We critically reviewed the arguments of the symptom-oriented researchers who propose to replace syndromes and diagnostic categories with symptoms as units of analysis in psychiatric research.

Three central arguments were examined: (a) current diagnostic categories lack reliability and validity; (b) using diagnostic categories leads to misclassification and confounding; and (c) symptom-oriented theories are clearer, easier to test, and more likely to lead to an explanation of psychopathology. These arguments are based on three assumptions respectively: (a) symptoms have higher reliability and validity; (b) underlying pathological processes are symptom-specific; and (c) elucidation of the process of symptom development will lead to (and must precede) the discovery of the causes of syndromes.

We found little evidence supporting these assumptions and arguments based on them.

There are no clear advantages in replacing syndromes with symptoms as units of analysis for psychiatric research.

Antidepressant therapy is not always effective and is slow to take effect. In theory, these shortfalls may be caused by induction of neuronal negative feedback via pre-synaptic 5-HTIA receptors. Pindolol, an antagonist at somatodentritic pre-synaptic 5-HTIA receptors has been investigated as a potential accelerator and augmenter of antidepressant response.

A Medline search was conducted in November 1997.

Six open-label studies and six controlled studies were identified for review Conclusions Open-label studies strongly suggest that pindolol may accelerate and augment antidepressant response, but controlled studies do not wholly support these findings: only three of six studies clearly demonstrate benefit. Larger, well-designed, controlled trials are needed to determine definitively the effectiveness of pindolol in this context.

The risk of suicide in postnatal women is low and those suicides that occur appear to be associated with severe psychiatric illness. No previous study has specifically studied the risk of suicide following post-partum psychiatric disorder.

We calculated standardised mortality ratios (SMRs) for suicide, unnatural deaths and deaths from natural causes for women admitted to psychiatric hospital in the first year after childbirth, using computerised cross-linkages between the Danish Psychiatric Case Register and the Danish registers of birth and causes of death for 1973–1993.

During the study period 1567 women were admitted to psychiatric hospital of whom 107 (6.8%) died. The SMRs (compared with 100) were 1719 (95% CI 1284–2254) for suicide, 1329 (95% CI 1038–1676) for all unnatural causes and 238 (95% CI 167–329) for natural causes. Suicides and deaths from all unnatural causes were most likely to occur in the first year after childbirth, the SMR for suicide within one year being 7216 (95% CI 3945–12 108).

Although postnatal women as a whole appear to have a low rate of suicide, severe post-partum psychiatric disorder is associated with a high rate of deaths from natural and unnatural causes, particularly suicide. The risk is especially high in the first postnatal year, when the suicide risk is increased 70-fold. Close clinical superivision at this time is indicated.

Data on eight specific fears representing DSM–III–R simple phobia were analysed to evaluate: (a) their prevalence and (b) the validity of subtypes of specific phobia defined by DSM–IV.

A modified version of the Composite International Diagnostic Interview was administered to a probability sample of 8098 community respondents. Correlates of responses to questions concerning these fears were analysed.

The most prevalent specific fears were of animals among women, and of heights among men. Slight evidence was found for specific phobia subtypes. Number of fears, independent of type, powerfully predicted impairment, comorbidity, illness course, demographic features, and family psychopathology.

Number of specific fears may mark a general predisposition to psychopathology. More detailed information is needed to resolve the question of specific phobia subtypes.

Hypochondriasis is generally considered difficult to manage. This study aimed to assess the effectiveness of cognitive therapy and to compare it with an equally credible, alternative treatment.

Forty-eight patients with hypochondriasis were initially randomly assigned to either cognitive therapy, behavioural stress management or a no treatment waiting list control group. At the end of the waiting period, patients in the control group were randomly assigned to one of the two treatments. Assessments were at pre-, mid- and post-treatment or waiting list and at three-, six- and 12-month post-treatment follow-up.

Comparisons with the waiting list group showed both treatments were effective. Comparisons between the treatments showed that cognitive therapy was more effective than behavioural stress management on measures of hypochondriasis, but not general mood disturbance at mid-treatment and at post-treatment. One year after treatment patients who had received either treatment remained significantly better than before treatment, and on almost all measures the two therapies did not differ from each other.

Cognitive therapy is a specific treatment for hypochondriasis. Behavioural stress management is also effective but its specificity remains to be demonstrated.

This study in North India compared acute brief psychosis – defined by acute onset, brief duration and no early relapse – with other remitting psychoses, over a 12-year course and outcome.

In a cohort of incident psychoses, we identified 20 cases of acute brief psychosis and a comparison group of 43 other remitting psychoses based on two-year follow-up. Seventeen people (85%) in the acute brief psychosis group and 36 (84%) in the comparison group were reassessed at five, seven and 12 years after onset, and were rediagnosed using ICD–10 criteria.

At 12-year follow-up, the proportion with remaining signs of illness was 6% (n=1) for acute brief psychosis versus 50% (n=18) for the comparison group (P=0.002). Using ICD–10 criteria, the majority in both groups were diagnosed as having schizophrenia.

Acute brief psychosis has a distinctive and benign long-term course when compared with other remitting psychoses. This finding supports the ICD– 10 concept of a separable group of acute and transient psychotic disorders. To effectively separate this group, however, the ICD–10 criteria need modification.

Formal thought disorder is a characteristic feature of psychosis, but little is known of its pathophysiology. We have investigated this in schizophrenia using positron emission tomography (PET).

Regional cerebral blood flow was measured using H215O and PET while six people with schizophrenia were describing a series of 12 ambiguous pictures which elicited different degrees of thought-disordered speech. In a within-subject design, the severity of positive thought disorder was correlated with cerebral blood flow across the 12 scans in each subject.

Verbal disorganisation (‘positive’ thought disorder) was inversely correlated with activity in the inferior frontal, cingulate and left superior temporal cortex, and positively correlated with activity in the parahippocampal/anterior fusiform region bilaterally, and in the body of the right caudate (P<0.001). The total amount of speech produced (independent of thought disorder) was positively correlated with activity in the left inferior frontal and left superior temporal cortex.

The severity of positive thought disorder was inversely correlated with activity in areas implicated in the regulation and monitoring of speech production. Reduced activity in these regions may contribute to the articulation of the linguistic anomalies that characterise positive thought disorder. The positive correlations between positive thought disorder and parahippocampal/anterior fusiform activity may reflect this regions role in the processing of linguistic anomalies.

5-HT2A receptor antagonism may be crucial to the action of atypical antipsychotics. Previous work has related 5-HT2A receptor blockade to clinical efficacy and protection from extrapyramidal side-effects.

We developed a SPET imaging protocol for assessing 5-HT2A receptor binding using the selective ligand 1231-5-1-R91150. Six healthy volunteers, five clozapine- and five risperidone-treated subjects with DSM–IV schizophrenia were studied. Multi-slice SPET was performed on each subject.

Cortex: cerebellum ratios were significantly lower in both clozapine-and risperidone-treated subjects compared with the healthy volunteers in all cortical regions. There was no difference in occupancy between the two drug-treated groups. No correlation was found between the percentage change in the Global Assessment Scale (GAS) and 5-HT2A receptor binding indices in the drug-treated groups.

Clozapine and risperidone potently block 5-HT2A receptors in vivo. The lack of relationship between receptor binding indices and change in GAS suggests that 5-HT2A receptor blockade may be unrelated to clinical improvement. Future studies will substantiate this finding by studying 5-HT2A receptor binding in large groups of patients treated with both typical and novel atypical antipsychotics.

In spite of the virtually ubiquitous nature of the initial 10-day placebo run-in period (IPR) in drug trials, there is little empirical data establishing its relevance.

Data from 593 subjects were examined retrospectively to determine whether or not the prognosis of subjects minimally improved during the IPR was different to those who were unimproved. The IPR period was single-blind and was followed by a six-week double-blind phase in all studies.

Twenty-six per cent of the subjects were minimally improved and 74% were unimproved. Approximately 10% of the subjects who were much improved were not followed systematically. Across a range of diagnosis, severity and chronicity subjects minimally improved (versus unimproved) after IPR had a more favourable prognosis whether assigned to drug or placebo.

Change during IPR appears to be a meaningful predictor. Stratification should be considered in future antidepressant studies.

We tested the validity of two screens for depression in older African–Caribbean adults, the 15-item Geriatric Depression Scale (GDS) and a new Caribbean Culture-Specific Screen for emotional distress (CCSS). Two independent criteria were used for validity: (a) a psychiatric diagnosis derived from GMS–AGECAT, and (b) a culturally sensitive assessment of mental disorder, derived from a tool developed with local African–Caribbean religious healers.

One hundred and sixty-four consecutive African–Caribbean primary care users, aged 60 years or older, were screened with the GDS and the CCSS. Diagnostic interviews were carried out on 80% of high scorers and 20% of low scorers.

The number of cases detected by the two separate diagnostic approaches was similar. However, the agreement between who was and who was not a case was only modest. At a cutoff of $5, the GDS was an adequate case detector for psychiatric depression, and, at a cut-off of $4, for ‘depressed/lost spirit’, as defined by culture-specific criteria. It performed as well as the new CCSS.

At a cut-off of $4 the 15-item GDS can be recommended as a case detector for significant forms of depression in older African–Caribbean people living in south London.

Physical symptoms and psychiatric disorder are associated. We aimed to investigate which comes first.

Data from the Medical Research Council National Survey of Health and Development, a population-based birth cohort study were used at two time points: 36 and 43 years. Six physical symptoms were reported at both time points. The Present State Examination and Psychiatric Symptom Frequency interviews were administered at 36 and 43 years respectively. Odds ratios corrected for a variety of confounders were used to describe the associations between physical symptoms and psychiatric disorder across these two time points.

Psychiatric disorder increased the odds of reporting symptoms 3–7-fold. The relationship strengthened when the outcome was defined as suffering from multiple symptoms. Population attributable risk of psychiatric disorder and subsyndromal disorder in causing multiple somatic symptoms was 40.3%. Prospectively, psychiatric disorder at 36 years was a predictor for five of the six physical symptoms. Three physical symptoms at 36 years predicted new onset of psychiatric symptoms at 43 years.

Psychiatric disorder is strongly related to physical symptoms. The direction of causality may operate in both directions. Assuming a causal relationship, psychiatric disorder (including subthreshold disorders) could account for at most 40% of cases of multiple physical symptoms.

Somatisation is a common and frustrating clinical problem in primary care.

Using structural diagnoses and functional measures, we examined the prevalence and associated features of somatisation disorder defined by three current nosologies and an abridged construct in subjects using primary care services.

Somatisation disorder, diagnosed according to the standard criteria, was found to have a very low prevalence (range 0.06–5%), while more than one-fifth of the sample (22%) met the criteria for the abridged diagnosis. There was poor agreement between succeeding versions of the DSM system for identifying cases of somatisation disorder, each system ending up with rather disparate sets of individuals as well as variable levels of psychopathology and disability.

According to these data, standard somatisation disorder diagnoses add little to the prediction of disability/psychopathology beyond the contributions of an abridged construct of somatisation.