Huw Strafford, Beata Fonferko-Shadrach, W. Owen Pickrell
et al.
Objective and Approach
Managers, clinicians and policymakers have limited information on near-real time patterns of attendance at emergency departments (ED). Our aim was to develop a surveillance system through extracting information from the presenting complaint to develop and feed activity dashboards, improve injury surveillance and support research.
We developed a Natural Language Processing (NLP) pipeline using the General Architecture for Text Engineering (GATE) system. The pipeline produced Unified Medical Language System (UMLS) codes from ED presenting complaint records and also implemented the Joint Action on Monitoring Injuries in Europe (JAMIE) Minimum Dataset. We classified complaints into 31 common problem presentations (Coughs, Bleeds, Mental Health, Dizziness...), along with categories from JAMIE (Mechanism of Injury, Location, Activity, Intent…).
Results
Preliminary results are from a pilot study of 86,416 presenting complaints from Morriston Hospital in 2014.
The most common terms used for grouped symptoms/categories were “Injury” (38.6%, n=28691), “Pain” (22.8%, n=19420), “Unwell” (7.2%, n=6117), “Bleeding” (6.7%, n=5760), and “Shortness of Breath” (3.8%, n=3255).
32.3% of records contained Mechanism of Injury codes, including, falls (15.2%, n=13174), struck by/against (6.9%, n=5960), cuts (5.6%, n=4833), road traffic collisions (2.5%, n=2124) and poisoning (1%, n=900). We are continuing to refine and validate the categories using precision and recall and with comparison with coded diagnoses.
Conclusions and Implications
NLP can be used to extract valuable additional detailed information from ED data. We will further develop the algorithms for all hospitals in Wales for data from 2023 and design dashboards to produce a near-real time surveillance system.
Abstract Previous research has demonstrated a positive association between gender equality in the family and fertility measures. However, this association may vary when different features of gender equality in the family, such as the division of housework and childcare, are analyzed. Furthermore, it has been argued that individuals may have different perceptions regarding housework and childcare, which can further explain how the division of these activities is associated with fertility outcomes. We performed a microlevel investigation of the association between the division of housework and childcare and fertility intentions with the aim of identifying patterns (or their absence) in the association. We hypothesized that a more equitable division of tasks is associated with increased fertility intentions. We used data from the first round of the Generations and Gender Survey (GGS) in ten countries: Germany, Belgium, Austria, France, Lithuania, Poland, Romania, Russia, Bulgaria, and Georgia. Our samples included individuals who were in heterosexual relationships and had one child. We employed binary logistic regression for analysis. Overall, our findings suggested that a more equitable division of housework was associated with an increased likelihood of intending to have a second child. Conversely, a less equitable division of childcare was associated with a decreased likelihood of intending to have a second child. Owing to the discussion of the diverse perceptions that individuals may hold regarding housework and childcare, this study contributes to the literature by proposing a theoretical explanation for the patterns observed in our results. The results align with those of previous studies indicating that childcare is generally viewed as enjoyable and rewarding, whereas housework tends to be negatively evaluated and linked to perceptions of unfairness and inequality. Consequently, a more equitable division of housework may reduce work‒family conflict and potentially be associated with increased fertility intentions. In contrast, if childcare is perceived as fulfilling, engaging in more childcare activities may lead to increased fairness rather than unfairness, leading to the association between less equal division of childcare and increased fertility intentions.
<b>Background</b>: Intergenerational social mobility, or the inheritance of status characteristics, is well-studied in Sweden. However, it accounts for just one aspect of the process of intergenerational reproduction of social inequality. The role of socially stratified fertility in this process remains underexplored. <b>Objective</b>: I address the gap in knowledge by replicating the approach pioneered by Skopek and Leopold (2020) in the context of Germany in order to study the relative contributions of the mobility component vis-à-vis the fertility component in the educational reproduction of Swedish cohorts born between 1930 and 1950. <b>Methods</b>: The approach involves estimating several components of a stylized population renewal model using retrospective data and performing counterfactual simulations. I utilize data from the Swedish samples of the Generations and Gender Survey, the European Social Survey, and the Survey of Health, Ageing, and Retirement in Europe. <b>Results</b>: My findings for Sweden reveal a relatively strong degree of intergenerational transmission of educational attainment, increasing for men and decreasing for women, coupled with an overall weak but stable educational gradient in fertility. Educational reproduction in Sweden is thus mostly driven by the mobility component. <b>Contribution</b>: This is the first study to obtain prospective estimates of educational reproduction based on retrospective data and to explore the relative role of mobility vis-à-vis fertility in the process of intergenerational reproduction of social inequality in Sweden.
Objectives
To develop a flexible platform for creating, reviewing and adjudicating annotation of unstructured text. Natural Language Processing models and statistical classifiers use the results for analysis of large databases of text, such as electronic health records, that are curated by the National Centre for Healthy Ageing (NCHA) Data Platform.
Approach
Automated approaches are essential for large scale extraction of structured data from unstructured documents. We applied the CogStack suite to annotate clinical text from hospital inpatient records based on the Unified Medical Language System (UMLS) for classifying dementia status. We trained a logistic regression classifier to determine dementia/non-dementia status within two cohorts based on frequency of occurrence of a set of terms provided by experts - one with confirmed dementia based on clinical assessment and the other confirmed non-dementia based on telephone cognitive interview. We used our annotation platform to review the accuracy of concepts assigned by CogStack.
Results
There were 368 people with clinically confirmed dementia and 218 screen-negative for dementia. Of these, 259 with dementia and 195 without dementia had documents in the inpatient electronic health record system, 84045 inpatient documents 16950 for the dementia and non-dementia cohort respectively. A set of key words pertaining to dementia was generated by a specialist neurologist and a health information manager, and matched to UMLS concepts. The NCHA data platform holds a copy of the inpatient text records (>13million documents) that has been annotated using CogStack. Annotated documents corresponding to the study cohort were extracted.
We tested true positive rates of annotation against 50 concepts judged by a neurologist and health information manager to be relevant to dementia patients by manually review of 100 documents.
Conclusion
Automated annotations must be validated. The platform we have developed allows efficient review and correction of annotations to allow models to be trained further or provide confidence that accuracy is sufficient for subsequent analysis. Implementation within our linked NCHA data platform will allow incorporation of text based data at scale.
Abstract Deciphering the mechanism behind the lack of growth during the prolonged low abundance phase of cyclic populations represents a major challenge due to the lack of data. However, long-term capture-mark-recapture datasets can help resolve this question through demographic analyses. We analysed the demography and movements of cyclic brown lemmings on Bylot Island, Canada, with a 16-yr (2004-2019) live-trapping program and looked for inversely density-dependent factors that could contribute to a prolonged low phase. We found that the proportion of females in the population was inversely density-dependent with a strong male-biased sex ratio at low densities but not at high densities. In contrast, survival of adult females was higher than adult males, but both had lower survival at low densities than at high ones. Distances moved by both adult males and females were density-dependent, and proportion of females in reproductive condition was weakly density-dependent. Individual body condition, measured as monthly change in body mass, was not density-dependent. Overall, the strong malebiased sex ratio at very low densities suggests a loss of reproductive potential due to the rarity of females and appears to be the most susceptible demographic factor that could contribute to the prolonged low phase in cyclic brown lemmings. What leads to this sexbias in the first place is still unclear, but indirect evidence suggests that this may be due to a higher vulnerability of nesting females to predation in winter than males.
Matthew A Jay, Natalie Byrom, Matthew A Jay
et al.
Introduction
There is widespread recognition of the need to use administrative data from justice, crime, prisons and legal services to improve the evidence base on determinants and outcomes of involvement with justice systems and to improve services. There is great opportunity to use data from justice to improve health and other outcomes for clients and service users but this is a relatively unexplored area and these datasets have been neglected partly due to access concerns. This 90 minute symposium presents research and learning from data initiatives in justice settings and will explore challenges from the perspectives of researchers and the legal profession.
Objectives and Approach
Talks will cover:
• Development of legal epidemiology and use of administrative data in this nascent field (Matthew A Jay). Examples from the presenter’s family justice research using linked data will also be presented.
• The UK Ministry of Justice’s (MoJ’s) Data First programme (Prof Betsy Stanko). Data First aims to unlock the potential of data already created by MoJ, by linking datasets from the justice system and beyond, and enabling accredited researchers to access the data ethically and responsibly.
• UK court reform and data collection: using data to monitor equality and access to justice in the move to on-line dispute resolution (Dr Natalie Byrom).
• Health Justice Partnerships (HJPs) in Australia (Prof Suzie Forell). HJPs bring legal help into health settings to address social issues affecting the health of patients. This will be a discussion of HJPs as an opportunity to explore the impact of legal assistance on health and the challenges and opportunities of relevant data generation.
Conclusion / Implications
Better use of data in justice spheres will be a difficult process requiring cross-disciplinary collaboration. But it is one that promises to bring the study and practice of law into the 21 st Century.
Man Yee Kan, Ekaterina Hertog, Kamila Kolpashnikova
<b>Background</b>: Previous research suggested that husbands' participation in housework is positively associated with fertility choices for both women and men. We tested this association by using data of four East Asian countries. <b>Objective</b>: This paper examines whether the positive association between gender-equal sharing of housework participation and fertility intention in China, Japan, South Korea, and Taiwan has strengthened between 2006 and 2012. <b>Methods</b>: We harmonize two datasets, the 2006 East Asian Social Survey and the 2012 International Social Survey Programme. We employ OLS and ordered logit models estimators to test the association between husband's housework participation and the ideal number of children. <b>Results</b>: In both 2006 and 2012, husband's participation in housework is associated with both own and partner's fertility intentions in 2006 and 2012. The association between the domestic division of labour and fertility has not changed between 2006 and 2012. <b>Conclusions</b>: Corroborating the findings of our earlier paper the results suggest that a more gender-equal domestic division of labour in East Asia is associated with higher fertility intentions in this region. The gender revolution framework offers a plausible explanation for the East Asian fertility trends between 2006 and 2012. The findings suggest that there is a stall in the pace of the gender revolution. <b>Contribution</b>: This paper provides a summary of the trends highlighted by the contributors to this special issue. This is also the first paper to look at the evolution of domestic division of labour and fertility preferences in four East Asian countries over time.
Background
Antidepressant management with Selective Serotonin Reuptake Inhibitors (SSRI) during pregnancy is associated with risks of congenital malformations and adverse pregnancy outcomes.
Main Aim
We examined the incidence of SSRI use during pregnancy using Australian administrative pharmaceutical data linked to medical service data detailing antenatal consultations to assess pregnancy risk at current levels of treatment.
Method
A published data source from the Australian Pharmaceutical and Medical Benefit Schemes was used to identify pregnant women starting new onset treatments with SSRI medications during the provision of antenatal services. A longitudinal profile of SSRI incidence in the pre, peri and post pregnancy periods were constructed to define incident SSRI use. The potential increased burden from adverse pregnancy outcomes was calculated using estimates of risks from the literature.
Results
From 2005-2011, 40,778 women were identified to have started de novo SSRI treatments within 10 years of pregnancy (n=172,951). The prenatal monthly incidence of SSRI prescriptions was 25 per 1000 women in 2005 and 7.5/1000 in 2011. During pregnancy, this incidence fell to 7.5 and 4/1000 women in years, and rose above perinatal levels to between 13 and 33/1000 women one year postnatally (X2 p<.001). At these incidence rates, an estimated 2400 women per year receive SSRI’s during pregnancy. In comparison, with non-depressed pregnant women this potentially results in 10 further cardiovascular defects, 94 pre-term births, and 1 still birth annually.
Conclusion
This pharmaco-surveillance study has demonstrated continued use of SSRIs in Australian women of childbearing age but diminishing usage during pregnancy. A small number of women continue to be prescribed SSRI during the prenatal period in Australia with risks of adverse pregnancy outcomes.
Abstract Many studies of fertility have reported a wide range of factors to be important determinants. These determinants include proximate factors and male and female background variables. However, most of the research has been based on an analysis of cross-sectional fertility elasticities. Therefore, there are limited attempts to examine the temporal behaviour of especially marital fertility in association to the proximate and background determinants. To fill this gap, this study analysed Demographic and Health Survey data for Kenya, Rwanda and Zimbabwe using the Oaxaca-Blinder decomposition technique to determine the magnitude of marital fertility changes in association with selected socioeconomic factors. The results showed evidence of significant marital fertility transitions characterised by stalling for all three countries. Marital fertility rates were more responsive changes in reproductive behaviours than compositional characteristics. Male variables like community level of education were positively associated with stalling especially in Kenya and Zimbabwe. Despite the small contributions overall, analysing male and female variables improves the understanding of the sources of marital fertility changes in patriarchal societies.
Сергей Андреевич Тимонин, Анна Васильевна Концевая, Мартин МакКи
et al.
За прошедшее десятилетие в России существенно увеличилось число медицинских организаций, выполняющих высокотехнологичные малоинвазивные операции при остром коронарном синдроме – чрескожные коронарные вмешательства (ЧКВ). В данной работе мы оцениваем степень влияния географических барьеров на доступность населения к этому наиболее эффективному способу лечения острого коронарного синдрома, в том числе инфаркта миокарда.
Нами была собрана информация о медицинских организациях, выполнявших ЧКВ в 2010 и 2015 гг., а также данные о численности населения муниципальных районов и городских округов в России. Методами сетевого анализа было оценено время транспортировки и расстояние от физического центра каждого муниципального района/городского округа до ближайшего ЧКВ- центра в России. Расчеты были выполнены с использованием графа улично-дорожной сети
В период с 2010 по 2015 г. число ЧКВ-центров в России увеличилось со 144 до 260. Медианное время доезда до ближайшего ЧКВ-центра составило 48 минут в 2015 г. (в 2010 г. ‒ 73 минуты). Две трети городского населения в 2015 г. могли быть доставлены в ближайший ЧКВ-центр за 60 минут и быстрее, а среди сельского населения такую возможность имела только одна пятая часть потенциальных пациентов. Создание 67 новых ЧКВ-центров в тех городских округах, где в настоящий момент имеется их дефицит, увеличит долю населения, находящегося в 60-минутной зоне доезда до ЧКВ- центра, до 62%, что эквивалентно 5,7 млн человек, для которых на данный момент невозможен своевременный доступ к данному виду вмешательства.
В период между 2010 и 2015 г. происходило значительное, но неравномерные сокращение времени доезда до ЧКВ-центров. Несмотря на позитивные тенденции, Россия еще не приблизилась к уровню доступа, наблюдающегося в сопоставимых странах с большой территорией и неравномерно размещенным населением, таких как Австралия и Канада. Открытие относительно небольшого числа новых ЧКВ-центров (отделений) могло бы существенно улучшить доступ к медицинской помощи, тем самым сократив имеющееся неравенство.
Introduction
Cholinesterase inhibitors (ChEIs) are medications used to treat cognitive symptoms associated with Alzheimer’s disease. Previous studies examining the determinants of continued use or withdrawal of ChEIs during the transition into a nursing home have lacked detailed clinical information needed to understand the range of factors associated with pharmacotherapeutic decision-making.
Objectives and Approach
Population-based clinical and administrative health databases were linked to examine patterns of ChEI use among 47,851 adults (aged 66+) with dementia newly admitted to nursing homes in Ontario between April 2011-March 2015. We examined whether resident frailty, among other factors, was associated with ChEI discontinuation in the following year. Frailty was calculated using a validated 72-item index derived from the Resident Assessment Instrument (RAI-MDS 2.0). Discontinuation was defined as a 30-day period when no dispensations occurred and no supply of ChEI was available. Subdistribution hazard models estimated the association between resident characteristics and discontinuation, accounting for competing risk of death.
Results
Over one-third (36.7%) of residents were receiving a ChEI at admission and this proportion was lower among those defined as frail (33.6%) vs. non-frail (40.7%) at admission. Among those on a ChEI at admission, 82.3% continued use and 17.7% discontinued during the following year. After accounting for resident characteristics, ChEI type and previous use, the incidence of discontinuation was 15% higher in frail residents vs. non-frail residents (hazard ratio (HR)= 1.15, 95\% confidence interval (CI) [1.01,1.30]). Residents with severe aggressive behaviours (HR=1.82, 95% CI [1.60, 2.07]), and higher levels of cognitive impairment (HR=1.29, 95% CI [1.10, 1.51]) were more likely to discontinue. Residents aged 85+ (HR=0.69, 95% CI [0.61, 0.77]) and those who were widowed (HR=0.84, 95% CI [0.77, 0.91]) were less likely to discontinue.
Conclusion/Implications
Most residents who entered on a ChEI continued treatment during follow-up. The availability of linked clinical and administrative data allowed for a novel exploration of predictors of ChEI discontinuation. Frailty, severity of cognitive impairment and aggressive behaviours were associated with ChEI discontinuation; whereas selected sociodemographic factors predicted continued use.
The Census Bureau is partnering with state public assistance agencies to acquire program participation data and estimate new statistics that deepen a state’s understanding of program participants and improve outreach efforts to those who are eligible but do not participate. In collaboration with the Economic Research Service and the Food and Nutrition Service within the United States Department of Agriculture, the Census Bureau obtains individual-level program participation administrative records (AR) data for three state programs, the Supplemental Nutrition Assistance Program (SNAP), Temporary Aid for Needy Families (TANF), and the Special Supplemental Nutrition Program for Women, Infants and Children (WIC). The Census Bureau constructs a unique data set for each state program by linking the AR data to survey response data for the same individuals. These linked data enable the Census Bureau to model which survey respondents are eligible for program participation and also to observe which eligible individuals participate in the program. The Census Bureau then estimates eligibility and participation rates by a variety of demographic and economic characteristics and by county. The individual-level data also enable the Census Bureau to construct a statistical profile of eligible individuals and families that do not participate to assist state agencies with their outreach programs. All statistical results provided back to state agencies in table reports and data visualizations are reviewed to insure that individual identities are protected and not disclosed. This paper will present results for several state programs that have partnered the Census Bureau.
ABSTRACT
Background
Monitoring sepsis outcomes over time is necessary to evaluate healthcare improvement interventions. However, it is a clinical diagnosis (infection with a systemic inflammatory response) that is not routinely recorded electronically. We aimed to create a proxy measure for sepsis mortality using routine data.
Objective
ICD-10 codes on hospital discharge summaries are often used to monitor disease outcome but the presence of a code for a specific infection, e.g. pneumonia, does not give any indication of severity of illness i.e. whether the patient had sepsis. There are ICD-10 codes that specifically indicate sepsis, most commonly A40 and A41. However, monitoring mortality using ICD-10 is susceptible to coding bias as practice differs between and within health boards. Using ICD-10 codes alone does not differentiate between changes in mortality and changes in coding practice.
As recommended in the “Sepsis 6” care bundle, all patients with suspected sepsis should have a blood culture taken urgently. A previous study showed that having a blood culture taken was associated with a 3-fold increase in risk of mortality, over other hospital inpatients, independent of having a positive or negative result. We tested whether having a blood culture taken might be a suitable proxy marker for sepsis and compared it to data using ICD-10 sepsis codes.
Method
Data on all patients that had a blood culture taken in hospital in Scotland between 2011 and 2013 were obtained from hospital laboratory systems. These data were linked to hospital discharge summary records (SMR01) and death records. Monthly 30-day mortality rates from date of blood culture were calculated. Patients with an A40 or A41 ICD-10 code were extracted from hospital discharge summary records (SMR01) and linked to deaths records. Monthly 30-day mortality rates from date of admission were calculated over the same period of time. Blood culture sampling and, more strikingly, the use of A40/A41 codes increased over the study period. Blood culture data would indicate no significant change in 30-day mortality over the study period but coding data would indicate a 20% relative reduction.
Conclusion
A40/A41 codes lack sensitivity as the numbers of patients with these codes is many times lower than estimates of sepsis incidence, and coding practice changed dramatically over time. Blood culture data lacks specificity as many patients that had a blood culture taken will end up with an alternative diagnosis. Combining a broader group of infection codes with blood culture data may be a more useful measure.