In an important paper in this issue Murphy et al. (pp. 503–512) report on incidence rates for depression over 40 years in the Stirling County Study. Incidence rates remained stable, contrary to most evidence that is principally based on prevalence or retrospective lifetime prevalence. Incidence rates were only a little higher in women than in men.

Incidence rates are assuming increasing importance in psychiatry as they do in other areas of epidemiology. Prevalence rates are complex, depending not only on incidence of new cases of the disorder, but its persistence, and in a recurrent disorder, recurrence. Incidence rates are unbiased by these factors, a particularly valuable attribute when considering risk factors. In depression, development and widespread use of antidepressants, newer patterns of care, and in the opposite direction, increased urbanization with its accompanying social pressures, could have produced considerable changes in outcome in the last 40 years. Changes in enumerated prevalences alone could reflect changed length of episodes and rates of recurrence, so giving a misleading picture as to what has happened to the disorder.

The Stirling County Study is one of the classics of epidemiology. Its originator, Alexander Leighton, is an author of the present paper with his wife, Jane Murphy, who has directed the study since the mid-1970s. Representative community samples were studied cross-sectionally in 1952, 1969 and 1992, and the previously studied samples restudied on follow-up at the later points. The design enables a separation of period effects, involving all subjects at one time point, from cohort effects. In this study similar temporal stability of prevalence has been found (Murphy et al. 2000b) to that now reported for incidence.

Background. Several empirical studies have found temporal lobe impairments in many patients with schizophrenia. The strength and consistency of this evidence, however, has not been evaluated and synthesized quantitatively. Hence, we ask to what extent are temporal cortices really defective in schizophrenia?

Methods. Meta-analytical methods were used to determine the magnitude of evidence in support of structural and physiological temporal–hippocampal system deficits in schizophrenia. We report effect sizes from studies since 1980 that used structural (CT, MRI) and functional (SPECT, PET) neuroimaging methods.

Results. Both structural and functional imaging literatures are distinguished by heterogeneity whereby most patients show normative temporal function and structure, a minority shows diminished values and some patients demonstrate augmented function and structure rather than a deficit.

Conclusions. The findings are hard to incorporate within single disease models that propose major involvement of the temporal system in schizophrenia, at least at the degree of resolution obtained with current imaging technology.

Background. The Stirling County Study provides a 40-year perspective on the epidemiology of psychiatric disorders in an adult population in Atlantic Canada. Across samples selected in 1952, 1970 and 1992 current prevalence of depression was stable. This paper concerns time trends in annual incidence as assessed through cohorts selected from the first two samples.

Methods. Consistent interview data were analysed by a computerized diagnostic algorithm. The cohorts consisted of subjects at risk for a first depression: Cohort-1 (N = 575) was followed 1952–1970; Cohort-2 (N = 639) was followed 1970–1992. Life-table methods were used to calculate incidence rates and proportional hazards procedures were used for statistical assessment.

Results. Average annual incidence of depression was 4·5 per 1000 for Cohort-1 and 3·7 for Cohort- 2. Differences by gender, age and time were not statistically significant. The stability of incidence and the similarity of distribution by gender and age in these two cohorts corresponds to findings about the two early samples. In contrast, current prevalence in the recent sample was distributed differently and showed an increase among women under 45 years.

Conclusions. The stability of the incidence of depression emphasizes the distinctive characteristics of current prevalence in the recent sample and suggests that the dominance of women in rates of depression may have occurred among those born after the Second World War. The results offer partial support for the interpretation of an increase in depression based on retrospective data in other recent studies but they indicate that the increase is specific to women.

Background. Data on the course of anxiety in late life are scarce. The present study sets out to investigate the course of anxiety, as measured by the HADS-A (Zigmond & Snaith, 1983) in community dwelling older persons, and to evaluate predictive factors for change over 3 years in anxiety symptoms following the vulnerability/stress model.

Method. Based on the first anxiety assessment, two cohorts were formed: subjects with and subjects without anxiety symptoms. In the non-anxious cohort (N = 1602) we studied risk factors for the development of anxiety symptoms; in the anxious cohort (N = 563) the same factors were evaluated on their predictive value for restitution of symptoms. Risk factors included vulnerability factors (demographics, health status, personality characteristics and social resources) and stressors (life events occurring in between both anxiety assessments). Logistic regression models estimated the effects of vulnerability factors, stress and their interaction on the likelihood of becoming anxious and chronicity of anxiety symptoms.

Results. It was indicated that the best predictors for becoming anxious were being female, high neuroticism, hearing/eyesight problems and life-events. Female sex and neuroticism also increased the likelihood of chronicity of anxiety symptoms in older adults, but life events were not related to chronicity. The main stressful event in late life associated with anxiety was death of one's partner. Vulnerability factors and stress added on to each other rather than their interaction being associated with development or chronicity of anxiety.

Conclusion. The vulnerability/stress model offers a useful framework for organizing risk factors for development and chronicity of anxiety symptoms in older persons, but no support was attained for the hypothesis that vulnerability and stress amplify each others effects. Finally, the results indicate to whom preventive efforts should be directed: persons high in neuroticism, women, and those who experience distressing life events.

Background. The aims of this study were to use longitudinal report data on physical and sexual abuse to examine the stability and consistency of abuse reports.

Methods. The study was based on the birth cohort of young people studied in the Christchurch Health and Development Study. At ages 18 and 21 years, these young people were questioned about their childhood exposure to physical punishment and sexual abuse. Concurrent with these assessments, sample members were also assessed on measures of psychiatric disorder and suicidal behaviour.

Results. Reports of childhood sexual abuse and physical punishment were relatively unstable and the values of kappa for test–retests of abuse reporting were in the region of 0·45. Inconsistencies in reporting were unrelated to the subject's psychiatric state. Latent class analyses suggested that: (a) those not abused did not falsely report being abused; and (b) those who were abused provided unreliable reports in which the probability of a false negative response was in the region of 50%. Different approaches to classifying subjects as abused led to wide variations in the estimated prevalence of abuse but estimates of the relative risk of psychiatric adjustment problems conditional on abuse exposure remained relatively stable.

Conclusions. There was substantial unreliability in the reporting of child abuse. This unreliability arose because those who were subject to abuse often provided false negative reports. The consequences of errors in reports appear to be: (a) that estimates of abuse prevalence based on a single report are likely to seriously underestimate the true prevalence of abuse; while (b) estimates of the relative risk of psychiatric adjustment problems conditional on abuse appear to be robust to the effects of reporting errors.

Background. The importance of genetic factors in the aetiology of manic-depressive illness (MDI) has been repeatedly confirmed and indicators of vulnerability to the illness in families with affective disorders are needed. Abnormal event-related potentials (ERP) may be markers of genetic vulnerability to mental illness. Long latency and low amplitude of P300 have consistently been reported in schizophrenic patients and their relatives. A few studies have also shown P300 deficits in MDI patients, but no ERP study has been performed on their relatives.

Methods. ERPs were recorded during an auditory oddball task in 19 relatives belonging to families with two or more bipolar patients and in controls with no familial or personal history of affective disorders. The relatives were selected as having no affective disorders on a lifetime basis, but eight had an anxiety disorder.

Results. In all relatives, a lower P300 amplitude and a longer P300 latency was found, with much longer reaction time and post-N200 duration till button-press than controls. A lack of P300 amplitude dominance in the right hemisphere was also found in relatives in comparison with controls. There also appeared to be a frontal predominance of ERP abnormalities in relatives.

Conclusion. We report the first evidence of deficits in reaction time and in P300 amplitude and latency, and a lack of P300 right-sided dominance, in relatives of manic-depressive patients. This pattern may constitute an endophenotypic marker of manic-depressive disorder.

Background. Impaired distractor inhibition may contribute to the selective attention deficits observed in depressed patients, but studies to date have not tested the distractor inhibition theory against the possibility that processes such as transient memory review processes may account for the observed deficits. A negative priming paradigm can dissociate inhibition from such a potentially confounding process called object review. The negative priming task also isolates features of the distractor such as colour and location for independent examination.

Method. A computerized negative priming task was used in which colour, identification and location features of a stimulus and distractor were systematically manipulated across successive prime and probe trials. Thirty-two unmedicated subjects with DSM-IV diagnoses of non-psychotic unipolar depression were compared with 32 age, sex and IQ matched controls.

Results. Depressed subjects had reduced levels of negative priming for conditions where the colour feature of the stimulus was repeated across prime and probe trials but not when identity or location was the repeated feature. When both the colour and location feature were the repeated feature across trials, facilitation in response was apparent.

Conclusions. The pattern of results supports studies that found reduced distractor inhibition in depressed subjects, and suggests that object review is intact in these subjects. Greater impairment in negative priming for colour versus location suggests that subjects may have greater impairment in the visual stream associated with processing colour features.

Background. Motor slowing in depression may be associated with a relative dopaminergic (DA) deficit. Bradykinesia in Parkinson's syndrome is associated with an abnormally short silent period (SP) using transcranial magnetic stimulation (TMS). We hypothesized that depression would also be associated with a short SP.

Methods. Sixteen patients with DSM-IV depression and 19 matched controls participated. SPs were elicited by exercising the contralateral abductor policis brevis (APB) during TMS.

Results. The SP was significantly increased in the patient group. No correlation was found between SP and depression score.

Conclusion. A long SP suggests increased motor cortical inhibition in depression. This finding is inconsistent with the hypothesis of behavioural motor slowing in depression being associated with Parkinsonian-like mechanisms including the dopaminergic deficit. There is a need for studies incorporating larger patient groups to investigate potential correlations between SP and depression indices.

Background. We have used proton magnetic resonance imaging and spectroscopy to measure hippocampus/amygdala volumes and anterior hippocampal metabolite concentrations (N-acetyl aspartate (NAA), creatine/phosphocreatine and choline) in subjects with temporal lobe epilepsy (TLE), schizophrenia and in normal controls.

Method. Four groups of right-handed patients were selected: 12 with TLE and psychosis (EP), 12 with TLE and no psychosis (ENP), 26 with schizophrenia, and 38 normal controls. Imaging and spectroscopy were performed with a 1.5T Signa GE scanner.

Results. The schizophrenia group showed a significant left-sided reduction in all metabolites. In the epilepsy groups NAA was reduced bilaterally. The NAA reduction in the EP group was greater than in the ENP group, especially on the left, although the result did not reach significance. Total hippocampus/amygdala volumes showed no significant differences in any of the groups when compared with normal controls. When compared with controls significant, specific regional volume reductions were present bilaterally in the EP group and in the left hippocampus/amygdala in schizophrenia. The regional volume reduction found in schizophrenia was also present in EP but not in ENP.

Conclusion. Spectroscopic abnormalities were more pronounced in the epilepsy groups and were bilateral, and abnormalities in schizophrenia were left sided. Specific regional hippocampus/amygdala volume reductions were more marked in the EP group and were bilateral. Left-sided regional volume reduction identified in the dominant hemisphere of schizophrenics was also present in EP patients, but not in ENP, suggesting that this region in the left temporal lobe may be significant in the aetiology of psychosis. This is further supported by the predominantly left-sided NAA reduction in schizophrenia. High resolution morphometric studies may identify specific regions of the brain associated with the development of psychosis.

Background. Catatonia is a psychomotor syndrome that can be characterized by behavioural, affective and motor abnormalities. In order to reveal further underlying pathophysiological mechanisms of psychomotor disturbances in catatonia we investigated neuropsychological function and regional cerebral perfusion (r-CBF) in a combined study.

Methods. Ten catatonic patients were investigated with Tc-99mECD brain SPECT and compared with 10 psychiatric (similar age, sex, medication and underlying psychiatric diagnosis but without catatonic syndrome) and 20 healthy controls. Neuropsychological measures included tests for general intelligence, attention, executive functions and right parietal visual–spatial abilities. Correlational analyses were performed between neuropsychological measures, catatonic symptoms and r-CBF.

Results. Catatonic patients showed a significant decrease of r-CBF in right lower and middle prefrontal and parietal cortex compared with psychiatric and healthy controls as well as significantly poorer performance in visual–spatial abilities associated with right parietal function. Correlational analysis revealed significant correlations between visual–spatial abilities and right parietal r-CBF only in psychiatric and healthy controls but not in catatonic patients. In contrast, attentional measures correlated significantly with motor symptoms, visual–spatial abilities and right parietal r-CBF in catatonia only but not in psychiatric or in healthy controls.

Conclusion. Findings are preliminary but suggest right lower prefronto-parietal cortical dysfunction in catatonia, which may be closely related to psychomotor disturbances.

Background. Many studies have suggested a possible aetiological role for obstetric complications in the development of schizophrenia. We focused on prenatal physical growth in schizophrenia, a contentious issue in the literature.

Methods. We compared gestational age at birth, birth weight (BW) and birth head circumference (BHC) between 312 schizophrenics and 517 controls, and between 187 schizophrenics and their matched healthy siblings. Information on obstetric histories was obtained from the Maternal and Child Health Handbooks (i.e. contemporaneous records).

Results. Gestational age at birth was significantly earlier in the schizophrenics than in the controls (P = 0·017). Pre-term birth (gestational age of 36 weeks or less) was more common in schizophrenics than in controls (8·0% v. 3·4%, P = 0·005, odds ratio 2·5). Low BW (2500 g or less) was more frequent in schizophrenics than in controls (9·6% v. 4·6%, P = 0·005, odds ratio 2·2). The schizophrenics had significantly lighter BW (P = 0·0003) and tended to have smaller BHC (P = 0·081) compared with controls. However, multiple regression analysis showed that there was no significant difference in BW or BHC between the schizophrenics and controls when gestational age and maternal weight were controlled. There was no significant difference in BW or BHC between schizophrenics and their siblings, although the schizophrenics tended to be born at earlier gestational age than their siblings.

Conclusions. Our results suggest that prematurity at birth is associated with a risk of developing schizophrenia in adulthood. When gestational age and maternal body weight were allowed for, there was no evidence that schizophrenics tend to have lower mean BW or smaller BHC.

Background. Little work has been published on the internal structure of the Mini-Mental State Examination (MMSE), one of the most widely used instruments for grading cognitive status in clinical settings and field research.

Methods. MMSE responses from a sample of older adults (50–98 years) in five US sites (N = 8556) were analysed.

Results. A five-factor solution was found to be most appropriate. The first factor (concentration) had large loadings with serial sevens and spell world backwards items. The second factor (language and praxis) had large loadings with naming, follow command and praxis items. The third factor (orientation) had loadings with orientation to time and place items. The fourth factor (memory) had large loadings with delayed recall items and the fifth (attention) had large loadings with immediate registration items.

Conclusions. We found that the MMSE is essentially unidimensional; nevertheless, evidence was revealed suggesting that the MMSE is a multidimensional assessment instrument. Dimensions revealed in this sample correspond directly to MMSE sections articulated by the developers of the instrument. These findings have not been reported in previous factor analyses of the MMSE. The findings support the construct validity of the MMSE as a measure of cognitive mental state among community dwelling older adults.

Background. Dementia screening instruments, such as the Cambridge Cognitive Examination (CAMCOG), measure a variety of cognitive functions. However, memory impairment generally is the first sign of Alzheimer's disease (AD). It seems logical, therefore, to use only memory-related items for the early detection of AD. We divided the CAMCOG into a memory section and a non-memory section, and tested the hypothesis that the memory section predicts AD better than the non-memory section. We also provide normative data for both sections.

Methods. Normal subjects (N = 169) and patients with incident AD (i.e. satisfying AD criteria between 1 and 3 years from baseline; N = 25) were participants in the Amsterdam Study of the Elderly (AMSTEL), a population-based longitudinal study on cognitive decline and dementia. Patients with prevalent AD (i.e. satisfying AD criteria at baseline; N = 155) were either recruited in a memory clinic or came from AMSTEL. Normal subjects were cognitively intact at baseline and remained so for at least 3 years. The CAMCOG was administered to all subjects. AD was diagnosed by DSM-III-R criteria.

Results. Logistic regression analysis showed that the memory section was related to prevalent AD, whereas in multivariate analysis the non-memory section was not (after correction for the memory score and demographic characteristics). A similar analysis showed that the memory section predicted incident AD, as did a higher score on the non-memory section. The MMSE did not predict incident AD better than age alone.

Conclusion. For the early detection of AD it is best to use the memory and non-memory sections separately instead of the total CAMCOG score.

Background. In primary care consultations patients with emotional distress tend to give verbal cues or symptom reports with psychological or psychiatric contents. This study examined the cue behaviour defined not only by psychological, but also by medical, social and life episodes related contents in patients with and without emotional distress, recognized and not by their GP. The GP's verbal behaviour in relation to patients' cue emission was also investigated.

Method. For the six participating GPs two groups of matched pairs of patients (N = 238) were created. The two groups comprised either patients considered by GPs as being without emotional distress or patients considered as emotionally distressed. Within each pair, one patient was a case (GHQ-12 score > 2) and the other was the matched control (GHQ-12 score < 3). The medical interviews with these patients were transcribed and classified according to the Verona Medical Interview Classification System (VR-MICS).

Results. GHQ positive patients of both groups gave more cues in terms of total proportion than their matched controls (GHQ negative patients). The proportion of cues given by patients was related also to GP's verbal behaviour, increasing with closed psychosocial questions and decreasing with the use of active interview techniques. Attribution of emotional distress was more frequent when patients were high attenders and had a past psychiatric history. The content of cues changed in relation to GP's attribution: recognized patients gave more cues and more often with psychological content, patients not recognized as distressed gave mainly cues related to their lifestyle and life episodes.

Conclusions. To improve the recognition of those emotionally distressed patients most likely to be missed GPs should increase their attention to cues related to life style and life episodes.

Background. The Australian National Survey of Mental Health and Well-being was designed to detect and describe psychiatric morbidity, associated disability, service use and perceived need for care. The survey employed a single-phase interview methodology, delivering a field questionnaire to a clustered probability sample of 10641 Australians. Perceived need was sampled with an instrument designed for this survey, the Perceived Need for Care Questionnaire (PNCQ). This questionnaire gathers information about five categories of perceived need, assigning each to one of four levels of perceived need. Reliability and validity studies showed satisfactory performance of the instrument.

Methods. Perceived need for mental health care in the Australian population has been analysed using PNCQ data, relating this to diagnostic and service utilization data from the above survey.

Results. The survey findings indicate that an estimated 13·8% of the Australian population have perceived need for mental health care. Those who met interview criteria for a psychiatric diagnosis and also expressed perceived need make up 9·9% of the population. An estimated 11·0% of the population are cases of untreated prevalence, a minority (3·6% of the population) of whom expressed perceived need for mental health care. Among persons using services, those without a psychiatric diagnosis based on interview criteria (4·4% of the population), showed high levels of perceived met need.

Conclusions. The overall rate of perceived need found by this methodology lies between those found in the USA and Canada. The findings suggest that service use in the absence of diagnosis elicited by survey questionnaires may often represent successful intervention. In the survey, untreated prevalence was commonly not accompanied by perceived need for mental health care.

Background. Few population studies have investigated the functional disabilities that accompany specific psychiatric diagnoses. This study assesses the nature and strength of current and residual impairments in various functional domains of life.

Methods. Data were derived from the Netherlands Mental Health Survey and Incidence Study (NEMESIS), a prospective study in the Dutch general population aged 18 to 64 (N = 7147). Psychiatric diagnoses were based on the Composite International Diagnostic Interview; functional disability was assessed on the basis of the Short-Form-36 and the number of disability days.

Results. Psychopathology was associated with increased disability in social, emotional and physical domains of life. Disability levels varied by psychiatric diagnosis, with mood disorders showing the poorest levels of functioning, especially for vitality and social functioning; alcohol-related disorders were associated with few disabilities. Co-morbidity strongly aggravated the disability. The effect of contextual factors on disability was limited, although somatic ill health, unemployment and adverse youth history increased the likelihood of functional disability. The findings indicate that psychopathology can also have residual debilitating effects.

Conclusions. Mental health care providers should be aware that the extent and the type of disability may vary with the different types of disorders and among different groups within the population. Since recovery from functional limitations may not be complete or may take more time than the remission of the psychiatric symptomatology, non-psychiatric follow-up care is needed. The high number of lost work days is relevant from an economic perspective. There is a need for illness-specific disability assessment instruments.

Background. The study examined the role of parental death and chronic depression with severe episodes in affecting risk of breast cancer. This avenue of research is in accord with oncology findings, which suggests that causative factors of breast cancer occur and develop over a period of 20 years or longer.

Methods. Participants consisted of 1213 women in the Baltimore Epidemiologic Catchment Area study surveyed in 1980 and followed through 1994–1995. They were assessed for depressive and anxious disorders, paternal death in childhood and relatively recent adverse life events prior to cancer hospitalization.

Results. In the course of the study, 29 women were hospitalized for breast cancer and 10 died of breast cancer. The psychosocial variables that predicted increased risk of breast cancer were maternal death in childhood (OR = 2·56, P < 0·001) and chronic depression with severe episodes (OR = 14·0, P < 0·001). Neither relatively recent life events nor other depressive and anxiety disorders were associated with increased risk. Maternal death and chronic depression with severe episodes were reported to have occurred at least 20 years prior to breast cancer hospitalization.

Conclusions. Maternal death and chronic and severe depression occurred at least 20 years prior to breast cancer hospitalization and could have been involved in the causation or facilitation of cancer development. The authors suggest that meta-analysis of other prospective studies are needed to create larger groups of individuals with these stresses to confidently establish these variables as risk factors.

Background. While neuropsychological dysfunction is common in geriatric depression, not all aspects of cognition are equally affected. It has been suggested that depressed patients are impaired only in tasks that make heavy demands on processing resources and that a resource decrement therefore underlies the neuropsychological decrements seen in geriatric depression. The present study examined whether processing resources in the form of working memory and information processing speed are decreased in depression and whether a decrease in these resources actually mediates neuropsychological impairment.

Methods. Measures of processing resources were administered to elderly depressed patients prior to treatment and to age-matched controls. Patients whose depression remitted were retested as were the controls. Subjects also received neuropsychological tests of episodic memory and visuospatial performance.

Results. Depressed patients performed significantly worse on measures of both processing speed and working memory. While performance on these measures improved in patients whose depression remitted, the amount of improvement was no greater than that seen in the controls with repeat testing. Hierarchical regression analyses showed that depression explained a significant amount of variance on the neuropsychological tasks. However, if the variance associated with processing resources was removed first, depression no longer accounted for a significant amount of neuropsychological variance.

Conclusions. Processing resources are decreased in elderly depressed patients and this decrease in resources appears to mediate impairments in several areas of neuropsychological functioning including episodic memory and visuospatial performance. The resource decrement persists after remission of the depression and thus may be a trait marker of geriatric depression.

Background. Self-report data obtained from members of advocate groups for patients with anxiety or mood disorders in 11 countries were used to study time to initial professional help-seeking after incident episodes.

Method. Data were taken from 3516 self-administered questionnaires completed by members of GAMIAN, an international consortium of mental health patient advocate groups. Reports about age at onset and age at first seeking treatment were obtained retrospectively.

Results. Approximately 40% of respondents reported that they sought treatment in the same year as the first onset of their disorder. The median delay in help-seeking was 8 years for the remainder of respondents. Synthetic cohort analysis suggests that delays have decreased in recent cohorts. However, time to initial help-seeking in all cohorts and all countries was found to be inversely related to age at onset.

Conclusions. Although caution is needed in generalizing the results beyond members of patient advocate groups, the key patterns found here were also found in previous analyses of general population surveys carried out in the US and Canada. The critical and consistent finding in all these studies is that presumably curable adolescents with early-onset disorders are, in effect, ignored by the treatment system in these countries. Aggressive outreach and intervention among early-onset cases might prove to be a cost-effective approach both to prevent the onset of secondary disorders and to improve success in treating primary disorders.

Background. Phobic disorder is one of the most prevalent psychiatric disorders in community populations and much attention has focused on the association of sociodemographic factors and social resources with the disorder. There has been little investigation of the more personal resources such as self-confidence, religiosity, social support and self-perceived health that may increase vulnerability.

Methods. We used a sample of 2914 community residents aged 18 or older who participated in the Duke University Epidemiologic Catchment Area Study to explore the relationship between both social and personal resources and the prevalence of DIS/DSM-III phobic disorder.

Results. The 1-month prevalence of any phobic disorder was 7·6%. In bivariate analyses, we found both race/ethnicity and gender differences in prevalence, with a higher prevalence of phobic disorder in African-Americans and females. Lower socio-economic status, rural residence and unmarried status were also associated with current prevalence. No association was found for social network and social interaction. Impaired subjective social support, low self-confidence, perceived poorer physical health and co-morbid psychiatric disorder were significantly associated with current prevalence in uncontrolled analyses, while associations between lack of a confidant as well as religiosity and phobic disorder were not. Female gender (OR = 1·7), perceived low self- confidence (OR = 2·0), and two interaction terms, age × co-morbid psychiatric disorder and race/ethnicity × perceived physical health were associated with phobic disorder in controlled analyses using logistic regression.

Conclusions. We conclude that both social and personal resources, particularly self-confidence, co-morbidity and perceived physical health are important correlates of phobic disorder.