Asthma, eczema and hay fever and are common diseases in childhood and responsible for a significant burden on families and health services. Atopic eczema, hay fever (allergic rhinitis) and atopic asthma often co-exist. Early diagnosis and appropriate management can reduce progression and severity of these diseases.
The Born in Bradford (BiB) birth cohort includes over 13,500 children born between 2007 and 2011, with around half born to women of Pakistani ethnicity. Linked primary care and hospital admission data are available for 97% of BiB children. Two sub-studies within BiB, the Allergy and Infection Study (ALL IN, n=2559) and Mechanisms of the Development of ALLergy (MeDALL, n=1814), have collected detailed parental questionnaire data at age 1 and 2 years (ALL IN) and at 4 years (MeDALL). The current data collection phase for the whole BiB cohort, Growing Up, at ages 7-11 years is ongoing and also includes questions on these outcomes.
The BiB data provide an opportunity to investigate trajectories of allergic disease and asthma through childhood, by ethnic group. The linked primary care and hospital electronic health records (EHR) will contribute a wealth of data which can be analysed with machine learning methods. There is uncertainty over the validity of routine data but the extensive BiB questionnaire data at different ages provide a rare opportunity to test this.
The aims of this proposed study are:
1) to link research and routine data to explore early life and childhood longitudinal trajectories and describe clinical phenotypes of asthma, eczema and hay fever;
2) to investigate ethnic inequalities in access to care and presentation of these diseases
3) to investigate early life risk factors for these diseases
Questionnaire data are available from the BiB ALL IN sub-study at age 1 year, including questions on pets, family history of asthma/eczema/hay fever, housing conditions (damp, heating, flooring, bedding etc.), and at 2 years (as for age 1 plus eczema, hay fever, food allergy). Detailed data relevant to asthma, eczema and hay fever are available for the MeDALL sub-study participants at 4 years, including skin prick testing for 2269.
BiB receive regular extracts of primary care EHR data on diagnoses and prescriptions for BiB children, which will be linked to the BiB maternal baseline questionnaire data, including socio-demographic and household characteristics. Linked hospital admissions data are available from the Bradford Royal Infirmary. We will also compare EHR data and questionnaire data from the Avon Longitudinal Study of Parents and Children (ALSPAC).
Latent class analysis or other cluster methods, such as k-means clustering, will be used to identify clinical phenotypes of asthma, eczema and hay fever. Longitudinal extensions of these cluster methods will be used to describe trajectories over age.
This study will provide important data on the validity of routine primary care EHR for asthma and allergic diseases, which is relevant as EHR are increasingly used for research studies. The comparison of questionnaire and EHR data will indicate whether there are ethnic differences in access to primary care for these diseases. Identification of clinical phenotypes of asthma, eczema and hay fever will inform appropriate treatment and management and the identification of factors associated with disease progression or severity could indicate potential prevention strategies.

Better Care South-West Partnership is a collaboration of NHS commissioners, primary, secondary, community and mental health care providers, local authorities, and academia. They look to address real-world health problems using the Bristol, North Somerset and South Gloucestershire (BNSSG) Systemwide health and social care dataset and have an ambition to use individual-level, linked routine care and administrative data to deliver a learning, Integrated Care System for the local population.
The Partnership represents a step change in using advanced analytics to deliver Better Care Loops across a care system and benefit patients and partner organisations. Its results will be scalable across the region and nationally.
Research Projects
· P-NEWS: personalised early warning scores for preventing unplanned critical care admission
· Precision antimicrobial prescribing: safeguarding patient outcomes and preserving future efficacy
· Using operation research methods to improve flow between acute and social care: modelling the responsiveness of system-level expenditure to changes in social care capacity
· Improving hospital efficiency by forecasting demand for hospital beds
· Underpinning infrastructure to the BNSSG Systemwide dataset

Vascular diseases such as hypertension, migraine, and atherosclerosis (hardening of arteries) involve changes in vascular cell function with reduced ability to contract and relax in response to different stresses. This results in chronic alterations in artery blood flow and maladaptive structural changes to the vascular wall leading to injury, tissue damage, and increased risk for life-threatening diseases such as stroke and heart attacks. Given that naturally occurring genetic variation contributes to changes in vascular function along with environmental risk from early life stages, it is now critical to evaluate the effects of these genetic associations at this stage using more systematic approaches. We plan to integrate these vascular phenotype data with high resolution molecular data to better understand how these genetic risk factors impact vascular disease risk at an early age.

The impact of the public health measures adopted to control the COVID-19 pandemic on young infants and family interactions is unknown. New information is vital to inform future policies and recovery for families and aid infant development. Infants may show more unsettled and
restless behaviours even if they are not aware of the situation, however, they may also show positive behaviours benefiting from more parental attention if parents are home. Understanding of both is important to manage further transitions in an ever changing home environment. Furthermore , young children will have been separated from wider family and friends. Online chats may provide a helpful substitute to retain attachments, but how young infants respond to such interactions is unknown . Using our existing methods to code indepeth parent and infant verbal and non-verbal behaviours we can compare parent and infant behaviours in interactions at this time to already collected and coded interactions of ALSPAC-G2 families pre-pandemic. We can also compare infant behaviours towards parents in the same room and a mimicked online interaction (where the one parent joins a chat from another room).

It is not clear why some people who are exposed to sars-cov-2 (the coronavirus) only develop mild symptoms while others go on to have life-threatening infections. Age and pre-existing medical problems are important, but it is likely that genetic factors also explain why some people are more susceptible to infection. One way to identify the most relevant genetic factors is to screen millions of points across the whole human genome and see which markers are more or less common in people with infection.

This type of analysis (called a genome-wide association study) needs very large studies to generate reliable results, and it is common to run the anlaysis in multiple different studies and then combine results. A global consortium has recently been set up to co-ordinate analysis of host genetic factors and coronavirus. This proposal plans to run genome-wide analysis in ALSPAC and share the results of this anlaysis which can then be combined with other studies globally. To do this it will be necessary to re-impute (re-process) the existing ALSPAC genetic data into an updated format so the results of analysis matches other studies.

Quantifying social contacts is essential for understanding how a disease will spread in a population. We have measured contact patterns using surveys that have formed the basis of predictive modelling presented at the UK Government Modelling advisory group on COVID-19. Here, we will measure social contact patterns during this unprecedented time of social distancing in the UK and compare to other studies and settings.

The on-going childhood obesity epidemic is accompanied by dramatic increases in childhood metabolic disorders, such as paediatric type 2 diabetes and a cluster of metabolic complications. However, children with obesity are not all equally prone to developing these metabolic disorders; research has shown that there is a subset of obese children who have a more ‘favourable’ health profile, including good insulin sensitivity and normal blood pressure, glucose regulation and blood lipid levels. This has been termed metabolically healthy obesity (MHO), which contrasts with metabolically unhealthy obesity (MUO), where excess body fat is accompanied by metabolic disorders. Since most of the research to date on MHO and MUO has focused on adults, the specific determinants of MHO and MUO in paediatric groups are still not clearly understood.

The concept of metabolically unhealthy and metabolically healthy profiles can also be applied to individuals of ‘normal’ weight, since normal-weight individuals can also have the metabolic and/or inflammatory abnormalities commonly observed in obese people. This group has been termed normal-weight metabolically unhealthy (NWMU). NWMU adults are also at increased risk of cardiometabolic diseases, however, little is known about the health profile and predictive characteristics of NWMU in children.

The relatively high proportion of metabolically healthy individuals in the obese (ranging from 4-60%) and metabolically unhealthy individuals in the lean population suggests that besides from total calories, diet quality could be an important influencing factor on metabolic health. However, few studies have looked at how dietary patterns in children might influence the sub-groups of obesity. Therefore, this study aims to identify which dietary patterns, lifestyle behaviours and socio-demographic factors in children are related to MHO and MUO, as well as NWMU in adolescents. Gaining a clearer understanding of the health profile, characteristics and potential determinants of MHO and MUO in paediatric groups could be valuable in developing more efficient and targeted treatment approaches for these groups of children with obesity.

Acute lymphoblastic leukaemia (ALL) is the most common cancer in children. Whilst a range of genetic abnormalities have been identified as crucial in disease initiation, these abnormalities alone are not sufficient for transformation. Maternal exposures during pregnancy such as smoking, or folate intake have the potential to impact on offspring DNA methylation. This epigenetic mark has also been shown to be altered in ALL. We have previously identified 5 CpG sites that have altered DNA methylation due to maternal smoking and folate intake and we aim to identify whether changes to DNA methylation at these sites affects risk of ALL.

School closures have been a central component of many countries’ response to contain COVID-19; however, we don’t know:
• What children do during unplanned school closures
• Whether children are infectious
• Whether other strategies could be equally effective.
Much of the policy is based on influenza, which does affect children more than SARS-CoV-2 [1].

Preliminary evidence suggests that children are able to become infected with SARS-CoV-2, but are either asymptomatic or show mild symptoms, with a minority of cases progressing to disease [2]. The role of healthy children in transmitting SARS-CoV-2 remains uncertain. It is of interest to define how many children do/don’t experience COVID-19 symptoms and have evidence of having had SARS-CoV-2 infection, this may have implications for transmission dynamics and policy decisions. Along with many European countries, the UK decided to close schools from March 23rd in an effort to slow SARS-CoV-2 transmission - only some vulnerable children and those of key workers remain in school. We do not know what children’s contact patterns are during this unplanned school closure, but there is evidence that contacts inside and outside the home might continue, especially for older children and where parents don’t agree with closures [3]. We propose to deploy a survey to G2 ALSPAC children to capture data on symptoms and contact patterns during the COVID-19 pandemic, as well determine evidence of SARS-CoV-2 infection through saliva antibody detection methods.