International Journal of Population Data Science最新文献

{"title":"Integrating Data, Driving Solutions: The Minnesota Information Linking for Kids (Minn-LInK) Project.","authors":"Sydney Johnson, Michael Hoffmeister, Kristine Piescher","doi":"10.23889/ijpds.v11i5.3721","DOIUrl":"https://doi.org/10.23889/ijpds.v11i5.3721","url":null,"abstract":"<p><p>Child well-being is complex and spans multiple disciplines and systems. The involvement of children with multiple public service systems (e.g., child protection, juvenile justice, public health, education, children's mental health) necessitates better service coordination and integration to ensure comprehensive service provision and delivery and promote child well-being. Through partnerships with state agencies, the Minnesota Linking Information for Kids (Minn-LInK) project links administrative data from a variety of systems to create a resource for interdisciplinary research regarding vulnerable populations, often focusing on cross-system involved children and families experiencing poverty. Minn-LInK currently houses over 20 years of state-wide individual level administrative data from multiple Minnesota systems, including educational data, child protection and children's mental health records, data on the receipt of public services such as food assistance or childcare assistance, juvenile justice and adult criminal justice data, and records of incarcerated adults, including sentence and offense information. This presentation will demonstrate how the longitudinal scope of Minn-LInK data allows researchers to consider complex questions about programs and policies that change over time and creates opportunities for ongoing follow-up of large-scale evaluations. We will also outline our plans for future expansion to increase capacity for new research questions, additional users, and greater responsivity to the needs to public systems and practitioners.</p>","PeriodicalId":36483,"journal":{"name":"International Journal of Population Data Science","volume":"11 5","pages":"3721"},"PeriodicalIF":2.2,"publicationDate":"2026-07-06","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://www.ncbi.nlm.nih.gov/pmc/articles/PMC13426778/pdf/","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"148654286","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"The Long Shadow of Trauma: Comparing the Prevalence and Risks of Chronic Health Conditions among orthopaedic and other injury types 1-5 Years Post-Injury Using Population-Level Data Linkage.","authors":"Asmare Gelaw, Belinda Gabbe, Christina Ekegren","doi":"10.23889/ijpds.v11i5.3473","DOIUrl":"https://doi.org/10.23889/ijpds.v11i5.3473","url":null,"abstract":"<p><strong>Background and objectives: </strong>Major trauma survivors experience a substantial long-term health burden, yet trajectories of chronic physical conditions across injury types remain poorly understood. This study used linked administrative datasets to determine (1) the prevalence of chronic conditions before injury, at injury, within one year, and one to five years postinjury; and (2) the risk of developing new chronic conditions one to five years after major trauma.</p><p><strong>Methods: </strong>We conducted a retrospective cohort study using linked records from the Victorian State Trauma Registry (VSTR), Victorian Admitted Episodes Dataset (VAED), Victorian Emergency Minimum Dataset (VEMD), and Victorian Registry of Births, Deaths and Marriages (BDM). Adults aged ≥18 years (N = 28,522) with major trauma between 2007-2016 who survived ≥1 year were included. Trauma was classified as orthopaedic injury, severe traumatic brain injury (s-TBI), spinal cord injury, or other trauma. Chronic conditions were identified from diagnostic codes, and cumulative prevalence and adjusted relative risks were estimated.</p><p><strong>Results: </strong>Overall, 69.3% developed at least one chronic condition within five years. Preinjury conditions were most common in s-TBI (19.3%) and least in other trauma (6.6%). New-onset conditions were highest in s-TBI (21.7%) and orthopaedic trauma (21.4%), most frequently arthritis, cancer, and cardiovascular disease. Adjusted analyses showed no significant difference in long-term risk between orthopaedic and other trauma groups.</p><p><strong>Conclusions: </strong>Chronic conditions are highly prevalent among trauma survivors. Linked population data highlight the need for long-term, integrated rehabilitation and chronic disease management after injury.</p>","PeriodicalId":36483,"journal":{"name":"International Journal of Population Data Science","volume":"11 5","pages":"3473"},"PeriodicalIF":2.2,"publicationDate":"2026-07-06","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://www.ncbi.nlm.nih.gov/pmc/articles/PMC13426545/pdf/","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"148654289","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Palliative care at end-of-life for people with liver cancer: A 10 year population-based study in Australia.","authors":"Rebecca Mitchell, Shenghan Cai, Jacob George, Rose Boutros, Amany Zekry, Yvonne Zurynski","doi":"10.23889/ijpds.v11i5.3578","DOIUrl":"https://doi.org/10.23889/ijpds.v11i5.3578","url":null,"abstract":"<p><p>Palliative care is essential to manage symptoms and end-of-life (EOL) quality, yet its utilisation in liver cancer remains poorly understood. This study examined sociodemographic and clinical characteristics associated with palliative care admission in the last five years of life for people with a death from liver cancer. A population-based retrospective cohort study of adults (≥18 years) who died from liver cancer between 2013 and 2022 in New South Wales, Australia was conducted. Multivariable logistic regression identified factors associated with palliative care admissions using linked hospital, cancer registry, notifiable conditions, and mortality records. There were 3,565 deaths and 55.3% of people had at least one palliative care admission. Females (OR 1.25; 95%CI1.06-1.49), individuals with anxiety-related disorders (OR 1.63; 95%CI1.18-2.25), drug-related dependence (OR 1.81; 95%CI1.29-2.54), who lived in rural locations (OR 1.39; 95%CI1.18-1.64), who had metastatic cancer at diagnosis (OR 1.23; 95%CI1.01-1.50), who had ≥4 hospital admissions (OR 1.43; 95%CI1.08-1.90) or who died in hospital (OR 4.64; 95%CI4.64-6.52) had a higher likelihood of a palliative care admission compared to no palliative care admissions. People admitted to intensive care (OR 0.60; 95%CI0.47-0.75) or who had mechanical ventilation (OR 0.46; 95%CI0.29-0.74) in the last 12 months prior to EOL were less likely to have a palliative care admission. This is the first Australian study to examine palliative care use in liver cancer on a population-level. Findings support earlier integration of palliative care to reduce high-acuity interventions. These insights have implications for service delivery, equity, and policy in EOL care planning.</p>","PeriodicalId":36483,"journal":{"name":"International Journal of Population Data Science","volume":"11 5","pages":"3578"},"PeriodicalIF":2.2,"publicationDate":"2026-07-06","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://www.ncbi.nlm.nih.gov/pmc/articles/PMC13426645/pdf/","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"148654359","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Accelerating AI-Driven Neuroimaging Research in Dementia.","authors":"Lewis Hotchkiss, Chloe Tulip, Emma Squires, Simon Thompson","doi":"10.23889/ijpds.v11i5.3568","DOIUrl":"https://doi.org/10.23889/ijpds.v11i5.3568","url":null,"abstract":"<p><p>Artificial intelligence (AI) is rapidly advancing neuroimaging research for earlier and more accurate detection of dementia, yet the complexity, size, and heterogeneity of neuroimaging data create substantial barriers to scalable analysis within secure environments. Neuroimaging datasets require specialised processing, harmonisation across cohorts, and significant computational capacity - challenges that exceed those associated with typical structured health data. To address these issues, the Dementias Platform UK (DPUK) has developed a comprehensive neuroimaging infrastructure within its Trusted Research Environment (TRE), enabling secure, efficient, and standardised access to large-scale imaging resources for AI-driven research. We implemented automated pipelines to convert raw scans into research-ready datasets, adopting the Brain Imaging Data Structure (BIDS) standard across more than 20 cohort and clinical datasets. Harmonised imaging-derived features were produced to support cross-study comparability, and a curated clinical imaging collection focused on dementia was created to facilitate targeted AI model development. All datasets and workflows are available through the DPUK Data Portal, which provides researchers with high-performance compute, scalable storage, and established neuroimaging and AI tools within a secure environment. The resulting harmonised imaging resource offers a rich platform for investigating neurodegenerative disease trajectories and developing generalisable AI models. In collaboration with Dementias Platform Australia, we are extending this work internationally by federating TREs and aligning imaging collections using shared harmonisation methods, enhancing data diversity while preserving governance and privacy standards. By integrating processing pipelines, standardisation tools, and advanced infrastructure, DPUK has enabled scalable, responsible, and high-quality AI research using population-scale neuroimaging data.</p>","PeriodicalId":36483,"journal":{"name":"International Journal of Population Data Science","volume":"11 5","pages":"3568"},"PeriodicalIF":2.2,"publicationDate":"2026-07-06","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://www.ncbi.nlm.nih.gov/pmc/articles/PMC13426816/pdf/","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"148654372","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Investigating Biases in Seniors' Drug Data Coverage in Canada.","authors":"Samuel A Stewart, Jeanyoung Park, Maia von Maltzahn, Emily Black, Jennifer E Isenor, Ingrid Sketris, Shanna Trenaman","doi":"10.23889/ijpds.v11i5.3631","DOIUrl":"https://doi.org/10.23889/ijpds.v11i5.3631","url":null,"abstract":"<p><strong>Background: </strong>While no province in Canada covers the cost of prescriptions for all of its residents, all have some form of provincial supplementary coverage (PSC) to cover some costs for particular subgroups (based on age, economics, and/or diagnosis). This PSC data can provide insights into the drug dispensation patterns of the population, but they often have significant biases that are often ignored in the literature.</p><p><strong>Objectives: </strong>The objective of this research is to explore the differences between Drug Information System (DIS) and PSC data in the province of Nova Scotia (NS). METHODS: Demographic data on all eligible Nova Scotians, and their prescriptions in both the DIS and NSSPP, were extracted from August 2016-December 2024. Residents were classified as recipients, non-recipients, and non-consumers, and their demographics and prescription patterns were explored.</p><p><strong>Results: </strong>There were over 342 000 residents eligible for NSSPP during the study window. Over 61 million prescriptions for these individuals were dispensed across the two databases, 38 million of which were paid for by NSSPP (63%). NSSPP paid for prescriptions for 58% of the population, though very few citizens had their prescriptions exclusively paid for by NSSPP (8.2%). Those that didn't receive prescriptions through NSSPP tended to be younger and more female.</p><p><strong>Conclusions: </strong>The data from NS show a clear potential for biases in using PSC data as a representative sample of seniors, with biases based on both age and gender. Future research will explore how potential patient disease patterns may influence coverage within the system.</p>","PeriodicalId":36483,"journal":{"name":"International Journal of Population Data Science","volume":"11 5","pages":"3631"},"PeriodicalIF":2.2,"publicationDate":"2026-07-06","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://www.ncbi.nlm.nih.gov/pmc/articles/PMC13426606/pdf/","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"148654375","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Unplanned health service use among people living with cancer and mental health conditions: A retrospective matched case-comparison study in Australia.","authors":"Ramya Walsan, Rebecca Mitchell, Reema Harrison","doi":"10.23889/ijpds.v11i5.3513","DOIUrl":"https://doi.org/10.23889/ijpds.v11i5.3513","url":null,"abstract":"<p><p>People living with mental health conditions experience persistent inequities across the cancer pathway, yet their patterns of unplanned service use remain poorly described. This study examined predictors of unplanned health service use among people with cancer, with and without mental health comorbidities, in Australia. A retrospective matched case-comparison study using data from three Australian states, linking non-admitted patient (NAP) oncology records to cancer registry, hospital admission, emergency department (ED) presentation, outpatient mental health, and mortality data. Adults aged ≥18 years living with cancer who had NAP oncology encounter between 2018-2021 were included. Individuals with mental health conditions were matched 1:1 to those without a mental health diagnosis using propensity scores. Negative binomial regression, adjusted for sociodemographic and clinical factors, examined unplanned ED presentations or hospitalisations within 12 months. Of 99,230 people living with cancer, 10,442 (10.5%) had a recorded mental health condition. After matching, individuals with mental health conditions had a 36% higher risk of unplanned ED use (IRR 95% CI 1.32-1.41), 31% higher risk of unplanned hospitalisations (IRR 95% CI 1.26-1.36) and 46% higher risk of cancer-related hospitalisations (IRR 95% CI 1.39-1.53) compared to those without. Virtual care use, digestive and lung cancers (vs breast) and physical comorbidities were also associated with increased unplanned use. Substantial inequities in unplanned acute care exist for people with cancer and mental health conditions. Strengthening integration between oncology and mental health services and refining virtual care delivery may reduce avoidable acute care use and support more equitable outcomes.</p>","PeriodicalId":36483,"journal":{"name":"International Journal of Population Data Science","volume":"11 5","pages":"3513"},"PeriodicalIF":2.2,"publicationDate":"2026-07-06","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://www.ncbi.nlm.nih.gov/pmc/articles/PMC13426559/pdf/","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"148654384","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"The nuance of public trust: How personal experiences shape and influence trust in Maritime health data use.","authors":"Amy Grant, Ted McDonald, Samuel A Stewart, Mary-Ann Standing, Elizabeth Lappin, Claire Keenan, Katie Aubrecht, Dorothy Barnard, Marc-André Bouchard, Sarah Bridges","doi":"10.23889/ijpds.v11i5.3541","DOIUrl":"https://doi.org/10.23889/ijpds.v11i5.3541","url":null,"abstract":"<p><strong>Background: </strong>Provincial data repositories in Canada provide secure access to de-identified administrative health data. Social license is implied, yet local evidence around public support is lacking. This study explored public knowledge and perceptions around health data use and conditions that would support acceptance.</p><p><strong>Methods: </strong>In collaboration with public partners, this qualitative descriptive study involved residents in two Maritime provinces. Focus groups were held, with level of support for health data use cases measured using Fist to Five Voting (a real time 5-point voting activity) before and after group discussion. Transcripts were analyzed using inductive thematic analysis.</p><p><strong>Results: </strong>Seven focus groups were held with 32 participants (75% White; 72% female; 44% urban; 50% 60+). Participants' level of support often shifted after group discussion, increasing for some use cases but decreasing for others, depending on perceived risks and benefits. Participants generally supported data being used by health care practitioners, governments, healthcare facilities/administrators, and university-based researchers to improve the health system or understand disease drivers. Participants were less supportive of research by private companies even if analyzed locally or reported in aggregate, but support increased when they perceived greater public benefit. Support frequently changed after hearing others' narratives, knowledge, and experience.</p><p><strong>Conclusion: </strong>We found broad but conditional support of health data use for research. People's views on acceptable uses/users of health data are shaped by lived realities, and the experiences and perspectives of others. Communication on non-standard uses must acknowledge nuance - and speak directly to the conditions under which support changes.</p>","PeriodicalId":36483,"journal":{"name":"International Journal of Population Data Science","volume":"11 5","pages":"3541"},"PeriodicalIF":2.2,"publicationDate":"2026-07-06","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://www.ncbi.nlm.nih.gov/pmc/articles/PMC13426841/pdf/","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"148654427","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Enhanced Depression Phenotyping Using Machine Learning and Electronic Medical Records for Disease Surveillance: A Validation Study in Inpatient Settings.","authors":"Na Lin, Elliot A Martin, Seungwon Lee, Allison Grothman, Na Li, Kamala Adhikari, Kiarash Riazi, Cathy A Eastwood, Danielle A Southern, Hude Quan","doi":"10.23889/ijpds.v11i5.3729","DOIUrl":"https://doi.org/10.23889/ijpds.v11i5.3729","url":null,"abstract":"<p><strong>Background: </strong>Accurate detection of inpatient depression is essential for surveillance and resource allocation. Traditional International Classification of Diseases (ICD) codes suffer from low sensitivity and delayed availability.</p><p><strong>Objective: </strong>To develop and validate machine learning (ML) phenotyping algorithms for depression using unstructured inpatient clinical notes from electronic medical records (EMRs), compared with ICD-10-CA codes and manual chart review.</p><p><strong>Methods: </strong>This retrospective cohort study included a derivation cohort (N = 3032 adults, Jan-June 2015) and a temporally independent external validation cohort (N = 10659 adults, 2017-2022) from Calgary acute care hospitals. Both cohorts were deterministically linked to DAD and EMR databases. Manual chart reviews defined the reference standard for depression status. We compared ICD-10-CA codes against three EMR-based algorithms: Keyword Search, Concept Model and Document-Concept Model. The latter two employed a supervised ML classifier (XGBoost) utilizing clinical concepts extracted via the cTAKES natural language processing (NLP) pipeline. Primary measures were prevalence, sensitivity, positive predictive value (PPV), specificity, negative predictive value (NPV), and F1 score.</p><p><strong>Results: </strong>In derivation cohort (prevalence: 18.1%), the Concept Model achieved 76.4% sensitivity and 76.7% F1 score, significantly outperforming Keyword Search (F1: 66.7%), Document-Concept Model (F1: 73.7%), and ICD-10-CA codes (sensitivity: 20.0%; F1: 32.6%). In external validation (prevalence: 13.1%), the Concept Model retained robust performance (sensitivity: 92.3%; F1: 80.1%), compared with ICD-10-CA (sensitivity: 6.1%; F1: 11.4%).</p><p><strong>Conclusion: </strong>An EMR-based Concept Model using NLP and ML markedly improves depression case detection over ICD coding in inpatient settings. The methods could potentially facilitate psychiatric surveillance and epidemiological research using real-world clinical data.</p>","PeriodicalId":36483,"journal":{"name":"International Journal of Population Data Science","volume":"11 5","pages":"3729"},"PeriodicalIF":2.2,"publicationDate":"2026-07-06","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://www.ncbi.nlm.nih.gov/pmc/articles/PMC13426532/pdf/","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"148654460","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Fuel Poverty Interventions and Healthcare Usage: Insights from Warm Wales.","authors":"Manar Alshams, Laura Bentley, Jerlyn Peh, Giles Greene, Alisha Davies, Ashley Akbari, Claire Newman, Owen Davies, Emma Taylor-Collins, Joanna Dundon","doi":"10.23889/ijpds.v11i5.3525","DOIUrl":"https://doi.org/10.23889/ijpds.v11i5.3525","url":null,"abstract":"<p><p>Warm Wales (WW) is a community-based programme addressing fuel poverty and its broader impact on health and well-being. Interventions combine tailored energy advice, practical support, and education with wellbeing and social prescribing projects to improve health outcomes. Fuel poverty is closely linked to poorer health, making evaluating such programmes using housing and health linked data crucial. To describe the demographic, socioeconomic, health, and housing characteristics of individuals supported by WW and examine the programme's impact on primary care interactions. Methodology: An observational longitudinal cohort study linking WW service data with routinely collected electronic health record and administrative data in the Secure Anonymised Information Linkage (SAIL) Databank. Descriptive statistics summarised the WW supported population. Difference in Differences (DiD) compared primary care interactions rates between cases and Coarsened Exact Matched controls. WW supported 4,083 individuals living across 1,317 properties (January 2021-December 2023). Most WW-supported properties were older (65% built before 1966 vs 60.7% in Welsh housing stock) and socially rented (62.3% vs 19.2% in Welsh housing stock 2021). The cohort mirrored Welsh population sex and ethnicity distributions but was notably younger (aged 18 and under 40.1% vs 25.8%), in the most deprived quintile (40.9% vs 24.0%), and predominantly urban (81.5% vs 73.1%). Health characteristics and DiD results are pending approvals. Warm Wales supports younger individuals living in most deprived areas and older housing. These characteristics are consistent with groups at risk of fuel poverty. DiD analyses are ongoing and will assess the intervention's impact on healthcare utilisation.</p>","PeriodicalId":36483,"journal":{"name":"International Journal of Population Data Science","volume":"11 5","pages":"3525"},"PeriodicalIF":2.2,"publicationDate":"2026-07-06","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://www.ncbi.nlm.nih.gov/pmc/articles/PMC13426600/pdf/","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"148654471","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Generating realistic synthetic data.","authors":"Zoe White","doi":"10.23889/ijpds.v11i5.3554","DOIUrl":"https://doi.org/10.23889/ijpds.v11i5.3554","url":null,"abstract":"<p><p>Synthetic data has many different uses such as testing pipelines and training people for using new methods and technologies, which could be used within data linkage. The synthetic data generators project objectives are to create high fidelity synthetic data. To have high fidelity data, the synthetic data generator would need to include all variables, e.g. forename, surname and date of birth, and to include realistic errors, for example the frequency and types of errors. This ensures the synthetic data behaves similarly to the real underlying data. An initial workshop with users gained the requirements for; variables, data size, realistic errors and bias. For example, swapped forename and middle names for error, or missing children in census sized data. We also carried out research into different synthetic data generators in Python and Pyspark, as well as meeting our user requirements. Our final approach was to create the functions and documentation for running the generators with errors and guidance on how to run the generator. Results, Conclusions and Implementation The synthetic data generator will be run end to end with user specifying where errors and biases are for a particular dataframe and compared to the genuine sensitive data. The usability of the code is highly important for ability to create synthetic data for developing code or in the training of new team members. Therefore, user testing of the synthetic data generator will be carried out to gather feedback, which will inform version 2 development or implemented within the code.</p>","PeriodicalId":36483,"journal":{"name":"International Journal of Population Data Science","volume":"11 5","pages":"3554"},"PeriodicalIF":2.2,"publicationDate":"2026-07-06","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://www.ncbi.nlm.nih.gov/pmc/articles/PMC13426838/pdf/","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"148654496","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}