Search

Background Major congenital anomalies (CAs) affect around 2% of live births and are a primary cause of infant mortality, childhood morbidity and long-term disability, often requiring hospitalisation and/or surgery. Children with CAs are at greater risk of lower educational attainment compared with their peers, which could be due to learning disabilities, higher rates of ill-health and school absences, or lack of adequate educational support. Our study will compare the educational attainment of children with CAs to those of their peers up to age 11 in England, using linked administrative health and education data. Methods We will analyse data from the ECHILD (Education and Child Health Insights from Linked Data) database. Children born in NHS-funded hospitals from 1st September 2003 to 31st August 2008 whose hospital records were linked to their educational records at three Key Stages (ages 4/5, 6/7 and 10/11 years) will be included. Children with different CAs, indicated by recorded hospital diagnosis codes, will be compared to children without CAs. We will compare the proportions of enrolled children who take the assessment, the proportions who reached national expected levels of attainment, and the mean standardised attainment scores for Maths and English at each Key Stage. We will describe variations in outcome by sex, ethnic minority background, region, and neighbourhood deprivation, and perform regression modelling to compare the attainment trajectories of children with and without CAs, controlling for sociodemographic factors. Ethics and dissemination Ethics approval has been obtained for the analyses of the ECHILD database. Our findings will provide information for parents regarding their children’s expected academic potential, and also enable the development of interventions to support those at risk of not doing well. We will disseminate our findings to academics, policy makers, service users and providers through seminars, peer-reviewed publications, conference abstracts and other media (lay summaries and infographics).

Background Children with epilepsy have higher rates of internalising mental health (MH) conditions compared to their peers, although available evidence is limited by small and selective samples. We used linked primary care and hospital admission records covering approximately 4.42% of English population to describe incidence of mental health problems in children with epilepsy compared to their peers. Methods We developed a cohort of young people aged 7-17 years old between 2010-2019, with minimum 1 year of follow-up until death, 18th birthday or 2020 using linked data from the Clinical Practice Research Datalink and Hospital Episodes Statistics. Epilepsy (exposure), depression and anxiety (outcomes) were identified using a combination of diagnostic codes, and primary care prescribing records. We estimated incident rate ratios (IRR) for anxiety/depression/any MH condition for children with epilepsy compared to their peers using negative binomial regression adjusted for sex, age, and area-level deprivation. Results Study cohort included 510,177 children and young people, of whom 3,011 (0.59%) had epilepsy. Children with epilepsy had higher incidence of any MH condition (IRR: 1.39, 95% CI: 1.11–1.73) and anxiety (IRR: 1.38, 95% CI: 1.06–1.77), with differences in depression being non-significant (IRR: 1.25, 95% CI: 0.92–1.66). Discussion Children with epilepsy have higher rates of MH problems than their peers. Further work will examine if this can be partially explained by increased opportunity for MH diagnosis due to more healthcare contacts. Our findings emphasise the importance of integrated care models that address both neurological and psychological aspects of epilepsy management.

Objective Children with Down syndrome (DS) have high rates of chronic health conditions (CHCs) which can affect special education needs and primary school placement (mainstream or special). We assessed timing and type of CHC by year-specific primary school placement (ages 5 to 11 years). Approach We followed two cohorts: first, a birth cohort of all children with DS born from 2003 to 2020 in England using hospital administrative data; second, a sub-cohort with linked hospital and education records who enrolled in state-funded primary schools in England at age 5 from 2009/10 to 2013/14. We calculated the cumulative incidence of CHCs and multimorbidity (defined as distinct CHCs affecting two or more body systems) from birth through age 11, accounting for variable follow-up and the competing risk of death. We assessed multimorbidity by school placement over time. Results The birth cohort included 11,478 children with DS (prevalence: 0.11%). By age 5, 6.2% of children had died, 80.8% had at least one recorded CHC, and 53.4% had multiple CHCs. Corresponding figures by age 11 were 6.9%, 87.4% and 65.9%, respectively. 85% of children with DS alive at school entry linked to a school record. 72% were enrolled in mainstream primary school at age 5, decreasing to 62% by age 7, and 54% by age 11. Conclusions Most children with DS have multiple hospital-recorded CHCs. Nearly three-quarters start mainstream primary school, but just over half remain by age 11. Implications Parents, clinicians, and educators will benefit from understanding relationships between multimorbidity and school placement.

Background Over a third of children in England receive either special educational need (SEN) provision/children’s social care (CSC) support at least once during school. We aimed to describe all-cause and adversity-related mortality for adolescents who received SEN or CSC provision in England. Approach We used linked data from hospital admissions, death registrations, social care and education in the ECHILD database to follow all adolescents in state schools from age 13-years between 2009/10-2014/15 until death or 2020. Adolescents were grouped according to provision ever recorded before age 14: None, SEN only, CSC only or CSC and SEN. We estimated all-cause and adversity-related mortality in children with SEN/CSC/both compared to peers. Results The cohort comprised 3,381,502 children (51% boys), of whom 30.1% (n=1,017,670, 62% boys) received SEN only, 4.2% (n=143,430, 36% boys) received CSC only and 8.7% (n=293,310, 57% boys) received SEN and CSC. In ten-year follow up, 6,590 deaths occurred (66% in boys). Mortality rates were 5.4 times higher for adolescents receiving both SEN and CSC, 2.4 times higher for CSC only, and 1.7 times higher for SEN only compared to peers. Rates of adversity-related deaths were highest for boys receiving SEN and CSC compared to peers. Conclusions & Implications 8-in-1000 adolescents who received both SEN and CSC before age 13 died by age 23 years, far more often than their peers. Policies need to consider support for young people with additional educational and social care needs as they leave school and transition to adulthood.

Objective We quantified planned and unplanned hospital admissions during adolescence for children with special educational needs (SEN) or placed into the care of the local authority (child looked after in out-of-home care; CLA). Approach We analysed linked hospital, education and social care records with whole nation coverage of England (ECHILD). Our cohort comprised pupils who started secondary school between 2007/8 and 2011/12 (aged 11), with follow-up until 2018/19 (up to 17-22 years old). Hospital admission rates were examined by sex across distinct groups with varying levels of support: i) no support, ii) school-based SEN only, iii) higher SEN (an Education and Health Care Plan) only, iv) CLA only, v) SEN and CLA, and vi) higher SEN and CLA. Negative binomial regression models were used to estimate rate ratios (RR) for each group relative to no support. Results The study population totalled 2,827,464 children; 65% (1,835,404) had no support, 30% (848,620) SEN only, 3% (93,620) higher SEN only, 1% (30,880) SEN and CLA, with the remaining groups accounting for

Background Transition from adolescent to adult health and education services is a vulnerable period for young people with learning disability (LD) or autism spectrum disorders (ASD), yet there is a lack of information on their risk of death in that period. Methods We developed an English national cohort of 11-year-olds enrolled in state schools between 2009/10-20014/15, followed-up until their 22nd birthday, death or 31st August 2020 using linked health, mortality and education data from the ECHILD database. LD and ASD were indicated using diagnoses from hospital admissions or if LD or ASD were a reason for special educational needs provision in education records before age 11. We estimated crude mortality rates and compared with unaffected peers using rate ratios (RR). Results The cohort comprised 3,080,516 young people (52% boys), of whom 4,589 (0.15%) died (crude mortality rate: 0.17/1000 person-years, 95% CI: 0.16-0.17/1000). 45,174 (1.5%) young people were autistic (85% boys; 15% with co-existing LD), of whom 113 died (mortality rate: 0.29/1000 person-years, 95%CI: 0.24-0.35/1000). Mortality was 1.7 times higher for autistic young people compared to their peers (RR: 1.7, 95% CI: 1.4-2.1). Among 40,264 (1.3% of all) young people with LD (63% boys; 17% with co-existing ASD) there were 812 deaths (mortality rate: 2.31/1000, 95%CI: 2.15-2.47/1000PY). They had 17 times higher risk of death (RR: 16.4, 95%CI: 15.2-17.7). Conclusions Young people with LD/ASD are more likely to die than their peers. Findings on disparities within these groups according to co-existing comorbidities, sex and socioeconomic factors will be reported.

Introduction Over 3000 infants suffer a brain injury around the time of birth every year in England. Although these injuries can have important implications for children and their families, our understanding of how these injuries affect children’s lives is limited.Methods and analysis The aim of the CHERuB study (Childhood Health and Educational outcomes afteR perinatal Brain injury) is to investigate longitudinal childhood health and educational outcomes after perinatal brain injury through the creation of a population-matched cohort study. This study will use the Department of Health and Social Care definition of perinatal brain injury which includes infants with intracranial haemorrhage, preterm white matter injury, hypoxic ischaemic encephalopathy, perinatal stroke, central nervous system infections, seizures and kernicterus. All children born with a perinatal brain injury in England between 2008 and 2019 will be included (n=54 176) and two matched comparator groups of infants without brain injury will be created: a preterm control group identified from the National Neonatal Research Data Set and a term/late preterm control group identified using birth records. The national health, education and social care records of these infants will be linked to ascertain their longitudinal childhood outcomes between 2008 and 2023. This cohort will include approximately 170 000 children. The associations between perinatal brain injuries and survival without neurosensory impairment, neurodevelopmental impairments, chronic health conditions and mental health conditions throughout childhood will be examined using regression methods and time-to-event analyses.Ethics and dissemination This study has West London Research Ethics Committee and Confidential Advisory Group approval (20/LO/1023 and 22/CAG/0068 issued 20/10/2022). Findings will be published in open-access journals and publicised via the CHERuB study website, social media accounts and our charity partners.

Background Neurodisability describes a broad set of conditions affecting the brain and nervous system which result in functional limitations. Children with neurodisability have more hospital admissions than their peers without neurodisability and higher rates of school absence. However, longitudinal evidence comparing rates of hospital admission and school absence in children with neurodisability to peers without neurodisability throughout school is limited, as is understanding about whether differences are greatest for planned care (e.g., scheduled appointments) or unplanned care. This study will describe rates of planned and unplanned hospital admissions and school absence due to illness and medical reasons throughout primary school (Reception to Year 6, ages 4 to 11 in England) for children with neurodisability and all other children, using linked individual-level health and education data. Methods We will use the ECHILD (Education and Child Insights from Linked Data) database, which links educational and health records across England. We will define a primary school cohort of children who were born in National Health Service-funded hospitals in England between 1st September 2003 and 31st August 2008, and who were enrolled in Reception (age 4/5) at state-funded schools. We will use hospital admissions records to identify children who have recorded indicators of neurodisability from birth up to the end of primary school (Year 6, age 10/11). Results We will describe rates of planned and unplanned hospital admissions and health-related school absence for three groups of children: those with a neurodisability indicator first recorded before beginning primary school, those with neurodisability first recorded during primary school, and those without a record of neurodisability before end of primary school. Conclusions We will further explore whether differences between these group vary across primary school years and by socioeconomic and demographic characteristics.

Introduction One-third of children in England have special educational needs (SEN) provision recorded during their school career. The proportion of children with SEN provision varies between schools and demographic groups, which may reflect variation in need, inequitable provision and/or systemic factors. There is scant evidence on whether SEN provision improves health and education outcomes.Methods The Health Outcomes of young People in Education (HOPE) research programme uses administrative data from the Education and Child Health Insights from Linked Data—ECHILD—which contains data from all state schools, and contacts with National Health Service hospitals in England, to explore variation in SEN provision and its impact on health and education outcomes. This umbrella protocol sets out analyses across four work packages (WP). WP1 defined a range of ‘health phenotypes’, that is health conditions expected to need SEN provision in primary school. Next, we describe health and education outcomes (WP1) and individual, school-level and area-level factors affecting variation in SEN provision across different phenotypes (WP2). WP3 assesses the impact of SEN provision on health and education outcomes for specific health phenotypes using a range of causal inference methods to account for confounding factors and possible selection bias. In WP4 we review local policies and synthesise findings from surveys, interviews and focus groups of service users and providers to understand factors associated with variation in and experiences of identification, assessment and provision for SEN. Triangulation of findings on outcomes, variation and impact of SEN provision for different health phenotypes in ECHILD, with experiences of SEN provision will inform interpretation of findings for policy, practice and families and methods for future evaluation.Ethics and dissemination Research ethics committees have approved the use of the ECHILD database and, separately, the survey, interviews and focus groups of young people, parents and service providers. These stakeholders will contribute to the design, interpretation and communication of findings.

Abstract Background There is limited understanding of the drivers of increasing infant accident and emergency (A&E) attendances and emergency hospital admissions across England. We examine variations in use of emergency hospital services among infants by local areas in England and investigate the extent to which infant and socio-economic factors explain these variations. Methods Birth cohort study using linked administrative Hospital Episode Statistics data in England. Singleton live births between 1-April-2012 and 31-March-2019 were followed up for 1 year; from 1-April-2013 (from the discharge date of their birth admission) until their first birthday, death or 31-March-2019. Mixed effects negative binomial models were used to calculate incidence rate ratios for A&E attendances and emergency admissions and mixed effects logistic regression models estimated odds ratio of conversion (the proportion of infants subsequently admitted after attending A&E). Models were adjusted for individual-level factors and included a random effect for local authority (LA). Results The cohort comprised 3,665,414 births in 150 English LAs. Rates of A&E attendances and emergency admissions were highest amongst: infants born

Summary: Background: Transition from paediatric to adult health care may disrupt continuity of care, and result in unmet health needs. We describe changes in planned and unplanned hospital admission rates before, during and after transition for young people with learning disability (LD), or autism spectrum disorders (ASD) indicated in hospital records, who are likely to have more complex health needs. Methods: We developed two mutually exclusive cohorts of young people with LD, and with ASD without LD, born between 1990 and 2001 in England using national hospital admission data. We determined the annual rate of change in planned and unplanned hospital admission rates before (age 10–15 years), during (16–18 years) and after (19–24 years) transition to adult care using multilevel negative binomial regression models, accounting for area-level deprivation, sex, birth year and presence of comorbidities. Findings: The cohorts included 51,291 young people with LD, and 46,270 autistic young people. Admission rates at ages 10–24 years old were higher for young people with LD (54 planned and 25 unplanned admissions per 100 person-years) than for autistic young people (17/100 and 16/100, respectively). For young people with LD, planned admission rates were highest and constant before transition (rate ratio [RR]: 0.99, 95% confidence interval [CI] 0.98–0.99), declined by 14% per year of age during (RR: 0.86, 95% CI: 0.85–0.88), and remained constant after transition (RR: 0.99, 95% CI: 0.99–1.00), mainly due to fewer admissions for non-surgical care, including respite care. Unplanned admission rates increased by 3% per year of age before (RR: 1.03, 95% CI: 1.02–1.03), remained constant during (RR: 1.01, 95% CI: 1.00–1.03) and increased by 3% per year after transition (RR: 1.03, 95% CI: 1.02–1.04). For autistic young people, planned admission rates increased before (RR: 1.06, 95% CI: 1.05–1.06), decreased during (RR: 0.95, 95% CI: 0.93–0.97), and increased after transition (RR: 1.05, 95%: 1.04–1.07). Unplanned admission rates increased most rapidly before (RR: 1.16, 95% CI: 1.15–1.17), remained constant during (RR: 1.01, 95% CI: 0.99–1.03), and increased moderately after transition (RR: 1.03, 95% CI: 1.02–1.04). Interpretation: Decreases in planned admission rates during transition were paralleled by small but consistent increases in unplanned admission rates with age for young people with LD and autistic young people. Decreases in non-surgical planned care during transition could reflect disruptions to continuity of planned/respite care or a shift towards provision of healthcare in primary care and community settings and non-hospital arrangements for respite care. Funding: National Institute for Health Research Policy Research Programme.

Introduction Administrative data are a valuable research resource, but are under-utilised in the UK due to governance, technical and other barriers (e.g., the time and effort taken to gain secure data access). In recent years, there has been considerable government investment in making administrative data "research-ready", but there is no definition of what this term means. A common understanding of what constitutes research-ready administrative data is needed to establish clear principles and frameworks for their development and the realisation of their full research potential. Objective To define the characteristics of research-ready administrative data based on a systematic review and synthesis of existing literature. Methods On 29th June 2021, we systematically searched seven electronic databases for (1) peer-reviewed literature (2) related to research-ready administrative data (3) written in the English language. Following supplementary searches and snowball screening, we conducted a thematic analysis of the identified relevant literature. Results Overall, we screened 2,375 records and identified 38 relevant studies published between 2012 and 2021. Most related to administrative data from the UK and US and particularly to health data. The term research-ready was used inconsistently in the literature and there was some conflation with the concept of data being ready for statistical analysis. From the thematic analysis, we identified five defining characteristics of research-ready administrative data: (a) accessible, (b) broad, (c) curated, (d) documented and (e) enhanced for research purposes. Conclusions Our proposed characteristics of research-ready administrative data could act as a starting point to help data owners and researchers develop common principles and standards. In the more immediate term, the proposed characteristics are a useful framework for cataloguing existing research-ready administrative databases and relevant resources that can support their development.

Abstract Background Infant mortality rates are commonly used to compare the health of populations. Observed differences are often attributed to variation in child health care quality. However, any differences are at least partly explained by variation in the prevalence of risk factors at birth, such as low birth weight. This distinction is important for designing interventions to reduce infant mortality. We suggest a simple method for decomposing inter-country differences in crude infant mortality rates into two metrics representing risk factors operating before and after birth. Methods We used data from 7 European countries participating in the EURO-PERISTAT project in 2010. We calculated crude and birth weight-standardised stillbirth and infant mortality rates using Norway as the standard population. We decomposed between-country differences in crude stillbirth and infant mortality rates into the within-country difference in crude and birth weight-standardised stillbirth and infant mortality rates (metric 1), reflecting prenatal risk factors, and the between-country difference in birth weight-standardised stillbirth and infant mortality rates (metric 2), reflecting risk factors operating after birth. We also calculated birth weight-specific mortality. Results Using our metrics, we showed that for England, Wales and Scotland risk factors before and after birth contributed equally to the differences in crude stillbirth and infant mortality rates relative to Norway. In Austria, Czech Republic and Switzerland the differences were driven primarily by metric 1, reflecting high rate of low birth weight. The highest values of metric 2 observed in Poland partially reflected high rates of congenital anomalies. Conclusions Our suggested metrics can be used to guide policy decisions on preventing infant deaths through reducing risk factors at birth or improving the care of babies after birth. Aggregate data tabulated by birth weight/gestational age should be routinely collected and published in high-income countries where birth weight is reported on birth certificates.

BackgroundNational birth cohorts derived from administrative health databases constitute unique resources for child health research due to whole country coverage, ongoing follow-up and linkage to other data sources. In England, a national birth cohort can be developed using Hospital Episode Statistics (HES), an administrative database covering details of all publicly funded hospital activity, including 97% of births, with longitudinal follow-up via linkage to hospital and mortality records. We present methods for developing a national birth cohort using HES and assess the impact of changes to data collection over time on coverage and completeness of linked follow-up records for children.MethodsWe developed a national cohort of singleton live births in 1998-2015, with information on key risk factors at birth (birth weight, gestational age, maternal age, ethnicity, area-level deprivation). We identified three changes to data collection, which could affect linkage of births to follow-up records: (1) the introduction of the "NHS Numbers for Babies (NN4B)", an on-line system which enabled maternity staff to request a unique healthcare patient identifier (NHS number) immediately at birth rather than at civil registration, in Q4 2002; (2) the introduction of additional data quality checks at civil registration in Q3 2009; and (3) correcting a postcode extraction error for births by the data provider in Q2 2013. We evaluated the impact of these changes on trends in two outcomes in infancy: hospital readmissions after birth (using interrupted time series analyses) and mortality rates (compared to published national statistics).ResultsThe cohort covered 10,653,998 babies, accounting for 96% of singleton live births in England in 1998-2015. Overall, 2,077,929 infants (19.5%) had at least one hospital readmission after birth. Readmission rates declined by 0.2% percentage points per annual quarter in Q1 1998 to Q3 2002, shifted up by 6.1% percentage points (compared to the expected value based on the trend before Q4 2002) to 17.7% in Q4 2002 when NN4B was introduced, and increased by 0.1% percentage points per annual quarter thereafter. Infant mortality rates were under-reported by 16% for births in 1998-2002 and similar to published national mortality statistics for births in 2003-2015. The trends in infant readmission were not affected by changes to data collection practices in Q3 2009 and Q2 2013, but the proportion of unlinked mortality records in HES and in ONS further declined after 2009.DiscussionHES can be used to develop a national birth cohort for child health research with follow-up via linkage to hospital and mortality records for children born from 2003 onwards. Re-linking births before 2003 to their follow-up records would maximise potential benefits of this rich resource, enabling studies of outcomes in adolescents with over 20 years of follow-up.

Objectives Acute respiratory infections (ARIs) are a global cause of childhood morbidity. We compared temporal trends and socioeconomic disparities for ARI hospitalisations in young children across Western Australia, England and Scotland.Design Retrospective population-based cohort studies using linked birth, death and hospitalisation data.Setting and participants Population birth cohorts spanning 2000–2012 (Western Australia and Scotland) and 2003–2012 (England).Outcome measures ARI hospitalisations in infants (

Introduction Child mortality is nearly twice as high in England as in Sweden. A comparison of mortality from potentially preventable causes could inform health system responses. This study focused on respiratory tract infection (RTI)-related deaths, amenable to healthcare interventions, and sudden unexpected deaths in infancy (SUDI), amenable to public health interventions. Objectives and Approach We developed nationally-representative birth cohorts of singleton live births in 2003-2012 using a hospital admissions database in England and the Medical Birth Register in Sweden. Children were followed-up from 31st day of life until their fifth birthday via linkage to hospital admission and mortality records. We compared child mortality using Cox proportional hazards models to estimate hazard ratios (HR) for England versus Sweden for RTI-related mortality at 31-364 days and 1-4 years, and for SUDI mortality at 31-364 days. Models were adjusted for birth characteristics (gestational age, birthweight, sex, congenital anomalies), and socio-economic factors (maternal age and socio-economic status). Results Of 3,928,483 children in England, there were 807 RTI-related deaths at 31-364 days (17% of all deaths in the age range), 691 deaths at 1-4 years (31%), and 1,166 SUDIs (24%) in England. Corresponding figures for 1,012,682 children in Sweden were 136 (18%), 118 (25%) and 189 (24%). Unadjusted HRs for RTI-related deaths in England versus Sweden were 1.50 (95% confidence interval: 1.25-1.80) at 31-364 days. Adjustment for birth characteristics reduced the HR to 1.16 (0.97-1.39), and for socio-economic factors to 1.11 (0.92-1.33). Corresponding figures for RTI-related mortality at 1-4 years were 1.58 (1.30-1.92), 1.32 (1.09-1.61) and 1.30 (1.07-1.59), respectively. Unadjusted HRs for SUDIs reduced from 1.59 (1.36-1.85) to 1.40 (1.20-1.63) after adjusting for birth characteristics, and to 1.19 (1.02-1.39) after adjusting for socio-economic factors. Conclusion/Implications Higher prevalence of adverse birth characteristics (such as prematurity, low birthweight, congenital anomalies) contributed to increased risks of RTI-related and SUDI mortality in England relative to Sweden. Therefore, preventive strategies should focus on maternal health and socio-economic circumstances before and during pregnancy to reduce RTI-related and SUDI mortality in England.

ABSTRACT Objectives We aimed to compare prevalence rates of complex congenital anomalies (CCAs) in the first two years of life in England and Scotland using two published CCA diagnosis code lists (phenotypes) developed to identify children who require high levels of healthcare. Approach We developed birth cohorts of singleton, hospital live births between 1998-2011 using birth registration, hospitalisation and mortality register datasets from England (n=8,157,213) and from Scotland (n=731,377) from. Children were followed up until 2 years of age. We compared prevalence rates of congenital anomalies using two code lists developed to identify CCAs: a UK chronic condition code list developed by Hardelid et al. and a US medical complexity code list developed by Feudtner et al. Children were classified as having a CCA if a diagnosis code was included in any admission during follow-up as a primary or subsidiary diagnosis. To assess the level of healthcare attention required, we compared the proportion of children who had at least one re-admission to hospital in the first two years of life. Results Rates of CCAs in England and Scotland were higher using the Hardelid phenotype (2.8% for both countries) than when using the Feudtner phenotype (1.8% in England and 1.5% in Scotland). Rates of CCAs increased in both England and Scotland between 1998 and 2011. The increase was more notable using the Hardelid phenotype (37.2% and 20.2%, in England and Scotland, respectively) compared to the Feudtner phenotype (35.6% and 13.0%). Prevalence rates using the Hardelid phenotype were similar to EUROCAT figures. The proportion of children with at least one re-admission to hospital was higher in the Feudtner code list (69.2% in England and 84.8% in Scotland), than for Hardelid’s code list (67.9% and 72.0% in England and Scotland, respectively). The proportions of children with any of the CCA phenotypes were higher than compared to children without CCAs (24.4% in England and 25.8% in Scotland). Conclusion Prevalence of CCAs and associated hospital use vary depending on choice of CCA code list. Our findings suggest that Feudtner’s code list identifies children with higher level of healthcare needs. The impact of coding methods for defining CCAs in administrative hospital records should be explored and externally validated when comparing prevalence of CCAs and associated hospital use between countries. The impact of coding method for defining CCAs in administrative hospital records should be explored and externally validated when comparing prevalence and admission rates between countries.

ABSTRACT Objectives Higher rates of infant mortality in the UK than in the Nordic countries are partly explained by wider socio-economic disparities in the UK. We examined the extent to which low birth weight mediates the association between socioeconomic status (SES) and infant mortality using causal mediation analysis. We used cohorts of live births identified in administrative hospital data for the whole of Scotland and Denmark to explore the contribution of prenatal factors, represented by low birth weight, to differences in infant mortality between the two countries. Approach We included live-born children born in Denmark (n=1,432,205) and Scotland (n=1,427,163) from 1981-2004. Follow up was to 12 months of age. Information on deaths in first year of life was obtained through linkage with cause of death registers. We determined the effect of socioeconomic status on all cause infant mortality by comparing the highest and lowest quintiles of area-based deprivation (based on Carstairs score in Scotland) or level of maternal education in Denmark. Causal mediation analysis was used for survival outcomes with adjustment for maternal age at birth, sex, birth year of the child, and records indicating congenital malformation. Results During the follow-up, there were 8,158(0.57%) deaths in Denmark and 8,271(0.58%) deaths in Scotland. Comparing with the very high SES group, the overall hazard ratios of death for each SES quintile (starting with the lowest) compared with the highest SES quintile were 1.58(95% Confidence interval: 1.47-1.71), 1.40(1.32-1.49),1.25(1.20-1.30), 1.11(1.09-1.14) in Denmark, and 1.50(1.36-1.65),1.35(1.25-1.45),1.22(1.16-1.28),1.10(1.08-1.13) in Scotland. The proportions of excess infant deaths mediated through low birth weight (starting with the lowest) compared with the highest SES quintile were 54.7%, 52.1%, 49.5%, 46.9% in Denmark, and 26.0%, 23.9%, 22.0%, 20.1% in Scotland. Conclusion Our result suggests that SES has similar effects on infant mortality in Denmark and Scotland but more of the effect of SES on infant mortality is mediated through low birth weight in Denmark. Public health preventive strategies for infant mortality in both countries need to address prenatal risk factors for low birth weight. The substantial direct effects of SES on infant mortality seen in Scotland, which were not mediated through low birth weight, may be explained by other birth characteristics or could reflect persisting SES disparities in the care of infants after birth.