Data Linkage Research Articles

How do large scale population studies inform vaccine evaluations in England?

Large scale population studies are important to monitor and evaluate aspects of a vaccination programme including vaccine coverage, real-world effectiveness, and post-licensure vaccine safety. These types of epidemiological studies often come under the remit of public health agencies, such as the UK Health Security Agency (UKHSA) in England, who are required to undertake surveillance of vaccine preventable diseases, including via seroepidemiological studies and data linkage studies using national level electronic healthcare data. An individual level national vaccine register with an accurate denominator can be the key to gaining insights into vaccine coverage, effectiveness and safety. During the COVID-19 pandemic, England's first vaccine register was developed. This enabled timely estimates of real-world vaccine effectiveness in the whole population of England, as well as enabling epidemiological investigations of rare potential risks from vaccines in specific populations. Population-based research studies, including prospective cohort studies are complementary to surveillance, and combined, enable more comprehensive assessments. As there was unprecedented investment into research studies and infrastructure during the pandemic, the scale of these studies meant they were able to contribute to vaccine programme evaluations in a way that had not been possible for previous vaccine programmes. In this review, we summarise the different large scale surveillance and research studies that have been used to evaluate and inform vaccine policy from the time of the first data linkage studies undertaken in England in the 1990s to the present day post-COVID-19 pandemic.

Long-term chronic conditions in individuals with mental and behavioural disorders: A data linkage study.

This study aimed to investigate the physical health conditions among health service users in the first year since recorded mental disorder diagnosis in Western Australia. Community mental health, emergency department (ED) and inpatient records of individuals aged ⩾ 18 years with a recorded mental disorder diagnosis in state-funded health services were analysed. We identified long-term physical health conditions recorded within the first year of the first recorded mental disorder diagnosis. Prevalence of physical comorbidity across time was estimated using multinomial logistic regression. Mean number of health service contacts in the first year of the recorded mental disorder diagnosis was obtained using generalised linear model. Altogether, 253,362 individuals were included. Within the first year of the first recorded mental disorder, the prevalence of at least one physical comorbidity ranged from 20.0% in 2006 to 14.5% in 2020. Cardiovascular disease was the most common comorbidity, but the most common combinations of comorbidities became more varied over time. The number of ED and inpatient contacts were higher in subgroups with a higher number of physical comorbidities (ED contacts: 2.4 [95% confidence intervals: 2.4, 2.4] for no comorbidities to 3.6 [3.4, 3.8] for ⩾ 3 comorbidities; inpatient contacts: 2.6 [2.6, 2.7] for no comorbidities to 4.5 [4.1, 4.9] for ⩾ 3 comorbidities). With a substantial proportion of individuals with mental disorders already having physical comorbidities on their first year of contact with state-funded health services, and the comorbidity combinations becoming more diverse, there is a need to implement more comprehensive joint mental and physical health services.

Abstract TP329: Hospital encounters in the 7 days and 30 days before first-ever stroke: An observational study using data linkage

Background/Aim: Up to 20% of people have a hospital encounter in the 30 days before a first ever stroke, suggesting opportunities for better primary prevention. The reasons for these healthcare encounters before stroke have been largely unexplored. We aimed to describe diagnoses in hospital encounters within 7 days and 30 days before a first-ever stroke. Methods: We used a statewide (Tasmania, ~500,000 population) linked dataset comprising emergency department [ED] presentations, hospital admissions and deaths in Australia from 2007-2020. An 8-year look back period was used to identify first ever strokes. ICD 10-AM codes in the principal diagnosis field were used to determine diagnoses in hospital encounters before stroke. We describe the distribution of diagnosis groups for each hospital encounters, separately for ED presentations and hospital admissions, in the 7 days and 30 days before stroke. Results: Out of 4907 first-ever stroke patients, 202 (4.12%) people had 234 hospital admissions within 7 days before stroke and 507 (10.33%) people had 724 hospital admissions within 30 days before stroke. Among 339 (7%) people, there were 351 ED presentations within the 7 days before stroke and among 676 (13.78%) people, there were 765 ED presentations in the 30 days before stroke. In event-level analyses, the most common diagnoses for hospital admissions were dialysis, diseases of circulatory system (e.g. carotid artery stenosis and unruptured cerebral aneurysm), abnormal symptoms or pathological findings (e.g. convulsions and headache), nervous system disorders (e.g. migraine) and respiratory diseases (Table 1). The most common diagnoses for ED presentations were circulatory system disorders, abnormal pathological findings and nervous system disorders in the 7 days and 30 days before stroke. Conclusion: Hospital encounters in the period immediately before stroke are common and associated with a broad range of health conditions. These may represent potentially high-risk groups requiring more intensive primary stroke prevention interventions.

The Impact of Severe Mental Illness on Treatment Retention and All-Cause Mortality of Individuals in Opioid Agonist Treatment: A Retrospective Data Linkage Study

The Impact of Severe Mental Illness on Treatment Retention and All-Cause Mortality of Individuals in Opioid Agonist Treatment: A Retrospective Data Linkage Study

Exploring the causal paths and longitudinal impact of telework on health: the TeleHealth survey–data linkage study protocol

BackgroundThe transition from traditional office work to telework has accelerated significantly since the late 20th century, especially in light of the COVID-19 pandemic. Despite its widespread adoption, the long-term health...

Accuracy of privacy preserving record linkage for real world data in the United States: a systemic review.

Examine the accuracy of privacy preserving record linkage (PPRL) matches in real world data (RWD). We conducted a systematic literature review to identify articles evaluating PPRL methods from January 1, 2013 to June 15, 2023. Eligible studies included original research reporting quantitative metrics such as precision and recall in health-related data sources. Covidence software was used to manage the review process. Five studies met our inclusion criteria. Tokenization and hash functions were used to hash and encrypt personally identifiable information (PII) including first and last names, dates of birth (DOB), and Social Security Numbers (SSNs) in a variety of RWD. All identified studies utilized deterministic matching. Combinations of tokenized or hashed PII that included "quasi-identifiers" like names and DOBs had consistently high precision (>95%) but lower recall, likely due to misspelled or inconsistently spelled names and name changes. SSN-based combinations demonstrated high precision but variable recall due to incomplete SSN data in RWD. Studies that employed algorithms in which at least one match was identified from a specified set of PII combinations provided high precision and high recall. The systematic review indicates that PPRL methods generally provide highly accurate patient data linkage while maintaining privacy. Researchers should carefully consider the completeness and stability of each PII element selected for PPRL and may want to employ a strategy that allows for patient records to be matched if they meet at least one of several combinations of PII.

Counting the cost of hospitalised injury using administrative data in Australia: three considerations from a retrospective cohort data linkage study

ObjectivesTo examine the relative differences in injury hospitalisation cost estimates from hospital administrative data when using (1) hospital reimbursement based on National Efficient Price (NEP) versus local hospital costings methods,...

Reproductive outcomes, prenatal healthcare utilization, and predictors Among the justice-involved young women who used drugs: a population-based data linkage study in Taiwan

Reproductive outcomes, prenatal healthcare utilization, and predictors Among the justice-involved young women who used drugs: a population-based data linkage study in Taiwan

Comparative effectiveness of antipsychotic treatment strategies for relapse prevention in first-episode schizophrenia in Finland: a population-based cohort study.

The best pharmacological treatment practices for relapse prevention in patients with first-episode schizophrenia are unclear. We aimed to assess different treatment strategies used before and after the first relapse, and their associations with subsequent relapse risk. In this population-based cohort study, we enrolled individuals (aged ≤45 years) with first-episode schizophrenia who were hospitalised and subsequently relapsed between 1996 and 2014 from the nationwide Finnish Hospital Discharge Register. Individuals who had not been taking antipsychotics within the year preceeding initial hospitalisation and who had a relapse within 5 years of discharge were included in the analyses. Treatment strategies were assessed during the 30 days before hospitalisation for the first relapse and 30 days after discharge and were categorised as either long-acting injectable, clozapine, non-clozapine oral antipsychotic monotherapy, non-clozapine oral antipsychotic polypharmacy, and antipsychotic non-use. Adjusted hazard ratios (aHRs) of the risk of second relapse based on treatment type were analysed with Cox regression models for 2 years after the first relapse, or until death or end of data linkage (Dec 31, 2017). People with lived experience of schizophrenia were not involved in the research and writing process. Between Jan 31, 1996 and Dec 31, 2017, 3000 individuals had their first psychosis relapse and were eligible for analysis. Mean age was 30·0 years (SD 7·6), 1069 (35·6%) of patients were women and 1931 (64·4%) men. No ethnicity data were available. 2148 (71·7%) had a second relapse within 2 years. Before first relapse, most individuals were either not using antipsychotics (n=1366 [45·5%]), or were using non-clozapine oral antipsychotic monotherapy (n=973 [32·4%]). Compared with continuing the same treatment strategy used before the first relapse, switching to clozapine was associated with the lowest risk of second relapse compared with continuing any non-clozapine oral antipsychotic monotherapy (aHR 0·66, 95% CI 0·49-0·89; relapse rate 73·2% with oral non-clozapine antipsychotic monotherapy continuation vs 57·1% with switch to clozapine). Switching to another non-clozapine oral antipsychotic monotherapy (0·99, 0·76-1·28) was approximately as unhelpful in preventing the next relapse as switching to antipsychotic non-use (1·07, 0·80-1·42). In patients with first-episode schizophrenia having their first psychosis relapse despite use of non-clozapine oral antipsychotics, continuation with the same antipsychotic modality or switch to another non-clozapine oral antipsychotic did not show evidence of being beneficial in relapse prevention, suggesting that clozapine should be started instead. This finding, together with existing knowledge of decreased risk of mortality associated with clozapine, challenges current treatment guidelines that recommend clozapine as a third-line treatment, resulting in treatment practices characterised by long delays to clozapine initiation. Sigrid Jusélius Foundation.

The Danish Drowning Cohort: Utstein-style data from fatal and non-fatal drowning incidents in Denmark

BackgroundEffective interventions to reduce drowning incidents require accurate and reliable data for scientific analysis. However, the lack of high-quality evidence and the variability in drowning terminology, definitions, and outcomes present significant challenges in assessing studies to inform drowning guidelines. Many drowning reports use inappropriate classifications for drowning incidents, which significantly contributes to the underreporting of drowning. In particular, non-fatal drowning incidents are underreported because many countries do not routinely collect this data.The Danish Drowning CohortThe Danish Drowning Cohort was established in 2016 to facilitate research to improve preventative, rescue, and treatment interventions to reduce the incidence, mortality, and morbidity of drowning. The Danish Drowning Cohort contains nationwide data on all fatal and non-fatal drowning incidents treated by the Danish Emergency Medical Services. Data are extracted from the Danish prehospital electronic medical record using a text-search algorithm (Danish Drowning Formula) and a manual validation process. The WHO definition of drowning, supported by the clarification statement for non-fatal drowning, is used as the case definition to identify drowning. All drowning patients are included, including unwitnessed incidents, non-conveyed patients, patients declared dead prehospital, or patients with obvious clinical signs of irreversible death. This method allows syndromic surveillance and monitors a nationwide cohort of fatal and non-fatal drowning incidents in near-real time to inform future prevention strategies. The Danish Drowning Cohort complies with the Utstein style for drowning reporting guidelines. The 30-day mortality is obtained through the Civil Personal Register to differentiate between fatal and non-fatal drowning incidents. In addition to prehospital data, new data linkages with other Danish registries via the patient’s civil registration number will enable the examination of various additional factors associated with drowning risk.ConclusionThe Danish Drowning Cohort contains nationwide prehospital data on all fatal and non-fatal drowning incidents treated by the Danish Emergency Medical Service. It is a basis for all research on drowning in Denmark and may improve preventative, rescue, and treatment interventions to reduce the incidence, mortality, and morbidity of drowning.Graphical

Improving UK data on avoidable perinatal brain injury: review of data dictionaries and consultation.

High quality data is important to understanding epidemiology and supporting improvement efforts in perinatal brain injury. It is not clear which data items relevant to brain injury are captured across UK sources of routinely collected data, nor what needs to be done to ensure that those sources are fit for purpose in improving care. We reviewed data dictionaries of four main UK perinatal data sources and consulted a multi-professional group (N = 27) with expertise in neonatal/maternity care, statistics, and clinical negligence. None of the data sources we reviewed currently captures, on its own, the range of items relevant to brain injury. Data items lack common definitions and ongoing linkage across the different sources. Our consultation identified the need for standardising the definition of avoidable perinatal brain injury, resolving inconsistencies in capturing data, improving linkage of data across existing data sources, and co-designing a strategy for meaningful use of data. Limited standardisation and linkage across UK data sources are key problems in using data to guide improvement efforts aimed at reducing risk of avoidable perinatal brain injury. A programme involving co-design with healthcare professionals and families to improve capture and use of data is now needed. Limited standardisation and linkage across UK data sources currently challenge the use of data as the basis of efforts to reduce risk of avoidable perinatal brain injury. A harmonisation programme involving consultation and co-design with healthcare professionals, families, and other specialists is needed to enable better capture and use of data in this key area. There is need to standardise the definition of avoidable perinatal brain injury, resolve inconsistencies in capturing data, improve linkage of data collected across existing data sources, and co-design a strategy for meaningful use of data.

Communicating with Stakeholders to Identify High-Impact Research Directions for Non-Targeted Analysis.

Non-targeted analysis (NTA) using high-resolution mass spectrometry without defined chemical targets has the potential to expand and improve chemical monitoring in many fields. Despite rapid advancements within the research community, NTA methods and data remain underutilized by many potential beneficiaries. To better understand barriers toward widespread adoption, the Best Practices for Non-Targeted Analysis (BP4NTA) working group conducted focus group meetings and follow-up surveys with scientists (n = 61) from various sectors (e.g., drinking water utilities, epidemiologists, n = 9) where NTA is expected to provide future value. Meeting participants included producers and end-users of NTA data with a wide range of familiarity with NTA methods and outputs. Discussions focused on identifying specific barriers that limit adoption and on setting NTA product development priorities. Stated priorities fell into four major categories: 1) education and training materials; 2) QA/QC frameworks and study design guidance; 3) accessible compound databases and libraries; and 4) NTA data linkages with chemical fate and toxicity information. Based on participant feedback, this manuscript proposes research directions, such as standardization of training materials, that BP4NTA and other institutions can pursue to expand NTA use in various application scenarios and decision contexts.

Linking American Spine Registry (ASR) and Medicare Data:an analysis of 8,755 Lumbar Fusion Cases.

Retrospective observational study. To evaluate whether the combined American Spine Registry and Medicare (ASR/CMS) data yields substantially different findings versus ASR data alone with regard to key parameters such as risk stratification, complication rates and readmission rates in lumbar surgery investigated through an analysis of 8,755 spondylolisthesis cases. Medicare data correlation has been effective for determining revision rates for other procedures such as total hip replacement. Our aim is to determine whether these findings are translatable in the realm of lumbar spinal surgery investigated through an analysis of 8,755 spondylolisthesis cases. The American Spine Registry (ASR) was queried for Medicare-eligible patients who underwent lumbar spinal fusion for lumbar spondylolisthesis. This cohort was analyzed based upon ASR data alone in comparison to the same patients in the combined ASR/Medicare (ASR/CMS) dataset. The primary outcome of interest was readmission at 30 and 90 days postoperatively. There were 8,755 Medicare-eligible cases with a diagnosis of spondylolisthesis within the ASR. The mean age was 72.7 years, 60.8% were female. Medical comorbidities were more frequently detected in the combined ASR/CMS dataset, reflected by a higher mean Charlson Comorbidity Index score (3.49 vs. 3.27, P<0.001). Hospital readmission rates were significantly higher in the combined ASR/CMS dataset at both 30 days (4.89% vs. 1.83%, P<0.001) and 90 days (7.68% vs. 2.66%, P<0.001), with notable increases in readmissions for infections and medical complications. Discharge disposition remained comparable across datasets, with most patients discharged to home or home health care. This study demonstrates that integrating patient-identified Medicare data with the ASR provides a more comprehensive assessment of outcomes for lumbar spinal fusion surgery as demonstrated through an analysis of 8,755 spondylolisthesis cases. These findings, establish the importance of multi-source data linkage to overcome the limitations of single-source registries, thereby enhancing data quality for clinical decision-making and quality improvement in spinal surgery.

P-1348. Acquired Carbapenemase-Producing Gram-Negative Organisms (CPOs) Across the Pond: Surveillance of Antibiotic Resistance in CPOs in England from 2020-2023

Abstract Background Acquired carbapenemase-producing Gram-negative organisms (CPOs) threaten viable antibiotic treatment options. This is the first national look at CPO phenotypic antimicrobial resistance (AMR) rates since becoming statutorily notifiable in England on 01/10/2020.Table 1.Summary of all big-5 CPO isolates reported from October 2020 to December 2023 Methods All acquired big-5 (NDM, KPC, OXA-48-like, VIM, IMP) CPO reported specimens and antibiotic susceptibilities were extracted from national laboratory surveillance databases between Oct-2020 to Dec-2023. Records were deduplicated by patient, specimen group (invasive/screening/other), bacterial species and CPO mechanism per year. Regional big-5 CPO population rates were calculated; 2023 AMR data were assessed by mechanism and species.Figure 1.Regional notifications per 100,000 population of acquired CPOs by big-5 carbapenemase mechanisms in England, 2023 Results In total, 13,022 episodes from 9,945 patients were reported; median age was 70y, 54% were male. The dominant mechanisms were OXA-48-like (n=4,884; 37.6%), NDM (n=3,927; 30.2%) and KPC (n=3,045; 23.4%); (Table 1) with large regional variation (Figure 1). NDMs became more dominant in 2023 than OXA-48-like. Screening samples accounted for 72% (n=9,378), and &amp;lt; 5% (n=591) were from invasive samples. Thirty-three patients with invasive samples had dual-mechanisms detected: 61% were NDM &amp; OXA-48-like; Klebsiella pneumoniae was the dominant species (patients=21). In 2023, the lowest reported resistance for the most frequently isolated species (n=5,535/5,711; Table 2) was for colistin (4-7%R) for all big-5 mechanisms in Escherichia coli, Enterobacter spp., Citrobacter spp., whilst higher for Klebsiella spp. and Pseudomonas spp. (10-17%R). Similarly, lower Enterobacterales resistance was observed for KPC and OXA-48-like to ceftazidime/avibactam (9-12%R) and amikacin (7-8%R), except for Klebsiella spp., which was higher (22%R). Ciprofloxacin and gentamicin had ≥ 35% resistance. Fosfomycin resistance in E. coli was low (≤ 9%). Proportion of big-5 CPO isolates (n=5,535) resistant to other antimicrobials by most frequently isolates species in England, 2023 Conclusion Continued national CPO and associated AMR surveillance is crucial, particularly for new antimicrobials, to inform guidance and local, empiric prescribing by region, given the geographical variation in CPO mechanism. Further enhancement via data linkage to ethnicity and clinical characteristics to identify risk factors, and tracking changes in patients’ antibiograms over time would help further target national guidance. Disclosures All Authors: No reported disclosures

Characteristics and outcomes associated with fidelity in the Family-Nurse Partnership in England: a data linkage cohort study

ObjectivesTo determine (1) which maternal and area characteristics are associated with reaching fidelity targets (the expected number of visits mothers should receive at each stage of the programme) in the...

Recommendations for Improving Surveillance of Congenital Anomalies in Europe Using Healthcare Databases.

Although accessing administrative data in healthcare databases may be a more time-efficient and cost-effective method of conducting surveillance, there is evidence suggesting that administrative data alone are not sufficient for population-based surveillance of congenital anomalies. To propose recommendations to maximise the potential use of healthcare databases for surveillance of congenital anomalies based on our data linkage experiences and results from the EUROlinkCAT study. EUROlinkCAT is a population-based cohort study of 99,416 children with anomalies born between 1995 and 2014. The congenital anomaly case records of children in 11 European congenital anomaly (EUROCAT) registries (eight countries) were linked to standardised administrative healthcare data (birth records, death records and hospital discharge records) to evaluate mortality and morbidity outcomes in these children. Overall, 97% of children with anomalies were successfully matched to their records in their national or regional administrative databases. Recommendations to improve surveillance of anomalies when using healthcare data were developed through establishing and analysing data from this cohort. The primary recommendation is to develop systems to report anomalies diagnosed in foetuses who undergo a termination and link these data to their mothers. Each liveborn baby must be assigned a permanent unique identification number at birth to enable accurate linkage across healthcare databases. Implementing and improving existing algorithms to discriminate between major anomalies and suspected or minor anomalies will improve accuracy in coding. Heterogeneity in coding anomalies will improve by avoiding the use of 'unspecified' or 'other specified' codes in hospital databases. Relaxation of country-specific regulations concerning the suppression of small numbers are necessary to enable data to be combined across European countries. Implementation of these recommendations will enable the information in electronic healthcare databases, in conjunction with Congenital Anomaly registries, to be fully exploited and hence will improve the surveillance of anomalies in children.

Time and cost of linking administrative datasets for outcomes assessment in a follow-up study of participants from two randomised trials

BackgroundFor the follow-up of participants in randomised trials, data linkage is thought a more cost-efficient method for assessing outcomes. However, researchers often encounter technical and budgetary challenges. Data requests often require a significant amount of information from researchers, and can take several years to process. This study aimed to determine the feasibility, direct costs and the total time required to access administrative datasets for assessment of outcomes in a follow-up study of two randomised trials.MethodsWe applied to access administrative datasets from New Zealand government agencies. All actions of study team members, along with their corresponding dates, were recorded prospectively for accessing data from each agency. Team members estimated the average time they spent on each action, and invoices from agencies were recorded. Additionally, we compared the estimated costs and time required for data linkage with those for obtaining self-reported questionnaires and conducting in-person assessments.ResultsEight agencies were approached to supply data, of which seven gave approval. The time from first enquiry to receiving an initial dataset ranged from 96 to 854 days. For 859 participants, the estimated time required to obtain outcome data from agencies was 1,530 min; to obtain completed self-reported questionnaires was 11,025 min; and to complete in-person assessments was 77,310 min. The estimated total costs were 20,827 NZD for data linkage, 11,735 NZD for self-reported questionnaires, and 116,085 NZD for in-person assessments. Using this data, we estimate that for a cohort of 100 participants, the costs would be similar for data linkage and in-person assessments. For a cohort of 5,000 participants, we estimate that costs would be similar for data linkage and questionnaires, but ten-fold higher for in-person assessments.ConclusionsObtaining administrative datasets demands a substantial amount of time and effort. However, data linkage is a feasible method for outcome ascertainment in follow-up studies in New Zealand. For large cohorts, data linkage is likely to be less costly, whereas for small cohorts, in-person assessment has similar costs but is likely to be faster and allows direct assessment of outcomes.

Number of children and dementia risk: a causal mediation analysis using data from the HUNT study linked with national registries in Norway

BackgroundChildlessness, as well as having a high number of children, has been reported to be associated with an elevated risk of dementia compared to having 2–3 children. The mechanisms underlying these relationships are not well understood and may be mediated by different midlife risk factors. We examined the mediating role of various factors on the relationship between the number of children and dementia risk. These factors include socioeconomic factors (e.g., occupational complexity), psychosocial (e.g.., social activities, loneliness, life satisfaction), lifestyle (e.g., smoking, physical inactivity, alcohol intake), and chronic diseases (e.g., obesity, diabetes, depression, hearing impairment and hypertension).MethodsUsing a historic cohort design, we included 9,745 participants born between 1931–48, with a mean age of 78.2 (SD = 6.4) years at the time of cognitive testing in the HUNT4 70 + sub-study (2017–2019). Further measures were obtained through data linkage between information from Statistics Norway and the HUNT1(1984–86), and HUNT2 (1995–97) Surveys. Causal mediation analyses using an inverse odd weighting approach were conducted to decompose the total effect of the number of children (0, 1, or 4 + children vs. 2–3) on the risk of dementia at age 70 + years into direct and indirect effects with mediators assessed at a mean age of 50.7 (SD = 6.4) years. The analyses were adjusted for age, sex, marital status at age 25 years, educational status, and religion assessed during HUNT3 (2006–2008).ResultsOverall, 15.7% were diagnosed with dementia. The proportions with dementia by the number of children were 22.3% among those with no children, 21.4% for those with one child, 13% for those with 2–3 children (specifically, 12.6% for those with 2 children and 13.4% for those with 3 children), and 19.9% for those with 4 + children. Compared to the reference group of individuals with 2–3 children, the dementia risk was higher among the groups with no children (relative risk (RR): 1.30, 95% confidence interval (CI) (1.12, 1.51)), those with one child (RR: 1.30, 95% CI (1.14, 1.47)) and those with 4 + children (RR: 1.12, 95% CI (1.01, 1.24)). The elevated risks of dementia were not mediated by the socioeconomic, psychosocial, lifestyle, or chronic diseases related factors that we tested. Sex-stratified analysis showed higher dementia risk for men without children and women with one or 4 + children compared to those with 2–3 children, with similar patterns across sexes. None of the mediators contributed to mediation in either group. None of the mediators appeared to contribute through mediation in either group.ConclusionsOur findings suggest that the number of children—specifically being childless, having one child, or having four or more children—may influence the risk of dementia. These relationships were not mediated by psychosocial, lifestyle, and socioeconomic factors, or markers of chronic diseases in adulthood considered in this study.

Open science 2.0: revolutionizing spatiotemporal data sharing and collaboration

The Spatial Data Lab (SDL) project is a collaborative initiative by the Center for Geographic Analysis at Harvard University, KNIME, Future Data Lab, China Data Institute, and George Mason University. Co-sponsored by the NSF IUCRC Spatiotemporal Innovation Center, SDL aims to advance applied research in spatiotemporal studies across various domains such as business, environment, health, mobility, and more. The project focuses on developing an open-source infrastructure for data linkage, analysis, and collaboration. Key objectives include building spatiotemporal data services, a reproducible, replicable, and expandable (RRE) platform, and workflow-driven data analysis tools to support research case studies. Additionally, SDL promotes spatiotemporal data science training, cross-party collaboration, and the creation of geospatial tools that foster inclusivity, transparency, and ethical practices. Guided by an academic advisory committee of world-renowned scholars, the project is laying the foundation for a more open, effective, and robust scientific enterprise.

Outpatient Palliative Care and End-of-Life Care Intensity: Linking Massachusetts Cancer Registry with All-Payer Claims.

Early palliative care is associated with better outcomes for patients with advanced-stage cancers. Using a novel data linkage, we assessed outpatient palliative care use before death and its association with end-of-life care intensity and variation across eight provider networks. We linked Massachusetts Cancer Registry and the All-Payer Claims Database for individuals with commercial insurance, Medicaid or Medicare Advantage diagnosed with colorectal, lung, prostate, and breast cancers from 2010 through 2013 who died by December 31, 2014. We characterized outpatient palliative care visits in the 6 months before death and identified end-of-life hospitalizations, emergency department visits, intensive care unit admissions, chemotherapy, no/late hospice enrollment, and in-hospital deaths. We used logistic regression to assess factors associated with outpatient palliative care and ordinal logistic regression with provider network fixed effects to assess the association of palliative care with a composite measure summing individual end-of-life intensity measures. Among 6,279 decedents, 11.3% had at least one outpatient palliative care visit. Palliative care use varied across provider networks from 6.0% to 19.3%. In adjusted analyses, younger age, longer duration from diagnosis to death, death in 2012-2014 vs. 2010, and provider network were associated with palliative care visits (all P < 0.05). End-of-life care intensity varied across provider networks. Patients with palliative care visits had lower adjusted odds of receiving intensive end-of-life care (adjusted odds ratio (OR) 0.62 per additional measure of end-of-life intensity, 95%CI 0.53, 0.72). Outpatient palliative care use varied substantially among regional provider networks and was associated with less intensive end-of-life care.