Finding undiagnosed patients with hepatitis C virus: an application of machine learning to US ambulatory electronic medical records.

AIMS: To develop and validate a machine learning (ML) algorithm to identify undiagnosed hepatitis C virus (HCV) patients, in order to facilitate prioritisation of patients for targeted HCV screening. METHODS: This retrospective study used ambulatory electronic medical records (EMR) from January 2015 to February 2020. A Gradient Boosting Trees algorithm was trained using patient records to predict initial HCV diagnosis and was validated on a temporally independent held-out cross-section of the data. The fold improvement in precision (proportion of patients identified by the algorithm who are HCV positive) over universal screening was examined and compared with risk-based screening. RESULTS: 21 508 positive (HCV diagnosed) and 28.2M unlabelled (lacking evidence of HCV diagnosis) patients met the inclusion criteria for the study. After down-sampling unlabelled patients to aid the algorithm's learning process, 16.2M unlabelled patients entered the analysis. Performance of the algorithm was compared with universal screening on the held-out cross-section, which had an incidence of HCV diagnoses of 0.02%. The algorithm achieved a 101.0 ×, 18.0 × and 5.1 × fold improvement in precision over universal screening at 5%, 20% and 50% levels of recall. When compared with risk-based screening, the algorithm required fewer patients to be screened and improved precision. CONCLUSIONS: This study presents strong evidence towards the use of ML on EMR data for the prioritisation of patients for targeted HCV testing with potential to improve efficiency of resource utilisation, thereby reducing the workload for clinicians and saving healthcare costs. A prospective interventional study would allow for further validation before use in a clinical setting.

Prediction of successful labor induction in persons with a low Bishop score using machine learning: Secondary analysis of two randomized controlled trials.

BACKGROUND: The objective of this paper was to identify predictors of a vaginal birth in individuals with singleton pregnancies and a Bishop Score <4, following Induction of Labor (IoL) using dinoprostone vaginal insert (DVI). Secondarily, we sought to understand the association between oxytocin use for labor augmentation and IoL outcomes. METHODS: We developed and internally validated a multivariate prediction model using machine learning (ML) applied to data from two Phase-III randomized controlled double-blind trials (NCT01127581, NCT00308711). The model was internally validated using 10-fold cross-validation. RESULTS: This study included 1107 participants. Despite unfavorable cervical status and inclusion of high-risk pregnancies, 72% of participants had vaginal births. The model's area under receiver operating characteristic curve was 0.73. The following factors increased the chance of vaginal birth: being parous; being between 37 and 41 weeks of gestation; having a lower Body Mass Index; having a lower maternal age; having fewer maternal comorbidities; and having a higher Bishop score. Parity alone correctly predicted the outcome in ~50% of cases, at a ~10% false-negative rate. Participants whose labors progressed without requiring oxytocin had a higher probability of vaginal birth than those requiring oxytocin for either induction or augmentation (81% vs 70% vs 77%, respectively). DISCUSSION: Even in high-risk pregnancies and with low Bishop scores, the use of DVI results in a high chance of vaginal birth. Parity is a critical predictor of success. The judicious use of oxytocin for labor induction or augmentation can increase the chance of vaginal birth. Our study validates the use of ML and predictive modeling for treatment response prediction when considering IoL.

Worker Stress, Burnout, and Wellbeing Before and During the COVID-19 Restrictions in the United Kingdom.

COVID-19 created a transformational shift in the working environment for much of the labour force, yet its impact on workers is unclear. This study uses longitudinal data to examine the wellbeing of 621 full-time workers assessed before (November 2019-February 2020) and during (May-June 2020) the first lockdown in the United Kingdom. We employ fixed effects analyses to investigate the impact of the restrictions and mandatory homeworking on cognitive, emotional, and psychological wellbeing. Within the sample, the rate of full-time homeworking increased from 2 to 74% between waves. We identify significant changes in 9 of the 15 measures assessed, with a general pattern of improvements in wellbeing during lockdown. Overall levels of stress, self-rated mental health, positive emotions and life and job satisfaction are not adversely affected by the restrictions. There is a reduction in the burnout symptoms of disengagement (-0.13 sd) and exhaustion (-0.20 sd) and in the frequency with which negative emotions are experienced at work (-0.15 sd). Workers feel more autonomous (+0.09 sd), closer to their co-workers (+0.10 sd), and more attached to their organisations (+0.19 sd). However, homelife satisfaction declines (-0.11 sd). These findings highlight the possibility that the COVID-19 pandemic and large-scale transition to homeworking was associated with unchanged or improved worker wellbeing. This study has important implications for governments and employers regarding a global shift to homeworking.

Machine learning using longitudinal prescription and medical claims for the detection of non-alcoholic steatohepatitis (NASH).

OBJECTIVES: To develop and evaluate machine learning models to detect patients with suspected undiagnosed non-alcoholic steatohepatitis (NASH) for diagnostic screening and clinical management. METHODS: In this retrospective observational non-interventional study using administrative medical claims data from 1 463 089 patients, gradient-boosted decision trees were trained to detect patients with likely NASH from an at-risk patient population with a history of obesity, type 2 diabetes mellitus, metabolic disorder or non-alcoholic fatty liver (NAFL). Models were trained to detect likely NASH in all at-risk patients or in the subset without a prior NAFL diagnosis (at-risk non-NAFL patients). Models were trained and validated using retrospective medical claims data and assessed using area under precision recall curves and receiver operating characteristic curves (AUPRCs and AUROCs). RESULTS: The 6-month incidences of NASH in claims data were 1 per 1437 at-risk patients and 1 per 2127 at-risk non-NAFL patients . The model trained to detect NASH in all at-risk patients had an AUPRC of 0.0107 (95% CI 0.0104 to 0.0110) and an AUROC of 0.84. At 10% recall, model precision was 4.3%, which is 60× above NASH incidence. The model trained to detect NASH in the non-NAFL cohort had an AUPRC of 0.0030 (95% CI 0.0029 to 0.0031) and an AUROC of 0.78. At 10% recall, model precision was 1%, which is 20× above NASH incidence. CONCLUSION: The low incidence of NASH in medical claims data corroborates the pattern of NASH underdiagnosis in clinical practice. Claims-based machine learning could facilitate the detection of patients with probable NASH for diagnostic testing and disease management.

Application of machine learning methods to bridge the gap between non-interventional studies and randomized controlled trials in ophthalmic patients with neovascular age-related macular degeneration.

PURPOSE: The effectiveness of intravitreal anti-vascular endothelial growth factor agents is usually lower in real world settings compared with randomized clinical trials (RCTs), often limiting the use of real-world evidence (RWE) in regulatory and healthcare decisions. The current analysis aimed to develop and validate an algorithm to explain the difference in outcomes between RWE studies and RCTs in patients with neovascular age-related macular degeneration. METHODS: The algorithm was developed using ranibizumab real world data (RWD) from the US and validated on Australian and UK RWD. A decision model was developed using machine learning principles, in which the model learns how to partition the most influential factors (out of 59 variables) so that they maximally relate to the change in visual acuity (VA) over 12 months. RESULTS: The algorithm identified baseline VA <73 Early Treatment Diabetic Retinopathy Study letters, presence of baseline subretinal fluid, and administration of three loading doses by Day 90 from drug initiation as the characteristics with the greatest impact on VA at month 12. When applying the different criteria, RWE outcomes became similar to those obtained in known RCTs. CONCLUSION: Machine learning techniques can be used to classify real world cohorts and identify subsets of patients who benefit to the same extent as that reported in RCTs. This methodology may support the translation of clinical trial findings to treatment performance in the clinical practice setting.

A Demonstration Study of the Quiet Time Transcendental Meditation Program.

This manuscript presents a demonstration study of Quiet Time (QT), a classroom-based Transcendental Meditation intervention. The aim of the study is to assess the feasibility of implementing and evaluating QT in two pilot settings in the United Kingdom and Ireland. This study contributes to the field by targeting middle childhood, testing efficiency in two settings operating under different educational systems, and including a large array of measures. First, teacher and pupil engagement with QT was assessed. Second, the feasibility of using a quasi-experimental design and a wide range of instruments to measure changes in pupil outcomes before and after the intervention was assessed. This allows us to obtain information about which instruments might be feasible to administer and most sensitive to change. The first setting included 89 students from a primary school in the United Kingdom: those in sixth grade received the QT intervention, while those in fifth grade practiced meditation using the Headspace application. The second setting included 100 fifth- and sixth-grade students from two schools in Ireland: one received the QT intervention, the other served as a control. Recruitment and retention rates were high in both settings, and the intervention was feasible and accepted by students, parents and teachers. Implementation fidelity was lower in the United Kingdom setting where delivery started later in the school year and the practice was affected by preparation for the Standard Assessment Tests. These results show that QT may be feasibly delivered in school settings, and suggest the use of a compact battery of tests to measure impact. We find suggestive evidence that the intervention affected executive function as children who practiced QT showed improved working memory in both settings. In the Irish setting, pupils in the QT group had improved ability to control responses. These results have implications for future studies by a) demonstrating that implementation fidelity is highly context dependent and b) providing suggestive evidence of the malleability of children's skills in middle childhood. The results of this demonstration study will be used to inform a larger RCT of the QT intervention.

Finding undiagnosed patients with hepatitis C infection: an application of artificial intelligence to patient claims data.

Hepatitis C virus (HCV) remains a significant public health challenge with approximately half of the infected population untreated and undiagnosed. In this retrospective study, predictive models were developed to identify undiagnosed HCV patients using longitudinal medical claims linked to prescription data from approximately ten million patients in the United States (US) between 2010 and 2016. Features capturing information on demographics, risk factors, symptoms, treatments and procedures relevant to HCV were extracted from patients' medical history. Predictive algorithms were developed based on logistic regression, random forests, gradient boosted trees and a stacked ensemble. Descriptive analysis indicated that patients exhibited known symptoms of HCV on average 2-3 years prior to their diagnosis. The precision was at least 95% for all algorithms at low levels of recall (10%). For recall levels >50%, the stacked ensemble performed best with a precision of 97% compared with 87% for the gradient boosted trees and just 31% for the logistic regression. For context, the Center for Disease Control recommends screening in an at-risk sub-population with an estimated HCV prevalence of 2.23%. The artificial intelligence (AI) algorithm presented here has a precision which is substantially higher than the screening rates associated with recommended clinical guidelines, suggesting that AI algorithms have the potential to provide a step change in the effectiveness of HCV screening.

Identification of potentially undiagnosed patients with nontuberculous mycobacterial lung disease using machine learning applied to primary care data in the UK.

Nontuberculous mycobacterial lung disease (NTMLD) is a rare lung disease often missed due to a low index of suspicion and unspecific clinical presentation. This retrospective study was designed to characterise the prediagnosis features of NTMLD patients in primary care and to assess the feasibility of using machine learning to identify undiagnosed NTMLD patients.IQVIA Medical Research Data (incorporating THIN, a Cegedim Database), a UK electronic medical records primary care database was used. NTMLD patients were identified between 2003 and 2017 by diagnosis in primary or secondary care or record of NTMLD treatment regimen. Risk factors and treatments were extracted in the prediagnosis period, guided by literature and expert clinical opinion. The control population was enriched to have at least one of these features.741 NTMLD and 112 784 control patients were selected. Annual prevalence rates of NTMLD from 2006 to 2016 increased from 2.7 to 5.1 per 100 000. The most common pre-existing diagnoses and treatments for NTMLD patients were COPD and asthma and penicillin, macrolides and inhaled corticosteroids. Compared to random testing, machine learning improved detection of patients with NTMLD by almost a thousand-fold with AUC of 0.94. The total prevalence of diagnosed and undiagnosed cases of NTMLD in 2016 was estimated to range between 9 and 16 per 100 000.This study supports the feasibility of machine learning applied to primary care data to screen for undiagnosed NTMLD patients, with results indicating that there may be a substantial number of undiagnosed cases of NTMLD in the UK.

The effect of local unemployment on health: A longitudinal study of Irish mothers 2001-2011.

The relationship between economic conditions and health is mixed, with some evidence from the most recent financial crisis finding a positive effect of periods of unemployment on heath behaviours. This study uses longitudinal data spanning the periods before, during and after the Irish crisis of 2007, to test the impact of economic expansion and contraction on mothers physical and mental health and health behaviours. Three waves of data from the Irish Lifeways Cohort Study for the period 2001-2011, and local area employment rates from the Irish Census, are used to capture the impact of a period of increased unemployment on health before, during and after the Irish recession, independent of individual employment status. The results from fixed effect linear probability models demonstrate that an additional unit of local area unemployment increases the probability of reporting poor self-rated health by 1-1.8 percentage points and of reporting poor mental well-being by 1.4 and 2.7 percentage points depending on the instrument used. There are decreases in the probability of consuming cigarettes by 3.3 percentage points, self-describing as a regular smoker by 2.9 percentage points and obesity by 0.9 percentage points. The probability of engaging in at least 20 minutes of strenuous or moderate exercise per week declines by 7.8 and 8.7 percentage points respectively, while the probability of engaging in at least 20 minutes of mild exercise rises by 10 percentage points. These results are largely consistent with the US literature, which is predominantly based on working men, thus demonstrating the universal impact of economic vulnerability on health.

Utilising artificial intelligence to determine patients at risk of a rare disease: idiopathic pulmonary arterial hypertension.

Idiopathic pulmonary arterial hypertension is a rare and life-shortening condition often diagnosed at an advanced stage. Despite increased awareness, the delay to diagnosis remains unchanged. This study explores whether a predictive model based on healthcare resource utilisation can be used to screen large populations to identify patients at high risk of idiopathic pulmonary arterial hypertension. Hospital Episode Statistics from the National Health Service in England, providing close to full national coverage, were used as a measure of healthcare resource utilisation. Data for patients with idiopathic pulmonary arterial hypertension from the National Pulmonary Hypertension Service in Sheffield were linked to pre-diagnosis Hospital Episode Statistics records. A non-idiopathic pulmonary arterial hypertension control cohort was selected from the Hospital Episode Statistics population. Patient history was limited to ≤5 years pre-diagnosis. Information on demographics, timing/frequency of diagnoses, medical specialities visited and procedures undertaken was captured. For modelling, a bagged gradient boosting trees algorithm was used to discriminate between cohorts. Between 2008 and 2016, 709 patients with idiopathic pulmonary arterial hypertension were identified and compared with a stratified cohort of 2,812,458 patients classified as non-idiopathic pulmonary arterial hypertension with ≥1 ICD-10 coded diagnosis of relevance to idiopathic pulmonary arterial hypertension. A predictive model was developed and validated using cross-validation. The timing and frequency of the clinical speciality seen, secondary diagnoses and age were key variables driving the algorithm's performance. To identify the 100 patients at highest risk of idiopathic pulmonary arterial hypertension, 969 patients would need to be screened with a specificity of 99.99% and sensitivity of 14.10% based on a prevalence of 5.5/million. The positive predictive and negative predictive values were 10.32% and 99.99%, respectively. This study highlights the potential application of artificial intelligence to readily available real-world data to screen for rare diseases such as idiopathic pulmonary arterial hypertension. This algorithm could provide low-cost screening at a population level, facilitating earlier diagnosis, improved diagnostic rates and patient outcomes. Studies to further validate this approach are warranted.

Maternal employment and childcare during infancy and childhood overweight.

OBJECTIVE: This paper examines the joint impact of maternal employment and childcare during infancy on childhood weight at ages three and five in the context of weak social support for early childhood care and education. METHOD: Using three waves of longitudinal data from the Growing-Up in Ireland survey (n = 8,393 age three, n = 8,039 age five), propensity score matching is used to address the endogeneity of employment and childcare decisions. Selection on observables is used to assess potential bias arising from selection on unobservables whereby unobserved characteristics of the mother or child may jointly influence child weight and maternal employment and childcare. RESULTS: Full-time maternal employment at nine months combined with either formal or informal childcare increases the likelihood of being overweight at three years by 8.1% and 5.9% respectively, but only for children of highly educated mothers. Similar results are observed for part-time employment coupled with informal (7.5%) or parental (8.0%) care. The results for mothers with lower levels of education are either not significant or favourable. While the majority of the effects dissipate by age five, there is some evidence that full-time maternal employment coupled with informal care increases the risk of being overweight at both ages three and five for children of higher-educated mothers. An assessment of selection bias finds that the estimates of full-time employment combined with formal childcare by well-educated mothers are a lower bound, such that the true effect on child weight may be understated. CONCLUSIONS: The findings for Ireland are consistent with studies from the United States and the United Kingdom, and are in contrast to findings from the rest of Europe, suggesting the role of institutional factors, such as the lack of subsidised, universal, high-quality childcare.

Impact of an early childhood intervention on the home environment, and subsequent effects on child cognitive and emotional development: A secondary analysis.

The objective of this study was to use secondary data from the Preparing for Life (PFL) trial to test (1) the impact of a prenatal-to-age-five intervention targeting women from a disadvantaged Irish community on the quality of the home environment; (2) whether any identified changes in the home environment explain the positive effects of the PFL program on children's cognitive and emotional development at school entry which have been identified in previous reports of the PFL trial (ES = .72 and .50, respectively). Pregnant women were randomized into a treatment (home visits, baby massage, and parenting program, n = 115) or control (n = 118) group (trial registration: ISRCTN04631728). The home environment was assessed at 6 months, 1½, and 3 years using the Home Observation for Measurement of the Environment (responsiveness, acceptance, organization, learning material, involvement, variety). Cognitive skills were assessed at 5 years using the British Ability Scales. Emotional problems were teacher-reported at 5 years using the Short Early Development Inventory. Latent growth modeling was used to model changes in the home environment, and mediation analyses to test whether those changes explained children outcomes. Compared to controls, treatment children were exposed to more stimulating environments in terms of learning material (B = -1.62, p = 0.036) and environmental variety (B = -1.58, p = 0.009) at 6 months, but these differences faded at 3 years. Treatment families were also more likely to accept suboptimal child behaviors without using punishment (acceptance score, B = 1.49, p = 0.048) and were more organized at 3 years (B = 1.08, p = 0.033). None of the changes mediated children's outcomes. In conclusion, we found that the program positively impacted different home environment dimensions, but these changes did not account for improvements in children's outcomes. Exploratory analyses suggest that the impact of improvements in the home environment on child outcomes may be limited to specific groups of children. Limitations of the study include the potential lack of generalizability to other populations, the inability to assess the individual treatment components, and sample size restrictions which precluded a moderated mediation analysis.

Does maternal attachment to her infant mediate the link between perceptions of infant crying at 6 months and parenting stress at 24 months? A structural equation modelling approach.

BACKGROUND: Parenting stress is influenced by many factors including maternal attachment and excessive infant crying, yet the nature of these relationships is not well understood. For example, excessive infant crying despite maternal soothing may impact maternal attachment to the child, leading to higher stress. This paper explored whether maternal perception of excessive infant crying at 6 months was associated with higher maternal parenting stress at 24 months, and whether maternal attachment mediated this relationship. METHODS: All families, present at 24 months in a randomized controlled trial of a 5-year early intervention programme targeting school readiness skills in disadvantaged area of Ireland, were included. At 6 months, infant crying was assessed using a maternal reported measure of duration of infant crying, and maternal attachment to the infant was assessed using the Condon Maternal Attachment Scale. Parenting stress was assessed at 24 months using the childrearing stress subscale from the Parenting Stress Index. Structural equation modelling was used to explore the direct and indirect effects of maternal perceptions of excessive infant crying on parenting stress, controlling for infant, maternal, and environmental characteristics, and focusing on the mediating role of maternal attachment. RESULTS: Reporting excessive infant crying at 6 months was associated with lower maternal attachment at 6 months, which led to higher parenting stress at 24 months. In addition, vulnerable adult attachment style, previous maternal mental health difficulties, low paternal education, paternal involvement with the child, and not being married were associated with higher parenting stress. CONCLUSION: Findings suggest that the association between maternal perceptions of excessive crying at 6 months and later parenting stress may be mediated through maternal attachment to the infant. Interventions based on improving maternal attachment could be investigated to determine the effectiveness of supporting mothers with low attachment.

Population-Based System of Parenting Support to Reduce the Prevalence of Child Social, Emotional, and Behavioural Problems: Difference-In-Differences Study.

The quality of parenting is recognised as an important determinant of children's mental health. Parenting interventions typically target high-risk families rather than adopting a universal approach. This study examined the population impact of the Triple P Positive Parenting Programme on the prevalence of children's social, emotional, and behavioural problems. A propensity score matching difference-in-differences method was used to compare intervention and comparison regions matched on socio-demographic characteristics in midlands Ireland. The pre-intervention sample included 1501 and 1495 parents of children aged 4-8 years in the intervention and comparison regions respectively. The post-intervention sample included 1521 and 1544 parents respectively. The primary outcome measure was parental reports on the Strengths and Difficulties Questionnaire. There were some significant reductions in the prevalence rates of social, emotional, and behavioural problems in the intervention regions compared to the comparison regions. Children in the intervention sample experienced lower total difficulties, emotional symptoms, and conduct problems than children in the comparison sample, and they were less at risk of scoring within the borderline/abnormal range for total difficulties, conduct problems, and hyperactivity. The programme reduced the proportion of children scoring within the borderline/abnormal range by 4.7% for total difficulties, 4.4% for conduct problems, and 4.5% for hyperactivity in the total population. This study demonstrated that a universal parenting programme implemented at multiple levels using a partnership approach may be an effective population health approach to targeting child mental health.

A Multicomponent Early Intervention Program and Trajectories of Behavior, Cognition, and Health.

OBJECTIVES: To investigate the developmental impact of a prenatal-to-age-5 multicomponent early intervention program targeting families living in low socioeconomic conditions. METHODS: Pregnant women from a disadvantaged Irish community were randomly assigned into a treatment group (home visits, baby massage, and parenting program; n = 115) or control group (n = 118). Children's behavioral problems (externalizing, internalizing), cognitive skills (general, vocabulary), and health service use (number of health clinic visits), were regularly assessed (6 months to 4 years of age). Children's developmental trajectories were modeled by using latent class growth analyses to test whether certain subgroups benefited more than others. RESULTS: High and low developmental trajectories were identified for each outcome. Treated children were more likely to follow the high-level trajectory for cognition (odds ratio = 2.89; 95% confidence interval = 1.55-5.50) and vocabulary skills (odds ratio = 2.02; 95% confidence interval = 1.08-3.82). There were no differences by treatment condition in the risk of belonging to a high externalizing or high health clinic visit trajectory. However, within the high externalizing trajectory, treated children had lower scores than controls (Hedges' g range (2-4 years) = 0.45-0.58; P < .05) and, within the high health clinic visit trajectory, only children in the control group experienced an increasing number of visits. CONCLUSIONS: This program revealed moderate positive impacts on trajectories of cognitive development and number of health clinic visits for all children, whereas positive impacts on externalizing behavior problems were restricted to children with the most severe problems.

Breastfeeding and externalising problems: a quasi-experimental design with a national cohort.

Evidence from correlational studies supporting the benefits of breastfeeding on children's externalising problems is mixed. Quasi-experimental approaches can help in better understanding possible 'effects'. We aimed to investigate the longitudinal impact of breastfeeding on externalising problems from childhood into adolescence. Participants included ~ 5000 full-term children, from the Growing Up in Ireland Child Cohort. Externalising problems (conduct problems and hyperactivity) were assessed using both the parent and teacher versions of the Strengths and Difficulties Questionnaire when children were age 9 and 13. Maternal reports were used to collect retrospective information on breastfeeding. Propensity score matching, and adjusting for multiple testing were used to compare the average treatment effects for children who were breastfed. Post matching results revealed statistically significant reductions in hyperactivity at age nine, using both maternal and teacher reports (difference score - 0.48, 95% CI - 0.85, - 0.11; and - 0.51, 95% CI - 0.90, - 0.12, respectively), for children who were breastfed between 6 and 12 months, but not thereafter. These effects were not maintained at age 13. Moreover, no effects of breastfeeding on conduct problems were found at any age, regardless of duration, using either maternal or teacher report. While some benefits of breastfeeding were found, compatible with a temporary modest reduction in hyperactivity, related to being breastfed for 26-50 weeks, these results must be viewed in the context of reliance of propensity score matching on observable characteristics. Additionally, our results are suggestive of a potential non-linear dose-response of breastfeeding on hyperactivity.

Personalized Medication Response Prediction for Attention-Deficit Hyperactivity Disorder: Learning in the Model Space vs. Learning in the Data Space.

Attention-Deficit Hyperactive Disorder (ADHD) is one of the most common mental health disorders amongst school-aged children with an estimated prevalence of 5% in the global population (American Psychiatric Association, 2013). Stimulants, particularly methylphenidate (MPH), are the first-line option in the treatment of ADHD (Reeves and Schweitzer, 2004; Dopheide and Pliszka, 2009) and are prescribed to an increasing number of children and adolescents in the US and the UK every year (Safer et al., 1996; McCarthy et al., 2009), though recent studies suggest that this is tailing off, e.g., Holden et al. (2013). Around 70% of children demonstrate a clinically significant treatment response to stimulant medication (Spencer et al., 1996; Schachter et al., 2001; Swanson et al., 2001; Barbaresi et al., 2006). However, it is unclear which patient characteristics may moderate treatment effectiveness. As such, most existing research has focused on investigating univariate or multivariate correlations between a set of patient characteristics and the treatment outcome, with respect to dosage of one or several types of medication. The results of such studies are often contradictory and inconclusive due to a combination of small sample sizes, low-quality data, or a lack of available information on covariates. In this paper, feature extraction techniques such as latent trait analysis were applied to reduce the dimension of on a large dataset of patient characteristics, including the responses to symptom-based questionnaires, developmental health factors, demographic variables such as age and gender, and socioeconomic factors such as parental income. We introduce a Bayesian modeling approach in a "learning in the model space" framework that combines existing knowledge in the literature on factors that may potentially affect treatment response, with constraints imposed by a treatment response model. The model is personalized such that the variability among subjects is accounted for by a set of subject-specific parameters. For remission classification, this approach compares favorably with conventional methods such as support vector machines and mixed effect models on a range of performance measures. For instance, the proposed approach achieved an area under receiver operator characteristic curve of 82-84%, compared to 75-77% obtained from conventional regression or machine learning ("learning in the data space") methods.

Parsing the Roles of the Frontal Lobes and Basal Ganglia in Task Control Using Multivoxel Pattern Analysis.

Cognitive control has traditionally been associated with pFC based on observations of deficits in patients with frontal lesions. However, evidence from patients with Parkinson disease indicates that subcortical regions also contribute to control under certain conditions. We scanned 17 healthy volunteers while they performed a task-switching paradigm that previously dissociated performance deficits arising from frontal lesions in comparison with Parkinson disease, as a function of the abstraction of the rules that are switched. From a multivoxel pattern analysis by Gaussian Process Classification, we then estimated the forward (generative) model to infer regional patterns of activity that predict Switch/Repeat behavior between rule conditions. At 1000 permutations, Switch/Repeat classification accuracy for concrete rules was significant in the BG, but at chance in the frontal lobe. The inverse pattern was obtained for abstract rules, whereby the conditions were successfully discriminated in the frontal lobe but not in the BG. This double dissociation highlights the difference between cortical and subcortical contributions to cognitive control and demonstrates the utility of multivariate approaches in investigations of functions that rely on distributed and overlapping neural substrates.

Breastfeeding, Cognitive and Noncognitive Development in Early Childhood: A Population Study.

BACKGROUND AND OBJECTIVES: There is mixed evidence from correlational studies that breastfeeding impacts children's development. Propensity score matching with large samples can be an effective tool to remove potential bias from observed confounders in correlational studies. The aim of this study was to investigate the impact of breastfeeding on children's cognitive and noncognitive development at 3 and 5 years of age. METHODS: Participants included ∼8000 families from the Growing Up in Ireland longitudinal infant cohort, who were identified from the Child Benefit Register and randomly selected to participate. Parent and teacher reports and standardized assessments were used to collect information on children's problem behaviors, expressive vocabulary, and cognitive abilities at age 3 and 5 years. Breastfeeding information was collected via maternal report. Propensity score matching was used to compare the average treatment effects on those who were breastfed. RESULTS: Before matching, breastfeeding was associated with better development on almost every outcome. After matching and adjustment for multiple testing, only 1 of the 13 outcomes remained statistically significant: children's hyperactivity (difference score, -0.84; 95% confidence interval, -1.33 to -0.35) at age 3 years for children who were breastfed for at least 6 months. No statistically significant differences were observed postmatching on any outcome at age 5 years. CONCLUSIONS: Although 1 positive benefit of breastfeeding was found by using propensity score matching, the effect size was modest in practical terms. No support was found for statistically significant gains at age 5 years, suggesting that the earlier observed benefit from breastfeeding may not be maintained once children enter school.

Can Early Intervention Improve Maternal Well-Being? Evidence from a Randomized Controlled Trial.

OBJECTIVE: This study estimates the effect of a targeted early childhood intervention program on global and experienced measures of maternal well-being utilizing a randomized controlled trial design. The primary aim of the intervention is to improve children's school readiness skills by working directly with parents to improve their knowledge of child development and parenting behavior. One potential externality of the program is well-being benefits for parents given its direct focus on improving parental coping, self-efficacy, and problem solving skills, as well as generating an indirect effect on parental well-being by targeting child developmental problems. METHODS: Participants from a socio-economically disadvantaged community are randomly assigned during pregnancy to an intensive 5-year home visiting parenting program or a control group. We estimate and compare treatment effects on multiple measures of global and experienced well-being using permutation testing to account for small sample size and a stepdown procedure to account for multiple testing. RESULTS: The intervention has no impact on global well-being as measured by life satisfaction and parenting stress or experienced negative affect using episodic reports derived from the Day Reconstruction Method (DRM). Treatment effects are observed on measures of experienced positive affect derived from the DRM and a measure of mood yesterday. CONCLUSION: The limited treatment effects suggest that early intervention programs may produce some improvements in experienced positive well-being, but no effects on negative aspects of well-being. Different findings across measures may result as experienced measures of well-being avoid the cognitive biases that impinge upon global assessments.