Summary. We consider the problem of comparing complex hierarchical models in which the number of parameters is not clearly defined. Using an information theoretic argument we derive a measure pD for the effective number of parameters in a model as the difference between the posterior mean of the deviance and the deviance at the posterior means of the parameters of interest. In general pD approximately corresponds to the trace of the product of Fisher's information and the posterior covariance, which in normal models is the trace of the ‘hat’ matrix projecting observations onto fitted values. Its properties in exponential families are explored. The posterior mean deviance is suggested as a Bayesian measure of fit or adequacy, and the contributions of individual observations to the fit and complexity can give rise to a diagnostic plot of deviance residuals against leverages. Adding pD to the posterior mean deviance gives a deviance information criterion for comparing models, which is related to other information criteria and has an approximate decision theoretic justification. The procedure is illustrated in some examples, and comparisons are drawn with alternative Bayesian and classical proposals. Throughout it is emphasized that the quantities required are trivial to compute in a Markov chain Monte Carlo analysis.

/pdf/bayesian-measures-of-model-complexity-and-fit-1ouzmtfu7x.pdf

Bayesian measures of model complexity and fit

Background Little is known about the general population prevalence or severity of DSM-IV mental disorders. Objective To estimate 12-month prevalence, severity, and comorbidity of DSM-IV anxiety, mood, impulse control, and substance disorders in the recently completed US National Comorbidity Survey Replication. Design and Setting Nationally representative face-to-face household survey conducted between February 2001 and April 2003 using a fully structured diagnostic interview, the World Health Organization World Mental Health Survey Initiative version of the Composite International Diagnostic Interview. Participants Nine thousand two hundred eighty-two English-speaking respondents 18 years and older. Main Outcome Measures Twelve-month DSM-IV disorders. Results Twelve-month prevalence estimates were anxiety, 18.1%; mood, 9.5%; impulse control, 8.9%; substance, 3.8%; and any disorder, 26.2%. Of 12-month cases, 22.3% were classified as serious; 37.3%, moderate; and 40.4%, mild. Fifty-five percent carried only a single diagnosis; 22%, 2 diagnoses; and 23%, 3 or more diagnoses. Latent class analysis detected 7 multivariate disorder classes, including 3 highly comorbid classes representing 7% of the population. Conclusion Although mental disorders are widespread, serious cases are concentrated among a relatively small proportion of cases with high comorbidity.

/pdf/prevalence-severity-and-comorbidity-of-12-month-dsm-iv-54fahxwd1s.pdf

Prevalence, Severity, and Comorbidity of 12-Month DSM-IV Disorders in the National Comorbidity Survey Replication

Objective To present estimates of the lifetime prevalence of DSM-IV mental disorders with and without severe impairment, their comorbidity across broad classes of disorder, and their sociodemographic correlates. Method The National Comorbidity Survey–Adolescent Supplement NCS-A is a nationally representative face-to-face survey of 10,123 adolescents aged 13 to 18 years in the continental United States. DSM-IV mental disorders were assessed using a modified version of the fully structured World Health Organization Composite International Diagnostic Interview. Results Anxiety disorders were the most common condition (31.9%), followed by behavior disorders (19.1%), mood disorders (14.3%), and substance use disorders (11.4%), with approximately 40% of participants with one class of disorder also meeting criteria for another class of lifetime disorder. The overall prevalence of disorders with severe impairment and/or distress was 22.2% (11.2% with mood disorders, 8.3% with anxiety disorders, and 9.6% behavior disorders). The median age of onset for disorder classes was earliest for anxiety (6 years), followed by 11 years for behavior, 13 years for mood, and 15 years for substance use disorders. Conclusions These findings provide the first prevalence data on a broad range of mental disorders in a nationally representative sample of U.S. adolescents. Approximately one in every four to five youth in the U.S. meets criteria for a mental disorder with severe impairment across their lifetime. The likelihood that common mental disorders in adults first emerge in childhood and adolescence highlights the need for a transition from the common focus on treatment of U.S. youth to that of prevention and early intervention.

Lifetime Prevalence of Mental Disorders in U.S. Adolescents: Results from the National Comorbidity Survey Replication-Adolescent Supplement (NCS-A)

Cluster analysis is the automated search for groups of related observations in a dataset. Most clustering done in practice is based largely on heuristic but intuitively reasonable procedures, and most clustering methods available in commercial software are also of this type. However, there is little systematic guidance associated with these methods for solving important practical questions that arise in cluster analysis, such as how many clusters are there, which clustering method should be used, and how should outliers be handled. We review a general methodology for model-based clustering that provides a principled statistical approach to these issues. We also show that this can be useful for other problems in multivariate analysis, such as discriminant analysis and multivariate density estimation. We give examples from medical diagnosis, minefield detection, cluster recovery from noisy data, and spatial density estimation. Finally, we mention limitations of the methodology and discuss recent development...

https://calvino.polito.it/~gasparini/teaching/fraley2002.pdf

Model-Based Clustering, Discriminant Analysis, and Density Estimation

Ten per cent of the world’s school-aged children are estimated to be carrying excess body fat (Fig. 1), with an increased risk for developing chronic disease. Of these overweight children, a quarter are obese, with a significant likelihood of some having multiple risk factors for type 2 diabetes, heart disease and a variety of other co-morbidities before or during early adulthood. The prevalence of overweight is dramatically higher in economically developed regions, but is rising significantly in most parts of the world. In many countries the problem of childhood obesity is worsening at a dramatic rate. Surveys during the 1990s show that in Brazil and the USA, an additional 0.5% of the entire child population became overweight each year. In Canada, Australia and parts of Europe the rates were higher, with an additional 1% of all children becoming overweight each year. The burden upon the health services cannot yet be estimated. Although childhood obesity brings a number of additional problems in its train – hyperinsulinaemia, poor glucose tolerance and a raised risk of type 2 diabetes, hypertension, sleep apnoea, social exclusion and depression – the greatest health problems will be seen in the next generation of adults as the present childhood obesity epidemic passes through to adulthood. Greatly increased rates of heart disease, diabetes, certain cancers, gall bladder disease, osteoarthritis, endocrine disorders and other obesityrelated conditions will be found in young adult populations, and their need for medical treatment may last for their remaining life-times. The costs to the health services, the losses to society and the burdens carried by the individuals involved will be great. The present report has been written to focus attention on the issue and to urge policy-makers to consider taking action before it is too late. Specifically, the report:

https://onlinelibrary.wiley.com/doi/pdf/10.1111/j.1467-789X.2004.00133.x

Obesity in children and young people: a crisis in public health.

Results: Although 3-month prevalence of any disorder averaged 13.3% (95% confidence interval [CI], 11.7%15.0%), during the study period 36.7% of participants (31% of girls and 42% of boys) had at least 1 psychiatric disorder. Some disorders (social anxiety, panic, depression, and substance abuse) increased in prevalence, whereas others, including separation anxiety disorder and attention-deficit/hyperactivity disorder (ADHD), decreased. Lagged analyses showed that children with a history of psychiatric disorder were 3 times more likely than those with no previous disorder to have a diagnosis at any subsequent wave (odds ratio, 3.7; 95% CI, 2.94.9; P.001). Risk from a previous diagnosis was high among both girls and boys, but it was significantly higher among girls. Continuity of the same disorder (homotypic) was significant for all disorders except specific phobias. Continuity from one diagnosis to another (heterotypic) was significant from depression to anxiety and anxiety to depression, from ADHD to oppositional defiant disorder, and from anxiety and conduct disorder to substance abuse. Almost all the heterotypic continuity was seen in girls. Conclusions: The risk of having at least 1 psychiatric disorder by age 16 years is much higher than point estimates would suggest. Concurrent comorbidity and homotypic and heterotypic continuity are more marked in girls than in boys.

https://jamanetwork.com/journals/jamapsychiatry/articlepdf/207725/YOA20769.pdf

Prevalence and Development of Psychiatric Disorders in Childhood and Adolescence

This DataWatch explores the roles of human service sectors (mental health, education, health, child welfare, and juvenile justice) in providing mental health services for children. The data are from the first wave of the Great Smoky Mountains Study of Youth, a population-based study of psychopathology and mental health service use among children. The results show somewhat higher rates of mental health service use than has been reported previously, while continuing to show a substantial amount of unmet need, even among children with both a psychiatric diagnosis and functional impairment. The findings point to a significant role for the education sector, suggesting that schools may function as the de facto mental health system for children and adolescents.

Children's mental health service use across service sectors

Background: The Great Smoky Mountains Study of youth focuses on the relationship between the development of psychiatric disorder and the need for and use of mental health services. Methods: A multistage, overlapping cohorts design was used, in which 4500 of the 11 758 children aged 9, 11, and 13 years in an 11-country area of the southeastern United States were randomly selected for screening for psychiatric symptoms. Children who scored in the top 25% on the screening questionnaire, together with a 1 in 10 random sample of the rest, were recruited for 4 waves of intensive, annual interviews (n=1015 at wave 1). In a parallel study, all American Indian children aged 9, 11, and 13 years were recruited (N=323 at wave 1). Results: The 3-month prevalence (±SE) of any DSMIII-R axis I disorder in the main sample, weighted to reflect population prevalence rates, was 20.3%±1.7%. The most common diagnoses were anxiety disorders (5.7%±1.0%), enuresis (5.1%±1.0%), tic disorders (4.2%±0.9%), conduct disorder (3.3%±0.6%), oppositional defiant disorder (2.7%±0.4%), and hyperactivity (1.9%±0.4%). Conclusions: The prevalence of psychiatric disorder in this rural sample was similar to rates reported in other recent studies. Poverty was the strongest demographic correlate of diagnosis, in both urban and rural children.

https://jamanetwork.com/journals/PSYCH/articlepdf/497702/archpsyc_53_12_012.pdf

The Great Smoky Mountains Study of Youth. Goals, design, methods, and the prevalence of DSM-III-R disorders.

BACKGROUND: Both the professional and the general media have recently published concerns about an 'epidemic' of child and adolescent depression. Reasons for this concern include (1) increases in antidepressant prescriptions, (2) retrospective recall by successive birth cohorts of adults, (3) rising adolescent suicide rates until 1990, and (4) evidence of an increase in emotional problems across three cohorts of British adolescents. METHODS: Epidemiologic studies of children born between 1965 and 1996 were reviewed and a meta-analysis conducted of all studies that used structured diagnostic interviews to make formal diagnoses of depression on representative population samples of participants up to age 18. The effect of year of birth on prevalence was estimated, controlling for age, sex, sample size, taxonomy (e.g., DSM vs. ICD), measurement instrument, and time-frame of the interview (current, 3 months, 6 months, 12 months). RESULTS: Twenty-six studies were identified, generating close to 60,000 observations on children born between 1965 and 1996 who had received at least one structured psychiatric interview capable of making a formal diagnosis of depression. Rates of depression showed no effect of year of birth. There was little effect of taxonomy, measurement instrument, or time-frame of interview. The overall prevalence estimates were: under 13, 2.8% (standard error (SE) .5%); 13-18 5.6% (SE .3%); 13-18 girls: 5.9% (SE .3%); 13-18 boys: 4.6% (SE .3%). CONCLUSIONS: When concurrent assessment rather than retrospective recall is used, there is no evidence for an increased prevalence of child or adolescent depression over the past 30 years. Public perception of an 'epidemic' may arise from heightened awareness of a disorder that was long under-diagnosed by clinicians. Language: en

Is there an epidemic of child or adolescent depression

Background: Few longitudinal studies of child and adolescent psychopathology have examined the links between specific childhood anxiety disorders and adolescent psychiatric disorder. In this paper we test the predictive specificity of separation anxiety disorder (SAD), overanxious disorder (OAD), generalized anxiety disorder (GAD), and social phobia.

Methods:  Data come from the Great Smoky Mountains Study (GSMS). A representative population sample of children – ages 9, 11, and 13 years at intake – was followed to age 19. Diagnoses of both childhood (before age 13 years) and adolescent psychiatric disorders (age 13 to 19 years) were available from 906 participants.

Results:  Childhood SAD predicted adolescent SAD, whereas OAD was associated with later OAD, panic attacks, depression and conduct disorder (CD). GAD was related only to CD. Social phobia in childhood was associated with adolescent OAD, social phobia, and attention-deficit/hyperactivity disorder (ADHD).

Conclusions:  Anxiety disorders in childhood are predictors of a range of psychiatric disorders in adolescence. It appears that children meriting a well-defined diagnosis are missed by the current rules for the diagnosis of GAD. Future studies should examine whether OAD deserves reconsideration as a nosological entity.

Alaattin Erkanli

Papers

Prevalence and Development of Psychiatric Disorders in Childhood and Adolescence

Children's mental health service use across service sectors

The Great Smoky Mountains Study of Youth. Goals, design, methods, and the prevalence of DSM-III-R disorders.

Is there an epidemic of child or adolescent depression

What do childhood anxiety disorders predict