The severe depressive disorders of late life are associated with high rates of medical morbidity and mortality, cognitive impairment, suicide, disability, complex treatment regimens, institutionalization and high costs to the community (Murphy, 1983; Murphy et al. 1988; Bruce & Leaf, 1989; NIH Consensus Development Panel, 1992; Alexopoulos et al. 1993a, b; Brodaty et al. 1993; Bruce et al. 1994; Forsell et al. 1994; Hickie et al. 1995; Blazer, 1996). Those disorders that are accompanied by cognitive impairment and/or concurrent medical morbidity have a particularly poor outcome (Bruce & Leaf, 1989; Alexopoulos et al. 1993b; Hickie et al. 1995, 1997a). Although psychosocial models of late-life depression place considerable importance on age-related psychological and social risk factors, those who survive into later life may actually be characterized by psychological resilience (Henderson, 1994; Blazer, 1997).

Current aetiological research in late-life depression, therefore, places particular emphasis on the potential role of biological risk factors. The potential importance of vascular risk factors is receiving renewed attention and may provide opportunities for specific prevention and intervention strategies in high-risk populations. This emphasis on possible vascular risk factors, and the wider importance of vascular pathologies in late-life neuropsychiatric disorders, mirrors the emphasis of much earlier clinico-pathological studies (Binswanger, 1894; Alzheimer, 1895). The specific focus on the importance of small progressive changes within the subcortical white matter, as distinct from more discrete cortical infarcts (Olszewski, 1962), is now supported by the emerging neuroimaging literature and theoretical constructs in late-life depression (Krishnan, 1991, 1993; Hickie et al. 1996, 1997b; Krishnan et al. 1997).

Background. Late-life depression is associated with increased subcortical white matter hyper-intensities. There is some evidence that they are associated with a poorer response to acute treatment. Neurological signs and neuropsychological dysfunction are further evidence of abnormalities in the brain, but they have not been studied in relation to therapy resistance.

Methods. A prospective study of 24 normal controls and 75 consecutive elderly (aged 65 to 85) patients with DSM-III-R major depression entered a naturalistic study of treatment. Assessment of response to monotherapy and then lithium augmentation or ECT created three outcome groups. Investigations included magnetic resonance brain imaging, neuropsychological and neurological examination.

Results. Response to monotherapy within 12 weeks was shown by 42·7%, a further 37·3% responded to lithium augmentation or ECT within 24 weeks and 20% had responded poorly to all treatments at 24 weeks. Subcortical hyperintensities were significantly increased in the more resistant patients. These included confluent deep white matter, multiple (>5) basal ganglia lesions and pontine reticular formation lesions. Most of the neuropsychological impairment was restricted to the resistant groups and was of a subcortico-frontal type. Extrapyramidal, frontal and pyramidal neurological signs characterized the resistant groups. The combination of extrapyramidal signs, pyramidal tract signs and impairment of motor hand sequencing strongly predicted resistance to 12 weeks of antidepressant monotherapy with 89% sensitivity and 95% specificity.

Conclusion. In late-life depression a poor response to antidepressant monotherapy can be expected in those patients with a frontal lobe syndrome, extrapyramidal signs or if MRI T2-weighted lesions are present in both the basal ganglia and the pontine reticular formation.

Background. A review of studies of cognition in the euthymic phase of unipolar and bipolar affective disorder reveals diverging results.

Methods. The study was designed as a controlled cohort study, with the Danish psychiatric case register of admissions used to identify patients and the Danish civil register to identify controls. Patients who were hospitalized between 19 and 25 years ago with an affective diagnosis and who at interviews fulfilled criteria for a primary affective unipolar or bipolar disorder, according to ICD-10, were compared with age- and gender-matched controls. Interviews and assessment of the cognitive function were made in the euthymic phase of the disorder. In all, 118 unipolar patients, 28 bipolar patients and 58 controls were included. Analyses were adjusted for differences in the level of education and for subclinical depressive and anxiety symptoms.

Results. Patients with recurrent episodes were significantly more impaired than patients with a single episode and more impaired than controls. Also, within patients the number of prior episodes seemed to be associated with cognitive outcome. There was no difference in the severity of the dysfunction between unipolar and bipolar patients.

Conclusions. Cognitive impairment in out-patients with unipolar and bipolar disorder appears to be associated with the number of affective episodes.

Background. Neuropsychological and imaging studies suggest that frontal dysfunction may occur in apparently normal chronic alcoholic subjects.

Methods. To investigate this issue further, we performed neuropsychological and fluorodeoxy-glucose-PET studies in 17 chronic alcoholics without patent neurological and psychiatric complications.

Results. Metabolic abnormalities were found in the mediofrontal and in the left dorsolateral prefrontal cortex, but not in the orbitofrontal cortex. Neuropsychological testing revealed significantly reduced verbal fluency and impaired performance on the Stroop test. The mediofrontal hypometabolism correlated with the reduction in verbal fluency and the time necessary to perform the interference condition of the Stroop test. The left dorsolateral prefrontal hypometabolism correlated with the number of errors on the Stroop test.

Conclusion. These data indicate that circumscribed frontal dysfunctions may occur in chronic alcoholic subjects before clinically obvious neurological complications, and may account for some of the alcohol-related neuropsychological and behavioural impairments.

Background. The aim of this study was to investigate mnemonic strategic deficits in schizophrenic patients.

Methods. Analogous tasks were used that required the self-generation of an efficient strategy and its implementation in two domains: visuospatial and verbal. The tasks were given to 20 IQ preserved schizophrenics and 20 matched normal controls. A number of different scores was derived from each task including strategy, short-term memory capacity and perseveration.

Results. Overall, the schizophrenic patients were significantly impaired in their ability to generate effective mnemonic strategies on both tasks. In addition, on the visuospatial task there was no difference between the groups on the memory scores, but the schizophrenic patients made significantly more perseverative errors than controls. They were disproportionately worse on the verbal strategy task, showing impairment on memory as well as on strategy scores and were also impaired at semantically classifying the words. Performance was similar to the deficit seen in patients with frontal lobe excisions and Parkinson's disease, in terms of the inability to generate an effective strategy. The deficit on the verbal task was similar to patients with temporal lobe excisions who show impaired verbal memory. However, the pattern differed in the sense that the temporal lobe patients were able to generate effective strategies, unlike the patients with schizophrenia.

Conclusions. High functioning schizophrenic patients are impaired in utilizing visuospatial and verbal mnemonic strategies. By comparing the results with those of neurosurgical excision patients, further evidence is provided for both frontal and temporal lobe involvement in schizophrenia.

Background. The current study investigated the fundamental nature of the cognitive processing deficit that has been demonstrated to be associated with dieting to lose weight. Previous work has characterized this deficit as being primarily one of a reduction in working-memory capacity. The present study investigated the particular components of the working-memory system affected during dieting.

Method. A sample of female subjects was classified as either low/medium restrained eaters (N=34), highly restrained eaters (N=18) or current dieters (N=19), based on their responses to a modified version of the Dutch Eating Behaviour Questionnaire (DEBQ). Each subject completed tasks that assessed the Visuo-Spatial Sketchpad (mental rotation), Phonological Loop (effect of phonological similarity on recall) and the Central Executive (Tower of London Task) components of working memory.

Results. Those subjects who reported themselves to be currently dieting displayed poorer recall on the Phonological Loop task and slower planning times on the Tower of London Task. Performance on both these tasks correlated significantly with a self-report measure of body shape concern.

Conclusions. These results support the hypothesis that the mediating variable in this deficit is that of preoccupying cognitions concerning food and body shape.

Background. Systematic comparisons of the psychological and physical responses of caregivers of Alzheimer's disease (AD) patients in the US and China have not been previously reported.

Methods. Informal caregivers of community dwelling AD patients in Shanghai, China and demographically-matched non-caregiving Chinese controls were compared with a sample of American caregivers residing in San Diego, California and demographically-matched American controls.

Results. Despite some demographic discrepancies, caregivers from both China and the US were similar. Caregivers from both cultures reported more depressive symptoms and more physical symptoms when compared with non-caregivers. Both groups of caregivers reported that patients required similar amounts of care and help with activities of daily living (ADLs). However, Shanghai caregivers reported less access to emotional support when compared with the San Diego sample. A conceptual model, guided by the stress process model of Pearlin et al. (1990), was used to explore multivariate relationships between caregiver characteristics and the physical and psychological health of our sample of AD caregivers in Shanghai, China. Results from a path analytical procedure revealed that the relationships among these variables and health outcome did not differ significantly from those observed in the US sample.

Conclusions. Although elderly family members are venerated in the stereotypical Chinese family unit, and informal caregiving of disabled family members is socially mandated, the negative health consequences of caregiving appear to be similar to those observed among caregivers in the US.

Background. Although high expressed emotion (EE) has been found to be an important predictor of poor prognosis in a wide range of conditions such as schizophrenia, anorexia and depression no complete explanation exists for individual differences in EE responses The aim of this paper is to investigate the role of intimacy in determining the level of EE in carers of people with dementia.

Methods. Ninety-nine carers of people with dementia who presented to Old Age Psychiatry Services in South and Central Manchester completed questionnaires to ascertain past and current levels of intimacy. Camberwell Family Interviews (CFIs) were carried out to ascertain levels of EE.

Results. Current intimacy was found to be strongly related to EE such that low current intimacy was associated with high EE and there was a significant difference between high and low intimacy groups on measures of criticism and hostility, though not warmth.

Conclusions. The association found between intimacy and EE indicates that high EE may be a characteristic of low intimacy relationships between the carer and the cared-for-person. Since the assessment of EE is time intensive, perhaps a measure of intimacy will provide a short-hand screen for identifying critical and hostile caring environments.

Background. Abnormal performance on the antisaccade task suggests that patients with schizophrenia have difficulty with the inhibition of reflexive attentional shifts. The aim of the study was to investigate whether deficits in the inhibition of reflexive attentional shifts were specific to the oculomotor modality or whether they could also occur when attentional shifts were made without eye movements (e.g. covert attentional shifts).

Methods. Fifteen medicated patients with chronic schizophrenia and 15 matched controls performed the antisaccade task and the covert orientating task (COVAT) where the probability of targets appearing at the same location of a peripheral cue was varied so that voluntary and reflexive orientating systems had the same goal (80% probability of target and cued condition) or opposite goals (20%probability of target at cued location). A condition where only reflexive orientating was initiated was also included (50% probability of target at cued location). For each of these conditions the stimulus onset asynchrony (SOA) varied between 150 and 350 ms.

Results. Patients with schizophrenia showed normal latency and accuracy for visually guided saccades but increased error rates and latency on the antisaccade task. For the COVAT, patients with schizophrenia were unable to use voluntary orientating strategies to inhibit reflexive shifts of covert attention. On conditions where only reflexive orientating was required or when the goals of the reflexive and voluntary orientating systems were the same, patients with schizophrenia showed normal performance.

Conclusions. These results suggest the reflexive orientating mode is normal in patients with chronic schizophrenia. However, these patients have a reduced ability to utilize the voluntary orientating mode to control or inhibit reflexive orientating. This impairment of voluntary control is evident for both overt and covert attentional shifts.

Background. A neuropsychological formulation of schizophrenia has suggested that problems with meta-representation underpin both positive and negative symptoms. This study tested Frith's account by asking patients experiencing an acute episode of psychosis to complete a set of tasks that involved Theory of Mind (ToM) skills.

Methods. Fourteen patients who fulfilled criteria for schizophrenia, 10 deluded patients who were suffering from psychotic disorders other than schizophrenia and 12 depressed patients completed second-order false belief tasks, a test which involved substitution of a co-referential term in a linguistic description of an event, and metaphor and irony tasks. The battery of tests was completed during the acute phase and following recovery. Selection of these patient groups allowed comparisons to be made between schizophrenia patients and non-schizophrenia patients and between patients with and without persecutory delusions.

Results. Schizophrenia patients, who had a multiplicity of positive and negative symptoms, performed significantly worse than non-schizophrenia patients on some of the ToM tasks during an acute episode. Patients with delusions of persecution and reference did not perform significantly worse than non-deluded patients on ToM tasks. There was no significant difference between groups in performance on any of the tasks at recovery.

Conclusions. The results provide at best weak support for Frith's account and it remains unclear whether the ToM deficits demonstrated are genuine deficits or are a result of information-processing overload. However, it is clear that difficulties interpreting interpersonal contexts, as shown by some schizophrenia patients, are state rather than trait characteristics.

Background. A significant number of patients with severe psychiatric disorders remain untreated in the community although health services are available. The factors related to non-treatment are not well understood.

Method. A door-to-door survey was conducted on an Indian urban population of 100 000 using standardized screening and clinical instruments as a part of a larger epidemiological study. Treatment status was determined from multiple sources of information.

Results. Nearly one-third of 261 schizophrenia patients were found never to have received treatment. They were older in age and ill for a longer duration than those who had been treated and were more symptomatic and severely disabled. They were more often uneducated and divorced and lived with larger extended/joint families. This last factor was considered as being the important factor in determining whether the patient received treatment.

Conclusion. The larger extended/joint family, which was able to compensate and cope with the dysfunctional member, seemed to be the crucial factor related to non-treatment of the schizophrenic patient.

Background. Cross-sectional studies of non-cognitive symptoms in dementia show that patients with psychotic symptoms tend to have more disturbed behaviour. However, it is not known whether individuals who experience psychiatric symptoms early in dementia are more prone to develop behavioural problems later in the illness.

Method. The behaviour of 86 community-dwelling subjects with dementia was intensively studied for 4 years or until death, using an informant interview which was administered every 4 months on a median of eight occasions. The extent to which psychiatric symptoms, age, sex and cognitive function predicted clinically significant physical aggression or motor hyperactivity was assessed.

Results. Physical aggression was predicted by sad appearance and motor hyperactivity was predicted by persecutory ideas. These associations were robust, remaining significant over 2, 3 and 4 years of follow-up and were independent of cognitive function, age, sex and duration of illness.

Conclusions. There may be two distinct longitudinal syndromes of non-cognitive symptoms in dementia. This suggests that important aberrant behaviours in late dementia may share pathophysiological mechanisms with psychiatric symptoms in early dementia.

Background. We report the prevalence, clinical characteristics, frequency of mental health treatment, demographic correlates, frequency of co-morbid psychiatric conditions, and general health ramifications of DSM-IV blood-injection-injury phobia in the general population.

Method. The Diagnostic Interview Schedule (version III-R), which included questions on blood-injection-injury phobia, was administered to 1920 subjects in the Baltimore ECA Follow-up Study.

Results. The estimated unweighted lifetime prevalence of blood-injection-injury phobia was 3·5%. The median age of onset was 5·5 years; 78% had had symptoms within the last 6 months. Subjects with blood-injection-injury phobia (cases) had higher lifetime histories of fainting and seizures than those without (non-cases). None reported seeking mental health treatment specifically for phobia. Prevalences were lower in the elderly and higher in females and persons with less education. Cases had significantly higher than expected lifetime prevalences of other psychiatric conditions, including marijuana abuse/dependence, major depression, obsessive–compulsive disorder, panic disorder, agoraphobia, social phobia and other simple phobia. Cases and non-cases did not differ with regard to usual health-care settings, regular care for specific medical conditions, numbers of out-patient visits or hospitalizations, or previous general anaesthesia or live births. However, diabetics with blood-injection-injury phobia had higher than expected rates of macrovascular complications.

Conclusion. Blood-injection-injury phobia is common, especially in females and those with less education, and it is associated with several co-morbid psychiatric conditions. No strong, broad general health ramifications of this phobia are apparent. However, diabetics with this phobia appear at particular risk for complications; this deserves further study.

Background. This paper reports on a two-stage community-based epidemiological study of selected minor psychiatric disorders conducted on an adult African population in South Africa.

Methods. Using a modified random cluster sampling method, 354 adults were identified as the first-stage sample, with the SRQ-20 being used as a first-stage screen. Clinical interviews based on DSM-IV checklists for generalized anxiety disorder, major depression and dysthymia were administered as the second-stage criterion to 81 subjects from the sample.

Results. The weighted prevalence for generalised anxiety and depressive disorders was 23·9% (95% CI 15·1%–32·7%), comprising: generalized anxiety 3·7%, major depression 4·8%, dysthymia 7·3%, and major depression and dysthymia 8·2%. Statistically significant associations were found between caseness and age, marital status, employment, income and educational level.

Conclusions. The results are discussed in relation to comparative local and international data as well as in the context of the current restructuring of the mental-health care system in South Africa from tertiary curative care to integrated primary mental-health care.

Background. There have been no previous studies of the outcome of different neurotic disorders in which a prospective group with original randomization to treatment have been followed up over a long period. Such studies are important in identifying the factors associated with good and poor outcome.

Methods. A 5-year follow-up assessment was made of a cohort of 210 psychiatric out-patients seen in general practice psychiatric clinics with a DSM-III diagnosis of generalized anxiety disorder (71), panic disorder (74) or dysthymic disorder (65) and randomized to drug treatment, cognitive and behaviour therapy, and self-help. A total of 182 of the patients (87%) were assessed after 5 years by examination of hospital and GP records using a standardized procedure and outcome determined with a four-point outcome scale.

Results. One hundred and seven (60%) of the patients had a favourable outcome but the remainder continued to be handicapped either intermittently or continuously throughout the 5-year period. Analysis of the value of initial data in predicting outcome using polychotomous step-wise logistic regression revealed that five variables were significant predictors of poor prognosis: older age; recurrent episodes; the presence of personality disorder at entry; general neurotic syndrome at entry; and symptom severity after 10 weeks. The initial DSM diagnosis and original treatment given, together with ten other variables, were of no predictive value.

Conclusions. The long-term outcome of neurotic disorder is better predicted by age, personality and recency of onset than by other clinical variables with the exception of initial response to treatment.

Background. Many studies have noted a strong association between poor social support and premature mortality. A limitation of such studies has been their failure to control adequately for confounders that damage both social supports and physical health.

Methods. A 50-year prospective multivariate study of 223 men was used to examine the possible causal relationships between social supports and health. Alcohol abuse, prior physical health and mental illness prior to age 50 were controlled. Relative social supports were quantified over the period from age 50 to 70.

Results. Adequacy of social supports from age 50 to 70 was powerfully correlated with physical health at age 70 (P<0·001). However, such social supports were also powerfully predicted by alcohol abuse (P<0·001), smoking (P<0·001) and indicators of major depressive disorder (P<0·01) assessed at age 50. When prior smoking, depression and alcohol abuse were controlled, then the association of physical health with social supports was very much attenuated. Some facets of social support like religion and confidantes were unassociated with health even at a univariate level. Surprisingly, in this sample friends seemed more important for sustained physical health than closeness to spouse and to children.

Conclusions. While social supports undoubtedly play a significant role in maintaining physical well-being in late life, much of the association between poor social supports and mortality may be mediated by alcoholism, smoking and pre-morbid psychopathology.

Background. The number of antidepressant drugs available in the market has grown rapidly in the last few years. The present paper underlines some of the pre-clinical and clinical problems that call close attention from the regulatory authorities when approving new drugs.

Methods. We present here a review of the literature.

Results. A wide heterogeneity in the action of the various antidepressants precludes any single theory about the pathogenesis and therapy of depression. Antidepressant activity, in fact, may be achieved by acting on a number of different monoaminergic mechanisms. The variety in the neurochemical effects of antidepressants is not reflected in clinical trials, which tend to stereotypy. In many cases clinical trials aim at demonstrating equivalence rather than differences in efficacy. Regulatory authorities should, therefore, pay attention in accepting the equivalence of effects of a new drug in relation to a reference one: most clinical trials of new antidepressant drugs do not have the power to detect clinically relevant differences.

Conclusions. Unconventional new pre-clinical tests are needed to generate antidepressants with a different mechanism of action. Clinical studies are needed to promote objective comparative evaluation of the cost, benefits and toxic effects of new antidepressants.

Background. Self-report instruments assessing the DSM personality disorders are characterized by overdiagnosis due to their emphasis on the measurement of personality traits rather than the impairment and distress associated with the criteria.

Methods. The ADP-IV, a Dutch questionnaire, introduces an alternative assessment method: each test item assesses ‘Trait’ as well as ‘Distress/impairment’ characteristics of a DSM-IV criterion. This item format allows dimensional as well as categorical diagnostic evaluations. The present study explores the validity of the ADP-IV in a sample of 659 subjects of the Flemish population.

Results. The dimensional personality disorder subscales, measuring Trait characteristics, are internally consistent and display a good concurrent validity with the Wisconsin Personality Disorders Inventory. Factor analysis at the item-level resulted in 11 orthogonal factors, describing personality dimensions such as psychopathy, social anxiety and avoidance, negative affect and self-image. Factor analysis at the subscale-level identified two basic dimensions, reflecting hostile (DSM-IV Cluster B) and anxious (DSM-IV Cluster C) interpersonal attitudes. Categorical ADP-IV diagnoses are obtained using scoring algorithms, which emphasize the Trait or the Distress concepts in the diagnostic evaluation. Prevalences of ADP-IV diagnoses of any personality disorder according to these algorithms vary between 2·28 and 20·64%.

Conclusions. Although further research in clinical samples is required, the present results support the validity of the ADP-IV and the potential of the measurement of trait and distress characteristics as a method for assessing personality pathology.

Background. A sentence verification task was developed to investigate semantic memory in schizophrenia.

Methods. The test consisted of three types of sentence (true, unlikely and nonsense) and seven different types of content (neutral, persecutory, grandiose, political, religious, relationships and somatic) representing common delusional themes present in schizophrenic patients. Sixty-three schizophrenic patients and 66 matched control subjects were asked to make true/false judgements to 143 sentences.

Results. Overall accuracy was similar across the two groups; sentences with some emotional themes and sentences of the unlikely type produced the most violations. Significant differences between the two subject groups were found specifically on nonsense sentences with persecutory and religious themes. Patients made significantly more incorrect responses (acceptance) to nonsense sentences that had an emotional content congruent with their delusional beliefs, past or present, and also on unlikely sentences (incorrect rejections) whose content was not congruent with their delusions. Further analysis of response bias in the patients showed, overall, that there were more incorrect rejections (a reflection of the large number of unlikely sentence errors) and more incorrect responses to sentences congruent with patients delusions. Furthermore, analysis of those patients currently experiencing delusions revealed more incorrect responses to sentences congruent with their delusional ideas compared with patients not currently deluded.

Conclusions. These findings are indicative of cognitive bias in schizophrenia towards certain emotional themes that may underlie illogical semantic connections and delusions.

Background. Cognitive models suggest that auditory hallucinations are experienced when mental events are misattributed to an external source; therefore, this study was designed to examine attributional biases in patients experiencing auditory hallucinations. The study also examined the role of metacognitive beliefs in the experience of auditory hallucinations, as some theories have implicated metacognition in the development and maintenance of auditory hallucinations.

Methods. Fifteen participants with a diagnosis of schizophrenia experiencing auditory hallucinations were compared with 15 non-hallucinating schizophrenics and 15 non-psychiatric control subjects on several measures, including an immediate source monitoring task and a questionnaire assessing metacognitive beliefs.

Results. Results indicated that patients experiencing hallucinations exhibited the predicted bias towards misattributing internal events to an external source, as measured by ratings of internality of responses in a word association task. All groups had lower perceived levels of internality and control for emotionally salient words, which provides further evidence for the importance of emotional content in hallucinations. Patients experiencing hallucinations were found to score higher than the other two groups on metacognitive beliefs about uncontrollability and danger and positive beliefs about worry. In addition, a logistic regression analysis showed that beliefs about uncontrollability and danger were predictive of whether subjects experienced auditory hallucinations or not.

Conclusions. These results offer considerable support to cognitive bias models of auditory hallucinations, particularly those that implicate metacognition.