Record Linkage Process Research Articles

Towards malaria elimination: a case–control study to assess associated factors to malaria relapses in the extra-Amazon Region of Brazil from 2008 to 2019

BackgroundMalaria is an infectious disease caused by the Plasmodium species and is a global burden. When not treated correctly, it can reemerge as a relapse or recrudescence. Malaria relapse cases can contribute to maintaining active transmission chains and can influence the patient to develop severe malaria, potentially leading to hospitalization or death. The objective of this study is to estimate the number of malaria relapse cases in the extra-Amazon region of Brazil and to investigate the associated factors.MethodsThis is a case–control study that analyses malaria infections caused by Plasmodium vivax, as reported in Notifiable Diseases Information System (Sinan) for the Brazilian extra-Amazon region (an area not endemic for the disease) from 2008 to 2019. For the identification of relapse cases, deduplication record linkage processes in R software were used. Malaria relapses were defined as the case group, and new malaria infections were defined as the control group. Logistic regression models were used to assess associated factors.ResultsOf the 711 malaria relapses, 589 (82.8%) were first relapses. Most relapses (71.6%) occurred between 30 and 120 days after the previous infection. Malaria relapses are spread throughout the extra-Amazon region, with a higher concentration near big cities. Driver occupation was found to be a common risk factor compared to other occupations, along with asymptomatic individuals. Other associated factors were: being infected in the Brazilian Amazon region, having follow-ups for malaria relapses, and having parasite density of the previous infection higher than 10,000 parasites per mm3.ConclusionsThis study provides evidence that allows malaria health surveillance services to direct their efforts to monitor cases of malaria in the highest risk segments identified in this study, particularly in the period between 30 and 120 days after being infected and treated. Relapses were associated to driver occupation, absence of symptoms, infection in endemic areas of Brazil, being detected through active surveillance or routine follow-up actions, and with parasitaemia greater than 10,000 parasites per mm3 in the previous infection. Improving cases follow-up is essential for preventing relapses.

Linkability measures to assess the data characteristics for record linkage.

Accurate record linkage (RL) enables consolidation and de-duplication of data from disparate datasets, resulting in more comprehensive and complete patient data. However, conducting RL with low quality or unfit data can waste institutional resources on poor linkage results. We aim to evaluate data linkability to enhance the effectiveness of record linkage. We describe a systematic approach using data fitness ("linkability") measures, defined as metrics that characterize the availability, discriminatory power, and distribution of potential variables for RL. We used the isolation forest algorithm to detect abnormal linkability values from 188 sites in Indiana and Colorado, and manually reviewed the data to understand the cause of anomalies. We calculated 10 linkability metrics for 11 potential linkage variables (LVs) across 188 sites for a total of 20680 linkability metrics. Potential LVs such as first name, last name, date of birth, and sex have low missing data rates, while Social Security Number vary widely in completeness among all sites. We investigated anomalous linkability values to identify the cause of many records having identical values in certain LVs, issues with placeholder values disguising data missingness, and orphan records. The fitness of a variable for RL is determined by its availability and its discriminatory power to uniquely identify individuals. These results highlight the need for awareness of placeholder values, which inform the selection of variables and methods to optimize RL performance. Evaluating linkability measures using the isolation forest algorithm to highlight anomalous findings can help identify fitness-for-use issues that must be addressed before initiating the RL process to ensure high-quality linkage outcomes.

Identifying “problematic” clusters in record linkage using graph measures

Manual review of outputs from record linkage processes remains a frequently necessary but time-consuming operation. This work outlines a potential means to reduce the manual review workload through the extraction of graph measures from the outputs of record linkage processes combined with machine learning classification methods. The outputs of a pairwise record linkage process can naturally be represented by undirected simple graphs with a vertex representing each record and an edge representing each comparison of records. One representation may be a graph which only includes edges representing pairwise comparisons that have a match score greater than a given threshold. A second representation may include match scores (suitably transformed) as the weight of the edges (pairwise record comparisons) connecting vertices (records). A third representation may include not just the overall match score as a weight on each of the edges (pairwise record comparisons) but also, for each compared attribute (e.g. names, date of birth, address, etc.), the contributing component to the overall match score as additional edge attributes. In this work, a "problematic" cluster is a set formed of three or more records which, in totality, belong to two or more true entities. Equivalently, a connected graph of three or more vertices (representing records) with at least one edge that represents a false positive link. Finally, this work evaluates the performance and generalisability of various combinations of graph measures for identifying problematic clusters across linkage processes featuring varying degrees of corruption and duplication.

The use of Supervised Learning to perform pairwise classification for Record Linkage over real world data

Supervised Learning (SL) is a subset of Machine Learning that uses labeled data to establish the learning process and perform tasks such as classification and regression. Record Linkage (RL) is the process for entity resolution and aims to determine if two records correspond to the same entity in the real world. A challenging aspect of RL involves estimating probabilities and weights for attributes and choosing the algorithm to perform the comparison. Our study aims to address this challenge and contribute in two primary ways: 1) by providing a labeled dataset comprising pairs of records, and 2) by evaluating an alternative classification method using an SL classifier. We generated more than 28.000 pairs, manually reviewed, derived from a massive Record Linkage Process on the Brazilian Public Health System (SUS). From that, we selected 5000 random stratified pairs for evaluation. The data contains demographic attributes such as sex, city, date of birth, zip code, and name. We modeled the data pairwise in a vector of latent space. Our methodology involved comparing Random Forest, LightGBM, XGBoost, and Neural Networks for classification tasks on these pairs, utilizing cross-validation to optimize hyperparameters. Our findings reveal that XGBoost had the highest performance, achieving 96% accuracy. Additionally, we conducted privacy evaluations using bloom filters, yielding an accuracy of 93%. The adoption of SL presents a promising space due to its capability for weight estimation and classification. In future endeavors, integrating these algorithms into a framework for the Record Linkage process could streamline procedures for researchers without compromising performance.

Algorithmic Accountability for Record Linkage

This session will examine record linkage approaches through the lens of principles for algorithmic accountability: responsibility, explainability, accuracy, auditability, and fairness.[1] It is easy to hypothesize cases where linkage algorithms could have disparate impacts, such as linking names more or less easily depending on their cultural background. How can researchers use careful linkage design and sensitivity testing to identify and, where necessary, mediate these impacts? What does it look like for record linkage processes to be explainable and auditable—and how can researchers invite engagement from broader groups of stakeholders in considering linkage methods? We will present on our experiences identifying and addressing these challenges in the context of linking state and local public sector administrative data sources. These data sources, which generally include populations engaging with particular programs or systems, have especially variable data quality and are often marred by structural biases. Our approach to linking these datasets emphasizes integrating policy and data context, interrogating linkage decisions for iterative improvement, and collaborating with a wide range of stakeholders for equitable, accountable data use. We will demonstrate examples of how linkage decisions and results have varied across contexts. [1] Diakopoulos, N., Friedler, S., Arenas, M., Barocas, S., Hay, M., Howe, B., Jagadish, H.V., Unsworth, K., Sahuguet, A., Venkatasubramanian, S., Wilson, C., Yu, C., and Zevenbergen, B. (2016). Principles for Accountable Algorithms and a Social Impact Statement for Algorithms. Fairness, Accountability, and Transparency in Machine Learning. https://www.fatml.org/resources/principles-for-accountable-algorithms.

A general framework for regression with mismatched data based on mixture modelling

Abstract The advent of the information age has revolutionized data collection and has led to a rapid expansion of available data sources. Methods of data integration are indispensable when a question of interest cannot be addressed using a single data source. Record linkage (RL) is at the forefront of such data integration efforts. Incentives for sharing linked data for secondary analysis have prompted the need for methodology accounting for possible errors at the RL stage. Mismatch error is a common consequence resulting from the use of nonunique or noisy identifiers at that stage. In this paper, we present a framework to enable valid postlinkage inference in the secondary analysis setting in which only the linked file is given. The proposed framework covers a variety of statistical models and can flexibly incorporate information about the underlying RL process. We propose a mixture model for linked records whose two components reflect distributions conditional on match status, i.e. correct or false match. Regarding inference, we develop a method based on composite likelihood and the expectation-maximization algorithm that is implemented in the R package pldamixture. Extensive simulations and case studies involving contemporary RL applications corroborate the effectiveness of our framework.

An open-source probabilistic record linkage process for records with family-level information: Simulation study and applied analysis.

Research with administrative records involves the challenge of limited information in any single data source to answer policy-related questions. Record linkage provides researchers with a tool to supplement administrative datasets with other information about the same people when identified in separate sources as matched pairs. Several solutions are available for undertaking record linkage, producing linkage keys for merging data sources for positively matched pairs of records. In the current manuscript, we demonstrate a new application of the Python RecordLinkage package to family-based record linkages with machine learning algorithms for probability scoring, which we call probabilistic record linkage for families (PRLF). First, a simulation of administrative records identifies PRLF accuracy with variations in match and data degradation percentages. Accuracy is largely influenced by degradation (e.g., missing data fields, mismatched values) compared to the percentage of simulated matches. Second, an application of data linkage is presented to compare regression model estimate performance across three record linkage solutions (PRLF, ChoiceMaker, and Link Plus). Our findings indicate that all three solutions, when optimized, provide similar results for researchers. Strengths of our process, such as the use of ensemble methods, to improve match accuracy are discussed. We then identify caveats of record linkage in the context of administrative data.

Evaluation of the reported data linkage process and associated quality issues for linked routinely collected healthcare data in multimorbidity research: a systematic methodology review

ObjectiveThe objective of this systematic review was to examine how the record linkage process is reported in multimorbidity research.MethodsA systematic search was conducted in Medline, Web of Science and Embase...

Evaluation of the reported data linkage process and associated quality issues for linked routinely collected healthcare data in Multimorbidity research: a systematic review.

ObjectivesThe objective for this systematic review was to examine how the record linkage process was reported and to understand challenges related to accessing, linking, and analysing linked routinely collected data used for multimorbidity research.
 ApproachA systematic search for relevant studies was conducted in three online databases (Medline, Web of Science and Embase) in May 2021 using predefined search terms, and inclusion and exclusion criteria. Studies using linked routinely collected data for multimorbidity research were included. Information was extracted on how the linkage process was reported, which conditions were studied together, which data sources were used, as well as challenges encountered during the linkage process or with the linked dataset.
 The findings from this study will feed into further guidance to understand and minimise bias due to linkage error in medical research.
 ResultsTwenty studies were included, of which seventeen looked at the relationship between two specified long-term conditions. Fourteen studies received the linked dataset from an external data linkage provider. Hospital Episode Statistics was the most common source of data (n=5). Eight studies reported variables used for the data linkage, while only two studies reported pre-linkage checks. The quality of the linkage was assessed by three studies, of which two reported linkage rate and one reported raw linkage figures. Only one study checked for bias by comparing patient characteristics of linked and non-linked records.
 ConclusionsThe linkage process was poorly reported in multimorbidity research, even though this might introduce bias and potentially lead to inaccurate inferences drawn from the results. There is therefore a need for increased awareness of linkage bias and transparency of the linkage processes, which could be achieved through better adherence to reporting guidelines.

A retrospective cohort study of outcomes in hospitalised COVID-19 patients during the first pandemic wave in Ireland.

AimDescribe the epidemiology, resource use and adverse outcomes of COVID-19 patients hospitalised during wave 1 of the COVID-19 pandemic in Ireland. Use this data to identify specific cohorts at high risk of adverse outcomes and to inform acute hospital requirements for future COVID-19 waves in Ireland.MethodsThe Health Service Executive’s (HSE) Hospital Inpatient Enquiry (HIPE) system produced a daily database of COVID-19 discharge episodes from the onset of the COVID-19 pandemic in Ireland. This study analysed data on episodes of COVID-19 hospitalisation recorded between February 29 and July 31, 2020. A deterministic record linkage process transformed records from episode to patient level. Logistic regression modelling identified factors associated with long length of stay (LLOS), intensive care unit (ICU) admission and inhospital mortality.ResultsMedian length of stay was 9 days; 12.8% of patients had ICU admission and 16.6% died in hospital. Male patients were more likely to have ICU admission and die in hospital. Likelihood of LLOS and inhospital mortality increased with age. Obesity, hypertension and diabetes were associated with ICU admission while chronic kidney disease and chronic obstructive pulmonary disease were associated with inhospital mortality. Nursing home residents were less likely to be admitted to ICU and more likely to die in hospital compared to patients admitted from home.ConclusionThis study provides patient-level epidemiological characterisation of hospitalisations during the first COVID-19 pandemic wave in Ireland. The higher risk of adverse outcomes in older age groups supports the age-based prioritisation of COVID-19 vaccinations currently used in Ireland.

A new hybrid record linkage process to make epidemiological databases interoperable: application to the GEMO and GENEPSO studies involving BRCA1 and BRCA2 mutation carriers

BackgroundLinking independent sources of data describing the same individuals enable innovative epidemiological and health studies but require a robust record linkage approach. We describe a hybrid record linkage process to link databases from two independent ongoing French national studies, GEMO (Genetic Modifiers of BRCA1 and BRCA2), which focuses on the identification of genetic factors modifying cancer risk of BRCA1 and BRCA2 mutation carriers, and GENEPSO (prospective cohort of BRCAx mutation carriers), which focuses on environmental and lifestyle risk factors.MethodsTo identify as many as possible of the individuals participating in the two studies but not registered by a shared identifier, we combined probabilistic record linkage (PRL) and supervised machine learning (ML). This approach (named “PRL + ML”) combined together the candidate matches identified by both approaches. We built the ML model using the gold standard on a first version of the two databases as a training dataset. This gold standard was obtained from PRL-derived matches verified by an exhaustive manual review. ResultsThe Random Forest (RF) algorithm showed a highest recall (0.985) among six widely used ML algorithms: RF, Bagged trees, AdaBoost, Support Vector Machine, Neural Network.Therefore, RF was selected to build the ML model since our goal was to identify the maximum number of true matches. Our combined linkage PRL + ML showed a higher recall (range 0.988–0.992) than either PRL (range 0.916–0.991) or ML (0.981) alone. It identified 1995 individuals participating in both GEMO (6375 participants) and GENEPSO (4925 participants).ConclusionsOur hybrid linkage process represents an efficient tool for linking GEMO and GENEPSO. It may be generalizable to other epidemiological studies involving other databases and registries.

A MULTIPLE IMPUTATION PROCEDURE FOR RECORD LINKAGE AND CAUSAL INFERENCE TO ESTIMATE THE EFFECTS OF HOME-DELIVERED MEALS.

Causal analysis of observational studies requires data that comprise of a set of covariates, a treatment assignment indicator, and the observed outcomes. However, data confidentiality restrictions or the nature of data collection may distribute these variables across two or more datasets. In the absence of unique identifiers to link records across files, probabilistic record linkage algorithms can be leveraged to merge the datasets. Current applications of record linkage are concerned with estimation of associations between variables that are exclusive to one file and not causal relationships. We propose a Bayesian framework for record linkage and causal inference where one file comprises all the covariate and observed outcome information, and the second file consists of a list of all individuals who receive the active treatment. Under certain ignorability assumptions, the procedure properly propagates the error in the record linkage process, resulting in valid statistical inferences. To estimate the causal effects, we devise a two-stage procedure. The first stage of the procedure performs Bayesian record linkage to multiply impute the treatment assignment for all individuals in the first file, while adjustments for covariates' imbalance and imputation of missing potential outcomes are performed in the second stage. This procedure is used to evaluate the effect of Meals on Wheels services on mortality and healthcare utilization among homebound older adults in Rhode Island. In addition, an interpretable sensitivity analysis is developed to assess potential violations of the ignorability assumptions.

Use of administrative record linkage to measure medical and social risk factors for early developmental vulnerability in Ontario, Canada.

BackgroundLinkage of demographic, health, and developmental administrative data can enrich population-based surveillance and research on developmental and educational outcomes. Transparency of the record linkage process and results are required to assess potential biases.ObjectivesTo describe the approach used to link records of kindergarten children from the Early Development Instrument (EDI) in Ontario to health administrative data and test differences in characteristics of children by linkage status. We demonstrate how socio-demographic and medical risk factors amass in their contribution to early developmental vulnerability and test the concordance of health diagnoses in both the EDI and health datasets of linked records.MethodsChildren with records in the 2015 EDI cycle were deterministically linked to a population registry in Ontario, Canada. We compared sociodemographic and developmental vulnerability data between linked and unlinked records. Among linked records, we examined the contribution of medical and social risk factors obtained from health administrative data to developmental vulnerability identified in the EDI using descriptive analyses.ResultsOf 135,937 EDI records, 106,217 (78.1%) linked deterministically to a child in the Ontario health registry using birth date, sex, and postal code. The linked cohort was representative of children who completed the EDI in age, sex, rural residence, immigrant status, language, and special needs status. Linked data underestimated children living in the lowest neighbourhood income quintile (standardized difference [SD] 0.10) and with higher vulnerability in physical health and well-being (SD 0.11) , social competence (SD 0.10), and language and cognitive development (SD 0.12). Analysis of linked records showed developmental vulnerability is sometimes greater in children with social risk factors compared to those with medical risk factors. Common childhood conditions with records in health data were infrequently recorded in EDI records.ConclusionsLinkage of early developmental and health administrative data, in the absence of a single unique identifier, can be successful with few systematic biases introduced. Cross-sectoral linkages can highlight the relative contribution of medical and social risk factors to developmental vulnerability and poor school achievement.

Comparative Study of Record Linkage Approaches for Big Data

Record linkage is a challenging task for Big Data. This paper, hence, attempts to shed light on record linkage approaches for Big Data by comparing three dimensions involving record linkage phases, dataset properties, and parallel processing approach for Big Data. The current state of art have only conducted comparative studies between record linkage approaches. There has been only one comparative study exploring the whole record linkage framework of the relational database. It is believed that the focus of the present study on the dimensions of the parallel processing approaches for Big Data and dataset properties was worth exploring. It was found that first, data exploration was almost a non-existing phase despite its importance of exploring the dataset being examined; second, techniques that handle data standardization and preparation phase of the first dimension were not extensively covered in the literature which can directly affect the results’ quality; third, the record linkage in unstructured data was not yet explored in literature; fourth, the MapReduce was used in about 50 % of the selected studies to handle the parallel processing of Big Data, but due to its limitations, more recent and efficient approaches had been used, such as Apache Spark and Apache Flink. Apache Spark is just recently adapted to resolve duplicates due to its supporting of in-memory computation, which makes the whole linkage process more efficient. Although the comparative study includes many recent studies supporting Apache Spark, adopting Apache Spark to solve the problem of record linkage is not yet well explored in literature, as more researches need to be conducted. In addition, Apache Flink is still rarely used to solve the record linkage problem of Big Data. Fifth, pruning techniques, used to eliminate unnecessary comparisons, are not adequately applied in the covered studies despite their effect on reducing the search space resulting in a more effective Record Linkage process.

Communicating With Patients About Software for Enhancing Privacy in Secondary Database Research Involving Record Linkage: Delphi Study.

BackgroundThere is substantial prior research on the perspectives of patients on the use of health information for research. Numerous communication barriers challenge transparency between researchers and data participants in secondary database research (eg, waiver of informed consent and knowledge gaps). Individual concerns and misconceptions challenge the trust in researchers among patients despite efforts to protect data. Technical software used to protect research data can further complicate the public's understanding of research. For example, MiNDFIRL (Minimum Necessary Disclosure For Interactive Record Linkage) is a prototype software that can be used to enhance the confidentiality of data sets by restricting disclosures of identifying information during the record linkage process. However, software, such as MiNDFIRL, which is used to protect data, must overcome the aforementioned communication barriers. One proposed solution is the creation of an interactive web-based frequently asked question (FAQ) template that can be adapted and used to communicate research issues to data subjects.ObjectiveThis study aims to improve communication with patients and transparency about how complex software, such as MiNDFIRL, is used to enhance privacy in secondary database studies to maintain the public's trust in researchers.MethodsA Delphi technique with 3 rounds of the survey was used to develop the FAQ document to communicate privacy issues related to a generic secondary database study using the MiNDFIRL software. The Delphi panel consisted of 38 patients with chronic health conditions. We revised the FAQ between Delphi rounds and provided participants with a summary of the feedback. We adopted a conservative consensus threshold of less than 10% negative feedback per FAQ section.ResultsWe developed a consensus language for 21 of the 24 FAQ sections. Participant feedback demonstrated preference differences (eg, brevity vs comprehensiveness). We adapted the final FAQ into an interactive web-based format that 94% (31/33) of the participants found helpful or very helpful. The template FAQ and MiNDFIRL source code are available on GitHub. The results indicate the following patient communication considerations: patients have diverse and varied preferences; the tone is important but challenging; and patients want information on security, identifiers, and final disposition of information.ConclusionsThe findings of this study provide insights into what research-related information is useful to patients and how researchers can communicate such information. These findings align with the current understanding of health literacy and its challenges. Communication is essential to transparency and ethical data use, yet it is exceedingly challenging. Developing FAQ template language to accompany a complex software may enable researchers to provide greater transparency when informed consent is not possible.

DaT18 Database: A Prosopographical Approach to the Study of the Social Structures of Religious Dissent in Mid-Eighteenth-Century Transylvania

"Drawing on the many records created by the Habsburg state during the confessional troubles in Transylvania from the 1740s to the 1760s, the DaT18 project merges digital instruments and prosopography to arrive at sketching the social pattern of the Orthodox leadership. This article briefly discusses the technical choices involved in building the relational database that my approach centres on, before talking in more detail about the challenges faced when transposing the information in the primary sources into digital format. First, the question of making use of structured vs. unstructured data, as most of the documents I work with already present some form of tabular layout, while the more narrative ones require different strategies to mitigate losses when converting them. Secondly, the difficult process of record linkage, with many of the persons only mentioned by their first name and no surname to help label each individual entered in more than one source. Lastly, the daunting task of estimating economic resources, since there was no reliable standard in an age that saw four different fiscal systems in use and many regional flavours within the same scheme. Keywords: prosopography, relational database, clerical careers, data structuring, Greek Orthodox Church."

An Automatic Blocking Keys Selection For Efficient Record Linkage

One of the important processes in the data quality field is record linkage (RL). RL (also known as entity resolution) is the process of detecting duplicates that refer to the same real-world entity in one or more datasets. The most critical step during the RL process is blocking, which reduces the quadratic complexity of the process by dividing the data into a set of blocks. By that way, matching is done only between the records in the same block. However, selecting the best blocking keys to divide the data is a hard task, and in most cases, it's done by a domain expert. In this paper, a novel unsupervised approach for an automatic blocking key selection is proposed. This approach is based on the recently proposed meta-heuristic bald eagles search (bes) optimization algorithm, where the problem is treated as a feature selection case. The obtained results from experiments on real-world datasets showed the efficiency of the proposition where the BES for feature selection outperformed existed approaches in the literature and returned the best blocking keys.

Record linkage of routine data with cohorts’ data of infants under European and Portuguese law

Abstract The usage and exchange of “big data” is at the forefront of the data science agenda where Record Linkage plays a prominent role in biomedical research. In an era of ubiquitous data exchange and big data, Record Linkage is almost inevitable, but raises ethical and legal problems, namely personal data and privacy protection. Record Linkage refers to the general merging of data information to consolidate facts about an individual or an event that are not available in a separate record. This article provides an overview of ethical challenges and research opportunities in linking routine data on health and education with cohort data from very preterm (VPT) infants in Portugal. Portuguese, European and International law has been reviewed on data processing, protection and privacy. A three-stage analysis was carried out: i) interplay of threefold law-levelling for Record Linkage at different levels; ii) impact of data protection and privacy rights for data processing, iii) data linkage process' challenges and opportunities for research. A framework to discuss the process and its implications for data protection and privacy was created. The GDPR functions as utmost substantial legal basis for the protection of personal data in Record Linkage, and explicit written consent is considered the appropriate basis for the processing sensitive data. In Portugal, retrospective access to routine data is permitted if anonymised; for health data if it meets data processing requirements declared with an explicit consent; for education data if the data processing rules are complied. Routine health and education data can be linked to cohort data if rights of the data subject and requirements and duties of processors and controllers are respected. A strong ethical context through the application of the GDPR in all phases of research need to be established to achieve Record Linkage between cohort and routine collected records for health and education data of VPT infants in Portugal. Key messages GDPR is the most important legal framework for the protection of personal data, however, its uniform approach granting freedom to its Member states hampers Record Linkage processes among EU countries. The question remains whether the gap between data protection and privacy is adequately balanced at three legal levels to guarantee freedom for research and the improvement of health of data subjects.

Examining the quality of record linkage process using nationwide Brazilian administrative databases to build a large birth cohort

BackgroundResearch using linked routine population-based data collected for non-research purposes has increased in recent years because they are a rich and detailed source of information. The objective of this study is to present an approach to prepare and link data from administrative sources in a middle-income country, to estimate its quality and to identify potential sources of bias by comparing linked and non-linked individuals.MethodsWe linked two administrative datasets with data covering the period 2001 to 2015, using maternal attributes (name, age, date of birth, and municipally of residence) from Brazil: live birth information system and the 100 Million Brazilian Cohort (created using administrative records from over 114 million individuals whose families applied for social assistance via the Unified Register for Social Programmes) implementing an in house developed linkage tool CIDACS-RL. We then estimated the proportion of highly probably link and examined the characteristics of missed-matches to identify any potential source of bias.ResultsA total of 27,699,891 live births were submited to linkage with maternal information recorded in the baseline of the 100 Million Brazilian Cohort dataset of those, 16,447,414 (59.4%) children were found registered in the 100 Million Brazilian Cohort dataset. The proportion of highly probably link ranged from 39.3% in 2001 to 82.1% in 2014. A substantial improvement in the linkage after the introduction of maternal date of birth attribute, in 2011, was observed. Our analyses indicated a slightly higher proportion of missing data among missed matches and a higher proportion of people living in an urban area and self-declared as Caucasian among linked pairs when compared with non-linked sets.DiscussionWe demonstrated that CIDACS-RL is capable of performing high quality linkage even with a limited number of common attributes, using indexation as a blocking strategy in larg e routine databases from a middle-income country. However, residual records occurred more among people under worse living conditions. The results presented in this study reinforce the need of evaluating linkage quality and when necessary to take linkage error into account for the analyses of any generated dataset.

Linking disease registries and nationwide healthcare administrative databases: the French renal epidemiology and information network (REIN) insight

BackgroundRecord linkage is increasingly used in health research worldwide. Combining the patient information available in healthcare, administrative and clinical databases broadens the research perspectives, particularly for chronic diseases. Recent guidelines highlight the need for transparency on the used record linkage processes and the extracted data to be used by researchers.MethodsTherefore, the aim of this study was to describe the deterministic iterative approach used to link the French Epidemiology and Information Network (REIN), a French national End-Stage Renal Disease registry, with the Système National des Données de Santé (SNDS), a French nationwide medico-administrative healthcare database.ResultsAmong the 22,073 patients included in the REIN registry who started renal replacement therapy between 2014 and 2015 in France, 19,223 (87.1%) were matched with patients in the SNDS database. Comparison of matched and unmatched patients confirmed the absence of any major selection bias. Then, the record linkage was evaluated using the comorbidity status (diabetes).ConclusionsThis fast and efficient method of record linkage with pseudonymized data and without unique and direct identifier might inspire other research teams. It also opens the path for new research on chronic kidney disease.