This issue contains two reviews, one on the question of ‘what is a mood stabilizer?’, and one on genetic influences on measures of the environment. Other sets of papers examine various aspects of genetics, bipolar disorder, and five individual papers examine a variety of topics.

The concept of mood stability is attractive to both patients and clinicians alike, and hence the term ‘mood stabilizer’ has widespread currency. However, its worldwide acceptance and use in clinical practice is at odds with the absence of official recognition by regulatory authorities. The ideal mood stabilizer is said to have efficacy in the treatment of acute manic and depressive episodes, and also be effective in the prevention of recurrences. However, in reality, few drugs with perhaps the exception of lithium, come close to this gold standard; yet many agents aspire to the title, and some have arguably achieved it prematurely. It is, therefore, important to reconsider the definition of a mood stabilizer and critically review which agents, if any, satisfy the necessary eligibility requirements by reference to reasonable criteria and comparator data. The term mood stabilizer is an important label. It needs to be applied judiciously because it confers clinical credibility and qualifies long-term use in maintenance and prophylaxis. It is also important with respect to developing guidelines for treatment and the further development of novel agents. Most importantly, however, it is a term that is innately appealing because of what it promises: for this reason alone it should encompass only those agents that can deliver.

Background. Traditional models of psychiatric epidemiology often assume that the relationship between individuals and their environment is unidirectional, from environment to person. Accumulating evidence from developmental and genetic studies has made this perspective increasingly untenable.

Method. Literature search using Medline, PsycINFO, article references and contact with experts to identify all papers examining the heritability of measures of environments of relevance to psychiatry/psychology.

Results. We identified 55 independent studies organized into seven categories: general and specific stressful life events (SLEs), parenting as reported by child, parenting reported by parent, family environment, social support, peer interactions, and marital quality. Thirty-five environmental measures in these categories were examined by at least two studies and produced weighted heritability estimates ranging from 7% to 39%, with most falling between 15% and 35%. The weighted heritability for all environmental measures in all studies was 27%. The weighted heritability for environmental measures by rating method was: self-report 29%, informant report 26%, and direct rater or videotape observation (typically examining 10 min of behavior) 14%.

Conclusion. Genetic influences on measures of the environment are pervasive in extent and modest to moderate in impact. These findings largely reflect ‘actual behavior’ rather than ‘only perceptions’. Etiologic models for psychiatric illness need to account for the non-trivial influences of genetic factors on environmental experiences.

Background. Previous research suggests that genetic influences on disordered eating may be greater in pubertal than pre-pubertal girls. Although these findings are consistent with pubertal activation of genetic influences on disordered eating, earlier studies were unable to directly test this hypothesis. The purpose of the present study therefore was to directly examine this possibility by investigating whether pubertal development moderates genetic influences on disordered eating.

Method. Participants were 510 female adolescent twins from the Minnesota Twin Family Study. Disordered eating was measured with the Total Score of the Minnesota Eating Behavior Survey, while pubertal status was assessed with the Pubertal Development Scale.

Results. Consistent with our hypothesis, model-fitting indicated significant increases in genetic influence on disordered eating with advancing pubertal development.

Conclusions. These findings suggest that puberty influences the expression of genes for disordered eating.

Background. Theory and evidence strongly suggest that perfectionism may be a risk factor for eating disorders. The purpose of the current study was to investigate a model that would explain the relationship between the cognitive diagnostic criterion for both anorexia nervosa and bulimia nervosa, namely undue influence of body weight or shape on self-evaluation, and dimensions of perfectionism. The model of particular interest was the common cause model, which hypothesizes that the phenotypes are caused by the same underlying genetic and environmental risk factors.

Method. Female twins (n=1002) from the Australian Twin Registry (ATR), aged 28 to 39 years, were interviewed using the Eating Disorder Examination (EDE). In addition, questions relating to the Equal Environment Assumption (EEA) and the Frost Multidimensional Perfectionism Scale, namely concern over mistakes (CM), personal standards (PS) and doubts about actions (DA), were assessed.

Results. There was no evidence of violation of the EEA for any of the four phenotypes. Univariate models showed all phenotypes to be influenced by both genetic and non-shared environmental action, where genetic estimates ranged from 25% to 39% of the variance. Multivariate analyses suggested the best explanation of covariation among the phenotypes was an independent pathways, rather than a common pathways, model.

Conclusions. Undue influence of body weight or shape on self-evaluation shared about 10% of its sources of genetic and environmental variance with perfectionism, thus suggesting that a common cause model does not represent the best explanation of the relationship between perfectionism and this cognitive diagnostic criterion for eating disorders.

Background. The DSM-IV cluster C Axis II disorders include avoidant (AVPD), dependent (DEPD) and obsessive-compulsive (OCPD) personality disorders. We aimed to estimate the genetic and environmental influences on dimensional representations of these disorders and examine the validity of the cluster C construct by determining to what extent common familial factors influence the individual PDs.

Method. PDs were assessed using the Structured Interview for DSM-IV Personality (SIDP-IV) in a sample of 1386 young adult twin pairs from the Norwegian Institute of Public Health Twin Panel (NIPHTP). A single-factor independent pathway multivariate model was applied to the number of endorsed criteria for the three cluster C disorders, using the statistical modeling program Mx.

Results. The best-fitting model included genetic and unique environmental factors only, and equated parameters for males and females. Heritability ranged from 27% to 35%. The proportion of genetic variance explained by a common factor was 83, 48 and 15% respectively for AVPD, DEPD and OCPD. Common genetic and environmental factors accounted for 54% and 64% respectively of the variance in AVPD and DEPD but only 11% of the variance in OCPD.

Conclusion. Cluster C PDs are moderately heritable. No evidence was found for shared environmental or sex effects. Common genetic and individual environmental factors account for a substantial proportion of the variance in AVPD and DEPD. However, OCPD appears to be largely etiologically distinct from the other two PDs. The results do not support the validity of the DSM-IV cluster C construct in its present form.

Background. Personality disorders (PDs) as assessed by questionnaires and personal interviews are heritable. However, we know neither how much unreliability of measurement impacts on heritability estimates nor whether the genetic and environmental risk factors assessed by these two methods are the same. We wish to know whether the same set of PD vulnerability factors are assessed by these two methods.

Method. A total of 3334 young adult twin pairs from the Norwegian Institute of Public Health Twin Panel (NIPHTP) completed a questionnaire containing 91 PD items. One to 6 years later, 1386 of these pairs were interviewed with the Structured Interview for DSM-IV Personality (SIDP-IV). Self-report items predicting interview results were selected by regression. Measurement models were fitted using Mx.

Results. In the best-fit models, the latent liabilities to paranoid personality disorder (PPD), schizoid personality disorder (SPD) and schizotypal personality disorder (STPD) were all highly heritable with no evidence of shared environmental effects. For PPD and STPD, only unique environmental effects were specific to the interview measure whereas both environmental and genetic effects were found to be specific to the questionnaire assessment. For SPD, the best-fit model contained genetic and environmental effects specific to both forms of assessment.

Conclusions. The latent liabilities to the cluster A PDs are highly heritable but are assessed by current methods with only moderate reliability. The personal interviews assessed the genetic risk for the latent trait with excellent specificity for PPD and STPD and good specificity for SPD. However, for all three PDs, the questionnaires were less specific, also indexing an independent set of genetic risk factors.

Background. Electrophysiological endophenotypes are far less explored in bipolar disorder as compared to schizophrenia. No previous twin study of event-related potentials (ERPs) in bipolar illness has been reported. This study uses a twin design and advanced genetic model fitting analyses aiming to (1) assess and quantify the relationship of a range of ERP components with bipolar disorder with psychotic features, and (2) examine the source of the relationship (due to genetic or environmental factors).

Method. P300, P50 suppression and mismatch negativity (MMN) were recorded in 10 discordant monozygotic (MZ) bipolar twin pairs, six concordant MZ bipolar twin pairs and 78 control twin pairs. Statistical analyses were based on structural equation modelling.

Results. Bipolar disorder was significantly associated with smaller P300 amplitude and decreased P50 suppression. Genetic correlations were the main source of the associations, estimated to be −0·33 for P300 amplitude and 0·46 for P50 ratio. Individual-specific environmental influences were not significant. MMN and P300 latency were not associated with the illness.

Conclusions. The results provide supporting evidence that P300 amplitude and P50 suppression ratio are ERP endophenotypes for bipolar disorder.

Background. Impairments in verbal learning and memory, executive functions and attention are manifest in some euthymic patients with bipolar disorder (BPD). However, evidence is sparse on their putative role as aetiologically important genetic vulnerability markers for the disorder. This population-based study examined the cognitive functions of affected and unaffected individuals in families with BPD. The aim was to discover whether any cognitive function would indicate genetic liability to the disorder and could thus be regarded as endophenotypes of BPD.

Method. A diagnostic interview and a neuropsychological test battery were administered to 32 familial bipolar I disorder patients, 40 of their unaffected first-degree relatives and 55 controls, all representing population-based samples.

Results. Unaffected first-degree relatives showed impairment in psychomotor performance speed and slight impairment in executive function. Bipolar patients were impaired in verbal learning and memory compared with unaffected relatives and controls. They also differed from controls in tasks of executive functions. There were no difference between the groups in simple attention and working memory tasks.

Conclusions. Impaired psychomotor performance speed and executive function may represent endophenotypes of BPD, reflecting possible underlying vulnerability to the disorder. Verbal memory impairments appear to be more related to the fully developed disorder.

Background. Cognitive behavioural therapy (CBT) has been implemented in bipolar (BP) disorder with varying degrees of success. The link between CBT for BP disorder and underlying theory is not clear. There have been attempts to identify a unique style of dysfunctional thinking in BP disorder analogous to that in unipolar (UP) disorder but this has suffered from a dearth of prospective studies controlling for mood and phase of illness. In this prospective study, we have examined whether dysfunctional thinking and the (dys)regulation of traumatic memories are trait vulnerability factors in BP versus UP mood disorders.

Method. BP-1 and UP groups were followed-up from acute episode to recovery, and compared with a healthy control group. Measures of dysfunctional thinking and linked personality dimensions were taken, together with measures assessing autobiographical memory retrieval (AMT). The presence and control of intrusive memories of traumatic events in BP and UP disorder were assessed.

Results. Controlling for mood symptoms and phase of illness, the BP (but not the UP) group was largely indistinguishable from controls on dysfunctional thinking. Intrusive memories of traumatic events were present in 45% of the BP and 48% of the UP groups; and those without intrusions were more overgeneral on the AMT in all phases of illness.

Conclusion. These findings are in line with the ‘affect regulation’ hypothesis for UP disorder and deliberate self-harm. This may help in understanding the mode of action of CBT in BP disorder and to further improve the therapy, primarily through the promotion of affective regulation, which is one component of the complex CBT intervention for BP disorder.

Background. Morphological changes in the corpus callosum (CC) have been described in bipolar disorder, but with inconsistencies among the reports. We investigated the CC areas by magnetic resonance imaging (MRI) in 12 first-episode patients with bipolar disorder and 12 controls.

Method. Twelve medication-naive patients with bipolar I disorder (six males, six females; aged 28·2±6·5 years) with manic or mixed episodes and 12 age- and gender-matched healthy controls (six males, six females; aged 26·8±7·6 years) were recruited to the study. MRI scans was obtained using a 1.5-T GE Signa Excite high-speed scanner. Anatomical measurements were conducted on a computer workstation with the software Scion Image Beta-3b for Windows. Statistical analysis was performed using an analysis of covariance (ANCOVA), the t test, χ2 and partial correlation analyses.

Results. Bipolar patients had significantly smaller areas of total CC, anterior body posterior body and isthmus compared with healthy control subjects by ANCOVA, with age, gender and intracranial volume (ICV) as covariates. There was a negative correlation between total CC, posterior body and isthmus areas and Young Mania Rating Scale (YMRS) scores.

Conclusion. The findings suggest that CC morphology may be associated with the pathophysiology of bipolar disorder.

Background. Depression is associated with smoking, but the causality of the relationship is debated. The authors examine smoking behaviour as a predictor of depression among the Finnish adult twin population.

Method. Based on responses to surveys in 1975 and 1981, the authors characterized the subjects as never smokers, persistent former smokers, quitters, recurrent smokers and persistent smokers. The Beck Depression Inventory (BDI) was applied in 1990 to measure depression (BDI score >9). Although the population consisted of twins, the authors first considered the subjects as individuals. Logistic regression models were computed for 4164 men and 4934 women. In order to control for family and genetic background, conditional logistic regression analyses were conducted among twin pairs discordant for depression. Bivariate genetic modelling was used to examine genetic and environmental components of the correlation between smoking and depression.

Results. Among the men, persistent smoking (OR 1·42, 95% CI 1·07–1·89) and smoking in 1975 but quitting by 1981 (OR 1·68, 95% CI 1·17–2·42) was associated with a higher risk of depression, while among the women only the quitters had an elevated risk (OR 1·38, 96% CI 1·01–1·87). The gender×smoking interaction showed persistent smoking to be a stronger risk for men. When family and genetic background were controlled, smoking remained a predictor of depression. Genetic modelling among the men suggested a modest correlation (rg=0·25) between genetic components of smoking and depression.

Conclusions. Smoking behaviour may be a gender-sensitive predictor of depression, the stronger association in men being partly accounted for by having underlying genes in common.

Background. This study sought to evaluate the acceptance of two brief psychological interventions for depressed individuals, contacted through a community survey, and to look for predictors of adherence at the patient level.

Method. The authors used data from the Outcomes of Depression International Network (ODIN) study, which included a randomized controlled trial in which depressed individuals from five European countries, and nine geographical areas were assigned to one of three groups: individual problem-solving treatment, group psychoeducation, or control group. In this analysis, we included all of the individuals who had been assigned to one of the psychological interventions. Compliance with intervention was defined in two different ways. Multiple logistic regression was used to see which variables might predict an individual's compliance with psychological treatment.

Results. Psychological intervention was offered to 236 subjects. Treatment was completed by 128 subjects and not by 108 (compliance definition A). Three variables were found to have an effect on compliance A: the presence of a confidant, the use of antidepressant medication during the previous 6 months, and the previous use of any social or health services. On the other hand, 164 subjects had agreed to at least start the treatment, and 72 had not (compliance definition B). The three factors associated with compliance B were presence of a confidant, previous use of services, and the ‘desire for change’ score.

Conclusions. Social support and previous use of services are the main predictors of compliance with a psychological treatment in depressed individuals from the community. Implications for clinical practice and community programs are discussed.

Background. Although a host of studies have now examined the relationship between quality of life (QoL) and non-seasonal depression, few have measured QoL in seasonal affective disorder (SAD). We report here on results from the Can-SAD trial, which assessed the impact of treatment with either antidepressant medication or light therapy upon QoL in patients diagnosed with SAD.

Method. This Canadian double-blind, multicentre, randomized controlled trial included 96 patients who met strict diagnostic criteria for SAD. Eligible patients were randomized to 8 weeks of treatment with either: (1) 10000 lux light treatment and a placebo capsule or (2) 100 lux light treatment (placebo light) and 20 mg fluoxetine. QoL was measured with the Quality of Life Enjoyment and Satisfaction Questionnaire (Q-LES-Q) and the Medical Outcomes Study (MOS) Short-Form General Health Survey (SF-20) at baseline and 8 weeks.

Results. Both intervention groups showed significant improvement in QoL over time with no significant differences being detected by treatment condition. Q-LES-Q scores increased significantly in seven of eight domains, with the average scores rising from 48·0 (S.D.=10·7) at baseline to 69·1 (S.D.=15·6) at week 8. Treatment-related improvement in QoL was strongly associated with improvement in depression symptoms.

Discussion. Patients with SAD report markedly impaired QoL during the winter months. Treatment with light therapy or antidepressant medication is associated with equivalent marked improvement in perceived QoL. Studies of treatment interventions for SAD should routinely include broader indices of patient outcome, such as the assessment of psychosocial functioning or life quality.

Background. We identified the need to develop a scientifically rigorous measure of health-related quality of life (HRQL) in dementia that would be appropriate for use at all stages of dementia severity and would be available in both self- and proxy-report versions.

Method. We used standard psychometric methods to eliminate items with poor psychometric properties (item-reduction field test) and to assess the acceptability, reliability and validity of the item-reduced instruments (psychometric evaluation field test). We developed and validated two versions of DEMQOL: a 28-item interviewer-administered questionnaire that is self-reported by the person with dementia (DEMQOL) and a 31-item interviewer-administered questionnaire that is proxy-reported by a caregiver (DEMQOL-Proxy).

Results. DEMQOL shows high reliability (internal consistency and test–retest) and moderate validity in people with mild/moderate dementia. DEMQOL-Proxy shows good acceptability and internal consistency and moderate evidence of validity in people with mild/moderate and severe dementia. Test–retest reliability and performance in people with severe dementia need further testing.

Conclusions. DEMQOL and DEMQOL-Proxy show psychometric properties that are comparable with the best available dementia-specific measures of HRQL. We recommend that DEMQOL and DEMQOL-Proxy are used together. Reliability and validity need to be confirmed in independent samples and responsiveness needs to be evaluated.

Background. Discriminating Alzheimer's disease (AD) and mild cognitive impairment (MCI) from depression is a challenge in psychogeriatric medicine. A study was set up to ascertain whether cued recall could be useful in differentiating early AD and MCI from depression among elderly individuals.

Method. The Visual Association Test (VAT) and the Memory Impairment Screen-plus (MIS-plus) were administered together with the Mini-Mental State Examination (MMSE) and the Geriatric Depression Scale (GDS) to 40 MCI patients, 35 mild AD patients, 46 depressed patients and 52 healthy control subjects.

Results. A one-way analysis of variance (ANOVA) followed by post-hoc Scheffé tests showed that AD patients had significantly lower cued recall scores (i.e. combined VAT and MIS-plus scores) than MCI patients, who in turn had lower scores than depressed patients. The scores of depressed patients and controls were not significantly different. Discriminant analysis revealed that 94% of the AD patients and 96% of the depressed patients could be classified correctly by means of the GDS and the cued recall sores. Receiver operating characteristic (ROC) curves identified an optimal cut-off score of 8 (maximum score 12) for differentiating AD and MCI patients from depressed elderly patients and controls. Applying this cut-off, a sensitivity of 83% (58%) and a specificity of 85% (85%) was obtained when differentiating AD (MCI) from depression.

Conclusions. Cued recall, operationalized by the combined scores of VAT and MIS-plus, is a useful method for differentiating AD patients from depressed individuals and healthy controls. Probably because of the great heterogeneity among MCI patients, the diagnostic power of cued recall decreases when applied to differentiate MCI from depression.