England Dataset Research Articles

A Genomics England haplotype reference panel and imputation of UK Biobank

We built a reference panel with 342 million autosomal variants using 78,195 individuals from the Genomics England (GEL) dataset, achieving a phasing switch error rate of 0.18% for European samples and imputation quality of r2 = 0.75 for variants with minor allele frequencies as low as 2 × 10−4 in white British samples. The GEL-imputed UK Biobank genome-wide association analysis identified 70% of associations found by direct exome sequencing (P < 2.18 × 10−11), while extending testing of rare variants to the entire genome. Coding variants dominated the rare-variant genome-wide association results, implying less disruptive effects of rare non-coding variants.

Embedding an engagement exercise into an administrative data research project: An example focussing on Gypsy, Traveller and Roma young people and their families

ObjectivesThe presentation reports findings from the ADR funded project ‘Testing and demonstrating the value of the Growing Up in England dataset: Gypsy, Traveller and Roma case study’. It will (1) address the rationale for embedding an engagement exercise in this project; (2) set out methodology and findings; and (3) draw broader implications for administrative data research.&#x0D; MethodsTwo engagement workshops were co-organised with the civil society organisation Friends, Families and Travellers (FFT). Recruitment was via FFT outreach networks and young people aged 14-29 and their broader family groups were invited to attend. Participants included Romany Gypsies, Traveller and Roma young people and family groups, ensuring diversity of experiences and perspectives. Informed consent was obtained from all participants (and from parents/carers for under 18s). The purpose of the engagement exercise was to hear directly from Gypsy, Traveller and Roma young people about issues raised by our analysis of the newly linked Growing Up in England (GUiE) dataset.&#x0D; ResultsFeedback from the workshops explores the educational experiences of Gypsy, Traveller and Roma young people and their families and provides important insights into barriers to educational progression and unmet needs for support. Additionally, feedback from the workshops provides new insights into the views, perspectives and experiences of Gypsy, Traveller and Roma young people and their families relating to administrative data collection and data use. One key concern highlighted by workshop participants is that the disclosure of Gypsy, Traveller and Roma identity in administrative data collected by schools, hospitals and other public services could result in discrimination or other forms of adverse or hostile treatment.&#x0D; ConclusionsThe findings from the engagement exercise have important implications for our GUiE project and for broader reflection on public attitudes to administrative data. Additionally, the exercise enabled us to address NSDEC recommendations on engagement. However, engaging with Gypsy, Traveller and Roma young people has involved specialist input, time, funding and additional ethics review, which all require careful consideration and planning.

Centre assessment grades in 2020: a natural experiment for investigating bias in teacher judgements

The COVID-19 pandemic meant that, in 2020, students in England were unable to sit their examinations and instead received predicted grades, or “centre assessment grades” (CAGs), from their teachers to allow them to progress. Using the Grading and Admissions Data for England (GRADE) dataset for students from 2018 to 2020, this study treats the use of CAGs as a natural experiment for causally understanding how teacher judgements of academic ability may be biased according to the demographic and socio-economic characteristics of their students. A variety of machine learning models were trained on the 2018–19 data and then used to generate predictions for what the 2020 students were likely to have received had their examinations taken place as usual. The differences between these predictions and the CAGs that students received were calculated and then averaged across students’ different characteristics, revealing what the treatment effects of the use of CAGs were likely to have been for different types of students. No evidence of absolute negative bias against students of any demographic or socio-economic characteristic was found, with all groups of students having received higher CAGs than the grades they were likely to have received had they sat their examinations. Some evidence for relative bias was found, with consistent, but insubstantial differences being observed in the treatment effects of certain groups. However, when higher-order interactions of student characteristics were considered, these differences became more substantial. Intersectional perspectives which emphasise interactions and sub-group differences should be used more widely within quantitative educational equalities research.

Swarm-based hybrid optimization algorithms: an exhaustive analysis and its applications to electricity load and price forecasting.

In this work, we intend to propose multiple hybrid algorithms with the idea of giving a choice to the particles of a swarm to update their position for the next generation. To implement this concept, Cuckoo Search Algorithm (CSA), Grey Wolf Optimization (GWO), Harris Hawks Optimization (HHO), and Whale Optimization Algorithm (WOA) have been utilized. Exhaustive possible combinations of these algorithms are developed and benchmarked against the base algorithms. These hybrid algorithms have been validated on twenty-four well-known unimodal and multimodal benchmarks functions, and detailed analysis with varying dimensions and population size is discussed for the same. Further, the efficacy of these algorithms has been tested on short-term electricity load and price forecasting applications. For this purpose, the algorithms have been combined with Artificial Neural Networks (ANNs) to evaluate their performance on the ISO New Pool England dataset. The results demonstrate that hybrid optimization algorithms perform superior to their base algorithms in most test cases. Furthermore, the results show that the performance of CSA-GWO is significantly better than other algorithms.

Exploring the potential of the newly linked Growing up in England dataset for building up new evidence for Gypsy, Traveller and Roma young people.

ObjectivesThis paper aims to test and demonstrate the potential of the newly linked Growing up in England (GUIE) dataset for building up new evidence, knowledge and understanding of education and socio-economic circumstance of an often ‘invisible’ group of children and young people - Gypsy, Roma and Travellers.&#x0D; ApproachThe paper utilises the first wave of the newly linked Growing up in England dataset, which consists of administrative education data for children and young people in schools and further education institutions in England linked to their 2011 Census records. It focuses on qualifications studied as well as attainment achieved of a cohorts of Gypsy, Traveller and Roma young people between age 15 (around the time of 2011 Census) and 19. The analysis explores the relationship between educational pathways and attainment for these young people and their household/family characteristics recorded in the 2011 Census.&#x0D; ResultsThis exploratory study shows that Gypsy, Traveller and Roma young people are more likely to be disadvantaged across multiple dimensions such as health and housing, and are less likely to achieve level 1, 2 and 3 educational qualifications compared to their peers.&#x0D; ConclusionsEvidence on outcomes for Gypsy, Traveller and Roma children and young people is scarce, in part due to limited available data on them. The newly linked GUIE dataset is therefore a rare opportunity to improve the evidence base for education and socio-economic outcomes for this group.

Genetic predispositions to psychiatric disorders and the risk of COVID-19

BackgroundWhether a genetic predisposition to psychiatric disorders is associated with coronavirus disease 2019 (COVID-19) is unknown.MethodsOur analytic sample consisted of 287,123 white British participants in UK Biobank who were alive on 31 January 2020. We performed a genome-wide association study (GWAS) analysis for each psychiatric disorder (substance misuse, depression, anxiety, psychotic disorder, and stress-related disorders) in a randomly selected half of the study population (“base dataset”). For the other half (“target dataset”), the polygenic risk score (PRS) was calculated as a proxy of individuals’ genetic predisposition to a given psychiatric phenotype using discovered genetic variants from the base dataset. Ascertainment of COVID-19 was based on the Public Health England dataset, inpatient hospital data, or death registers in UK Biobank. COVID-19 cases from hospitalization records or death records were considered “severe cases.” The association between the PRS for psychiatric disorders and COVID-19 risk was examined using logistic regression. We also repeated PRS analyses based on publicly available GWAS summary statistics.ResultsA total of 143,562 participants (including 10,868 COVID-19 cases) were used for PRS analyses. A higher genetic predisposition to psychiatric disorders was associated with an increased risk of any COVID-19 and severe COVID-19. The adjusted odds ratio (OR) for any COVID-19 was 1.07 (95% confidence interval [CI] 1.02–1.13) and 1.06 (95% CI 1.01–1.11) among individuals with a high genetic risk (above the upper tertile of the PRS) for substance misuse and depression, respectively, compared with individuals with a low genetic risk (below the lower tertile). Slightly higher ORs were noted for severe COVID-19, and similar result patterns were obtained in analyses based on publicly available GWAS summary statistics.ConclusionsOur findings suggest a potential role of genetic factors in the observed phenotypic association between psychiatric disorders and COVID-19. Our data underscore the need for increased medical surveillance for this vulnerable population during the COVID-19 pandemic.

Accessory Genome Dynamics of Local and Global Staphylococcus pseudintermedius Populations.

Staphylococcus pseudintermedius is a major bacterial colonizer and opportunistic pathogen in dogs. Methicillin-resistant S. pseudintermedius (MRSP) continues to emerge as a significant challenge to maintaining canine health. We sought to determine the phylogenetic relationships of S. pseudintermedius across five states in the New England region of the United States and place them in a global context. The New England dataset consisted of 125 previously published S. pseudintermedius genomes supplemented with 45 newly sequenced isolates. The core genome phylogenetic tree revealed many deep branching lineages consisting of 142 multi-locus sequence types (STs). In silico detection of the mecA gene revealed 40 MRSP and 130 methicillin-susceptible S. pseudintermedius (MSSP) isolates. MRSP were derived from five structural types of SCCmec, the mobile genetic element that carries the mecA gene conferring methicillin resistance. Although many genomes were MSSP, they nevertheless harbored genes conferring resistance to many other antibiotic classes, including aminoglycosides, macrolides, tetracyclines and penams. We compared the New England genomes to 297 previously published genomes sampled from five other states in the United States and 13 other countries. Despite the prevalence of the clonally expanding ST71 found worldwide and in other parts of the United States, we did not detect it in New England. We next sought to interrogate the combined New England and global datasets for the presence of coincident gene pairs linked to antibiotic resistance. Analysis revealed a large co-circulating accessory gene cluster, which included mecA as well as eight other resistance genes [aac (6′)-Ie-aph (2″)-Ia, aad (6), aph (3′)-IIIa, sat4, ermB, cat, blaZ, and tetM]. Furthermore, MRSP isolates carried significantly more accessory genes than their MSSP counterparts. Our results provide important insights to the evolution and geographic spread of high-risk clones that can threaten the health of our canine companions.

Association between health-related quality of life and menopausal status and symptoms in women living with HIV aged 45–60 years in England: An analysis of the PRIME study

Objectives:The aim of this study was to compare the health-related quality of life between mid-life women with HIV and the general population and to investigate the association between health-related quality of life and menopausal (1) status and (2) symptoms among women with HIV.Methods:Cross-sectional data of women with HIV aged 45–60 years from the Positive Transitions Through the Menopause Study. Health-related quality of life was assessed using the Euroqol questionnaire with utility scores categorizing health as perfect (score = 1.00), sub-optimal (0.75–0.99) or poor (< 0.75). Scores were compared between Positive Transitions Through the Menopause study participants and women (aged 45–59 years) from the Health Survey for England. Associations between health-related quality of life and menopausal status/symptoms in Positive Transitions Through the Menopause participants were assessed using a multivariable two-part regression model, the results of which are combined to produce a single marginal effect.Results:In total, 813 women from the Positive Transitions Through the Menopause study were included (median age 49 (interquartile range: 47–53) years); the majority were of Black African ethnicity (72.2%). Overall, 20.9%, 43.7% and 35.3% of women were pre-, peri- and post-menopausal, respectively, and 69.7% experienced mild/moderate/severe menopausal symptoms. Approximately, 40% reported perfect health, 22.1% sub-optimal health and 39.0% poor health, similar to women from the Health Survey for England (perfect health: 36.9%, sub-optimal health: 25.2%, poor health: 37.9%). In multivariable models, we found an association between health-related quality of life and peri-menopausal status (marginal effect: 0.07 (0.02, 0.12)); however, the association with post-menopausal status was attenuated (marginal effect: 0.01 (–0.05, 0.06)). There remained a strong association between lower utility scores and moderate (marginal effect: 0.16 (0.11, 0.20)) and severe (marginal effect: 0.32 (0.27, 0.39)) menopausal symptoms.Conclusion:There were no differences in health-related quality of life between women with HIV (Positive Transitions Through the Menopause participants) and women from the Health Survey for England dataset. Among Positive Transitions Through the Menopause participants, health-related quality of life was reduced in peri-menopausal women and those with increasingly severe menopausal symptoms. Our findings highlight the importance of proactive assessment of menopausal status and symptoms to optimize health-related quality of life in women living with HIV as they reach mid-life and beyond.

POS-421 KIDNEYNETWORK: A NEW METHOD TO PREDICT KIDNEY DISEASE GENES USING KIDNEY DERIVED GENE EXPRESSION DATA IDENTIFIES A NEW CANDIDATE GENE FOR MILD ADPKD / PCLD

Genetic testing of patients with suspected hereditary kidney disease using exome-based panel sequencing can reveal pathogenic variants in kidney-related genes. However, we are still faced with unsolved cases. Potentially harmful variants can reside in other genes that are either not annotated for kidney disease or in genes of unknown function. This makes it difficult to prioritize and interpret the relevance of variants in these genes for kidney diseases. As such, there is a clear need for methods that predict the phenotypic consequences of gene expression in a way that is as unbiased as possible. We have recently done this using public RNA sequencing (RNA-seq) data and shown that our method substantially improves the diagnostic yield of clinical exome sequencing in rare disease (Deelen et al., 2019). We have now extended upon this work by developing a kidney-specific gene network -KidneyNetwork- similar to www.genenetwork.nl using a combined approach of non-specific human RNA-seq data and human kidney-derived RNA-seq data. We calculated gene co-expression in 878 publicly available kidney RNA-seq samples and combined this with the calculated co-expression in the general human RNA-seq dataset. These expression patterns were used to predict genes involved in kidney related phenotypes as established in the HPO database. We used ‘leave-one-out’ cross validations to determine the prediction accuracy (AUC) of our predictions. The KidneyNetwork was applied to prioritize variants in 13 undiagnosed kidney patients for whom WES data were available. In KidneyNetwork, we observe a significantly improved prediction accuracy of kidney-related pathways as compared to GeneNetwork. Moreover, the total number of significantly predicted pathways has increased (Figure 1). For each HPO term, all genes are assigned a z-score representing their potential involvement in the respective pathway. Genes with a known association to that pathway are expected to have high scores and genes unknown for their involvement with high scores are potential new candidate disease genes. Based on the HPO terms “Polycystic kidney dysplasia” and “Hepatic cysts”, combined with information on potentially harmful variants in one of the undiagnosed patients with mild ADPKD/PCLD, we propose “gene X” (proper name will be revealed at the conference) as a new candidate gene. Gene X bears resemblance to other genes implicated in this phenotype. Two patients included in the Genomics England dataset with a similar phenotype have suspected deleterious variants in gene X. The improvement of KidneyNetwork relative to www.genenetwork.nl, has enabled us to more accurately predict candidate genes involved in renal disease. By analysing 13 undiagnosed patients, for which we found gene X (name will be revealed at the conference) to be a highly promising candidate gene, we have shown the added benefit of KidneyNetwork for the research community and the clinic.

Energy-Net: A Deep Learning Approach for Smart Energy Management in IoT-Based Smart Cities

Although intelligent load forecasting is essential for optimal energy management (EM) in smart cities, there is a lack of current research exploring EM in well-regulated Internet-of-Things (IoT) networks. This article develops a new deep learning (DL) model for efficient forecasting of short-term energy consumption while maintaining effective communication between energy providers and users. The proposed Energy-Net stack comprises multiple stacked spatiotemporal modules, where each module consists of a temporal transformer (TT) submodule and a spatial transformer (ST) submodule. The TT models the temporal relationships in load data; and the ST submodule extracts hidden spatial information by integrating convolutional layers and includes an improved self-attention mechanism. The experimental evaluation on IHPEC and independent system operator New England (ISO-NE) data set demonstrates the superiority of Energy-Net over recent cutting-edge DL models with root mean-square error (RMSE) of 0.354 and 0.535, respectively. The computational complexity of Energy-Net is appropriate for dependable resource-constrained IoT devices (i.e., fog nodes or edge nodes) linked to a joint IoT-cloud server that interacts with connected smart grids to handle EM tasks.

Pre-pandemic psychiatric disorders and risk of COVID-19: a UK Biobank cohort analysis

BackgroundPsychiatric morbidities have been associated with a risk of severe infections through compromised immunity, health behaviours, or both. However, data are scarce on the association between multiple types of pre-pandemic psychiatric disorders and COVID-19. We aimed to assess the association between pre-pandemic psychiatric disorders and the subsequent risk of COVID-19 using UK Biobank.MethodsFor this cohort analysis, we included participants from UK Biobank who were registered in England and excluded individuals who died before Jan 31, 2020, (the start of the COVID-19 outbreak in the UK) or had withdrawn from UK Biobank. Participants diagnosed with a psychiatric disorder before Jan 31 were included in the group of individuals with pre-pandemic psychiatric disorders, whereas participants without a diagnosis before the outbreak were included in the group of individuals without pre-pandemic psychiatric disorders. We used the Public Health England dataset, UK Biobank hospital data, and death registers to collect data on COVID-19 cases. To examine the relationship between pre-pandemic psychiatric disorders and susceptibility to COVID-19, we used logistic regression models to estimate odds ratios (ORs), controlling for multiple confounders and somatic comorbidities. Key outcomes were all COVID-19, COVID-19 specifically diagnosed in inpatient care, and COVID-19-related deaths. ORs were also estimated separately for each psychiatric disorder and on the basis of the number of pre-pandemic psychiatric disorders. As a positive disease control, we repeated analyses for hospitalisation for other infections.FindingsWe included 421 014 UK Biobank participants in our study and assessed their COVID-19 status between Jan 31 and July 26, 2020. 50 809 participants were diagnosed with psychiatric disorders before the outbreak, while 370 205 participants had no psychiatric disorders. The mean age at outbreak was 67·80 years (SD 8·12). We observed an elevated risk of COVID-19 among individuals with pre-pandemic psychiatric disorders compared with that of individuals without such conditions. The fully adjusted ORs were 1·44 (95% CI 1·28–1·62) for All COVID-19 cases, 1·55 (1·34–1·78) for Inpatient COVID-19 cases, and 2·03 (1·59–2·59) for COVID-19-related deaths. We observed excess risk, defined as risk that increased with the number of pre-pandemic psychiatric disorders, across all diagnostic categories of pre-pandemic psychiatric disorders. We also observed an association between psychiatric disorders and elevated risk of hospitalisation due to other infections (OR 1·74, 95% CI 1·58–1·93).InterpretationOur findings suggest that pre-existing psychiatric disorders are associated with an increased risk of COVID-19. These findings underscore the need for surveillance of and care for populations with pre-existing psychiatric disorders during the COVID-19 pandemic.FundingNational Natural Science Foundation of China.

PLS-CNN-BiLSTM: An End-to-End Algorithm-Based Savitzky–Golay Smoothing and Evolution Strategy for Load Forecasting

This paper proposes an effective deep learning framework for Short-Term Load Forecasting (STLF) of multivariate time series. The proposed model consists of a hybrid Convolutional neural network-Bidirectional Long Short-Term Memory (CBiLSTM) based on the Evolution Strategy (ES) method and the Savitzky–Golay (SG) filter (SG-CBiLSTM). The adopted methodology incorporates the virtue of different prepossessing blocks to enhance the performance of the CBiLSTM model. In particular, a data-augmentation strategy is employed to synthetically improve the feature representation of the CBiLSTM model. The augmented data is forwarded to the Partial Least Square (PLS) method to select the most informative features above the predefined threshold. Next, the SG algorithm is computed for smoothing the load to enhance the learning capabilities of the underlying system. The structure of the SG-CBiLSTM for the ISO New England dataset is optimized using the ES technique. Finally, the CBiLSTM model generates output forecasts. The proposed approach demonstrates a remarkable improvement in the performance of the original CBiLSTM model. Furthermore, the experimental results strongly confirm the high effectiveness of the proposed SG-CBiLSTM model compared to the state-of-the-art techniques.

Comparison of Early and Late Post-operative Outcomes after Supra-inguinal Bypass for Aortoiliac Occlusive Disease

The choice for surgical revascularisation for aortoiliac occlusive disease is often tempered by patient comorbidities. This study compares peri-operative outcomes and the association between choice of operation and one year major adverse limb event (MALE) free survival and five year mortality. The Vascular Study Group of New England (VSGNE) dataset for supra-inguinal bypass operations from 2009 to 2015 was queried. This study excluded cases with bypass other than aortofemoral (AFB), axillofemoral (AXB), and femorofemoral (FFB), and those with endovascular interventions or femoral endarterectomy. Cases combined with other procedures, indications other than occlusive disease, and missing pathology were also excluded. Patients were divided into three groups: AFB, AXB, and FFB. Thirty day post-operative death (POD) and adverse events were compared using univariable and multivariable analyses. One year MALE free survival was compared between groups with log rank test and Kaplan-Meier plot. Proportional hazard Cox regression was used for adjusted comparison of MALE free and five year survival. In total, 1,602 cases were included: 207 (12.9%) AXB; 872 (54.4%) AFB; 523 (32.6%) FFB. AXB patients were older with more comorbidities. Post-operative complications and POD rates were significantly higher for AXB (p<.05). On adjusted analyses, AXB increased the hazard of one year MALE (hazard ratio [HR] 1.76, 95% confidence interval [CI] 1.12-2.78; p=.014) and five year mortality (HR 1.54; 95% CI 1.11-2.41; p=.009). Both FFB and AFB had similar one year MALE free survival but significantly better one year MALE free survival than AXB. After adjusting for confounding variables, and while acknowledging limitations related to the VSGNE data set, FFB led to significantly lower rates of post-operative complications than AXB. FFB may serve as the extra-anatomical operation of choice in high risk patients with extensive disease, who cannot undergo AFB, provided that anatomy permits. AFB should be performed preferentially in low risk patients with appropriate anatomy. Owing to its higher complications rates, the study suggests that AXB should be limited to patients with no other option for revascularisation.

An Examination of Who Is Eligible and Who Is Receiving Bariatric Surgery in England: Secondary Analysis of the Health Survey for England Dataset

BackgroundOver 2 million people in England were estimated to be eligible for bariatric surgery in 2006. In 2014, clinical guidelines were updated, widening potential eligibility, meanwhile, obesity prevalence rose. However, numbers receiving surgery decreased, and concerns exist of inequalities in access between population groups. This study is aimed at estimating the number of adults eligible for surgery in England and to compare demographics with those that receive surgery.MethodsBMI and comorbidity status were used to determine eligibility for bariatric surgery within participants of the 2014 Health Survey for England dataset (6938 adults), based on the National Institute of Health and Care Excellence guidelines. Results were scaled up using national population estimates. The demographics of eligible participants were compared against 2014/2015 hospital episode statistics for sex and age group using a chi-squared analysis.ResultsOf the total population of England, 7.78% (95% CI 7.1–8.6%), or 3,623,505 people, could have been eligible for bariatric surgery in 2014; nearly a million more than if previous guidelines applied. Eligibility peaked at ages 45–54, with most in the 35–64 age group (58.9%). 58.4% of those eligible were women. Patients receiving surgery were far more likely to be female than male (76.1%) and the distribution skewed towards younger ages when compared with those eligible.ConclusionBariatric surgery may benefit many people in England; significant investment is required so that service provision is adequate for demand. Differences between demographics of those eligible and receiving surgery may be explainable; however, the potential health inequality should be investigated.

Capital and Liquidity Interaction in Banking

We study how banks’ capital level affects the extent to which they engage in liquidity transformation. We first construct a simple model to develop testable hypotheses on this link. Then we test our predictions and establish the causality using a confidential Bank of England dataset that includes arguably exogenous changes in banks’ capital requirement add-ons. We find that banks engage in less liquidity transformation when their capital increases, which suggests that capital and liquidity requirements are at least to some extent substitutes. We also find that this substitution is mostly driven by small banks. These results have interesting implications for the optimal joint calibration of capital and liquidity requirements and for the proportionality of prudential regulations.

Girls' video gaming behaviour and undergraduate degree selection: A secondary data analysis approach

Girls' uptake of physical science, technology, engineering and mathematics (PSTEM) degrees continues to be poor. Identifying and targeting interventions for girl groups that are likely to go into STEM degrees may be a possible solution. This paper, using a self-determination theory and self-socialisation framework, determines whether one girl group's, “geek girls”, video gaming behaviour is associated with their choice of undergraduate degree by using two secondary datasets: a cross-sectional study of the Net Generation (n = 814) and the Longitudinal Study of Young People in England (LSYPE) dataset (n = 7342). Chi-square analysis shows that girls who were currently studying a PSTEM degree were more likely to be gamers and engage in multiplayer gamers. Further, using logistic regressions, girls who were heavy gamers (>9 h/wk) at 13–14 years were found to be more likely to pursue a PSTEM degree but this was influenced by their socio-economic status. Similar associations with boys and PSTEM degrees were not found or were weak. Therefore, girls were self-socialising or self-determining their identity groups through gaming. This research can provide the basis for whether encouraging gaming in adolescent girls can help them onto PSTEM pathways.

Developing and Applying Geographical Synthetic Estimates of Health Literacy in GP Clinical Systems.

Background: Low health literacy is associated with poorer health. Research has shown that predictive models of health literacy can be developed; however, key variables may be missing from systems where predictive models might be applied, such as health service data. This paper describes an approach to developing predictive health literacy models using variables common to both “source” health literacy data and “target” systems such as health services. Methods: A multilevel synthetic estimation was undertaken on a national (England) dataset containing health literacy, socio-demographic data and geographical (Lower Super Output Area: LSOA) indicators. Predictive models, using variables commonly present in health service data, were produced. An algorithm was written to pilot the calculations in a Family Physician Clinical System in one inner-city area. The minimum data required were age, sex and ethnicity; other missing data were imputed using model values. Results: There are 32,845 LSOAs in England, with a population aged 16 to 65 years of 34,329,091. The mean proportion of the national population below the health literacy threshold in LSOAs was 61.87% (SD 12.26). The algorithm was run on the 275,706 adult working-age people in Lambeth, South London. The algorithm could be calculated for 228,610 people (82.92%). When compared with people for whom there were sufficient data to calculate the risk score, people with insufficient data were more likely to be older, male, and living in a deprived area, although the strength of these associations was weak. Conclusions: Logistic regression using key socio-demographic data and area of residence can produce predictive models to calculate individual- and area-level risk of low health literacy, but requires high levels of ethnicity recording. While the models produced will be specific to the settings in which they are developed, it is likely that the method can be applied wherever relevant health literacy data are available. Further work is required to assess the feasibility, accuracy and acceptability of the method. If feasible, accurate and acceptable, this method could identify people requiring additional resources and support in areas such as medical practice.

Proceedings of Patient Reported Outcome Measure’s (PROMs) Conference Sheffield 2016: advances in patient reported outcomes research

Table of contentsS1 Using computerized adaptive testingTim CroudaceS2 Well-being: what is it, how does it compare to health and what are the implications of using it to inform health policyJohn BrazierO1 “Am I going to get better?”—Using PROMs to inform patients about the likely benefit of surgeryNils Gutacker, Andrew StreetO2 Identifying Patient Reported Outcome Measures for an electronic Personal Health RecordDan Robotham, Samantha Waterman, Diana Rose, Safarina Satkunanathan, Til WykesO3 Examining the change process over time qualitatively: transformative learning and response shiftNasrin Nasr, Pamela EnderbyO4 Developing a PROM to evaluate self-management in diabetes (HASMID): giving patients a voiceJill Carlton, Donna Rowen, Jackie Elliott, John Brazier, Katherine Stevens, Hasan Basarir, Alex LabeitO5 Development of the Primary Care Outcomes Questionnaire (PCOQ)Mairead Murphy, Sandra Hollinghurst, Chris SalisburyO6 Developing the PKEX score- a multimodal assessment tool for patients with shoulder problemsDominic Marley, James Wilson, Amy Barrat, Bibhas RoyO7 Applying multiple imputation to multi-item patient reported outcome measures: advantages and disadvantages of imputing at the item, sub-scale or score levelInes Rombach, Órlaith Burke, Crispin Jenkinson, Alastair Gray, Oliver Rivero-AriasO8 Integrating Patient Reported Outcome Measures (PROMs) into routine primary care for patients with multimorbidity: a feasibility studyIan Porter, Jaheeda Gangannagaripalli, Charlotte Bramwell, Jose M. ValderasO9 eRAPID: electronic self-report and management of adverse-events for pelvic radiotherapy (RT) patientsPatricia Holch, Susan Davidson, Jacki Routledge, Ann Henry, Kevin Franks, Alex Gilbert, Kate Absolom & Galina VelikovaO10 Patient reported outcomes (PROMs) based recommendation in clinical guidance for the management of chronic conditions in the United KingdomIan Porter, Jose M.ValderasO11 Cross-sectional and longitudinal parameter shifts in epidemiological data: measurement invariance and response shifts in cohort and survey data describing the UK’s Quality of LifeJan R. BoehnkeO12 Patient-reported outcomes within health technology decision making: current status and implications for future policyAndrew Trigg, Ruth HowellsO13 Can social care needs and well-being be explained by the EQ-5D? Analysis of Health Survey for England datasetJeshika Singh, Subhash Pokhrel, Louise LongworthO14 Where patients and policy meet: exploring individual-level use of the Long-Term Conditions Questionnaire (LTCQ)Caroline Potter, Cheryl Hunter, Laura Kelly, Elizabeth Gibbons, Julian Forder, Angela Coulter, Ray Fitzpatrick, Michele Peters

Group‐based chaos genetic algorithm and non‐linear ensemble of neural networks for short‐term load forecasting

This study presents a non-linear ensemble of partially connected neural networks for short-term load forecasting. Partially connected neural networks are chosen as individual predictors due to their good generalisation capability. A group-based chaos genetic algorithm is developed to generate diverse and effective neural networks. A novel pruning method is employed to develop partially connected neural networks. To further enhance prediction accuracy, an artificial neural network-based non-linear ensemble of partially connected neural network predictors is developed. The proposed non-linear ensemble neural network is evaluated on a PJM market dataset and an ISO New England dataset with promising results of 1.76 and 1.29% error, respectively, demonstrating its capability as a promising predictor.

Associations of mutually exclusive categories of physical activity and sedentary time with markers of cardiometabolic health in English adults: a cross-sectional analysis of the Health Survey for England.

BackgroundBoth physical activity and sedentary behaviour have been individually associated with health, however, the extent to which the combination of these behaviours influence health is less well-known. The aim of this study was to examine the associations of four mutually exclusive categories of objectively measured physical activity and sedentary time on markers of cardiometabolic health in a nationally representative sample of English adults.MethodsUsing the 2008 Health Survey for England dataset, 2131 participants aged ≥18 years, who provided valid accelerometry data, were included for analysis and grouped into one of four behavioural categories: (1) ‘Busy Bees’: physically active & low sedentary, (2) ‘Sedentary Exercisers’: physically active & high sedentary, (3) ‘Light Movers’: physically inactive & low sedentary, and (4) ‘Couch Potatoes’: physically inactive & high sedentary. ‘Physically active’ was defined as accumulating at least 150 min of moderate-to-vigorous physical activity (MVPA) per week. ‘Low sedentary’ was defined as residing in the lowest quartile of the ratio between the average sedentary time and the average light-intensity physical activity time. Weighted multiple linear regression models, adjusting for measured confounders, investigated the differences in markers of health across the derived behavioural categories. The associations between continuous measures of physical activity and sedentary levels with markers of health were also explored, as well as a number of sensitivity analyses.ResultsIn comparison to ‘Couch Potatoes’, ‘Busy Bees’ [body mass index: −1.67 kg/m2 (p < 0.001); waist circumference: −1.17 cm (p = 0.007); glycated haemoglobin: −0.12 % (p = 0.003); HDL-cholesterol: 0.09 mmol/L (p = 0.001)], ‘Sedentary Exercisers’ [body mass index: −1.64 kg/m2 (p < 0.001); glycated haemoglobin: −0.11 % (p = 0.009); HDL-cholesterol: 0.07 mmol/L (p < 0.001)] and ‘Light Movers’ [HDL-cholesterol: 0.11 mmol/L (p = 0.004)] had more favourable health markers. The continuous analyses showed consistency with the categorical analyses and the sensitivity analyses indicated robustness and stability.ConclusionsIn this national sample of English adults, being physically active was associated with a better health profile, even in those with concomitant high sedentary time. Low sedentary time independent of physical activity had a positive association with HDL-cholesterol.Electronic supplementary materialThe online version of this article (doi:10.1186/s12889-016-2694-9) contains supplementary material, which is available to authorized users.