Bayesian experimental design is a fast growing area of research with many real-world applications. As computational power has increased over the years, so has the development of simulation-based design methods, which involve a number of algorithms, such as Markov chain Monte Carlo, sequential Monte Carlo and approximate Bayes methods, and which have enabled more complex design problems to be solved. TheBayesian framework provides a unified approach for incorporating prior information and/or uncertainties regarding the statistical model with a utility function which describes the experimental aims. In this paper, we provide a general overview on the concepts involved in Bayesian experimental design, and focus on describing some of the more commonly-used Bayesian utility functions and methods for their estimation, as well as a number of algorithms that are used to search over the design space to find the optimal Bayesian design. We also discuss other computational strategies for further research in Bayesian optimal design.

BackgroundThere is limited evidence on the prevalence and identification of antenatal mental disorders.AimsTo investigate the prevalence of mental disorders in early pregnancy and the diagnostic accuracy of depression-screening (Whooley) questions compared with the Edinburgh Postnatal Depression Scale (EPDS), against the Structured Clinical Interview DSM-IV-TR.MethodCross-sectional survey of women responding to Whooley questions asked at their first antenatal appointment. Women responding positively and a random sample of women responding negatively were invited to participate.ResultsPopulation prevalence was 27% (95% CI 22–32): 11% (95% CI 8–14) depression; 15% (95% CI 11–19) anxiety disorders; 2% (95% CI 1–4) obsessive–compulsive disorder; 0.8% (95% CI 0–1) post-traumatic stress disorder; 2% (95% CI 0.4–3) eating disorders; 0.3% (95% CI 0.1–1) bipolar disorder I, 0.3% (95% CI 0.1–1%) bipolar disorder II; 0.7% (95% CI 0–1) borderline personality disorder. For identification of depression, likelihood ratios were 8.2 (Whooley) and 9.8 (EPDS). Diagnostic accuracy was similar in identifying any disorder (likelihood ratios 5.8 and 6).ConclusionsEndorsement of Whooley questions in pregnancy indicates the need for a clinical assessment of diagnosis and could be implemented when maternity professionals have been appropriately trained on how to ask the questions sensitively, in settings where a clear referral and care pathway is available.Declaration of interestL.M.H. chaired the National Institute for Health and Care Excellence CG192 guidelines development group on antenatal and postnatal mental health in 2012–2014.

BackgroundIncreasing evidence suggests that postnatal paternal depression is associated with adverse emotional, behavioural and cognitive outcomes in children. Despite this, few studies have determined the prevalence of fathers’ depressive symptoms during the first few years of their children’s lives and explored what factors are related to these symptoms. We estimated the prevalence and examined associated risk factors of paternal depressive symptoms in a nationally representative sample of fathers with children aged between 9 months and 7 years old from the Millennium cohort study. The risk factors examined were maternal depressive symptoms, marital conflict, child temperament, child gender, paternal education, fathers’ ethnic background, fathers’ employment status, family housing, family income and paternal age.MethodsSecondary data analysis was conducted using the UK Millennium cohort study, which consisted of data from England, Scotland, Wales and Northern Ireland of families with infants born in the year 2000/2001. Data from four sweeps were used from when children in the cohort were aged 9 months, 3, 5 and 7 years old (n = 5155–12,396).ResultsThe prevalence of paternal depressive symptoms over time was 3.6 % at 9 months, 1.2 % at 3 years old, 1.8 % at 5 years and 2.0 % at 7 years (using Kessler cut-off points to categorise high depressive symptoms vs low depressive symptoms). Linear regression trends (using continuous measures of depressive symptoms) indicated that both paternal and maternal depressive symptoms decreased over time, suggesting similar patterns of parents’ depressive symptoms after the birth of a child, but the decrease was more evident for mothers. Paternal depressive symptoms were consistently associated with fathers’ unemployment, maternal depressive symptoms and marital conflict. Socioeconomic factors such as rented housing when child was 9 months and low family income when child was 5 and 7 years were also associated with higher paternal depressive symptoms.ConclusionsPaternal depressive symptoms decreased among fathers when their children were aged between 9 months to 3 years old. Paternal unemployment, high maternal depressive symptoms and high marital conflict were important risk factors for paternal depressive symptoms. In light of our findings, we would recommend a more family centred approach to interventions for depression in the postnatal period.

No abstract