All Danish psychiatric patients admitted to hospital for the first time in the period 1969–88 with a diagnosis of schizophrenia given at least once during the period were included in the study. More than 8500 patients were examined in the Danish nationwide psychiatric register. First-admission rates decreased by approximately 50% irrespective of mode of calculation and increase in SMR. The use of in-patient care was unchanged despite a reduction in available beds of more than 40% in the study period.

A locus for schizophrenia within the pseudoautosomal region of chromosomes X and Y has been suggested by Crow on the basis of epidemiological data. The present report replicates this finding in a sample of 38 French multiply affected families with schizophrenia. Sibship and pairwise analysis, with or without weighted-pair correction, with three different systems of family classifications, showed there to be an excess of same-sex pairs in paternally derived sibships, as predicted by the pseudoautosomal hypothesis.

Forty-four schizophrenic patients were followed up for five years after their first admission to hospital for a first episode of illness. Thirteen (30%) of 43 patients had not relapsed; 28 of the 30 patients who did relapse did so within the first 42 months. The relapses occurred despite antipsychotic drug therapy. Also, 24% of patients had at least one course of ECT. Only 19% of the patients at five years were in open employment; unemployment was strongly associated with relapse. Eighteen per cent had neither relapses nor schizophrenic symptoms at follow-up. Poor outcome at five years was associated with greater psychological distress among relatives at first admission. At five years 43% of relatives continued to show case level psychological stress.

A modified version of the revised Personality Diagnostic Questionnaire (PDQ–R), based on DSM–III–R personality disorders (PDs), was completed by 60 psychiatric patients. An informant's version was also completed by 60 relatives or friends nominated by each subject. Discrete DSM–III–R PDs were rare; the mean number of PDs per subject was 4.5. Cluster analysis showed that only antisocial PD was a basis for classification of patients, while most patients formed two groups which were mainly distinguished by quantitative differences related to the total scores of positive PD criteria. A shorter version of the questionnaire can be used as a screening test for co-morbid PDs (STCPD) which can predict the number of co-morbid DSM–III–R PDs. The total scores of positive PD criteria from the STCPD were usually (and significantly) higher than the corresponding scores from informants' questionnaires but when an informant's total score exceeded that of the patient, this indicated a subject's under-reporting.

The clinical and social characteristics of new long-stay (NLS) patients at Friern and Claybury hospitals are described, together with their accumulation rates within health districts in northeast London, and the associations between accumulation rates and social deprivation. There is a fourfold variation between local districts in annual accumulation rates of NLS patients (between 2.5 and 11 per 100 000 population); 0.55 of this variation is accounted for by the Jarman scores of social deprivation, and 0.81 by local rates of unemployment. Other recent British studies support this finding that measures of social deprivation can statistically explain a large proportion of the variation in treated rates of psychiatric morbidity, and may be useful in predicting needs for psychiatric services.

Two critically important areas of research are the long-term course of illness and predictors of outcome in chronic schizophrenia. Although these two areas have been actively investigated throughout the 20th century, there is still a lack of agreement about even basic features of the long-term outcome of schizophrenia as well as a lack of validating evidence for putative predictors of outcome. Determining the long-term outcome of schizophrenia has implications for the pathophysiology and public health policy of this illness. For example, a course of illness that is marked by progressive deterioration may suggest a degenerative disease process, whereas a course that remains stable or improves over time may be more consistent with non-degenerative hypotheses such as neurodevelopmental, infectious, or toxic processes. In terms of public health policy, the allocation of resources related to issues such as community-based versus hospital-based care will be heavily influenced by knowledge of the level of chronicity and impairment associated with schizophrenia throughout the lifetime of affected individuals. Validation of predictors of outcome will have immediate clinical relevance, as well as providing support for pathophysiological hypotheses.

Expressed emotion (EE) in the partners of 25 pregnant women with a history of psychosis or severe depression and in 13 pregnant control subjects without any previous psychiatric disorder was assessed in the ninth month of pregnancy. At this time, no patient presented as a case according to RDC. Eleven subjects with a history of psychiatric disorder experienced a further episode of illness in the six months following delivery. Partners of women who became ill had made fewer critical and positive comments about their wives during the pregnancy than the partners of women who remained well. Poor self-rated social adjustment in the partners was also predictive of recurrence of illness after delivery.

Because of an association between sexual aneuploidies and schizophrenia, and because schizophrenic siblings have been found to be more often of the same than of the opposite sex, the susceptibility locus for schizophrenia is thought to lie within the pseudoautosomal region of the sex chromosomes. We analysed 33 sibships comprising 18 pairs, 13 trios, and 2 quartets of affected siblings, and found support for non-random segregation of alleles at the DXYS14 locus in affected siblings. These findings are consistent with the pseudoautosomal hypothesis for schizophrenia and favour a genetic linkage between DXYS 14 and the disease.

There is overwhelming evidence that the outcome for people with schizophrenia in Western industrialised countries is inferior to that of those living in the Third World. Extended family structures, greater opportunities for social reintegration, and more positive constructions of mental illness have been offered as possible explanations for this effect. The Asian community in the UK retains many of these features as well as strong links with native cultures of origin. The issue arises as to whether similar differences in outcome may be observed in the UK. An exploratory study was undertaken, examining the early progress of schizophrenia in a first-episode sample (n=137), and based on systematic examination of case-note data. A lower rate of relapse/readmission in the first 12 months after discharge was found in the Asian (16%) as compared with white (30%) and Afro-Caribbean (49%) patients. Available evidence suggested that speed of access to care, living with a family, and employment may account for this effect. Medication compliance may have contributed to differences in relapse between white and Afro-Caribbeans but was not a factor influencing the low rate among Asians. The limitations and strengths of case-note studies are discussed at length, and it is concluded that a prospective study is warranted and would be highly instructive.

A sample of 60 Spanish schizophrenic patients was studied to ascertain the relationship between their relatives' expressed emotion (EE) and relapse at follow-up. The relatives' EE and patients' relapse were operationalised following Leff & Vaughn's criteria. At nine months a significant association was not found between the relatives' EE and relapse, but this association became significant on reclassifying the relatives' EE scores after decreasing to four points the cut-off point for critical comments. At 24 months no association was found between EE and relapse. There was a tendency for patients who interrupted their medication or who did not work to relapse more frequently, particularly among the high-EE group.

In a double-blind, parallel group trial, 78 subjects with social phobia received moclobemide (a new reversible inhibitor of monoamine oxidase A) phenelzine, or placebo. After eight weeks, both active drugs – phenelzine somewhat more than moclobemide – were clinically and statistically significantly more effective than placebo, as assessed by rating scales. There was some further improvement between weeks 8 and 16, particularly in the moclobemide group; at week 16, 82% of the moclobemide and 91% of the phenelzine-treated patients were almost asymptomatic. Moclobemide was, however, much better tolerated than phenelzine. Patients withdrawn from active drugs had relapsed by week 24, providing additional support for the efficacy of the active drugs.

Kraepelin's assumption that affective psychoses always have a ‘good‘ outcome was found to be untrue — it could not differentiate affective from schizophrenic disorders. Long-term investigations have shown that a relevant proportion of affective disorders have an ‘unfavourable outcome’ (Angst, 1987; Marneros & Deister, 1990; Marneros et al, 1990b), while some schizophrenic disorders can have a good prognosis (Bleuler, 1972; Ciompi & Müller, 1976; Huber et al, 1979; Möller & von Zerssen, 1986; Marneros et al, 1989a, 1990b, 1991b).

There has been almost a threefold increase in the incidence of suicide in Ireland during the past 20 years. When each case occurring between 1971 and 1988 was reviewed, the major causes of death were found to be drowning and hanging. Deliberate overdoses with tricyclic antidepressants accounted for less than 4% of all suicides, over the 17 years studied, and during this time, overdose with tricyclics has made a proportionately smaller contribution to the total number of suicides. Greater attention should be paid to identifying and treating the majority of suicidal patients who are at risk by other methods.

Of 253 patients in their first schizophrenic episode, 52 behaved in a way threatening to the lives of others before their admission to hospital. These 52 patients were studied from data collected at the time of their initial presentation. Despite a history of illness in excess of 1 year in 24 cases, and evidence that violence was motivated by psychotic symptoms in 23 cases, fewer than half of the patients were admitted to hospital as a direct result of their dangerous behaviour. Life-threatening behaviour was more common where the patient had been ill for longer, and where there were delusions of being poisoned.

Evidence for a pseudoautosomal locus for a schizophrenia susceptibility gene was sought by two forms of analysis of 25 multiply affected families. Firstly, in the sample as a whole there was an excess of same-sex over mixed-sex siblings compared with that expected. Secondly, linkage analysis was performed in six of the families. The genotypes were studied for DXYS14, a highly polymorphic marker in the telomeric pseudoautosomal region. No evidence for positive linkage was found with two-point analysis under eight different genetic models for the mode of transmission. A non-parametric, sibling-pair analysis also failed to detect linkage. Our findings provide no evidence for linkage within the pseudoautosomal region; same-sex concordance must arise from some other mechanism.

Iron status and akathisia were assessed in 105 long-stay in-patients who fulfilled DSM–III–R criteria for schizophrenia, all but three of whom were receiving antipsychotic medication. Chronic akathisia was diagnosed in 23% and pseudoakathisia in 20%. No significant correlation was found between serum iron concentration and the severity of akathisia. There was no significant difference in serum iron concentration between patients with chronic akathisia and those without. However, serum iron and percentage saturation were significantly raised in patients with pseudoakathisia compared with patients with chronic akathisia, and tended to be higher than in patients with akathisia. These findings do not support an association between low serum iron and chronic akathisia.

Plasma melatonin levels were measured at three-hourly intervals over 24 hours in 11 women with untreated anorexia nervosa, and in nine healthy women of normal weight. The circadian rhythm was unaltered but the nocturnal secretion of melatonin was significantly greater in anorectics. It is possible that this was related to nocturnal hypoglycaemia.

It is accepted that schizophrenia runs in families, but whether this relates to genetic or psychosocial transmission is an unanswered question. Kendler (1988) has articulated four testable hypotheses: (a) a general liability to any psychiatric illness (b) a liability to schizotypal functioning — oddness, suspiciousness etc. (c) a liability to broadly defined schizophrenic psychosis, or any functional, non-affective psychosis and (d) a specific liability to narrowly defined schizophrenia, e.g. using DSM-III-R criteria (American Psychiatric Association, 1987). Kendler suggests that neither hypothesis (a) nor (d) is correct, and that the familial predisposition is neither completely non-specific nor highly specific; available results strongly support the second hypothesis and also provide some support in favour of the third.

Forty-two schizophrenic patients and their close relatives took part in an Italian replication study of expressed emotion (EE). The patients were selected from the psychiatric ward of a general hospital in Milan and were subsequently followed up for nine months. All patients attended a community service clinic as out-patients, and all but one were prescribed neuroleptics for the duration of the study. Relatives were assigned to the high-EE group if they scored 4 or 5 on the emotional overinvolvement (EOI) scale, or showed hostility, or made six or more critical comments. On this basis, 18 (42%) families were rated as low EE and 24 (57%) as high EE. At follow-up, the admission rate for the 9-month period was significantly higher for the high-EE group (P<0.05). Furthermore, significantly fewer patients were readmitted from families showing high warmth (P<0.05). The presence of high warmth appeared to be associated with a lower admission rate, even in high-EE families.

Standard of living reflects the objective dimension of how well the basic needs of life are met, while quality of life is the patient's own subjective view of well-being and satisfaction with her/his life. Sixty-one schizophrenic out-patients completed self-report inventories and participated in interviews about quality of life and standard of living. When living standards were met by a well functioning social service system, patients' perceptions of their quality of life and their standard of living appeared to be independent. Subsequent analyses revealed that ‘inner experiences' was one quality-of-life domain frequently reported as unsatisfactory. Moreover, differences in quality of life were found across patients' age, education, and work status.