Pub Date : 2022-08-25DOI: 10.23889/ijpds.v7i3.1816
Dave T. Rowley, C. Ellis
In December 2021, we published statistical research on Administrative Data Based Population Estimates (ABPEs) for Scotland’s population in 2016, 2017 and 2018. This work was developed as part of a project to consider how administrative data could be used to support Scotland’s Census. Following the governance process, administrative datasets were processed and de-identified, before being transferred to Scotland’s National Safe Haven for linking and analysis. The datasets used include data from health, the electoral register, vital events registrations, and education. The methodology used several linking variables so data could be linked, even without exact agreement between records. Records from across the data sources were resolved into individuals using these links. Business rules then indicated which individuals to include in Scotland’s Integrated Demographic Dataset (SIDD). The ABPEs were then produced from this and compared with the official mid-year population estimates (MYEs) to determine success. On aggregate, the population estimates from the ABPEs are very similar to the MYEs, differing by less than 0.5 per cent in each year. When broken down further, larger differences occur with ABPEs having more males and fewer people aged over 65 when compared with the official statistics. A notable difference between the two is for males aged between 30 and 65 in deprived areas, with ABPEs up to 20% higher than the MYEs. These differences by deprivation are smaller for other age ranges and for females. The ABPEs tend to be higher than official estimates for urban areas, and lower for rural areas. Differences for each local authority area range from 5 per cent below to 4 per cent above official estimates. It is therefore possible to produce Scottish population estimates purely from administrative sources that roughly agree with MYEs. Further investigation will help understand the differences for particular groups, and will be explored in future years by comparing ABPEs with the 2022 Census.
{"title":"Administrative Data Based Population Estimates for Scotland.","authors":"Dave T. Rowley, C. Ellis","doi":"10.23889/ijpds.v7i3.1816","DOIUrl":"https://doi.org/10.23889/ijpds.v7i3.1816","url":null,"abstract":"In December 2021, we published statistical research on Administrative Data Based Population Estimates (ABPEs) for Scotland’s population in 2016, 2017 and 2018. This work was developed as part of a project to consider how administrative data could be used to support Scotland’s Census. \u0000Following the governance process, administrative datasets were processed and de-identified, before being transferred to Scotland’s National Safe Haven for linking and analysis. The datasets used include data from health, the electoral register, vital events registrations, and education. The methodology used several linking variables so data could be linked, even without exact agreement between records. Records from across the data sources were resolved into individuals using these links. Business rules then indicated which individuals to include in Scotland’s Integrated Demographic Dataset (SIDD). The ABPEs were then produced from this and compared with the official mid-year population estimates (MYEs) to determine success. \u0000On aggregate, the population estimates from the ABPEs are very similar to the MYEs, differing by less than 0.5 per cent in each year. When broken down further, larger differences occur with ABPEs having more males and fewer people aged over 65 when compared with the official statistics. A notable difference between the two is for males aged between 30 and 65 in deprived areas, with ABPEs up to 20% higher than the MYEs. These differences by deprivation are smaller for other age ranges and for females. The ABPEs tend to be higher than official estimates for urban areas, and lower for rural areas. Differences for each local authority area range from 5 per cent below to 4 per cent above official estimates. \u0000It is therefore possible to produce Scottish population estimates purely from administrative sources that roughly agree with MYEs. Further investigation will help understand the differences for particular groups, and will be explored in future years by comparing ABPEs with the 2022 Census.","PeriodicalId":36483,"journal":{"name":"International Journal of Population Data Science","volume":" ","pages":""},"PeriodicalIF":0.0,"publicationDate":"2022-08-25","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"45401581","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2022-08-25DOI: 10.23889/ijpds.v7i3.1813
E. Duku, Molly M. Pottruff, M. Janus
ObjectivesThe Early Development Instrument (EDI) is a valid and reliable population-level tool measuring child developmental vulnerability in Kindergarten. The objective of this study was to derive and validate new EDI-based development “cumulative vulnerability” risk indicators using a cumulative risk index approach (Rutter, 1979). ApproachThe EDI has two main outcome measures: individual domain scores and vulnerability (scoring below a 10% cutpoint). To account for more complexity, we derived two new “cumulative vulnerability” measures. The Mean EDI Domain Score (MEDS) is the mean of the domain scores, and the Total EDI Vulnerability Index (TEVI) is an ordinal summative measure using domain vulnerability indicators. In Study I, we examined the relationship of the MEDS and TEVI measures with neighbourhood-level SES. In Study II, we examined the predictive/explanatory power of the MEDS and TEVI measures with Grade 3 provincial assessments in Ontario, Canada. ResultsStudy I used EDI Kindergarten data from twelve provincial and territorial data collections between 2008 and 2013 in Canada (316,015 children) aggregated to 2,038 customized neighbourhoods. The two new cumulative vulnerability measures worked as expected, with positive association between MEDS and neighbourhood SES (r=0.58), and a negative association between TEVI and neighbourhood SES (r=-0.57). Study II used data from 61,039 Kindergarten children matched between the EDI and Grade 3 EQAO datasets. The predictive/explanatory power of Mean EDI Domain Scores (MEDS; R2=0.11 to 0.15) was twice that of new ordinal summative measure (TEVI; R2=0.06 to 0.08). Interestingly, the predictive power of the TEVI was similar to that of the composite EDI outcome measure, overall vulnerability (vulnerable on one or more domains). ConclusionThe MEDS and TEVI work as expected and can be used for research and reporting purposes. More specifically, the TEVI can also be used as a severity metric evaluating the impact of multiple developmental vulnerabilities. It is recommended that further research be conducted to validate the measures with other datasets.
{"title":"Creating and Evaluating Two Cumulative Developmental Vulnerability Risk Measures.","authors":"E. Duku, Molly M. Pottruff, M. Janus","doi":"10.23889/ijpds.v7i3.1813","DOIUrl":"https://doi.org/10.23889/ijpds.v7i3.1813","url":null,"abstract":"ObjectivesThe Early Development Instrument (EDI) is a valid and reliable population-level tool measuring child developmental vulnerability in Kindergarten. The objective of this study was to derive and validate new EDI-based development “cumulative vulnerability” risk indicators using a cumulative risk index approach (Rutter, 1979). \u0000ApproachThe EDI has two main outcome measures: individual domain scores and vulnerability (scoring below a 10% cutpoint). To account for more complexity, we derived two new “cumulative vulnerability” measures. The Mean EDI Domain Score (MEDS) is the mean of the domain scores, and the Total EDI Vulnerability Index (TEVI) is an ordinal summative measure using domain vulnerability indicators. In Study I, we examined the relationship of the MEDS and TEVI measures with neighbourhood-level SES. In Study II, we examined the predictive/explanatory power of the MEDS and TEVI measures with Grade 3 provincial assessments in Ontario, Canada. \u0000ResultsStudy I used EDI Kindergarten data from twelve provincial and territorial data collections between 2008 and 2013 in Canada (316,015 children) aggregated to 2,038 customized neighbourhoods. The two new cumulative vulnerability measures worked as expected, with positive association between MEDS and neighbourhood SES (r=0.58), and a negative association between TEVI and neighbourhood SES (r=-0.57). Study II used data from 61,039 Kindergarten children matched between the EDI and Grade 3 EQAO datasets. The predictive/explanatory power of Mean EDI Domain Scores (MEDS; R2=0.11 to 0.15) was twice that of new ordinal summative measure (TEVI; R2=0.06 to 0.08). Interestingly, the predictive power of the TEVI was similar to that of the composite EDI outcome measure, overall vulnerability (vulnerable on one or more domains). \u0000ConclusionThe MEDS and TEVI work as expected and can be used for research and reporting purposes. More specifically, the TEVI can also be used as a severity metric evaluating the impact of multiple developmental vulnerabilities. It is recommended that further research be conducted to validate the measures with other datasets.","PeriodicalId":36483,"journal":{"name":"International Journal of Population Data Science","volume":" ","pages":""},"PeriodicalIF":0.0,"publicationDate":"2022-08-25","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"45574634","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2022-08-25DOI: 10.23889/ijpds.v7i3.1827
Marta Wilk, C. Dezateux, S. Liverani, G. Harper
ObjectivesHousehold overcrowding is associated with adverse health outcomes, including increased risk of infectious diseases, mental health problems, and poor educational attainment. We investigated inequalities in overcrowding in an urban, ethnically diverse, and disadvantaged London population by pseudonymously linking electronic health records (EHR) to Energy Performance Certificates (EPC) data. ApproachWe used pseudonymised Unique Property Reference Numbers to link EHRs for 1,066,156 currently registered patients from 321,318 households in north-east London to EPC data. We measured household occupancy and derived the bedroom standard overcrowding definition (number of rooms relative to occupants’ sex and ages) to estimate overcrowding prevalence. We examined associations with: household composition (adults only, single adult+children, ≥2 working-age adults+children, ≥1 retirement-age adults+children, three-generational household); ethnic background (White, South Asian, Black, Mixed, Other, missing); and Index of Multiple Deprivation (IMD) quintile. We used multivariable logistic regression to estimate the adjusted odds (aOR) and 95% Confidence Intervals (CI) of overcrowding. ResultsOverall, 243,793 (22.9%) people were overcrowded. People living in households with children, or three-generational households were more likely (aOR [95% CI] 3.79 [3.74 - 3.84]; 6.53 [6.41 - 6.66] respectively), and single adults or retirement age adults with children less likely (0.36 [0.35 - 0.38]; 0.36 [0.23 - 0.57] respectively), to be overcrowded. Overcrowding was more likely among people from Asian or Black ethnic backgrounds (1.24 [1.22 - 1.25] and 1.17 [1.15 - 1.19] respectively). There was a dose-response relationship between IMD quintile and overcrowding: OR 0.20 [0.20 - 0.21] in the least deprived compared to most deprived quintile. ConclusionOne in five people in north-east London live in overcrowded households with marked inequalities by ethnicity, household generational composition, and deprivation. Up-to-date estimates of household overcrowding can be derived from linked housing and health records and used to evaluate the impact of economic policies on health and housing inequalities.
{"title":"Who lives in overcrowded households in north-east London? Cross-sectional study of linked electronic health records and Energy Performance Certificate register data.","authors":"Marta Wilk, C. Dezateux, S. Liverani, G. Harper","doi":"10.23889/ijpds.v7i3.1827","DOIUrl":"https://doi.org/10.23889/ijpds.v7i3.1827","url":null,"abstract":"ObjectivesHousehold overcrowding is associated with adverse health outcomes, including increased risk of infectious diseases, mental health problems, and poor educational attainment. We investigated inequalities in overcrowding in an urban, ethnically diverse, and disadvantaged London population by pseudonymously linking electronic health records (EHR) to Energy Performance Certificates (EPC) data. \u0000ApproachWe used pseudonymised Unique Property Reference Numbers to link EHRs for 1,066,156 currently registered patients from 321,318 households in north-east London to EPC data. \u0000We measured household occupancy and derived the bedroom standard overcrowding definition (number of rooms relative to occupants’ sex and ages) to estimate overcrowding prevalence. We examined associations with: household composition (adults only, single adult+children, ≥2 working-age adults+children, ≥1 retirement-age adults+children, three-generational household); ethnic background (White, South Asian, Black, Mixed, Other, missing); and Index of Multiple Deprivation (IMD) quintile. We used multivariable logistic regression to estimate the adjusted odds (aOR) and 95% Confidence Intervals (CI) of overcrowding. \u0000ResultsOverall, 243,793 (22.9%) people were overcrowded. People living in households with children, or three-generational households were more likely (aOR [95% CI] 3.79 [3.74 - 3.84]; 6.53 [6.41 - 6.66] respectively), and single adults or retirement age adults with children less likely (0.36 [0.35 - 0.38]; 0.36 [0.23 - 0.57] respectively), to be overcrowded. Overcrowding was more likely among people from Asian or Black ethnic backgrounds (1.24 [1.22 - 1.25] and 1.17 [1.15 - 1.19] respectively). There was a dose-response relationship between IMD quintile and overcrowding: OR 0.20 [0.20 - 0.21] in the least deprived compared to most deprived quintile. \u0000ConclusionOne in five people in north-east London live in overcrowded households with marked inequalities by ethnicity, household generational composition, and deprivation. Up-to-date estimates of household overcrowding can be derived from linked housing and health records and used to evaluate the impact of economic policies on health and housing inequalities.","PeriodicalId":36483,"journal":{"name":"International Journal of Population Data Science","volume":" ","pages":""},"PeriodicalIF":0.0,"publicationDate":"2022-08-25","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"45600204","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2022-08-25DOI: 10.23889/ijpds.v7i3.2019
Shayda Kashef, Marylouise Cowan, Elizabeth Waind
ObjectivesBroad public acceptability of research using data and statistics entails demonstrating trustworthiness and understanding public needs to maximise public benefit. Insight into public attitudes is crucial for informing how to operate in a trustworthy way and within the public interest. ApproachIn 2022, two UK-wide public dialogues were undertaken to explore attitudes regarding the use of data and statistics in research. The first, explored views towards the creation of a more joined-up, efficient and trustworthy national data research infrastructure. The second aimed at understanding public perceptions of ‘public good’ of data and statistics. Both took a deliberative approach and involved diverse participation from members of the UK public. Both also sought the expertise of stakeholders within the data and statistics communities to inform the process. ResultsThe first dialogue found that the public want: more proactive transparency around data research processes, with greater efforts made to raise awareness about it and the security processes in place; meaningful and inclusive public involvement and engagement; a more standardised, centralised and unified approach to data research across the UK; and for sensitive data to be made safely and securely available to researchers for projects in the public benefit, regardless of whether those researchers sit within academia, government or the private sector. The second dialogue launched in March; results will be publicly available by the middle of summer. The findings of this dialogue are intended to influence policy and process relating to the use of data and statistics. ConclusionThe first dialogue emphasises that the public are supportive of data research. And whilst reassured by security processes, they also do not want them to unduly hinder public benefit. Attendees at the IPDLN Conference will be some of the first to hear about the findings from the second dialogue.
{"title":"Public attitudes to population data research in 2022.","authors":"Shayda Kashef, Marylouise Cowan, Elizabeth Waind","doi":"10.23889/ijpds.v7i3.2019","DOIUrl":"https://doi.org/10.23889/ijpds.v7i3.2019","url":null,"abstract":"ObjectivesBroad public acceptability of research using data and statistics entails demonstrating trustworthiness and understanding public needs to maximise public benefit. Insight into public attitudes is crucial for informing how to operate in a trustworthy way and within the public interest.\u0000ApproachIn 2022, two UK-wide public dialogues were undertaken to explore attitudes regarding the use of data and statistics in research. The first, explored views towards the creation of a more joined-up, efficient and trustworthy national data research infrastructure. The second aimed at understanding public perceptions of ‘public good’ of data and statistics. Both took a deliberative approach and involved diverse participation from members of the UK public. Both also sought the expertise of stakeholders within the data and statistics communities to inform the process.\u0000ResultsThe first dialogue found that the public want:\u0000\u0000more proactive transparency around data research processes, with greater efforts made to raise awareness about it and the security processes in place;\u0000meaningful and inclusive public involvement and engagement;\u0000a more standardised, centralised and unified approach to data research across the UK; and\u0000for sensitive data to be made safely and securely available to researchers for projects in the public benefit, regardless of whether those researchers sit within academia, government or the private sector.\u0000\u0000The second dialogue launched in March; results will be publicly available by the middle of summer. The findings of this dialogue are intended to influence policy and process relating to the use of data and statistics.\u0000ConclusionThe first dialogue emphasises that the public are supportive of data research. And whilst reassured by security processes, they also do not want them to unduly hinder public benefit. Attendees at the IPDLN Conference will be some of the first to hear about the findings from the second dialogue.","PeriodicalId":36483,"journal":{"name":"International Journal of Population Data Science","volume":" ","pages":""},"PeriodicalIF":0.0,"publicationDate":"2022-08-25","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"43071232","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2022-08-25DOI: 10.23889/ijpds.v7i3.2069
J. Das-Munshi, A. Dregan, R. Stewart, M. Hotopf, I. Bakolis, L. Bécares, J. Ocloo, R. Stuart, E. Impara
BackgroundThe association of COVID-19 infection with death in people with severe mental illnesses (SMI), and the relationship to multimorbidities/ underlying health conditions ethnicity is unclear. Health records linked to COVID-19 tests data could help to inform this knowledge gap. ObjectiveTo determine the risk of death in people with SMI following COVID-19 infection compared to reference groups and assess whether excess mortality is accounted through underlying health conditions or further elevated in minority ethnic groups. Design, setting and participantsNationally representative cohort study using primary care data from the Clinical Practice Research Database (CPRD), with participants followed from the start of the pandemic in 2020, for 1.5 years, covering England, Wales and Northern Ireland. For consenting practices, CPRD data was linked to COVID-19 data Public Health England (PHE) Second Generation Surveillance System (SGSS), PHE COVID-19 Hospitalisation in England Surveillance System (CHESS), and Intensive Care National Audit and Research Centre (ICNARC) data on COVID-10 intensive care admissions. The cohort comprised 795,836 individuals, with 7,493 individuals with SMI and a positive COVID-19 test (“SMI/COVID-19”). Comparison groups were: 2,325 individuals with SMI/ testing negative for COVID-19 (“SMI/ non COVID-19”), 657,414 individuals from a non-SMI group/ testing positive for COVID-19 (“non-SMI/ COVID-19”), and 128,604 individuals from a non-SMI group/ testing negative for COVID-19 (“non-SMI/ non-COVID-19”). ExposuresSMI defined as the presence of schizophrenia, schizoaffective disorder, bipolar disorder, or affective disorders with psychosis, according to the International Classification of Mental Disorders (ICD-10). COVID-19 diagnoses identified through confirmed laboratory tests and clinical diagnoses. OutcomesAll-cause mortality ResultsA higher proportion of SMI patients with COVID-19 were obese (37% versus 22% in the non-SMI/non-COVID-19 group), current smokers (27% versus 23% in the non-SMI/non-COVID-19 group), had underlying health conditions, and were Black Caribbean/ Black African (5% versus 1% in the non-SMI/non-COVID-19 group). Relative to the non-SMI/ non-COVID-19 group, the SMI/ COVID-19 group had an elevated risk of death (age and sex-adjusted hazard ratio (aHR) 5.03 (95%CI: 4.61-5.54)). This was elevated to a lesser extent, in the SMI/ non COVID-19 group (aHR: 1.93 (95%CI: 1.54-2.41)) and in the non-SMI/ COVID-19 group (aHR: 2.85 (95%CI: 2.72-2.98). Excess risk persisted after adjusting for tobacco use, weight and comorbidities. Mortality trends were similar across groups by ethnicity. Risk of death was highest for the SMI/ COVID-19 group during the first wave of infection in the UK, however excess mortality was still evident and substantially elevated at the second wave also. ConclusionsPeople living with SMI are at an increased risk of death compared to population controls; this excess risk is further elevated
{"title":"Severe mental illnesses and mortality following COVID-19 infection: Data linkage study using the Clinical Practice Research Database (CPRD).","authors":"J. Das-Munshi, A. Dregan, R. Stewart, M. Hotopf, I. Bakolis, L. Bécares, J. Ocloo, R. Stuart, E. Impara","doi":"10.23889/ijpds.v7i3.2069","DOIUrl":"https://doi.org/10.23889/ijpds.v7i3.2069","url":null,"abstract":"BackgroundThe association of COVID-19 infection with death in people with severe mental illnesses (SMI), and the relationship to multimorbidities/ underlying health conditions ethnicity is unclear. Health records linked to COVID-19 tests data could help to inform this knowledge gap. \u0000ObjectiveTo determine the risk of death in people with SMI following COVID-19 infection compared to reference groups and assess whether excess mortality is accounted through underlying health conditions or further elevated in minority ethnic groups. \u0000Design, setting and participantsNationally representative cohort study using primary care data from the Clinical Practice Research Database (CPRD), with participants followed from the start of the pandemic in 2020, for 1.5 years, covering England, Wales and Northern Ireland. For consenting practices, CPRD data was linked to COVID-19 data Public Health England (PHE) Second Generation Surveillance System (SGSS), PHE COVID-19 Hospitalisation in England Surveillance System (CHESS), and Intensive Care National Audit and Research Centre (ICNARC) data on COVID-10 intensive care admissions. The cohort comprised 795,836 individuals, with 7,493 individuals with SMI and a positive COVID-19 test (“SMI/COVID-19”). Comparison groups were: 2,325 individuals with SMI/ testing negative for COVID-19 (“SMI/ non COVID-19”), 657,414 individuals from a non-SMI group/ testing positive for COVID-19 (“non-SMI/ COVID-19”), and 128,604 individuals from a non-SMI group/ testing negative for COVID-19 (“non-SMI/ non-COVID-19”). \u0000ExposuresSMI defined as the presence of schizophrenia, schizoaffective disorder, bipolar disorder, or affective disorders with psychosis, according to the International Classification of Mental Disorders (ICD-10). COVID-19 diagnoses identified through confirmed laboratory tests and clinical diagnoses. \u0000OutcomesAll-cause mortality \u0000ResultsA higher proportion of SMI patients with COVID-19 were obese (37% versus 22% in the non-SMI/non-COVID-19 group), current smokers (27% versus 23% in the non-SMI/non-COVID-19 group), had underlying health conditions, and were Black Caribbean/ Black African (5% versus 1% in the non-SMI/non-COVID-19 group). Relative to the non-SMI/ non-COVID-19 group, the SMI/ COVID-19 group had an elevated risk of death (age and sex-adjusted hazard ratio (aHR) 5.03 (95%CI: 4.61-5.54)). This was elevated to a lesser extent, in the SMI/ non COVID-19 group (aHR: 1.93 (95%CI: 1.54-2.41)) and in the non-SMI/ COVID-19 group (aHR: 2.85 (95%CI: 2.72-2.98). Excess risk persisted after adjusting for tobacco use, weight and comorbidities. Mortality trends were similar across groups by ethnicity. Risk of death was highest for the SMI/ COVID-19 group during the first wave of infection in the UK, however excess mortality was still evident and substantially elevated at the second wave also. \u0000ConclusionsPeople living with SMI are at an increased risk of death compared to population controls; this excess risk is further elevated","PeriodicalId":36483,"journal":{"name":"International Journal of Population Data Science","volume":" 23","pages":""},"PeriodicalIF":0.0,"publicationDate":"2022-08-25","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"41252215","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2022-08-25DOI: 10.23889/ijpds.v7i3.2095
J. Wilton, Jalud Abdulmenan, M. Chong, A. Becerra, Mike Coss, Marsha Taylor, O. Djurdjev, D. Rasali, H. Sbihi, M. Krajden, A. Flatt, Seyed Ali Mussavi Rizi, N. Janjua
ObjectivesThe COVID-19 pandemic has necessitated access to large health system datasets to inform the public health response. To meet this need, the Provincial Health Services Authority and the British Columbia (BC) Ministry of Health collaborated to create a population-based platform that integrates COVID-19 datasets with sociodemographic and administrative health data. ApproachA BC COVID Data Library proof-of-concept was created as a cloud-based, dynamic platform composed of de-identified datasets. The BC COVID-19 Cohort (BCC19C) represents a subset composed of people accessing COVID-19 health services (e.g., testing, vaccination) and linked health histories. Provincial COVID-19 datasets are updated daily and include COVID-19 lab tests, case surveillance, vaccinations and hospitalizations/deaths. These can be linked to administrative data holdings for the BC population, which are updated weekly/monthly and include vital statistics, medications, hospital admissions, medical visits, among others. A patient matching algorithm creates unique patient keys that allows the same individual to be linked across datasets. ResultsThe BCC19C has been used provincially to 1) support ongoing surveillance, reporting, and modelling of COVID-19; 2) describe and characterize the epidemiology of COVID-19; and 3) inform acute care planning, public health interventions and health care services in BC. Ongoing and completed BCC19C analyses include assessment of vaccine safety, vaccine effectiveness, and characteristics associated with infection and severe outcomes; use of medical visit data for syndromic surveillance and monitoring of unintended outcomes of the pandemic (e.g., mental health visits); and characterization of long-COVID. Availability of linked administrative data holdings has been crucial for identifying non-COVID control groups, measuring sociodemographics and co-morbidities, and complementing COVID-19 datasets for more complete capture of health outcomes (e.g., deaths, hospitalizations). ConclusionsThe large scope/breadth and timeliness of the linkable datasets integrated within the COVID Data Library and the BCC19C has supported the public health response in BC. Additional linkage to other data sources will further strengthen this data platform.
{"title":"A large linked data platform to inform the COVID-19 response in British Columbia: The BC COVID-19 Cohort.","authors":"J. Wilton, Jalud Abdulmenan, M. Chong, A. Becerra, Mike Coss, Marsha Taylor, O. Djurdjev, D. Rasali, H. Sbihi, M. Krajden, A. Flatt, Seyed Ali Mussavi Rizi, N. Janjua","doi":"10.23889/ijpds.v7i3.2095","DOIUrl":"https://doi.org/10.23889/ijpds.v7i3.2095","url":null,"abstract":"ObjectivesThe COVID-19 pandemic has necessitated access to large health system datasets to inform the public health response. To meet this need, the Provincial Health Services Authority and the British Columbia (BC) Ministry of Health collaborated to create a population-based platform that integrates COVID-19 datasets with sociodemographic and administrative health data. \u0000ApproachA BC COVID Data Library proof-of-concept was created as a cloud-based, dynamic platform composed of de-identified datasets. The BC COVID-19 Cohort (BCC19C) represents a subset composed of people accessing COVID-19 health services (e.g., testing, vaccination) and linked health histories. Provincial COVID-19 datasets are updated daily and include COVID-19 lab tests, case surveillance, vaccinations and hospitalizations/deaths. These can be linked to administrative data holdings for the BC population, which are updated weekly/monthly and include vital statistics, medications, hospital admissions, medical visits, among others. A patient matching algorithm creates unique patient keys that allows the same individual to be linked across datasets. \u0000ResultsThe BCC19C has been used provincially to 1) support ongoing surveillance, reporting, and modelling of COVID-19; 2) describe and characterize the epidemiology of COVID-19; and 3) inform acute care planning, public health interventions and health care services in BC. Ongoing and completed BCC19C analyses include assessment of vaccine safety, vaccine effectiveness, and characteristics associated with infection and severe outcomes; use of medical visit data for syndromic surveillance and monitoring of unintended outcomes of the pandemic (e.g., mental health visits); and characterization of long-COVID. Availability of linked administrative data holdings has been crucial for identifying non-COVID control groups, measuring sociodemographics and co-morbidities, and complementing COVID-19 datasets for more complete capture of health outcomes (e.g., deaths, hospitalizations). \u0000ConclusionsThe large scope/breadth and timeliness of the linkable datasets integrated within the COVID Data Library and the BCC19C has supported the public health response in BC. Additional linkage to other data sources will further strengthen this data platform.","PeriodicalId":36483,"journal":{"name":"International Journal of Population Data Science","volume":" ","pages":""},"PeriodicalIF":0.0,"publicationDate":"2022-08-25","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"42305491","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2022-08-25DOI: 10.23889/ijpds.v7i3.1791
T. Whiffen
ObjectivesShielding was introduced as part of the UK government’s response to the SARS-CoV-2 pandemic to protect Clinically Extremely Vulnerable (CEV) people from infection and serious illness. Various research questions emerged in relation to non-clinical vulnerabilities of those shielding which could be addressed by utilising available health and administrative data. ApproachThe Shielded Patient List (SPL) was linked with various datasets on the UK Secure Electronic Research Platform (UKSERP) including the Pupil Level Annual School Census (PLASC), National Survey and Ordnance Survey data for Wales. Some of these were anonymised datasets contained in the Secure Anonymised Information Linkage (SAIL) databank. Algorithms were applied to determine household composition and whether private outdoor space was available for the shielding group. Results were then extracted for Wales broken down by local authority. ResultsResults from the various strands of research related to shielding will be presented covering provision of outdoor space, household characteristics and composition. ConclusionThese analyses demonstrate how population-level data resources can be leveraged quickly to answer newly-emerging policy questions as part of the response to the SARS-CoV-2 pandemic.
{"title":"The Impact of Shielding Policy in Wales.","authors":"T. Whiffen","doi":"10.23889/ijpds.v7i3.1791","DOIUrl":"https://doi.org/10.23889/ijpds.v7i3.1791","url":null,"abstract":"ObjectivesShielding was introduced as part of the UK government’s response to the SARS-CoV-2 pandemic to protect Clinically Extremely Vulnerable (CEV) people from infection and serious illness. Various research questions emerged in relation to non-clinical vulnerabilities of those shielding which could be addressed by utilising available health and administrative data. \u0000ApproachThe Shielded Patient List (SPL) was linked with various datasets on the UK Secure Electronic Research Platform (UKSERP) including the Pupil Level Annual School Census (PLASC), National Survey and Ordnance Survey data for Wales. Some of these were anonymised datasets contained in the Secure Anonymised Information Linkage (SAIL) databank. Algorithms were applied to determine household composition and whether private outdoor space was available for the shielding group. Results were then extracted for Wales broken down by local authority. \u0000ResultsResults from the various strands of research related to shielding will be presented covering provision of outdoor space, household characteristics and composition. \u0000ConclusionThese analyses demonstrate how population-level data resources can be leveraged quickly to answer newly-emerging policy questions as part of the response to the SARS-CoV-2 pandemic.","PeriodicalId":36483,"journal":{"name":"International Journal of Population Data Science","volume":" ","pages":""},"PeriodicalIF":0.0,"publicationDate":"2022-08-25","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"42192338","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2022-08-25DOI: 10.23889/ijpds.v7i3.1997
Nasir Rajah, L. Calderwood, B. D. De Stavola, K. Harron, G. Ploubidis, R. Silverwood
ObjectivesThere is growing interest in whether linked administrative data have the potential to aid analyses subject to missing data in cohort studies. We aimed to identify predictors of cohort non-response in linked administrative data and examine whether inclusion of these variables in principled methods for missing data handling can help restore sample representativeness. ApproachUsing linked 1958 National Child Development Study (NCDS) and Hospital Episode Statistics (HES) data, we applied a multi-stage data-driven approach to identify HES variable which are predictive of non-response at the age 55 sweep of NCDS. We then included these variables as auxiliary variables in multiple imputation (MI) analyses to see if they helped restore sample representativeness in terms of early life variables which were essentially fully observed in NCDS (mother’s husband’s social class at birth, cognitive ability at age 7) and relative to external population data (educational qualifications at age 55, marital status at age 55). ResultsWe took as our starting point 57 variables derived from HES data based on the presence or number of different types of appointments/admissions, diagnostic codes and treatment codes. After application of our multi-stage data-driven approach we identified five HES variables that were predictive of non-response at age 55 in NCDS. For example, cohort members who had been treated for adult mental illness were almost 3 times as likely to be non-respondents (risk ratio 2.81; 95% confidence interval 2.05, 3.86). Inclusion of these variables in MI analyses did help restore sample representativeness. However, there was no additional gain in sample representativeness relative to analyses using only previously identified survey predictors of non-response (i.e. NCDS rather than HES variables). ConclusionIn our applications, inclusion of HES predictors of NCDS non-response in analyses did not improve sample representativeness beyond that possible using survey variables alone. Whilst this finding may not extend to other analyses or NCDS sweeps, it highlights the utility of survey variables in handling non-response.
{"title":"Using linked Hospital Episode Statistics data to aid the handling of non-response and restore sample representativeness in the 1958 National Child Development Study.","authors":"Nasir Rajah, L. Calderwood, B. D. De Stavola, K. Harron, G. Ploubidis, R. Silverwood","doi":"10.23889/ijpds.v7i3.1997","DOIUrl":"https://doi.org/10.23889/ijpds.v7i3.1997","url":null,"abstract":"ObjectivesThere is growing interest in whether linked administrative data have the potential to aid analyses subject to missing data in cohort studies. We aimed to identify predictors of cohort non-response in linked administrative data and examine whether inclusion of these variables in principled methods for missing data handling can help restore sample representativeness. \u0000ApproachUsing linked 1958 National Child Development Study (NCDS) and Hospital Episode Statistics (HES) data, we applied a multi-stage data-driven approach to identify HES variable which are predictive of non-response at the age 55 sweep of NCDS. We then included these variables as auxiliary variables in multiple imputation (MI) analyses to see if they helped restore sample representativeness in terms of early life variables which were essentially fully observed in NCDS (mother’s husband’s social class at birth, cognitive ability at age 7) and relative to external population data (educational qualifications at age 55, marital status at age 55). \u0000ResultsWe took as our starting point 57 variables derived from HES data based on the presence or number of different types of appointments/admissions, diagnostic codes and treatment codes. After application of our multi-stage data-driven approach we identified five HES variables that were predictive of non-response at age 55 in NCDS. For example, cohort members who had been treated for adult mental illness were almost 3 times as likely to be non-respondents (risk ratio 2.81; 95% confidence interval 2.05, 3.86). Inclusion of these variables in MI analyses did help restore sample representativeness. However, there was no additional gain in sample representativeness relative to analyses using only previously identified survey predictors of non-response (i.e. NCDS rather than HES variables). \u0000ConclusionIn our applications, inclusion of HES predictors of NCDS non-response in analyses did not improve sample representativeness beyond that possible using survey variables alone. Whilst this finding may not extend to other analyses or NCDS sweeps, it highlights the utility of survey variables in handling non-response.","PeriodicalId":36483,"journal":{"name":"International Journal of Population Data Science","volume":" ","pages":""},"PeriodicalIF":0.0,"publicationDate":"2022-08-25","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"41603735","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2022-08-25DOI: 10.23889/ijpds.v7i3.1950
R. Urquhart, P. Awadalla, P. Bhatti, T. Dummer, S. Gravel, J. Vena, R. Alvi, P. Broet, C. Kendell, Victoria A. Kirsh, G. Lettre, Kimberly Skead, G. Shen-Tu, E. Sweeney, D. Turner
ObjectivesWe will enrich the cancer research ecosystem in Canada through linking cancer registry and administrative health data to the Canadian Partnership for Tomorrow’s Health (CanPath) cohort and biobank. CanPath is Canada’s largest population health study, including 1% of the Canadian population, which seeks to investigate cancer development. ApproachWe are achieving record-level linkage of the CanPath harmonized dataset to provincial cancer registry data, and hospitalization and ambulatory care data from the Canadian Institutes of Health Information (CIHI). The CanPATH harmonized dataset includes comprehensive genetics, environment, lifestyle, and behaviour data. Our linkage activities will result in interprovincial data sharing, with centrally-held linked data, a first in Canadian history. We will demonstrate the CanPath-cancer registry-CIHI linkage potential by investigating the impact of the COVID-19 pandemic on healthcare utilization and outcomes among those with cancer. ResultsThe linkage is ongoing and anticipated to be completed by September 2022. Linked data will be made available through the CanPath Data Safe Haven, a cloud-based solution that meets the legal requirements of the data sharing agreements and provincial privacy policies, and is accessible to researchers through secure access. The CanPath Data Safe Haven will be a federated data platform for Canadian researchers to access, analyze, and contribute research in a collaborative environment. By linking these datasets, this project will: address concerns related to accessibility of cancer data in Canada; bring more value to existing data; support an enhanced understanding of the impacts of cancer on marginalized populations; and create a more integrated approach to cancer data access and management. ConclusionCanPath will be the first program in Canadian history to combine the wealth of cohort resources with cancer registry and administrative health data in a central location at a national scale. We will provide a single point of access for researchers to conduct novel investigations into cancer development and outcomes.
{"title":"Harnessing the power of data linkage to enrich the cancer research ecosystem in Canada.","authors":"R. Urquhart, P. Awadalla, P. Bhatti, T. Dummer, S. Gravel, J. Vena, R. Alvi, P. Broet, C. Kendell, Victoria A. Kirsh, G. Lettre, Kimberly Skead, G. Shen-Tu, E. Sweeney, D. Turner","doi":"10.23889/ijpds.v7i3.1950","DOIUrl":"https://doi.org/10.23889/ijpds.v7i3.1950","url":null,"abstract":"ObjectivesWe will enrich the cancer research ecosystem in Canada through linking cancer registry and administrative health data to the Canadian Partnership for Tomorrow’s Health (CanPath) cohort and biobank. CanPath is Canada’s largest population health study, including 1% of the Canadian population, which seeks to investigate cancer development. \u0000ApproachWe are achieving record-level linkage of the CanPath harmonized dataset to provincial cancer registry data, and hospitalization and ambulatory care data from the Canadian Institutes of Health Information (CIHI). The CanPATH harmonized dataset includes comprehensive genetics, environment, lifestyle, and behaviour data. Our linkage activities will result in interprovincial data sharing, with centrally-held linked data, a first in Canadian history. We will demonstrate the CanPath-cancer registry-CIHI linkage potential by investigating the impact of the COVID-19 pandemic on healthcare utilization and outcomes among those with cancer. \u0000ResultsThe linkage is ongoing and anticipated to be completed by September 2022. Linked data will be made available through the CanPath Data Safe Haven, a cloud-based solution that meets the legal requirements of the data sharing agreements and provincial privacy policies, and is accessible to researchers through secure access. The CanPath Data Safe Haven will be a federated data platform for Canadian researchers to access, analyze, and contribute research in a collaborative environment. By linking these datasets, this project will: address concerns related to accessibility of cancer data in Canada; bring more value to existing data; support an enhanced understanding of the impacts of cancer on marginalized populations; and create a more integrated approach to cancer data access and management. \u0000ConclusionCanPath will be the first program in Canadian history to combine the wealth of cohort resources with cancer registry and administrative health data in a central location at a national scale. We will provide a single point of access for researchers to conduct novel investigations into cancer development and outcomes.","PeriodicalId":36483,"journal":{"name":"International Journal of Population Data Science","volume":" ","pages":""},"PeriodicalIF":0.0,"publicationDate":"2022-08-25","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"41610896","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2022-08-25DOI: 10.23889/ijpds.v7i3.1864
S. Guthrie, Tara Alexander
ObjectiveTo understand the pathway and diverse levels of functional impairment for people with traumatic brain injury (TBI) and spinal cord injury (SCI) as they transition from inpatient rehabilitation (IR) hospital setting to ongoing care and support under the National Disability Insurance Scheme (NDIS) in Australia. ApproachThe Australasian Rehabilitation Outcomes Centre (AROC) has data on almost every inpatient rehabilitation episode of care since 2002, including TBI and SCI. AROC data is de-identified with a statistical linkage key (SLK-581). The NDIS dataset contains identified administrative data on participants of the scheme from which the SLK-581 was derived. Datasets were restricted to TBI and SCI records and the SLK with key dates used to link records together. The linkage was done in multiple passes with different levels of information with each link being validated using secondary information relating to date of injury, date of admission and geographical location. ResultsOver the period 2012-2019, approximately 2,000 records from AROC episodes were linked to an NDIS participant following data validation and individual review of borderline matches We will compare the functional independence of the individual upon leaving rehabilitation with their need for support under the NDIS. Functional independence in rehabilitation is measured by clinicians using the Functional Independence Measure (FIM), a tool that requires clinicians to be trained and credentialed in its use as it is part of the funding model for IR in Australia. Need for support under the NDIS is measured by the funded supports available to a participant under the plan. We expect to demonstrate a correlation between FIM scores and funded supports and identify and analyse any unexpected results. ConclusionThese results will inform resource allocation within the NDIS. This project demonstrates how de-identified research datasets can be linked with administrative datasets to draw new and powerful insights into government service delivery and population health while maintaining privacy. Challenges to accurate linkage can be overcome through iterative and non-deterministic approaches.
{"title":"Transition between inpatient rehabilitation and National Disability Insurance Scheme for Traumatic Brain Injury and Spinal Cord Injury.","authors":"S. Guthrie, Tara Alexander","doi":"10.23889/ijpds.v7i3.1864","DOIUrl":"https://doi.org/10.23889/ijpds.v7i3.1864","url":null,"abstract":"ObjectiveTo understand the pathway and diverse levels of functional impairment for people with traumatic brain injury (TBI) and spinal cord injury (SCI) as they transition from inpatient rehabilitation (IR) hospital setting to ongoing care and support under the National Disability Insurance Scheme (NDIS) in Australia. \u0000ApproachThe Australasian Rehabilitation Outcomes Centre (AROC) has data on almost every inpatient rehabilitation episode of care since 2002, including TBI and SCI. AROC data is de-identified with a statistical linkage key (SLK-581). The NDIS dataset contains identified administrative data on participants of the scheme from which the SLK-581 was derived. Datasets were restricted to TBI and SCI records and the SLK with key dates used to link records together. The linkage was done in multiple passes with different levels of information with each link being validated using secondary information relating to date of injury, date of admission and geographical location. \u0000ResultsOver the period 2012-2019, approximately 2,000 records from AROC episodes were linked to an NDIS participant following data validation and individual review of borderline matches We will compare the functional independence of the individual upon leaving rehabilitation with their need for support under the NDIS. Functional independence in rehabilitation is measured by clinicians using the Functional Independence Measure (FIM), a tool that requires clinicians to be trained and credentialed in its use as it is part of the funding model for IR in Australia. Need for support under the NDIS is measured by the funded supports available to a participant under the plan. We expect to demonstrate a correlation between FIM scores and funded supports and identify and analyse any unexpected results. \u0000ConclusionThese results will inform resource allocation within the NDIS. This project demonstrates how de-identified research datasets can be linked with administrative datasets to draw new and powerful insights into government service delivery and population health while maintaining privacy. Challenges to accurate linkage can be overcome through iterative and non-deterministic approaches.","PeriodicalId":36483,"journal":{"name":"International Journal of Population Data Science","volume":" ","pages":""},"PeriodicalIF":0.0,"publicationDate":"2022-08-25","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"44035140","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
京公网安备 11010802042870号
京ICP备2023020795号-1
扫码关注我们
求助内容：
应助结果提醒方式：