Predicting sojourn times across dementia disease stages, institutionalization, and mortality

INTRODUCTION

Inferring the timeline from mild cognitive impairment (MCI) to severe dementia is pivotal for patients, clinicians, and researchers. Literature is sparse and often contains few patients. We aim to determine the time spent in MCI, mild-, moderate-, severe dementia, and institutionalization until death.

METHODS

Multistate modeling with Cox regression was used to obtain the sojourn time. Covariates were age at baseline, sex, amyloid status, and Alzheimer's disease (AD) or other dementia diagnosis. The sample included a register (SveDem) and memory clinics (Amsterdam Dementia Cohort and Memento).

RESULTS

Using 80,543 patients, the sojourn time from clinically identified MCI to death across all patient groups ranged from 6.20 (95% confidence interval [CI]: 5.57-6.98) to 10.08 (8.94-12.18) years.

DISCUSSION

Generally, sojourn time was inversely associated with older age at baseline, males, and AD diagnosis. The results provide key estimates for researchers and clinicians to estimate prognosis.

A genetically informed prediction model for suicidal and aggressive behaviour in teens

Suicidal and aggressive behaviours cause significant personal and societal burden. As risk factors associated with these behaviours frequently overlap, combined approaches in predicting the behaviours may be useful in identifying those at risk for either. The current study aimed to create a model that predicted if individuals will exhibit suicidal behaviour, aggressive behaviour, both, or neither in late adolescence. A sample of 5,974 twins from the Child and Adolescent Twin Study in Sweden (CATSS) was broken down into a training (80%), tune (10%) and test (10%) set. The Netherlands Twin Register (NTR; N = 2702) was used for external validation. Our longitudinal data featured genetic, environmental, and psychosocial predictors derived from parental and self-report data. A stacked ensemble model was created which contained a gradient boosted machine, random forest, elastic net, and neural network. Model performance was transferable between CATSS and NTR (macro area under the receiver operating characteristic curve (AUC) [95% CI] AUC_{CATSS(test set)} = 0.709 (0.671-0.747); AUC_NTR = 0.685 (0.656-0.715), suggesting model generalisability across Northern Europe. The notable exception is suicidal behaviours in the NTR, which was no better than chance. The 25 highest scoring variable importance scores for the gradient boosted machines and random forest models included self-reported psychiatric symptoms in mid-adolescence, sex, and polygenic scores for psychiatric traits. The model's performance is comparable to current prediction models that use clinical interviews and is not yet suitable for clinical use. Moreover, genetic variables may have a role to play in predictive models of adolescent psychopathology.

Genome-wide Association Meta-analysis of Childhood and Adolescent Internalizing Symptoms

OBJECTIVE

To investigate the genetic architecture of internalizing symptoms in childhood and adolescence.

METHOD

In 22 cohorts, multiple univariate genome-wide association studies (GWASs) were performed using repeated assessments of internalizing symptoms, in a total of 64,561 children and adolescents between 3 and 18 years of age. Results were aggregated in meta-analyses that accounted for sample overlap, first using all available data, and then using subsets of measurements grouped by rater, age, and instrument.

RESULTS

The meta-analysis of overall internalizing symptoms (INToverall) detected no genome-wide significant hits and showed low single nucleotide polymorphism (SNP) heritability (1.66%, 95% CI = 0.84-2.48%, neffective = 132,260). Stratified analyses indicated rater-based heterogeneity in genetic effects, with self-reported internalizing symptoms showing the highest heritability (5.63%, 95% CI = 3.08%-8.18%). The contribution of additive genetic effects on internalizing symptoms appeared to be stable over age, with overlapping estimates of SNP heritability from early childhood to adolescence. Genetic correlations were observed with adult anxiety, depression, and the well-being spectrum (|rg| > 0.70), as well as with insomnia, loneliness, attention-deficit/hyperactivity disorder, autism, and childhood aggression (range |rg| = 0.42-0.60), whereas there were no robust associations with schizophrenia, bipolar disorder, obsessive-compulsive disorder, or anorexia nervosa.

CONCLUSION

Genetic correlations indicate that childhood and adolescent internalizing symptoms share substantial genetic vulnerabilities with adult internalizing disorders and other childhood psychiatric traits, which could partially explain both the persistence of internalizing symptoms over time and the high comorbidity among childhood psychiatric traits. Reducing phenotypic heterogeneity in childhood samples will be key in paving the way to future GWAS success.

Borderline personality disorder: associations with psychiatric disorders, somatic illnesses, trauma, and adverse behaviors

In one of the largest, most comprehensive studies on borderline personality disorder (BPD) to date, this article places into context associations between this diagnosis and (1) 16 different psychiatric disorders, (2) eight somatic illnesses, and (3) six trauma and adverse behaviors, e.g., violent crime victimization and self-harm. Second, it examines the sex differences in individuals with BPD and their siblings. A total of 1,969,839 Swedish individuals were identified from national registers. Cumulative incidence with 95% confidence intervals (CI) was evaluated after 5 years of follow-up from BPD diagnosis and compared with a matched cohort. Associations were estimated as hazard ratios (HR) with 95% CIs from Cox regression. 12,175 individuals were diagnosed with BPD (85.3% female). Individuals diagnosed with BPD had higher cumulative incidences and HRs for nearly all analyzed indicators, especially psychiatric disorders. Anxiety disorders were most common (cumulative incidence 95% CI 33.13% [31.48-34.73]). Other notable findings from Cox regressions include psychotic disorders (HR 95% CI 24.48 [23.14-25.90]), epilepsy (3.38 [3.08-3.70]), violent crime victimization (7.65 [7.25-8.06]), and self-harm (17.72 [17.27-18.19]). HRs in males and females with BPD had overlapping CIs for nearly all indicators. This indicates that a BPD diagnosis is a marker of vulnerability for negative events and poor physical and mental health similarly for both males and females. Having a sibling with BPD was associated with an increased risk for psychiatric disorders, trauma, and adverse behaviors but not somatic disorders. Clinical implications include the need for increased support for patients with BPD navigating the health care system.

Association and Familial Coaggregation of Type 1 Diabetes and Eating Disorders: A Register-Based Cohort Study in Denmark and Sweden

OBJECTIVE

To ascertain the association and coaggregation of eating disorders and childhood-onset type 1 diabetes in families.

RESEARCH DESIGN AND METHODS

Using population samples from national registers in Sweden (n = 2,517,277) and Demark (n = 1,825,920), we investigated the within-individual association between type 1 diabetes and eating disorders and their familial coaggregation among full siblings, half siblings, full cousins, and half cousins. On the basis of clinical diagnoses, we classified eating disorders into any eating disorder (AED), anorexia nervosa (AN) and atypical AN, and other eating disorder (OED). Associations were determined with hazard ratios (HRs) with 95% CIs from Cox regressions.

RESULTS

Swedish and Danish individuals with a type 1 diabetes diagnosis had a greater risk of receiving an eating disorder diagnosis (HR [95% CI] Sweden: AED 2.02 [1.80-2.27], AN 1.63 [1.36-1.96], OED 2.34 [2.07-2.63]; Denmark: AED 2.19 [1.84-2.61], AN 1.78 [1.36-2.33], OED 2.65 [2.20-3.21]). We also meta-analyzed the results: AED 2.07 (1.88-2.28), AN 1.68 (1.44-1.95), OED 2.44 (2.17-2.72). There was an increased risk of receiving an eating disorder diagnosis in full siblings in the Swedish cohort (AED 1.25 [1.07-1.46], AN 1.28 [1.04-1.57], OED 1.28 [1.07-1.52]); these results were nonsignificant in the Danish cohort.

CONCLUSIONS

Patients with type 1 diabetes are at a higher risk of subsequent eating disorders; however, there is conflicting support for the relationship between having a sibling with type 1 diabetes and an eating disorder diagnosis. Diabetes health care teams should be vigilant about disordered eating behaviors in children and adolescents with type 1 diabetes.

Attention-deficit/hyperactivity disorder symptoms and dietary habits in adulthood: A large population-based twin study in Sweden

Associations between adult attention-deficit/hyperactivity disorder (ADHD) symptoms and dietary habits have not been well established and the underlying mechanisms remain unclear. We explored these associations using a Swedish population-based twin study with 17,999 individuals aged 20-47 years. We estimated correlations between inattention and hyperactivity/impulsivity with dietary habits and fitted twin models to determine the genetic and environmental contributions. Dietary habits were defined as (a) consumption of food groups, (b) consumption of food items rich in particular macronutrients, and (c) healthy and unhealthy dietary patterns. At the phenotypic level, inattention was positively correlated with seafood, high-fat, high-sugar, high-protein food consumptions, and unhealthy dietary pattern, with correlation coefficients ranging from 0.03 (95%CI: 0.01, 0.05) to 0.13 (95% CI: 0.11, 0.15). Inattention was negatively correlated with fruits, vegetables consumptions and healthy dietary pattern, with correlation coefficients ranging from -0.06 (95%CI: -0.08, -0.04) to -0.07 (95%CI: -0.09, -0.05). Hyperactivity/impulsivity and dietary habits showed similar but weaker patterns compared to inattention. All associations remained stable across age, sex and socioeconomic status. Nonshared environmental effects contributed substantially to the correlations of inattention (56-60%) and hyperactivity/impulsivity (63-80%) with dietary habits. The highest and lowest genetic correlations were between inattention and high-sugar food (r_A = .16, 95% CI: 0.07, 0.25), and between hyperactivity/impulsivity and unhealthy dietary pattern (r_A = .05, 95% CI: -0.05, 0.14), respectively. We found phenotypic and etiological overlap between ADHD and dietary habits, although these associations were weak. Our findings contribute to a better understanding of common etiological pathways between ADHD symptoms and various dietary habits.

Predicting mental health problems in adolescence using machine learning techniques

BACKGROUND

Predicting which children will go on to develop mental health symptoms as adolescents is critical for early intervention and preventing future, severe negative outcomes. Although many aspects of a child's life, personality, and symptoms have been flagged as indicators, there is currently no model created to screen the general population for the risk of developing mental health problems. Additionally, the advent of machine learning techniques represents an exciting way to potentially improve upon the standard prediction modelling technique, logistic regression. Therefore, we aimed to I.) develop a model that can predict mental health problems in mid-adolescence II.) investigate if machine learning techniques (random forest, support vector machines, neural network, and XGBoost) will outperform logistic regression.

METHODS

In 7,638 twins from the Child and Adolescent Twin Study in Sweden we used 474 predictors derived from parental report and register data. The outcome, mental health problems, was determined by the Strengths and Difficulties Questionnaire. Model performance was determined by the area under the receiver operating characteristic curve (AUC).

RESULTS

Although model performance varied somewhat, the confidence interval overlapped for each model indicating non-significant superiority for the random forest model (AUC = 0.739, 95% CI 0.708-0.769), followed closely by support vector machines (AUC = 0.735, 95% CI 0.707-0.764).

CONCLUSION

Ultimately, our top performing model would not be suitable for clinical use, however it lays important groundwork for future models seeking to predict general mental health outcomes. Future studies should make use of parent-rated assessments when possible. Additionally, it may not be necessary for similar studies to forgo logistic regression in favor of other more complex methods.