Background. Hypothalamic–pituitary–adrenal (HPA) axis function in patients with chronic depression has previously been shown to be normal when measured using the dexamethasone suppression test (DST). We examined patients with chronic depression using the sensitive dexamethasone/corticotropin releasing hormone (dex/CRH) test and the dexamethasone suppression test (DST) to establish whether HPA axis abnormalities are present in this group. We also compared the sensitivity of the two tests and compared the results with previous studies in depression that have not specifically selected chronic patients.

Method. Twenty-nine patients with the chronic subtype of major depressive disorder and 28 matched controls underwent examination of HPA axis function.

Results. Neither the cortisol response to the DST or the dex/CRH test differed significantly between the patient and control groups. There was a trend in favour of more patients than controls having an abnormal response to the dex/CRH test (P = 0.052). Neither the patients with an abnormally enhanced response, nor the magnitude of response could be predicted by any illness or demographic variable.

Conclusion. The HPA axis is not overtly abnormal in chronic depression. This contrasts with previous findings in acute depression and bipolar disorder and may suggest that the HPA axis abnormalities present in acute depression resolve, but are not accompanied by symptom resolution. Alternatively, a subgroup of depressives with less HPA dysfunction may progress to chronicity. This has implications for treatment and prognosis. The dex/CRH is a more sensitive test of HPA axis function than the DST in patients with chronic depression.

Background. A self-rated measure of health anxiety should be sensitive across the full range of intensity (from mild concern to frank hypochondriasis) and should differentiate people suffering from health anxiety from those who have actual physical illness but who are not excessively concerned about their health. It should also encompass the full range of clinical symptoms characteristic of clinical hypochondriasis. The development and validation of such a scale is described.

Method. Three studies were conducted. First, the questionnaire was validated by comparing the responses of patients suffering from hypochondriasis with those suffering from hypochondriasis and panic disorder, panic disorder, social phobia and non-patient controls. Secondly, a state version of the questionnaire was administered to patients undergoing cognitive-behavioural treatment or wait-list in order to examine the measure's sensitivity to change. In the third study, a shortened version was developed and validated in similar types of sample, and in a range of samples of people seeking medical help for physical illness.

Results. The scale was found to be reliable and to have a high internal consistency. Hypochondriacal patients scored significantly higher than anxiety disorder patients, including both social phobic patients and panic disorder patients as well as normal controls. In the second study, a ‘state’ version of the scale was found to be sensitive to treatment effects, and to correlate very highly with a clinician rating based on an interview of present clinical state. A development and refinement of the scale (intended to reflect more fully the range of symptoms of and reactions to hypochondriasis) was found to be reliable and valid. A very short (14 item) version of the scale was found to have comparable properties to the full length scale.

Conclusions. The HAI is a reliable and valid measure of health anxiety. It is likely to be useful as a brief screening instrument, as there is a short form which correlates highly with the longer version.

Background. Inhibitory functioning is assumed to be deficient in some psychiatric disorders, most notably in patients with schizophrenia and obsessive–compulsive disorder (OCD). In order to investigate inhibitory functioning, priming tasks are commonly used. The present paradigm offers the opportunity to isolate specific distractor features (identity, location) for independent examination.

Methods. Negative priming (NP) for stimulus identity and location was examined in patients with schizophrenia (N = 16), patients with OCD (N = 15) and matched controls (N = 16). All groups performed a referent size-selection task in which they were instructed successively to select the larger one of two cardinal numbers. The deeper processing of both stimuli was expected to yield large NP effects that allow the detection of subtle group differences.

Results. Large NP effects were found for stimulus identity in all three groups. Schizophrenic patients differed from normal controls with respect to the amount of incidental location priming. Subgroup analyses of OCD patients showed NP impairments for checkers when the response–stimulus interval (RSI) was short, but large identity NP when the RSI was long. OCD non-checkers showed normal NP patterns with short RSI, but reduced identity NP with longer RSI.

Conclusions. Schizophrenic patients do not show the ability to use spatial selective attention in order to guide their actions as shown by normal controls. Information processing was differentially impaired in OCD subgroups (checkers and non-checkers) dependent on RSI variation. This result supports those studies that found reduced NP in OCD patients and points to the necessity of differential subgroup studies.

Background. An increasing number of individuals with diverse health complaints are currently seeking help in the field of environmental medicine. Multiple chemical sensitivity (MCS) or idiopathic environmental intolerances (IEI) is defined as an acquired disorder with multiple recurrent symptoms associated with environmental chemicals in low concentrations that are well tolerated by the majority of people. Their symptoms are not explained by any known psychiatric or somatic disorder.

Method. Within a 2-year period we examined 264 of 267 consecutive patients prospectively presenting to a university based out-patient department for environmental medicine. Patients underwent routine medical examination, toxicological analysis and the structured clinical interview for DSM-IV psychiatric disorders (SCID).

Results. Seventy-five per cent of the patients met DSM-IV criteria for at least one psychiatric disorder and 35% of all patients suffered from somatoform disorders. Other frequent diagnoses were affective and anxiety disorders, and dependence or substance abuse. In 39% a psychiatric disorder, in 23% a somatic condition and in 19% a combination of the two were considered to provide sufficient explanation of the symptoms. Toxic chemicals were regarded as the most probable cause in only five cases. The suspected diagnosis of MCS/IEI could not be sustained in the vast majority of cases.

Conclusion. This investigation confirms previous findings that psychiatric morbidity is high in patients presenting to specialized centres for environmental medicine. Somatoform disorders are the leading diagnostic category, and there is reason to believe that certain ‘environmental’ or MCS patients form a special subgroup of somatoform disorders. In most cases, symptoms can be explained by well-defined psychiatric and medical conditions other than MCS, which need specific treatment. Further studies should focus on provocation testing in order to find positive criteria for MCS and on therapeutic approaches that consider psychiatric aspects.

Background. Hypofrontality has been demonstrated in mood disorders by functional brain imaging methods such as positron emission tomography. However, the neurobiological basis of hypofrontality has not been well clarified. Near-infrared spectroscopy (NIRS) is a non-invasive technique for continuous monitoring of alterations in oxygenated (oxyHb) and deoxygenated (deoxyHb) haemoglobin using near-infrared light, which penetrates biological tissues.

Methods. We used NIRS during cognitive and physiological tasks to investigate alterations of haemoglobin oxygenation in the frontal region of euthymic patients with mood disorders (major depressive disorder (MD) and bipolar disorder (BP)) and in controls.

Results. The increase of oxyHb during a verbal fluency task was significantly less in the MD and the BP groups than in the controls. The MD group showed a significantly smaller decrease of oxyHb during hyperventilation than the controls. The BP group also showed a similar trend.

Conclusions. These findings suggest that the hypofrontality in mood disorders may be associated with a poor response in the cerebral blood vessels to neuronal and chemical stimuli.

Background. Exclusion criteria are present in almost all diagnostic categories in both DSM-IV and ICD-10. These exclusion criteria state that one diagnosis is not made if it is ‘due to’ another disorder. However, there is little empirical evidence demonstrating that the hierarchy imposed by the exclusion criteria is meaningful. The current study examines associations between ten common mental disorder pairs to determine whether they are higher for disorder pairs that are classified in DSM-IV and/or ICD-10 as having a hierarchical relationship, than for those that are not hierarchically related.

Method. Data were analysed from the Australian National Survey of Mental Health and Wellbeing, a large (N = 10641) epidemiological survey of mental disorders. Bivariate odds ratios between disorder pairs were calculated from logistic regression analyses. Multivariate odds ratios were also calculated from separate logistic regression analyses in which the sample was restricted to likely positive cases, and co-morbid mental disorders and neuroticism scores were both controlled.

Results. The odds ratios between disorder pairs related according to the exclusion criteria were higher than those of unrelated disorders, for both DSM-IV and ICD-10. When constraints were placed on the data to control for the effects of co-morbid disorders and neuroticism the same pattern was evident in DSM-IV, but not in ICD-10.

Conclusion. The patterns of association between disorder pairs found in the exclusion criteria for DSM-IV appear to more closely mirror the patterns of association found in epidemiological data. While this does not guarantee that the exclusion criteria are valid it is argued that identifying significant associations should be the first step in establishing meaningful exclusion criteria.

Background. Obesity is a growing health problem, but most people find dieting unsuccessful. Three studies examine possible reasons for the difficulty and the extent to which dieting-related reductions in cognitive function are associated with mood and well-being.

Method. In Study One, 49 female dieters were compared with a control group of 31 matched non-dieters on measures of well-being, mood, eating behaviour (Dutch Eating Behaviour Questionnaire), and attention. Study Two examined two measures of restraint to examine why previous studies find high restrainers are prone to react to emotion. Study Three experimentally manipulated mood using music and the standard Velten Induction Procedure to examine attention in restrainers and emotional eaters.

Results. Dieting was found to be associated with deficits in sustained attention. This finding was further supported by the demonstration of a significant impairment in performance following a negative mood induction in high emotional eaters whereas high restrainers were relatively unaffected by the mood challenge.

Conclusions. We suggest that different aspects of eating behaviour have dissociable effects on cognitive-affective function. Trait tendencies to restrained eating are associated with attentional deficits, but are not further affected by mood disruption. It is the long-term tendency to eat when emotional that combines with current emotional state to trigger cognitive deficits.

Background. Much of the difficulty in helping the homeless mentally ill arises as a consequence of their resistance to engagement. A refused intervention can seldom influence a client's problems and engagement status can be argued as being an important independent predictor of outcome. No instrument could be identified which systematically measured the factors involved. This paper describes the development and psychometric properties of a new scale, the Homeless Engagement and Acceptance Scale (HEAS).

Method. Staff from an established project for the homeless mentally ill helped to identify relevant questions used to develop a five-item rating scale for completion by an informant. After piloting, the instrument was tested in a study in which subjects were assessed twice over 12 months by informants. Item analysis was undertaken and predictive validity was assessed.

Results. Item analysis indicated a good facility index signifying all items were able to differentiate subjects according to the characteristic being measured, and a high discrimination index demonstrating that all items were measuring the same concept. Predictive validity and internal consistency coefficients were both good. The 3 month HEAS score was found to be a significant predictor of accommodation status and adequacy of a support network at 12 months.

Conclusions. The good psychometric properties and predictive validity of the scale suggest the HEAS is likely to be a useful tool in assessing engagement status. One of the five questions (Q4) can be omitted for those who are not homeless and the scale termed the Engagement and Acceptance Scale (EAS).

Background. Knowledge of the level of psychosocial impairment associated with psychosis is important in evaluating the impact of the illness on those affected. When such knowledge is derived from community-based epidemiological surveys, it can help in providing a public health perspective for service planning with information derived from representative samples of patients.

Methods. A two-phase epidemiological survey of persons with psychosis in four predominantly urban areas of Australia. First phase screening for psychosis (N = 5710) was followed by a semi-structured interview of a stratified random sample (N = 980) to assess psychopathology (lifetime and current) and psychosocial disability.

Results. Most of the interviewees were unemployed and had never married. There was widespread impairment in sexual and social relationships and in the performance of activities of daily living. Over half expressed dissatisfaction with life in general. Persons with affective psychoses were often as disabled as those with schizophrenia and diagnostic categorizations were not important in the conferment of risk for disability. Rather, poor pre-morbid work or social adjustment and poor course of illness were potent risk factors for diverse forms of disability in persons with psychosis.

Conclusion. A large proportion of persons with experience of psychosis living in the community suffers from significant levels of psychosocial disability. Disablement seems to reflect, in part, a diathesis of poor pre-morbid functioning and less than optimal response to treatment of the disorder.

Background. Reduced serotonin (5-HT) function and deficits on neuropsychological tasks have been separately reported in antisocial populations. We investigated whether these impairments are independent or associated factors underlying impulsivity in aggressive personality disordered (PD) offenders and healthy controls and whether there are associated changes in quantitative brain measures.

Methods. This study reports on the findings from a sample of 51 PD offenders and 24 controls, recruited from maximum security psychiatric hospitals, who were characterized using the Special Hospital Assessment of Personality and Socialisation (SHAPS). Subjects underwent assessment of 5-HT function (prolactin response to D-fenfluramine challenge), neuropsychological testing and had a diagnostic MRI scan. Of this sample 19 controls and 24 patients also had quantitative measurement of frontal and temporal lobe volumes on magnetic resonance imaging (MRI).

Results. Non-psychopathic (low-impulsive) aggressive PDs had enhanced 5-HT function compared with controls and highly impulsive aggressive psychopaths. Primary and secondary psychopaths had poorer executive/frontal, but not memory/temporal neuropsychological function than controls and non-psychopaths. There were no significant group differences in frontal or temporal lobe brain volumes. Although impulsivity and aggression are correlated constructs impulsivity appeared to be related to both executive function and 5-HT function, while aggression only correlated inversely with executive/frontal and memory/temporal function. 5-HT did not directly correlate with frontal or temporal volume or function.

Conclusion. Impulsivity appears to be contributed to by both impaired neuropsychological function and 5-HT function. Impaired neuropsychological function alone makes a contribution to aggression. Treatment needs to take account of the neuropsychological and biochemical deficits in this challenging population.

Background. Categoric, overgeneral autobiographical memory is more common in depressed adults than controls and predicts persistence of depression. This cross-sectional study investigated whether, compared with non-depressed psychiatric cases and community controls, first episode major depressive disorder (MDD) in adolescents is associated with categoric overgeneral memory retrieval.

Methods. Ninety-six clinically referred adolescents (aged 12–17 years) with MDD, 26 non-depressed psychiatric cases and a sample of 33 community controls were recruited. All subjects were assessed using the Kiddie-Schedule for Schizophrenia and Affective Disorders, and completed Williams' cued Autobiographical Memory Test and the Mood and Feelings Questionnaire. Hamilton Depression Rating Scales were completed with MDD subjects, as an index of depression severity.

Results. Adolescents with current first episode MDD retrieved more categoric overgeneral memories than controls, but not than non-depressed psychiatric cases. Adolescents in full remission from a recent episode of MDD retrieved more categoric memories to positive cues than controls. Categoric memory in MDD was related to observer-rated and self-reported severity, but not to the pattern of co-morbid diagnoses. There were negative correlations between IQ and categoric memories in both clinical cases and controls. A positive correlation between categoric memory to negative cues and self-reported depressive symptoms was found in clinical cases (but not controls).

Conclusions. In adolescents, increased categoric overgeneral memory is associated with, but not specific to first episode MDD. Positive categoric memories are also increased in fully remitted MDD as compared to controls.

Background. Repeated, self-damaging behaviour occurring in the context of borderline personality disorder (BPD) may reflect impairments in decision-making and planning cognition. However, there has been no systematic neuropsychological examination of these particular cognitive functions in patients diagnosed with BPD. Such investigations may improve our understanding of the possible role of brain dysfunction in BPD and improve the characterization of the psychological difficulties associated with this disorder.

Method. Forty-two psychiatric patients with a diagnosis of DSM-III-R BPD (41 of whom gave a history of self-harm), without a history of specified ‘psychoses’ or current major affective disorder, were clinically assessed before completing computerized tasks of decision-making and planning previously shown to be sensitive to frontal lobe dysfunction, and tests of spatial and pattern visual recognition memory previously shown to be sensitive to frontal lobe damage and temporal lobe damage respectively. The performance of the BPD patient group was compared with that of a non-clinical control group consisting of 42 subjects.

Results. The performance of the BPD patients on the decision-making task was characterized by a pattern of delayed and maladaptive choices when choosing between competing actions, and by impulsive, disinhibited responding when gambling on the outcome of their decisions. BPD patients also showed impairments on the planning task. There was no evidence of impaired visual recognition memory. Additional analyses suggested only limited effects of current medication and history of previous substance use disorder.

Conclusions. These findings suggest that BPD is associated with complex impairments in dissociable cognitive processes mediated by circuitry encompassing the frontal lobes. These impairments may mediate some of the behavioural changes evident in BPD. Further work is needed to examine the specificity of these findings.

Background. In contrast with the large amount of research on the familial transmission of antisocial behaviour, few studies have investigated similarity between spouses for such behaviour. In addition, none of these studies have examined child conduct disorder (CCD) and adult antisocial behaviour (AAB) separately.

Method. We studied 519 pairs of spouses who completed the Diagnostic Interview Schedule. In each pair, one spouse belonged to a random subsample of persons who had participated in a large population survey and was re-interviewed. Association between spouses for lifetime symptoms and DSM-III criteria of CCD, AAB, antisocial personality disorder and co-morbid psychiatric diagnoses was examined with bivariate and multivariate logistic regression analyses.

Results. We observed a moderate association between spouses for the presence of CCD (OR=4·02, 95% CI=2·03–7·96), and a strong association for the presence of AAB (OR=20·1, 95% CI=5·97–67·5). This similarity for AAB was independent of the similarity for CCD and persisted after adjustment for spousal similarity for disorders co-morbid with AAB. An examination of the relationship between marital status and the presence of CCD and/or AAB in the general population sample (from which originated our sample of couples) suggested that the spousal similarity for AAB was more likely attributable to assortative mating rather than marital contamination.

Conclusion. Our finding of a strong similarity between spouses for AAB has significant implications for both clinicians and researchers. It also suggests that adult antisocial behaviour should be considered as a distinct diagnostic entity, an approach which diverges from DSM-IV diagnostic criteria.

Background. This paper reports population data on DSM-IV generalized anxiety disorder from the Australian National Survey of Mental Health and Well-Being.

Methods. The data were obtained from a nationwide household survey of adults using a stratified multi-stage sampling process. A response rate of 78·1% resulted in 10641 persons being interviewed. Diagnoses were made using the Composite International Diagnostic Interview. The interview was computerized and conducted by trained lay interviewers.

Results. Prevalence in the total sample was 2·8% for 1-month GAD and 3·6% for 12-month GAD. Persons over 55 years of age were less likely to have GAD than those in the younger age groups. Logistic regression analysis also showed that a diagnosis of GAD was significantly associated with being of younger to middle age, being separated divorced or widowed, not having tertiary qualifications or being unemployed. Co-morbidity with another affective, anxiety, substance use or personality disorders was common, affecting 68% of the sample with 1-month DSM-IV GAD. GAD was associated with significant disablement, and 57% of the sample with DSM-IV GAD had consulted a health professional for a mental health problem in the prior 12 months.

Conclusions. The survey provides population data on DSM-IV GAD and its correlates. GAD is a common disorder that is accompanied by significant morbidity and high rates of co-morbidity with affective and anxiety disorders, and is associated with marital status, education, employment status, but not sex. Changes to DSM-IV diagnostic criteria did not appear to affect the prevalence rate compared to previous population surveys.

Background. Research is needed to evaluate the efficacy of prevention and treatment for post-partum depression.

Method. Subjects were screened with the Edinburgh Post-natal Depression Scale (EPDS) at the obstetric clinic. Mothers at risk (N = 258) (EPDS scores [ges ]9) were randomly assigned to a prevention/treatment group or a control group. The prevention group received one cognitive-behavioural prevention session during hospitalization. At 4 to 6 weeks post-partum, subjects were screened again with the EPDS, after drop-out rates (refusals plus no return of the second EPDS) of 25.4% (33/130) in the intervention group and 10.9% (14/128) in the control group. Mothers with probable depression (EPDS scores [ges ]11) were assessed using the Hamilton Depression Rating Scale (HDRS) and the Beck Depression Inventory (BDI). Mothers with major depression continued in the treatment group (N = 18) or in the control group (N = 30). Treated subjects received a cognitive-behavioural programme of between five and eight weekly home-visits.

Results. Compared with the control group, women in the prevention group had significant reductions in the frequency of probable depression (30.2% v. 48.2%). Recovery rates based on HDRS scores of <7 and BDI scores of <4 were also significantly greater in the treated group than in the control group.

Conclusions. The study suggests that this programme for prevention and treatment of post-partum depression is reasonably well-accepted and efficacious.

Background. Early detection of cognitive decline in the elderly is important because this may precede progression to Alzheimer's disease. The aim of this study was to see whether sensitive neuropsychological tests could identify pre-clinical cognitive deficits and to characterize the cognitive profile of a subgroup with poor memory.

Methods. A neuropsychological test battery was administered to a community-dwelling sample of 155 elderly volunteers who were screened with CAMCOG at enrolment (mean age 74·7 years). The battery included tests of episodic memory, semantic and working memory, language and processing speed.

Results. Episodic memory test z scores below 1 S.D. from the cohort mean identified 25 subjects with ‘non-robust’ memory performance. This group was compared to the remaining ‘robust memory’ group with a General Linear Model controlling for age, IQ, education and gender. Test performance was significantly different in all tests for episodic and semantic memory, but not in tests for working memory, processing speed and language. CANTAB paired associates learning and spatial recognition tests identified the highest percentages of those in the ‘non-robust memory’ group. Processing speed partialled out the age effect on memory performance for the whole cohort, but the ‘non-robust memory’ group's performance was not associated with age or processing speed.

Conclusions. Sensitive neuropsychological tests can detect performance below the norm in elderly people whose performance on MMSE and CAMCOG tests is well within the normal range. Age-related decline in memory performance in a cohort of the elderly may be largely due to inclusion within the cohort of individuals with undetected pre-clinical Alzheimer's disease or isolated memory impairment.

Background. Sustained attention has been found to be impaired in individuals suffering from schizophrenia and their close relatives. This has led to the hypothesis that impaired sustained attention is an indicator of vulnerability to schizophrenia.

Methods. The Edinburgh High Risk Study used the Continuous Performance Test-Identical Pairs version (CPT-IP) to assess sustained attention in 127 high risk participants, 30 controls and 15 first-episode schizophrenic patients. A second assessment was completed by 59 high risk and 18 control participants 18 months to 2 years after the first.

Results. No differences in attentional capacity were found between the high risk and control groups and there was no association between genetic liability to schizophrenia and poor performance on the CPT-IP. Additionally, no association between occurrence of psychotic symptoms in the high risk group and impaired attentional capacity was found.

Conclusions. The results suggest that deficits in sustained attention are not indicative of a genetic vulnerability to schizophrenia, and are not associated with the occurrence of psychotic symptoms.

Background. Although it is well known that generalized anxiety disorder (GAD) is highly co-morbid with other mental disorders, little is known about the extent to which earlier disorders predict the subsequent first onset and persistence of GAD. These associations are examined in the current report using data from four community surveys in the World Health Organization (WHO) International Consortium in Psychiatric Epidemiology (ICPE).

Method. The surveys come from Brazil, Canada, the Netherlands and the United States. The Composite International Diagnostic Interview (CIDI) was used to assess DSM-III-R anxiety, mood and substance use disorders in these surveys. Discrete-time survival analysis was used to examine the associations of retrospectively reported earlier disorders with first onset of GAD. Logistic regression analysis was used to examine the associations of the disorders with persistence of GAD.

Results. Six disorders predict first onset of GAD in all four surveys: agoraphobia, panic disorder, simple phobia, dysthymia, major depression and mania. With the exception of simple phobia, only respondents with active disorders have elevated risk of GAD. In the case of simple phobia, in comparison, respondents with a history of remitted disorder also have consistently elevated risk of GAD. Simple phobia is also the only disorder that predicts the persistence of GAD.

Conclusions. The causal processes linking temporally primary disorders to onset of GAD are likely to be state-dependent. History of simple phobia might be a GAD risk marker. Further research is needed to explore the mechanisms involved in the relationship between simple phobia and subsequent GAD.

Background. A key challenge in trauma care is the prevention of psychopathology. However, no definitive method of identifying individuals at risk of developing psychopathology exists. The Aberdeen Trauma Screening Index (ATSI) is a brief screening tool developed for use in a clinical setting by non-mental health professionals to facilitate the early identification of individuals most at risk of psychopathology 3-months post-accident.

Methods. The ATSI derived from a prospective study of a 150 out of an initial pool of 213 consecutive admissions to the Orthopaedic Trauma Unit and the Accident and Emergency Department of Aberdeen Royal Infirmary. Potential predictors were identified by a comprehensive assessment conducted within 1-week post-accident. Outcome at 3-months post-accident was measured using ‘caseness’ according to the General Health Questionnaire (GHQ-28).

Results. The ATSI is based on a final model comprising only seven variables with a sensitivity of 79% and specificity of 65%. A predictive index score (0–100) was produced to ensure the practical utility of the ATSI in a clinical setting. A ROC curve was constructed to illustrate the relationship between sensitivity and the specificity values with their corresponding threshold scores. On the basis of a prevalence rate of 55% ‘caseness’, as identified in the present study, a cut-off point of 45 provides the optimal outcome with a sensitivity value of 70% and a specificity value of 71%.

Conclusions. The ATSI can accurately identify those most at risk of developing psychopathology 3-months post-accident in a sample of accidentally injured adult subjects recruited as consecutive admissions to an urban hospital in the North East of Scotland. However, to establish the generalizability of these findings, it is important that the ATSI be validated in both similar and diverse populations.

Background. Despite an abundance of clinical research on premenstrual and menstrual symptoms, few epidemiological data provide estimates of the prevalence, incidence, co-morbidity, stability and correlates of premenstrual dysphoric disorder (PMDD) in the community.

Aims. To describe the prevalence, incidence, 12 co-morbidity factors and correlates of threshold and subthreshold PMDD in a community sample of young women.

Methods. Findings are based on prospective–longitudinal community survey of 1488 women aged 14–24, who were followed-up over a period of 48 months (follow-up, N = 1251) as part of the EDSP sample. Diagnostic assessments were based on the Composite International Diagnostic Interview (CIDI) and its 12-month PMDD diagnostic module administered by clinical interviewers. Diagnoses were calculated using DSM-IV algorithms, but daily ratings of symptoms, as required, were not available.

Results. The baseline 12-month prevalence of DSM-IV PMDD was 5·8%. Application of the diagnostic exclusion rules with regard to concurrent major depression and dysthymia decreased the rate only slightly (5·3%). An additional 18·6% were ‘near-threshold’ cases, mostly because they failed to meet the mandatory impairment criterion. Over the follow-up period only few new PMDD cases were observed: cumulative lifetime incidence was 7·4%. PMDD syndrome was stable across 48 months with <10% complete remissions among baseline PMDD cases. The 12-month and lifetime co-morbidity rates were high (anxiety disorders 47·4%, mood disorders 22·9%; somatoform 28·4%), only 26·5% had no other mental disorder. Particularly high odds ratios were found with nicotine dependence and PTSD. In terms of correlates increased rates of 4-weeks impairment days, high use of general health and mental health services, and increased rates of suicide attempts were found.

Conclusion. In this sample of adolescents and young adults, premenstrual symptoms were widespread. However, DSM-IV PMDD was considerably less prevalent. PMDD is a relatively stable and impairing condition, with high rates of health service utilization, increased suicidality and substantial co-morbidity.