Administrative Data Sources Research Articles

Pretrial risk assessment instruments in practice: The role of judicial discretion in pretrial reform

AbstractResearch SummaryWe explored the extent to which the implementation of a pretrial risk assessment instrument (PRAI) corresponded to changes in the pretrial processing of defendants using multiple administrative data sources from a large county in the southeastern United States. Our findings revealed little evidence of reductions in detention lengths or increases in the use of nonfinancial forms of release following the tool's adoption. This was largely attributable to the exercise of judicial discretion, as judges frequently departed from the tool's recommendation using alternatives that were more punitive and often included financial conditions—particularly for Black and Latino defendants. Furthermore, the exercise of discretion was linked to increased rates of pretrial failure.Policy ImplicationsPRAIs were adopted on a massive scale with the understanding that they are evidence‐based and geared toward efficiently and equitably reducing pretrial populations; however, we are lacking the evaluative work to determine their impacts. Our findings suggest that PRAIs may not only undermine reform efforts, but may worsen disparities, if communities fail to complete the up‐front work of discussing their expectations for pretrial decision making, including the conditions under which financial constraints may be justifiable.

A population-based study of administrative data linkage to measure melanoma surgical and pathology quality.

Continuous quality improvement is important for cancer systems. However, collecting and compiling quality indicator data can be time-consuming and resource-intensive. Here we explore the utility and feasibility of linked routinely collected health data to capture key elements of quality of care for melanoma in a single-payer, universal health care setting. This pilot study utilized a retrospective population-based cohort from a previously developed linked administrative data set, with a 65% random sample of all invasive cutaneous melanoma cases diagnosed 2007-2012 in the province of Ontario. Data from the Ontario Cancer Registry was utilized, supplemented with linked pathology report data from Cancer Care Ontario, and other linked administrative data describing health care utilization. Quality indicators identified through provincial guidelines and international consensus were evaluated for potential collection with administrative data and measured where possible. A total of 7,654 cases of melanoma were evaluated. Ten of 25 (40%) candidate quality indicators were feasible to be collected with the available administrative data. Many indicators (8/25) could not be measured due to unavailable clinical information (e.g. width of clinical margins). Insufficient pathology information (6/25) or health structure information (1/25) were less common reasons. Reporting of recommended variables in pathology reports varied from 65.2% (satellitosis) to 99.6% (body location). For stage IB-II or T1b-T4a melanoma patients where SLNB should be discussed, approximately two-thirds met with a surgeon experienced in SLNB. Of patients undergoing full lymph node dissection, 76.2% had adequate evaluation of the basin. We found that use of linked administrative data sources is feasible for measurement of melanoma quality in some cases. In those cases, findings suggest opportunities for quality improvement. Consultation with surgeons offering SLNB was limited, and pathology report completeness was sub-optimal, but was prior to routine synoptic reporting. However, to measure more quality indicators, text-based data sources will require alternative approaches to manual collection such as natural language processing or standardized collection. We recommend development of robust data platforms to support continuous re-evaluation of melanoma quality indicators, with the goal of optimizing quality of care for melanoma patients on an ongoing basis.

Longitudinal Associations of Mental Disorders With Dementia

Mental disorders are an underappreciated category of modifiable risk factors for dementia. Developing an evidence base about the link between mental disorders and dementia risk requires studies that use large, representative samples, consider the full range of psychiatric conditions, ascertain mental disorders from early life, use long follow-ups, and distinguish between Alzheimer disease and related dementias. To test whether mental disorders antedate dementia across 3 decades of observation. This population-based administrative register study of mental disorders and Alzheimer disease and related dementias included all individuals born in New Zealand between 1928 and 1967 who resided in the country for any time during the 30-year observation period between July 1988 and June 2018. Data were from the New Zealand Integrated Data Infrastructure, a collection of whole-of-population administrative data sources linked at the individual level. Data were analyzed from October 2020 to November 2021. Diagnoses of mental disorders were ascertained from public-hospital records. Diagnoses of dementia were ascertained from public-hospital records, mortality records, and pharmaceutical records. Of 1 711 386 included individuals, 866 301 (50.6%) were male, and individuals were aged 21 to 60 years at baseline. Relative to individuals without a mental disorder, those with a mental disorder were at increased risk of developing subsequent dementia (relative risk [RR], 4.24; 95% CI, 4.07-4.42; hazard ratio, 6.49; 95% CI, 6.25-6.73). Among individuals with dementia, those with a mental disorder developed dementia a mean of 5.60 years (95% CI, 5.31-5.90) earlier than those without a mental disorder. Associations held across sex and age and after accounting for preexisting chronic physical diseases and socioeconomic deprivation. Associations were present across different types of mental disorders and self-harm behavior (RRs ranged from 2.93 [95% CI, 2.66-3.21] for neurotic disorders to 6.20 [95% CI, 5.67-6.78] for psychotic disorders), and were evident for Alzheimer disease (RR, 2.76; 95% CI, 2.45-3.11) and all other dementias (RR, 5.85; 95% CI, 5.58-6.13). In this study, mental disorders were associated with the onset of dementia in the population. Ameliorating mental disorders in early life might also ameliorate neurodegenerative conditions and extend quality of life in old age.

Implementation of the BC Congenital Anomalies Surveillance System (BCCASS).

Congenital anomalies (CAs) can cause lifelong morbidity and accounted for 23.2% of infant deaths from 2003 to 2007. In British Columbia (BC), surveillance of CAs has been irregular since the early 2000s. To enhance CAs surveillance in BC, the Public Health Agency of Canada has provided funding for the implementation of the BC Congenital Anomalies Surveillance System (BCCASS). BCCASS is a population-based surveillance system. The system leverages existing administrative data sources that capture information regarding vital events, disease status, drug prescription, and healthcare utilization. The system uses a series of algorithms to capture specific CAs diagnoses, some of which are further validated with the support of the Provincial Advisory Committee. This Advisory Committee is a multi-stakeholder coalition that includes the BC Office of the Provincial Health Officer, subject matter experts, data partners, users, and academia, and acts to provide support, expertise, and strategic guidance to BCCASS. Through BCCASS, prevalence and historical trends for 35 CAs in BC are available. Information pertaining to maternal place of residence, risk, and protective factors can be used for association studies such as links to environmental hazards and cluster analysis. BCCASS is a cost-effective and sustainable system that leverages existing data sources necessary to understand the overall burden of CAs across the BC population. This is fundamental to support data-driven decisions around policy development, program planning, and evaluation of preventive measures. Strong coalitions with stakeholders are instrumental to ensure successful implementation and expansion in the future.

Administrative data as a source of information about employed persons

The pace of changes taking place in the contemporary world creates demand for current data at the lowest possible level of territorial aggregation, enabling the assessment of the impact of these changes on local labour markets. Data resources gathered in registers and administrative systems directly or indirectly related to employment seem to provide an opportunity to satisfy these new expectations. However, to be able to use administrative resources for statistical purposes, it is necessary to make allowances for the fact that the administrators of registers and administrative systems collect data to satisfy their own needs and that official statistics reuses them. The aim of the article is to show the analytical opportunities which administrative data sources provide and to demonstrate that they can complement or even replace traditional methods of collecting data on employed persons. The experimental adaptation of data from these sources for statistical purposes proves that, for example, data on employed persons, calculated on the basis of the resources of the Social Insurance Institution (ZUS) and the Agricultural Social Insurance Fund (KRUS), correspond almost fully in terms of definitions with data collected by official statistics. The readiness to come to terms with slight methodological differences related to the count of employed persons in agriculture or persons employed for the purpose of vocational training or agents will make it possible to provide users with this important labour market statistics faster, more frequently, and in aggregations different than those recently offered by official statistics, while limiting or even eliminating many reporting obligations imposed on the national economy entities. This article describes official statistics’ current methods of researching employed persons and compares them with the experimental method of calculating the number of these persons on the basis of administrative data sources. Due to the fact that data calculated by means of the experimental method are characterised by a multitude of new information, we decided to present here only selected examples of their analyses and contribution to expanding the knowledge about employed persons in the Polish economy. We would also like to emphasise that data used for this purpose refer to all employed persons (in all entities, including the smallest ones with the maximum of nine employed persons). Other advantages of the administrative data sources are the monthly availability of data and the possibility of aggregating them down to gmina (the lowest local administrative unit) level.

Risk Stratification for Congenital Heart Surgery for ICD-10 Administrative Data (RACHS-2)

BackgroundAs the cardiac community strives to improve outcomes, accurate methods of risk stratification are imperative. Since adoption of International Classification of Disease-10th Revision (ICD-10) in 2015, there is no published method for congenital heart surgery risk stratification for administrative data. ObjectivesThis study sought to develop an empirically derived, publicly available Risk Stratification for Congenital Heart Surgery (RACHS-2) tool for ICD-10 administrative data. MethodsThe RACHS-2 stratification system was iteratively and empirically refined in a training dataset of Pediatric Health Information Systems claims to optimize sensitivity and specificity compared with corresponding locally held Society of Thoracic Surgeons-Congenital Heart Surgery (STS-CHS) clinical registry data. The tool was validated in a second administrative data source: New York State Medicaid claims. Logistic regression was used to compare the ability of RACHS-2 in administrative data to predict operative mortality vs STAT Mortality Categories in registry data. ResultsThe RACHS-2 system captured 99.6% of total congenital heart surgery registry cases, with 1.0% false positives. RACHS-2 predicted operative mortality in both training and validation administrative datasets similarly to STAT Mortality Categories in registry data. C-statistics for models for operative mortality in training and validation administrative datasets—adjusted for RACHS-2—were 0.76 and 0.84 (95% CI: 0.72-0.80 and 0.80-0.89); C-statistics for models for operative mortality—adjusted for STAT Mortality Categories—in corresponding clinical registry data were 0.75 and 0.84 (95% CI: 0.71-0.79 and 0.79-0.89). ConclusionsRACHS-2 is a risk stratification system for pediatric cardiac surgery for ICD-10 administrative data, validated in 2 administrative–registry-linked datasets. Statistical code is publicly available upon request.

Canadian Transfer Research: Past Achievements, Current Challenges, and Future Directions

The study of transfer in Canadian post-secondary education (PSE) is a fractured terrain, with vast inter-provincial differences and deep schisms between participating communities. At the time of writing, there exists no comprehensive review of the vast, fragmented literature examining the predictors and outcomes of Canadian transfer students, thus complicating the advancement of this subfield. Through this piece, we provide a bird’s eye view of a diverse group of 100+ academic articles, policy and institutional research reports on this topic, published from 1968-2020. We use this review to trace the evolution of this complex sub-field in Canada, outlining the major findings, methodological approaches, gaps, and future challenges that Canadian researchers will face as they attempt to improve our understanding of transfer. Drawing on examples from the American and European contexts, we illustrate the need for more robust administrative data sources and methodological sophistication in Canadian transfer research. We also cite inclusive inter-sector collaboration as a key strategy to improve research in this field.

Divergences between healthcare-associated infection administrative data and active surveillance data in Canada.

Although Canada has both a national active surveillance system and administrative data for the passive surveillance of healthcare-associated infections (HAI), both have identified strengths and weaknesses in their data collection and reporting. Active and passive surveillance work independently, resulting in results that diverge at times. To understand the divergences between administrative health data and active surveillance data, a scoping review was performed. Medline, Embase and Cumulative Index to Nursing and Allied Health Literature along with grey literature were searched for studies in English and French that evaluated the use of administrative data, alone or in comparison with traditional surveillance, in Canada between 1995 and November 2, 2020. After extracting relevant information from selected articles, a descriptive summary of findings was provided with suggestions for the improvement of surveillance systems to optimize the overall data quality. Sixteen articles met the inclusion criteria, including twelve observational studies and four systematic reviews. Studies showed that using a single source of administrative data was not accurate for HAI surveillance when compared with traditional active surveillance; however, combining different sources of data or combining administrative with active surveillance data improved accuracy. Electronic surveillance systems can also enhance surveillance by improving the ability to detect potential HAIs. Although active surveillance of HAIs produced the most accurate results and remains the gold-standard, the integration between active and passive surveillance data can be optimized. Administrative data can be used to enhance traditional active surveillance. Future studies are needed to evaluate the feasibility and benefits of potential solutions presented for the use of administrative data for HAI surveillance and reporting in Canada.

The impact of fetal sex on risk factors for gestational diabetes and related adverse pregnancy outcomes.

To investigate whether fetal sex affects the impact of classical GDM risk factors on the diagnosis of gestational diabetes (GDM) as well as on related adverse pregnancy outcomes. This retrospective observational study concerned 206,917 singleton live births born to 170,126 women aged 15-45 over the years 2010-2018 in Tuscany, Italy. GDM was identified by administrative data-sources in 21,613 pregnancies (10.5%) by assessing, through multiple logistic models, whether fetal sex modified the risk of GDM driven by maternal risk factors, and whether it modified the risk of adverse outcomes such as prematurity (birth ≤ 37th gestational week), large for gestational age (LGA), unplanned caesarean sections, or 5-min-Apgar-index ≤ 7 in pregnancies with GDM. GDM was diagnosed in 21,613 pregnancies (10.5%). Male fetal sex predicted a higher adjusted risk of GDM: OR = 1.05(95% CI: 1.01-1.07); p < 0.0009. In pregnancies with female sex, pre-pregnancy obesity amplified the risk of GDM: OR = 1.09(95% CI: 1.01-1.19); p = 0.04. In pregnancies with GDM, carrying a female fetus increased the risk of LGA associated with pregestational obesity OR = 1.45(95% CI: 1.15-1.81); p = 0.001, and in primiparous pregnancies, it protected mothers from the risk of unplanned caesarean sections OR = 0.80(95%CI: 0.67-0.92); p = 0.001. While male fetal sex is associated with rise in the risk of GDM, giving birth to a girl amplifies the excess GDM risk driven by pregestational obesity, thus increasing the risk of LGA in pregnancies with GDM. Additionally, female fetal sex in pregnancies with GDM seems to protect from the risk of unplanned caesarean sections in primiparous pregnancies.

Experiments on Multiple Requests for Consent to Data Linkage in Surveys

Abstract It is increasingly common for researchers to link survey data to administrative data. If several administrative data sources are of interest, respondents are required to give consent to each of them, meaning that multiple consent questions have to be included in one survey. Existing literature suggests that individual consent varies widely between data sources and over time, but little is known about how respondents process multiple consent requests in a single survey. Using an online access panel in Great Britain, we conducted a set of experiments in two surveys to explore multiple consent requests (covering five domains or data sources). In the first study, we experimentally varied the format of the request, testing three versions: (1) a sequence of pages (with one response per domain), (2) all five requests on the same page (with one response per domain), and (3) a single request (with one joint request covering all five domains). We also varied the order of the domains. We find that average consent rates do not differ by format, but asking a less-sensitive or easier-to-comply request first yields slightly higher average consent rates than asking a more sensitive request first. We repeated the order experiment in a second study, using an independent sample from the same panel, and adding two more order conditions. We find average consent rates are not affected much by order, but the consent to individual domains is affected by order. However, we fail to replicate the pattern of consents found in the first study. We conclude that the order in which multiple consent requests are asked does matter, but in complicated ways that depend on the particular outcomes in which one is interested. Objective knowledge and subjective comprehension of the consent process, and confidence in the decision are largely unaffected by format or order.

Gender differences in preferences of adolescents: Evidence from a large-scale classroom experiment

In this study, we estimate unadjusted and adjusted gender gaps in time preference, risk attitudes, altruism, trust, trustworthiness, cooperation, and competitiveness using data on 1088 high school students from 53 classes. These data, collected by running incentivized experiments in Hungarian classrooms, are linked to an administrative data source on the students’ standardized test scores, grades and family background. After taking into account class fixed effects, we find that females are significantly more altruistic, but are less present biased, less risk tolerant, less trusting, less trustworthy, and less competitive than males. At the same time we do not observe significant gender differences in patience, time inconsistency and cooperation at the 5% significance level. We also show that most of these initial gender differences do not change even if we control for age, family background, cognitive skills and school grades in a regression framework. We risk over-control when we include the time spent on each task as well as the other preference domains in our regressions, but the gender gap remains significant in social preferences (altruism, trust and trustworthiness), present bias and competition.

Does health service funding go where the need is? A prototype spatial access analysis for new urban contracts data

BackgroundMuch of spatial access research measures the proximity to health service locations. We advance this research by focusing on whether health service funding is within walkable reach of neighborhoods with high hardship. This is made possible by a new administrative data source: financial contracts data for those human services that are delivered by nonprofits under contract with the government.MethodsIn a prototypical spatial access study we apply a classic 2-step floating area catchment model for walkable network access to analyze 2018 data about contracted nonprofit health services funded by the Chicago Department of Public Health (CDPH). CDPH collected the data for the purpose of this study.ResultsWe find that the common container approach of aggregating contract amounts by provider headquarter locations in a given area (ignoring satellite service sites) underestimates the share of funding that goes to Chicago neighborhoods with higher hardship. Once service sites and spatial access are taken into account, a larger share of CDPH funds was found to be within walkable reach of Chicago’s high hardship areas. This was followed by low hardship areas (which could be driven by more headquarter locations there that do serve areas throughout the city). Medium hardship areas trail both, perhaps warranting closer attention. We explore these results by program type and neighborhood with a spatial decision support system developed for the health department.ConclusionsThe typical approach for analyzing human service contracts based on headquarters is misleading -- in fact, we find that results are reversed when service sites and walkable access are taken into account. This prototype provides an alternative framework for avoiding these misleading results.

How Many Cancer Patients Need Palliative Care? A Population-Based Study

BackgroundThe number of cancer patients potentially amenable to palliative care is conventionally estimated from cancer deaths, as reported in the death certificates. However, a more representative population should also include cancer patients who die from causes other than cancer, as they may develop other life-limiting chronic conditions leading to terminal prognosis. AimThis study aimed at refining the assessment of the number of cancer patients potentially in need of palliative care, by linked hospital and death data. DesignRetrospective study. Setting/ParticipantsResidents in the Emilia Romagna Region in Italy, who died between 2009 and 2017. ResultsWe identified a potential palliative care population of 157,547 cancer patients. The use of different administrative data sources enhanced the sensitivity of our selection. Starting from a standard estimate of 129,212 patients based on cancer as the primary cause of death, we showed that the additional use of hospital records identified a further 11.4% of possible palliative care patients 14,687. Also considering cancer as secondary cause of death, the estimate further increased by 10.6% (13,648 new cases). Notably, the proportion of cancer patients selected by the additional data sources were characterized by more advanced age and higher prevalence of comorbidity. ConclusionHealthcare services addressing the issue of estimating palliative care needs of cancer patients at a population level should consider that relying on the death certificate alone may lead to underestimating these needs of about 22%.

‘What about the dads?’ Linking fathers and children in administrative data: A systematic scoping review

Research has shown that paternal involvement positively impacts on child health and development. We aimed to develop a conceptual model of dimensions of fatherhood, identify and categorise methods used for linking fathers with their children in administrative data, and map these methods onto the dimensions of fatherhood. We carried out a systematic scoping review to create a conceptual framework of paternal involvement and identify studies exploring the impact of paternal exposures on child health and development outcomes using administrative data. We identified four methods that have been used globally to link fathers and children in administrative data based on family or household identifiers using address data, identifiable information about the father on the child's birth registration, health claims data, and Personal Identification Numbers. We did not identify direct measures of paternal involvement but mapping linkage methods to the framework highlighted possible proxies. The addition of paternal National Health Service numbers to birth notifications presents a way forward in the advancement of fatherhood research using administrative data sources.

Estimating Prevalence of Abortion Using List Experiments: Findings from a Survey of Women in Delaware and Maryland.

Widespread underreporting of abortion persists in survey data. The list experiment, a measurement tool designed to elicit truthful responses to sensitive questions, may alleviate underreporting. Using The Statewide Survey of Women of Reproductive Age in Delaware and Maryland (n=2,747), we estimate the prevalence of abortion in Maryland and Delaware using a double list experiment. We find 21% (95% confidence interval [CI]: 16.8%-25.3%) of respondents aged 18 to 44 ever had an abortion and we identify disparities in abortion prevalence by age, race, education, income, marital status, and insurance status. Respondents who were Black (37.0%; 95% CI: 27.1%-46.8%), had less than a college degree (24.8%; 95% CI: 18.3%-31.3%), were in a cohabiting relationship (39.0%; 95% CI: 29.1%-48.9%), were living in households with incomes less than $50,000 (28.6%; 95% CI: 19.7%-37.5%), and were currently covered by Medicaid (42.8%; 95% CI: 27.6%-58.0%) were more likely than their counterparts to have ever had an abortion. List experiments yield estimates of abortion substantially higher than those obtained from direct questions. Findings demonstrate external validity through consistency with estimates from administrative data sources and gold standard abortion provider survey data.

The First Results for the Population Characteristics of the 2021 Census in Czechia

The first results of the 2021 Census in Czechia were released in January 2022. The nationwide census was prepared in a combined form using both a census questionnaire and suitable administrative data sources. The aim of this paper is to analyse the data on population characteristics that were published until August 2022 – sex, age, marital status, educational attainment, citizenship, ethnicity, mother tongue, religious belief, and fertility. Where useful the variables are combined with sex/age structure. A basic comparison with data from the 2011 Census was also done. The population of Czechia increased by less than 1% and has grown older between the 2011 and the 2021 Census. The share of the population with higher levels of education continued to increase and 4.7% of the population in the 2021 Census were foreign nationals.

The First Results for the Population Characteristics of the 2021 Census in Czechia

The first results of the 2021 Census in Czechia were released in January 2022. The nationwide census was prepared in a combined form using both a census questionnaire and suitable administrative data sources. The aim of this paper is to analyse the data on population characteristics that were published until August 2022 – sex, age, marital status, educational attainment, citizenship, ethnicity, mother tongue, religious belief, and fertility. Where useful the variables are combined with sex/age structure. A basic comparison with data from the 2011 Census was also done. The population of Czechia increased by less than 1% and has grown older between the 2011 and the 2021 Census. The share of the population with higher levels of education continued to increase and 4.7% of the population in the 2021 Census were foreign nationals

A Comprehensive Study on Environmental Pollution and Pregnancy Outcomes

Vehicular traffic contributes significantly to the pollution of air and neighborhood noise. Since there is increasing affirmation that ambient air pollution hurts reproductive health, which is less known regarding the relationship between traffic noise and pregnancy outcomes. Many researchers have investigated all the possible adverse effects of ambient air pollution on birth outcomes during the last decade or so. We examined these studies, which were discovered by thoroughly searching the major scientific databases. Overall, research suggests that air pollution and noise pollution significantly impact birth outcomes, while the quality of the deposition varies. Researchers utilized connected administrative health data sources to discover pregnancy outcomes with detailed covariate data (gender, contemporary lifestyle, total no of pregnancies, birth month and year, and so on), earnings and education, and mother housing history. Using the deterministic model (CadnaA) and air pollutants exposure, they assessed noise exposure during pregnancy by assessing temporally adjusted land-use regression. Noise exposure is related to reduced birth weight. We also looked into various processes to prevent air pollution and noise. As a result of the policy change and technical innovation, the government, authorities, and industry have been at the forefront of combating pollution. Pollution levels in and around people's homes and workplaces are typically beyond their control. However, few things may be done to mitigate the impacts of contaminated air.

Use of Administrative Data for Waste Statistics in the Czech Republic

The Czech Statistical Office (CZSO) changed the calculation of the total volume of waste and also changed the definition of municipal waste compared to the previous methodology. This was made possible by the wider use of the existing administrative data source, the Integrated Environmental Reporting System (ISPOP). The change in the definition of municipal waste was a response to recent Eurostat activities, which led to a more precise definition. The original method no longer meets this definition.

The Wisconsin Longitudinal Study: Overview, Data Linkages, and Future Plans

The WLS is a study of Wisconsin high school class of 1957 graduates, with follow-ups in 1964, 1975, 1993, 2004, 2011, and 2020. The data reflect the life course of the graduates (and their siblings), initially covering education, switching to family, career, and social participation in midlife, and physical and mental health, cognitive status, caregiving, and social support as respondents age. The WLS is linked to multiple administrative data sources including: parent earnings from state tax records (1957-60) and Social Security earnings and benefits for respondents; 1940 Census data; characteristics of high schools and colleges, employers, industries, and communities of residence; voting records from 2000-2018; Medicare claims; and the National Death Index. Efforts are underway to expand the racial/ethnic and educational composition of the WLS by supplementing the original sample with a new cohort of age-matched adults drawn from Wisconsin’s Black, Hispanic, Asian-American, and Native American communities.