Depression is a common, disabling condition for which psychological treatments, in particular cognitive behavioural therapies are recommended. Promising results in recent randomized trials have renewed interest in behavioural therapy. This systematic review sought to identify all randomized trials of behavioural therapy for depression, determine the effect of such interventions and examine any moderators of such effect.

Randomized trials of behavioural treatments of depression versus controls or other psychotherapies were identified using electronic database searches, previous reviews and reference lists. Data on symptom-level, recovery/dropout rate and study-level moderators (study quality, number of sessions, severity and level of training) were extracted and analysed using meta-analysis and meta-regression respectively.

Seventeen randomized controlled trials including 1109 subjects were included in this meta-analysis. A random-effects meta-analysis of symptom-level post-treatment showed behavioural therapies were superior to controls [standardized mean difference (SMD) −0.70, 95% CI −1.00 to −0.39, k=12, n=459], brief psychotherapy (SMD −0.56, 95% CI −1.0 to −0.12, k=3, n=166), supportive therapy (SMD −0.75, 95% CI −1.37 to −0.14, k=2, n=45) and equal to cognitive behavioural therapy (SMD 0.08, 95% CI −0.14 to 0.30, k=12, n=476).

The results in this study indicate behavioural therapy is an effective treatment for depression with outcomes equal to that of the current recommended psychological intervention. Future research needs to address issues of parsimony of such interventions.

Internet-based cognitive behaviour therapy (CBT) is a promising new approach for the treatment of depressive symptoms. The current study had two aims: (1) to determine whether, after 1 year, an internet-based CBT intervention was more effective than a waiting-list control group; and (2) to determine whether the effect of the internet-based CBT differed from the group CBT intervention, 1 year after the start of treatment.

A total of 191 women and 110 men (mean age=55 years, s.d.=4.6) with subthreshold depression were randomized into internet-based treatment, group CBT (Lewinsohn's Coping with Depression Course), or a waiting-list control condition. The main outcome measure was treatment response after 1 year, defined as the difference in pretreatment and follow-up scores on the Beck Depression Inventory (BDI). Missing data were imputed using the multiple imputation procedure of data augmentation. Analyses were performed using multiple imputation inference.

In the waiting-list control group, we found a pretreatment to follow-up improvement effect size of 0.69, which was 0.62 in the group CBT condition and 1.22 with the internet-based treatment condition. Simple contrasts showed a significant difference between the waiting-list condition and internet-based treatment (p=0.03) and no difference between both treatment conditions (p=0.08).

People aged over 50 years with subthreshold depression can still benefit from internet-based CBT 1 year after the start of treatment.

Genetic epidemiology data suggest that younger age of onset is associated with family history (FH) of depression. The present study tested whether the presence of FH for depression or anxiety in first-degree relatives determines younger age of onset for depression.

A sample of 1022 cases with recurrent major depressive disorder (MDD) was recruited at the Max Planck Institute and at two affiliated hospitals. Patients were assessed using the Schedules for Clinical Assessment in Neuropsychiatry and questionnaires including demographics, medical history, questions on the use of alcohol and tobacco, personality traits and life events. Survival analysis and the Cox proportional hazard model were used to determine whether FH of depression signals earlier age of onset of depression.

Patients who reported positive FH had a significantly earlier age of onset than patients who did not report FH of depression (log-rank=48, df=1, p<0.0001). The magnitude of association of FH varies by age of onset, with the largest estimate for MDD onset before age 20 years (hazard ratio=2.2, p=0.0009), whereas FH is not associated with MDD for onset after age 50 years (hazard ratio=0.89, p=0.5). The presence of feelings of guilt, anxiety symptoms and functional impairment due to depressive symptoms appear to characterize individuals with positive FH of depression.

FH of depression contributes to the onset of depression at a younger age and may affect the clinical features of the illness.

Vagus nerve stimulation (VNS) therapy is associated with a decrease in seizure frequency in partial-onset seizure patients. Initial trials suggest that it may be an effective treatment, with few side-effects, for intractable depression.

An open, uncontrolled European multi-centre study (D03) of VNS therapy was conducted, in addition to stable pharmacotherapy, in 74 patients with treatment-resistant depression (TRD). Treatment remained unchanged for the first 3 months; in the subsequent 9 months, medications and VNS dosing parameters were altered as indicated clinically.

The baseline 28-item Hamilton Depression Rating Scale (HAMD-28) score averaged 34. After 3 months of VNS, response rates (⩾50% reduction in baseline scores) reached 37% and remission rates (HAMD-28 score <10) 17%. Response rates increased to 53% after 1 year of VNS, and remission rates reached 33%. Response was defined as sustained if no relapse occurred during the first year of VNS after response onset; 44% of patients met these criteria. Median time to response was 9 months. Most frequent side-effects were voice alteration (63% at 3 months of stimulation) and coughing (23%).

VNS therapy was effective in reducing severity of depression; efficacy increased over time. Efficacy ratings were in the same range as those previously reported from a USA study using a similar protocol; at 12 months, reduction of symptom severity was significantly higher in the European sample. This might be explained by a small but significant difference in the baseline HAMD-28 score and the lower number of treatments in the current episode in the European study.

The association between personality traits and the first lifetime onset of clinically significant depression has not been studied in older adults.

Experienced psychiatrists conducted interviews and chart reviews at baseline and throughout the 15-year follow-up period. Survival analyses were conducted on the presence/absence of a DSM-III-R mood disorder at follow-up.

There were 59 cases of first lifetime episodes of depression. Analyses showed that Neuroticism [hazard ratio (HR) per one point increase in the Maudsley Personality Inventory (MPI)=1.05, 95% confidence interval (CI) 1.02–1.08] but not Extroversion (HR 1.02, 95% CI 0.97–1.06) amplified risk for mood disorder.

This prospective study on a randomly sampled birth cohort of older adults showed that Neuroticism confers risk for a first lifetime episode of clinically significant depression. Findings have implications for understanding the etiology of late-life depression (LLD) and could also aid in the identification and treatment of people at risk.

To examine the effect of time on suicide after bereavement among widowed persons.

The data were extracted from Swiss mortality statistics for the period 1987–2005. The time between bereavement and subsequent death, specifically by suicide, was determined by linkage of individual records of married persons. The suicide rates and the standardized mortality ratios in the first week/month/year of widowhood were calculated based on person-year calculations.

The annualized suicide rates in widowed persons were highest in the first week after bereavement: 941 males and 207 females per 100 000. The corresponding standardized mortality ratios were approximately 34 and 19 respectively. In the first month(s) after bereavement, the rates and the ratios decreased, first rapidly, then gradually. Except in older widows, they did not reach the baseline levels during the first year after bereavement.

The suicide risk of widowed persons is increased in the days, weeks and months after bereavement. Widowed persons are a clear-cut risk group under the aegis of undertakers, priests and general practitioners.

Psychotherapy's equivalence paradox is that treatments tend to have equivalently positive outcomes despite non-equivalent theories and techniques. We replicated an earlier comparison of treatment approaches in a sample four times larger and restricted to primary-care mental health.

Patients (n=5613) who received cognitive–behavioural therapy (CBT), person-centred therapy (PCT) or psychodynamic therapy (PDT) at one of 32 NHS primary-care services during a 3-year period (2002–2005) completed the Clinical Outcomes in Routine Evaluation – Outcome Measure (CORE-OM) at the beginning and end of treatment. Therapists indicated which approaches were used on an End of Therapy form. We compared outcomes of groups treated with CBT (n=1045), PCT (n=1709), or PDT (n=261) only or with one of these plus one additional approach (e.g. integrative, supportive, art), designated CBT+1 (n=1035), PCT+1 (n=1033), or PDT+1 (n=530), respectively.

All six groups began treatment with equivalent CORE-OM scores, and all averaged marked improvement (overall pre/post effect size=1.39). Neither treatment approach nor degree of purity (‘only’ v. ‘+1’) had a statistically significant effect. Distributions of change scores were all similar.

Replicating the earlier results, the theoretically different approaches tended to have equivalent outcomes. Caution is warranted because of limited treatment specification, non-random assignment, incomplete data, and other issues. Insofar as these routine treatments appear effective for patients who complete them, those who fail to complete (or to begin) treatment deserve attention by researchers and policymakers.

Insufficient evidence exists for a viable choice between long- and short-term psychotherapies in the treatment of psychiatric disorders. The present trial compares the effectiveness of one long-term therapy and two short-term therapies in the treatment of mood and anxiety disorders.

In the Helsinki Psychotherapy Study, 326 out-patients with mood (84.7%) or anxiety disorder (43.6%) were randomly assigned to three treatment groups (long-term psychodynamic psychotherapy, short-term psychodynamic psychotherapy, and solution-focused therapy) and were followed up for 3 years from start of treatment. Primary outcome measures were depressive symptoms measured by self-report Beck Depression Inventory (BDI) and observer-rated Hamilton Depression Rating Scale (HAMD), and anxiety symptoms measured by self-report Symptom Check List Anxiety Scale (SCL-90-Anx) and observer-rated Hamilton Anxiety Rating Scale (HAMA).

A statistically significant reduction of symptoms was noted for BDI (51%), HAMD (36%), SCL-90-Anx (41%) and HAMA (38%) during the 3-year follow-up. Short-term psychodynamic psychotherapy was more effective than long-term psychodynamic psychotherapy during the first year, showing 15–27% lower scores for the four outcome measures. During the second year of follow-up no significant differences were found between the short-term and long-term therapies, and after 3 years of follow-up long-term psychodynamic psychotherapy was more effective with 14–37% lower scores for the outcome variables. No statistically significant differences were found in the effectiveness of the short-term therapies.

Short-term therapies produce benefits more quickly than long-term psychodynamic psychotherapy but in the long run long-term psychodynamic psychotherapy is superior to short-term therapies. However, more research is needed to determine which patients should be given long-term psychotherapy for the treatment of mood or anxiety disorders.

Clinical guidelines advise against prescribing more than one antipsychotic with limited exceptions. Despite this, surveys continue to report high antipsychotic polypharmacy rates. The aim of the study was to investigate the effectiveness of a multi-faceted intervention in reducing prescribing of antipsychotic polypharmacy on general adult psychiatry wards, compared with guidelines alone.

A pragmatic cluster randomized controlled trial recruited 19 adult psychiatric units (clusters) from the South West of England. Participants were all ward doctors and nurses. The multi-faceted intervention comprised: an educational/CBT workbook; an educational visit to consultants; and a reminder system on medication charts.

The odds of being prescribed antipsychotic polypharmacy in those patients prescribed antipsychotic medication was significantly lower in the intervention than control group when adjusted for confounders (OR 0.43, 95% CI 0.21–0.90, p=0.028). There was considerable between-unit variation in polypharmacy rates and in the change in rates between baseline and follow-up (5 months after baseline).

The intervention reduced levels of polypharmacy prescribing compared to guidelines alone although the effect size was relatively modest. Further work is needed to elicit the factors that were active in changing prescribing behaviour.

The aim was to reduce non-attendance for first-time consultations at psychiatric out-patient clinics.

The study was a pragmatic randomized controlled trial; the setting was seven inner-city UK out-patient clinics in Leeds. The participants were 764 subjects of working age with an appointment to attend a psychiatric out-patient clinic for the first time. The intervention was an ‘orientation statement’ letter delivered 24–48 h before the first appointment compared with standard care. The primary outcome measure was attendance at the first appointment; secondary outcomes included hospitalization, transfer of care, continuing attendance, discharge, presentation at accident and emergency and death by 1 year.

Follow-up was for 763 out of 764 subjects (>99%) for primary and for 755 out of 764 subjects (98.8%) of secondary outcome data. The orientation statement significantly reduced the numbers of people failing to attend [79 out of 388 v. 101 out of 376 subjects, relative risk 0.76, 95% confidence interval (CI) 0.59–0.98, number needed to treat 16, 95% CI 10–187].

Prompting people to go to psychiatric out-patient clinics for the first time encourages them to attend. Pragmatic trials within a busy working environment are possible and informative.

The ACE project involved 62 participants with a first episode of psychosis randomly assigned to either a cognitive behaviour therapy (CBT) intervention known as Active Cognitive Therapy for Early Psychosis (ACE) or a control condition known as Befriending. The study hypotheses were that: (1) treating participants with ACE in the acute phase would lead to faster reductions in positive and negative symptoms and more rapid improvement in functioning than Befriending; (2) these improvements in symptoms and functioning would be sustained at a 1-year follow-up; and (3) ACE would lead to fewer hospitalizations than Befriending as assessed at the 1-year follow-up.

Two therapists treated the participants across both conditions. Participants could not receive any more than 20 sessions within 14 weeks. Participants were assessed by independent raters on four primary outcome measures of symptoms and functioning: at pretreatment, the middle of treatment, the end of treatment and at 1-year follow-up. An independent pair of raters assessed treatment integrity.

Both groups improved significantly over time. ACE significantly outperformed Befriending by improving functioning at mid-treatment, but it did not improve positive or negative symptoms. Past the mid-treatment assessment, Befriending caught up with the ACE group and there were no significant differences in any outcome measure and in hospital admissions at follow-up.

There is some preliminary evidence that ACE promotes better early recovery in functioning and this finding needs to be replicated in other independent research centres with larger samples.

Predicting cognitive deficits in early psychosis may well be crucial to identify those individuals most in need of receiving intensive intervention. As yet, however, the identification of potential pretreatment predictors for cognitive performance has been hampered by inconsistent findings across studies. We aimed to examine the associations of functional and clinical pretreatment variables with cognitive functioning after a first psychotic episode.

One hundred and thirty-one patients experiencing first-episode psychosis were assessed for psychopathology, pre-morbid functioning, duration of illness, age of onset, and family history of psychosis and neurocognitive functioning. Multiple regression analyses were conducted for six basic cognitive dimensions known to be affected in this population: verbal learning, verbal memory, verbal comprehensive abilities, executive functioning, motor dexterity and sustained attention.

Pre-morbid functioning was the main predictor for five out of the six basic cognitive domains. Pre-morbid social adjustment difficulties were associated with worse performance in executive functioning, motor dexterity and sustained attention. Academic functioning was associated with verbal comprehension, and verbal learning and memory. Gender, age of onset, duration of untreated psychosis, and family history of psychosis had no or limited value as predictors of neurocognitive outcome.

Poor pre-morbid functioning was related to a worse performance in the six basic cognitive dimensions evaluated; however, this accounted for only a small amount of the explained variance. Cognitive impairment is a prominent feature in patients with early psychosis regardless of favorable prognostic features such as short duration of illness, female gender, later age of onset, and non-family history of psychosis.

Patients with schizophrenia have been found to display abnormalities in social cognition. The aim of the study was to test whether patients with schizophrenia and unaffected first-degree relatives of schizophrenic patients display behavioural signs of social brain dysfunction when making social judgements.

Eighteen patients with schizophrenia, 24 first-degree unaffected relatives and 28 healthy comparison subjects completed a task which involves trustworthiness judgements of faces. A second task was completed to measure the general ability to recognize faces.

Patients with schizophrenia rated faces as more trustworthy, especially those that were judged to be untrustworthy by healthy comparison subjects. Siblings of schizophrenia patients display the same bias, albeit to a lesser degree.

The pattern of more positive trustworthiness judgements parallels the results from studies of patients with abnormalities in brain areas involved in social cognition. Because patients and siblings did not differ significantly from controls in their general ability to recognize faces, these findings cannot be dismissed as abnormalities in face perception by itself.

Cognitive impairment and negative symptoms are two of the primary features of schizophrenia associated with poor social functioning. We examined the relationships between clinical characteristics, specific cognitive abilities and social skills performance in middle-aged and older out-patients with schizophrenia and normal comparison subjects.

One hundred and ninety-four middle-aged and older schizophrenia out-patients and 60 normal comparison subjects were administered a standardized, performance-based measure of social skills using role-plays of various social situations [Social Skills Performance Assessment (SSPA)] and measures of current level of social contact (the Lehman Quality of Life Interview), psychiatric symptom severity [the Positive and Negative Syndrome Scale (PANSS) and the Hamilton Depression Rating Scale (HAMD)], insight [the Birchwood Insight Scale (IS)] and cognitive functioning [the Mattis Dementia Rating Scale (DRS)].

Patients demonstrated worse social skills compared with normal subjects. Better performance on the SSPA was associated with having less severe positive and negative symptoms, fewer social contacts, and better attention, initiation/freedom from perseveration, visuospatial ability, abstraction ability and memory. After controlling for demographic, clinical and insight-related factors, abstraction ability was the strongest predictor of social skills performance, followed by frequency of social contact.

Social functioning (as measured through direct observation of social skills performance) was related to cognitive ability in out-patients with schizophrenia. Addressing such cognitive impairment may help to improve social functioning and result in greater overall quality of life.