of its potential for producing dependence are discussed. The consensus opinion is taken to be that a risk of psychological dependence does exist, but only one case has appeared in the medical literature to date of physiological dependence on chlormethiazole, with withdrawal symptoms. This case is re-examined. A second case is described in which the features of increasing dosage and withdrawal were followed by confusion, disorientation, delusions, hallucinations and muscular spasm and coarse jerking. This is taken as confirmation of the potential of chlormethiazole to produce physiological dependence. Some personal opinions of the author regarding indications for use and incidence of abuse are briefly mentioned.

This investigation is an attempt to test the common supposition that postpartum emotional disturbance is related to hormone changes. A group of 27 normal pregnant women were assessed three times before delivery and sixteen times in the six weeks following delivery. During the first two interviews baseline data on personality and other personal variables were obtained. On each occasion blood was taken and three measures of clinical status and mood were completed. Plasma LH, FSH, total oestrogen and progesterone results are presented in detail and the results of prolactin assays mentioned more briefly. An attempt to correlate hormone findings and clinical findings is described. This failed to produce any strong evidence that hormones are related to mood at this time, although hormone changes were correlated weakly with a few specific symptoms. Some of the unexpected clinical findings and technical difficulties of the study are discussed, with special reference to possible further research in this area.

We studied the relationship between side effects, clinical outcome and the drug plasma levels in 28 female depressed patients treated with protriptyline. After 3½ weeks treatment, patients with plasma levels within a median range (630 to 900 nmol/1) showed better responses to the drug than patients with plasma levels outside this range.

There were no statistically significant correlations between plasma levels and side effect scores or ‘corrected’ side effect scores (scores after subtracting pre-treatment values) for the group at any time after starting the treatment. But we found positive correlations between plasma levels and ‘corrected’ side effect scores for the neurotic subgroup after 14 and 21 days of treatment. Other correlations between plasma levels and side effect scores were non-significant.

In a group of British-born primiparous women, admission of a separation experience from parents in childhood was found to be associated with unmarried status, teenage pregnancy, having had psychiatric treatment, housing problems and a high score on a ‘Malaise’ inventory. Within the separated group the unmarried and teenage status were associated only with separations occurring in the context of a disrupted childhood, but were not related to the age of the separations. The Malaise score, however, was raised both for disrupted separations and for those occurring below the age of 5. It is noted that the factors associated with the separation in the mothers are those which have been found to be more common in the mothers of children with behaviour disorders.

Are research interview techniques adequate in assessing signs and symptoms? This question is investigated by obtaining two sets of interview schedule ratings on 49 patients, one by a research psychiatrist applying the interview and one by the patient's psychiatrist using all available information. The latter was considered a clinical standard with which a cross-sectional research interview could be contrasted.

These two data sets were subjected to several types of analysis commonly undertaken with research interview ratings. Results indicated that the research interview adequately represents symptom data, but is seriously lacking in the assessment of observed behaviour. The effect of this difference depends on the goals of the study and the nature of the data analysis. If overall group findings are desired, or if analysis relies primarily on symptom data, results with a research interview may be similar to results based on a far more extensive evaluation. On the other hand, if conclusions are to be drawn on individual patients, or if data analysis relies heavily on observed behaviour, then data derived from research interviews are questionable.

In a recent article, Everitt (1975) discussed several problems with multivariate techniques. However, two useful applications of multivariate techniques were not covered. The present paper describes the use of factor analysis to reduce a large array of outcome variables to a statistically manageable number, and multivariate analysis of variance to determine the relative effectiveness of several treatment regimes where a single outcome variable cannot be specified. It is concluded that the advantages of a multivariate approach outweigh the disadvantages, provided the researcher is careful in interpreting and reporting his results.