Predicting low cognitive ability at age 5 years using perinatal data and machine learning.

BACKGROUND: There are no early, accurate, scalable methods for identifying infants at high risk of poor cognitive outcomes in childhood. We aim to develop an explainable predictive model, using machine learning and population-based cohort data, for this purpose. METHODS: Data were from 8858 participants in the Growing Up in Ireland cohort, a nationally representative study of infants and their primary caregivers (PCGs). Maternal, infant, and socioeconomic characteristics were collected at 9-months and cognitive ability measured at age 5 years. Data preprocessing, synthetic minority oversampling, and feature selection were performed prior to training a variety of machine learning models using ten-fold cross validated grid search to tune hyperparameters. Final models were tested on an unseen test set. RESULTS: A random forest (RF) model containing 15 participant-reported features in the first year of infant life, achieved an area under the receiver operating characteristic curve (AUROC) of 0.77 for predicting low cognitive ability at age 5. This model could detect 72% of infants with low cognitive ability, with a specificity of 66%. CONCLUSIONS: Model performance would need to be improved before consideration as a population-level screening tool. However, this is a first step towards early, individual, risk stratification to allow targeted childhood screening. IMPACT: This study is among the first to investigate whether machine learning methods can be used at a population-level to predict which infants are at high risk of low cognitive ability in childhood. A random forest model using 15 features which could be easily collected in the perinatal period achieved an AUROC of 0.77 for predicting low cognitive ability. Improved predictive performance would be required to implement this model at a population level but this may be a first step towards early, individual, risk stratification.

Big data, machine learning, and population health: predicting cognitive outcomes in childhood.

The application of machine learning (ML) to address population health challenges has received much less attention than its application in the clinical setting. One such challenge is addressing disparities in early childhood cognitive development-a complex public health issue rooted in the social determinants of health, exacerbated by inequity, characterised by intergenerational transmission, and which will continue unabated without novel approaches to address it. Early life, the period of optimal neuroplasticity, presents a window of opportunity for early intervention to improve cognitive development. Unfortunately for many, this window will be missed, and intervention may never occur or occur only when overt signs of cognitive delay manifest. In this review, we explore the potential value of ML and big data analysis in the early identification of children at risk for poor cognitive outcome, an area where there is an apparent dearth of research. We compare and contrast traditional statistical methods with ML approaches, provide examples of how ML has been used to date in the field of neurodevelopmental disorders, and present a discussion of the opportunities and risks associated with its use at a population level. The review concludes by highlighting potential directions for future research in this area. IMPACT: To date, the application of machine learning to address population health challenges in paediatrics lags behind other clinical applications. This review provides an overview of the public health challenge we face in addressing disparities in childhood cognitive development and focuses on the cornerstone of early intervention. Recent advances in our ability to collect large volumes of data, and in analytic capabilities, provide a potential opportunity to improve current practices in this field. This review explores the potential role of machine learning and big data analysis in the early identification of children at risk for poor cognitive outcomes.

Co-design of a school-based physical activity intervention for adolescent females in a disadvantaged community: insights from the Girls Active Project (GAP).

BACKGROUND: Globally, adolescents' physical activity (PA) participation rates are low, particularly among lower socioeconomic groups, with females consistently the least active. The aim of this study was to co-design, with adolescent females, a school-based PA intervention in a single-sex, females-only designated disadvantaged post-primary school in Ireland. This involved using the Behaviour Change Wheel (BCW) and Public and Patient Involvement (PPI). This paper outlines the novel methodological approach taken. METHODS: The three stages 1) understand the behaviour, 2) identify intervention options, and 3) identify content and implementation options of the BCW guide is described. A student PPI Youth Advisory Group (YAG) (n = 8, aged 15-17) was established. Mixed-methods were used with students (n = 287, aged 12-18) and teachers (n = 7) to capture current self-reported PA levels and to identify factors influencing adolescent females' PA behaviour in their school setting. The intervention options, content and implementation options were identified through discussion groups with the YAG. The Template for Intervention Description and Replication (TIDieR) checklist was used to specify details of the intervention. RESULTS: Just 1.4% of the students in this sample reported meeting the recommended PA guidelines. Students identified having more 'time' as the strongest predictor to becoming more active in school (Mean = 4.01, 95% CI 3.91 to 4.12). Social influences, environmental context and resources, behavioural regulation, beliefs about capabilities, goals, and reinforcement emerged from the qualitative data as factors influencing PA behaviour at school. The BCW co-design process resulted in the identification of seven intervention functions, four policy categories and 21 Behaviour Change Techniques. The Girls Active Project (GAP) intervention, a peer-led, after-school PA programme was proposed. CONCLUSIONS: This paper describes how the BCW, a comprehensive, evidence-based, theory-driven framework was used in combination with PPI to co-design a school-based intervention aimed to increase adolescent females' PA levels. This approach could be replicated in other settings to develop targeted behavioural interventions in populations with specific demographic characteristics.

The predictive value of the ages and stages questionnaire in late infancy for low average cognitive ability at age 5.

AIM: This retrospective, longitudinal study examined the predictive value of the ages and stages questionnaire (ASQ) in late infancy for identifying children who progressed to have low cognitive ability at 5 years of age. METHODS: The ASQ was performed on 755 participants from the Irish BASELINE birth cohort at 24 or 27 months of age. Intelligence quotient was measured at age 5 with the Kaufmann Brief Intelligence Test, Second Edition, and low cognitive ability was defined as a score more than 1 standard deviation below the mean. The ASQ's predictive value was examined, together with other factors associated with low cognitive ability at 5 years. RESULTS: When the ASQ was performed at 24 or 27 months, the overall sensitivity for identifying low cognitive ability at 5 years was 20.8% and the specificity was 91.1%. Using a total score cut-off point increased the sensitivity to 46.6% and 71.4% at 24 and 27 months, but specificity fell to 74.1% and 67.2%, respectively. After adjusting for ASQ performance, maternal education and family income were strongly associated with cognitive outcomes at 5 years. CONCLUSION: The ASQ did not detect the majority of children with low cognitive ability at age 5. Alternative methods need investigation.

Embracing multiple aims in healthcare improvement and innovation.

BACKGROUND: Healthcare and other industries have to manage and coordinate a number of different aims and longer-term ambitions. They must maintain quality, manage costs, support and retain the workforce, manage regulatory demands and consider wider societal objectives. These aims are all legitimate, but they are not necessarily aligned, neither in their time frame nor in their nature. Conflicts between aims have a profound influence on the implementation of safety and quality improvement and wider innovation system change. Healthcare leaders understand that these aims may conflict, but the extent and nature of such conflicts have been underestimated. OBJECTIVE: This paper aims to support medical and nursing leadership and executives in the complex task of managing multiple aims in relation to improvement and innovation. METHODS: We drew on our experience and the wider industrial and healthcare literature to find examples of studies and improvement projects with multiple aims and examples of innovation and change in which conflicting aims were apparent. We sought to identify principles that would enable the management of parallel aims and practical strategies that might facilitate implementation. RESULTS: We argue that almost all improvement and innovation in healthcare should address parallel aims, actively seek to articulate these aims and manage potential conflicts between them. We propose four underlying principles to support a more productive approach to the identification and management of parallel aims: embrace multiple aims, consider both short- and long-term aims and ambitions, consider the wider societal context and appreciate that all changes take place within an evolving, dynamic context. In terms of practical actions, we identified five key strategies: (i) identify and monitor the parallel aims and accept that some will conflict; (ii) slow down to accommodate the natural flexibility of the system; (iii) think both the short term and the long term; (iv) expect and endeavour to anticipate some unintended consequences and (v) resist downgrading the project to partial implementation. CONCLUSIONS: We have argued that most improvement and innovation, unlike controlled trials, should consider multiple aims. We set out some broad principles and practices to reduce conflict and suggest avenues to manage conflicts and support positive synergies. We suggest that if this is not done, conflicts are much more likely to arise, which will be detrimental to the change process. Articulating the multiple aims and actively seeking to manage them in parallel will promote a more flexible and productive approach to innovation and change.

Experiences of residents, family members and staff in residential care settings for older people during COVID-19: A mixed methods study.

AIM: The aim of this study was to explore the COVID-19 pandemic as it was experienced by people on the front line in residential care settings for older people in the Republic of Ireland (ROI). BACKGROUND: The COVID-19 pandemic had a disproportionate effect in residential care settings for older people in Ireland. METHODS: A two-phased mixed methods study was conducted, consisting of an online survey administered shortly after the first wave of the virus to staff, residents and family members and one-to-one interviews with family members shortly after wave 2 of the virus. RESULTS: Isolation, loss of connectedness as well as a reduction in the level/quality of care provision led to significant adverse impacts for both residents and their families. Staff reported high levels of stress, trauma and burnout. Family input to care was suspended, with adverse consequences. CONCLUSION: The pandemic had an extremely adverse impact on residents, family members and staff in care settings for older people. IMPLICATIONS FOR NURSING MANAGEMENT: Strategies to ensure that residents' physical, emotional and social needs and staffs' professional and personal needs are appropriately supported during future waves of the pandemic should now be implemented.

Exploring the prevalence and profile of epilepsy across Europe using a standard retrospective chart review: Challenges and opportunities.

OBJECTIVE: This study aimed to determine the prevalence of epilepsy in four European countries (Austria, Denmark, Ireland, and Romania) employing a standard methodology. The study was conducted under the auspices of ESBACE (European Study on the Burden and Care of Epilepsy). METHODS: All hospitals and general practitioners serving a region of at least 50 000 persons in each country were asked to identify patients living in the region who had a diagnosis of epilepsy or experienced a single unprovoked seizure. Medical records were accessed, where available, to complete a standardized case report form. Data were sought on seizure frequency, seizure type, investigations, etiology, comorbidities, and use of antiseizure medication. Cases were validated in each country, and the degree of certainty was graded as definite, probable, or suspect cases. RESULTS: From a total population of 237 757 in the four countries, 1988 (.8%) patients were identified as potential cases of epilepsy. Due to legal and ethical issues in the individual countries, medical records were available for only 1208 patients, and among these, 113 had insufficient clinical information. The remaining 1095 cases were classified as either definite (n = 706, 64.5%), probable (n = 191, 17.4%), suspect (n = 153, 14.0%), or not epilepsy (n = 45, 4.1%). SIGNIFICANCE: Although a precise prevalence estimate could not be generated from these data, the study found a high validity of epilepsy classification among evaluated cases (95.9%). More generally, this study highlights the significant challenges facing epidemiological research methodologies that are reliant on patient consent and retrospective chart review, largely due to the introduction of data protection legislation during the study period. Documentation of the epilepsy diagnosis was, in some cases, relatively low, indicating a need for improved guidelines for assessment, follow-up, and documentation. This study highlights the need to address the concerns and requirements of recruitment sites to engage in epidemiological research.

Genetic overlap between autoimmune diseases and non-Hodgkin lymphoma subtypes.

Epidemiologic studies show an increased risk of non-Hodgkin lymphoma (NHL) in patients with autoimmune disease (AD), due to a combination of shared environmental factors and/or genetic factors, or a causative cascade: chronic inflammation/antigen-stimulation in one disease leads to another. Here we assess shared genetic risk in genome-wide-association-studies (GWAS). Secondary analysis of GWAS of NHL subtypes (chronic lymphocytic leukemia, diffuse large B-cell lymphoma, follicular lymphoma, and marginal zone lymphoma) and ADs (rheumatoid arthritis, systemic lupus erythematosus, and multiple sclerosis). Shared genetic risk was assessed by (a) description of regional genetic of overlap, (b) polygenic risk score (PRS), (c)"diseasome", (d)meta-analysis. Descriptive analysis revealed few shared genetic factors between each AD and each NHL subtype. The PRS of ADs were not increased in NHL patients (nor vice versa). In the diseasome, NHLs shared more genetic etiology with ADs than solid cancers (p = .0041). A meta-analysis (combing AD with NHL) implicated genes of apoptosis and telomere length. This GWAS-based analysis four NHL subtypes and three ADs revealed few weakly-associated shared loci, explaining little total risk. This suggests common genetic variation, as assessed by GWAS in these sample sizes, may not be the primary explanation for the link between these ADs and NHLs.

Occupational exposure to ionizing radiation and risk of lymphoma subtypes: results of the Epilymph European case-control study.

BACKGROUND: Evidence linking risk of lymphoma and B-cell lymphoma subtypes to ionizing radiation is inconclusive, particularly at low exposure levels. METHODS: We investigated risk of lymphoma (all subtypes), B-cell lymphomas, and its major subtypes, associated with low-level occupational exposure to ionizing radiation, in 2346 lymphoma cases and 2463 controls, who participated in the multicenter EpiLymph case-control study. We developed a job-exposure matrix to estimate exposure to ionizing radiation, distinguishing between internal and external radiation, and we applied it to the lifetime occupational history of study subjects, We calculated the Odds Ratio (OR) and its 95% confidence interval (95% CI) for lymphoma (all subtypes combined), B-cell lymphoma, and its major subtypes using unconditional, polytomous logistic regression adjusting for age, gender, and education. RESULTS: We did not observe an association between exposure metrics of external and internal radiation and risk of lymphoma (all subtypes), nor with B-cell lymphoma, or its major subtypes, at the levels regularly experienced in occupational settings. An elevated risk of diffuse large B cell lymphoma was observed among the most likely exposed study subjects with relatively higher exposure intensity, which would be worth further investigation. CONCLUSIONS: Further investigation is warranted on risk of B cell lymphoma subtypes associated with low-level occupational exposure to external ionizing radiation, and to clarify whether lymphoma should be included among the cancer outcomes related to ionizing radiation.

Impact of TeamSTEPPS on patient safety culture in a Swiss maternity ward.

OBJECTIVE: To assess the impact of implementation of the TeamSTEPPS teamwork improvement concept on patient safety culture. DESIGN: Pre-post culture assessment using the Hospital Survey on Patient Safety Culture, at baseline and one year after implementation of TeamSTEPPS. SETTING: Two maternity wards within the same 480-bed multisite teaching hospital. INTERVENTION: Implementation of the TeamSTEPPS teamwork improvement concept. MAIN OUTCOME MEASURES: Analysis of variation of the percentage of positive responses (score) in both wards (intervention and control) was conducted. RESULTS: There was a significant increase in scores in three dimensions of patient safety culture in the intervention ward: Supervisor/Manager Expectations and Actions Promoting Safety increased from 48.7% in 2015 to 70.8% in 2016 (P < 0.005); Teamwork Within Units increased from 35.5% in 2015 to 54.5% in 2016 (P < 0.005); Nonpunitive Response to Errors increased from 16.7% in 2015 to 32.3% in 2016 (P < 0.005). Other dimensions showed no significant changes. In the control ward, there was a significant decrease in scores in one dimension. A secondary analysis of differences in differences still shows significant improvement in one dimension (Supervisor/Manager Expectations and Actions Promoting Safety P < 0.005). CONCLUSION: After implementing the TeamSTEPPS teamwork concept, patient safety culture significantly improved for three of twelve dimensions in the intervention group. When controlling for differences in baseline scores between implementation and control wards, a significant improvement remains in one dimension. This suggests that TeamSTEPPS could be considered when seeking to enhance patient safety culture, especially in high-risk environments such as maternity wards.

'We know what we should eat but we don't ': a qualitative study in Irish secondary schools.

Healthy eating opportunities for young people need to be provided in the school setting. Links between formal education and local policies and practices, and active involvement of students are emphasized in whole school approaches to health promotion. In many jurisdictions, schools struggle to provide nutritious food that is acceptable to students. The aim of this study was to conduct an in-depth exploration of school food and students' food choice with students, teachers and principals in six Irish secondary schools. Students conducted their own focus groups with peers, and a researcher conducted focus groups with teachers and one-to-one interviews with principals. Students, teachers and principals expressed dissatisfaction with food provision within their respective schools and reported unhealthy dietary behaviours among the student body as the norm. Divergent views, however, emerged regarding education, knowledge and the primary drivers of food choice. Teachers and principals saw a role for more education modules and students believed environmental infrastructures, practices and policies needed to be addressed to improve behaviours. This consultation with students, teachers and principals highlight that, in addition to education, the social and environmental aspects of food choice and eating at school deserve attention in future research and policy.

Public health digitalization in Europe.

BACKGROUND: As digitalization is progressively permeating all aspects of society, how can be it fruitfully employed to sustain the public health goals of quality, accessibility, efficiency and equity in health care and prevention? In this paper, we reflect on the potential of applying digital tools to public health and discuss some key challenges. METHODS: We first define 'digitalization' in its broader sense, as well as applied to public health. We then build a conceptual framework where key public health domains are associated to available digital technologies in a matrix that help to identify digital features that bolster public health action. We also provide illustrative data and evidence on the application of digital solutions on selected public health areas. In the second part, we identify the strategic pillars for a successful European strategy for public health digitalization and we outline how the approach being pursued by the European Public Health Association (EUPHA) applies to digital health. RESULTS: From a public health perspective, digitalization is being touted as providing several potential benefits and advantages, including support for the transition from cure to prevention, helping to put people and patients at the center of care delivery, supporting patient empowerment and making healthcare system more efficient, safer and cheaper. These benefits are enabled through the following features of digital technologies: (i) Personalization and precision; (ii) Automation; (iii) Prediction; (iv) Data analytics and (v) Interaction. CONCLUSION: A successful European strategy for public health digitalization should integrate the following pillars: political commitment, normative frameworks, technical infrastructure, targeted economic investments, education, research, monitoring and evaluation. EUPHA acknowledges digitalization is an asset for public health and is working both to promote the culture of "public health digitalization", as well as to enable its planning, implementation and evaluation at the research, practice and policy level.

A comparison of physical activity, physical fitness levels, BMI and blood pressure of adults with intellectual disability, who do and do not take part in Special Olympics Ireland programmes: Results from the SOPHIE study.

People with an intellectual disability are less physically active, live more sedentary lives, have lower fitness levels and are more likely to be overweight or obese than the general population. No evidence exists on the impact of participation in Special Olympics Ireland (SOI) on physical activity and physical fitness levels. Adults with intellectual disabilities (16-64 years) were recruited from services and SOI clubs. Physical measures included waist circumference, height, weight, blood pressure, heart rate and 6-min walking test. Self-report questionnaires gathered data on physical activity levels. Actigraph (GT3X) accelerometers were used to gain an objective measure of physical activity. SOI participants accumulated more moderate to vigorous physical activity per day, had higher fitness levels and more positive health profile scores than those not taking part in SOI. SOI has the potential to make a positive difference to people's physical health and subsequently their overall health and well-being.