Search results for "record linkage"
showing 10 items of 20 documents
Evaluation of Record Linkage Methods for Iterative Insertions
2009
Summary Objectives: There have been many developments and applications of mathematical methods in the context of record linkage as one area of interdisciplinary research efforts. However, comparative evaluations of record linkage methods are still underrepresented. In this paper improvements of the Fellegi-Sunter model are compared with other elaborated classification methods in order to direct further research endeavors to the most promising methodologies. Methods: The task of linking records can be viewed as a special form of object identification. We consider several non-stochastic methods and procedures for the record linkage task in addition to the Fellegi-Sunter model and perform an e…
Developing and validating a novel multisource comorbidity score from administrative data: a large population-based cohort study from Italy
2017
ObjectiveTo develop and validate a novel comorbidity score (multisource comorbidity score (MCS)) predictive of mortality, hospital admissions and healthcare costs using multiple source information from the administrative Italian National Health System (NHS) databases.MethodsAn index of 34 variables (measured from inpatient diagnoses and outpatient drug prescriptions within 2 years before baseline) independently predicting 1-year mortality in a sample of 500 000 individuals aged 50 years or older randomly selected from the NHS beneficiaries of the Italian region of Lombardy (training set) was developed. The corresponding weights were assigned from the regression coefficients of a Weibull sur…
Bagging, bumping, multiview, and active learning for record linkage with empirical results on patient identity data
2011
Record linkage or deduplication deals with the detection and deletion of duplicates in and across files. For this task, this paper introduces and evaluates two new machine-learning methods (bumping and multiview) together with bagging, a tree-based ensemble-approach. Whereas bumping represents a tree-based approach as well, multiview is based on the combination of different methods and the semi-supervised learning principle. After providing a theoretical background of the methods, initial empirical results on patient identity data are given. In the empirical evaluation, we calibrate the methods on three different kinds of training data. The results show that the smallest training data set, …
A Cohort Study of Childhood Cancer Incidence after Postnatal Diagnostic X-Ray Exposure
2009
Ionizing radiation is an established cause of cancer, yet little is known about the health effects of doses from diagnostic examinations in children. The risk of childhood cancer was studied in a cohort of 92.957 children who had been examined with diagnostic X rays in a large German hospital during 1976-2003. Radiation doses were reconstructed using the individual dose area product and other exposure parameters, together with conversion coefficients developed specifically for the medical devices and standards used at the radiology department. Newly diagnosed cancers occurring between 1980 and 2006 were determined through record linkage to the German Childhood Cancer Registry. The median ra…
Deterministic Linkage as a Preceding Filter for Other Record Linkage Methods
2015
Deterministic record linkage (RL) is frequently regarded as a rival to more sophisticated strategies like probabilistic RL. We investigate the effect of combining deterministic linkage with other linkage techniques. For this task, we use a simple deterministic linkage strategy as a preceding filter: a data pair is classified as ‘match' if all values of attributes considered agree exactly, otherwise as ‘nonmatch'. This strategy is separately combined with two probabilistic RL methods based on the Fellegi–Sunter model and with two classification tree methods (CART and Bagging). An empirical comparison was conducted on two real data sets. We used four different partitions into training data a…
Active learning strategies for the deduplication of electronic patient data using classification trees.
2012
Graphical abstractDisplay Omitted Highlights? Active learning for medical record linkage is used on a large data set. ? We compare a simple active learning strategy with a more sophisticated variant. ? The active learning method of Sarawagi and Bhamidipaty (2002) 6] is extended. ? We deliver insights into the variations of the results due to random sampling in the active learning strategies. IntroductionSupervised record linkage methods often require a clerical review to gain informative training data. Active learning means to actively prompt the user to label data with special characteristics in order to minimise the review costs. We conducted an empirical evaluation to investigate whether…
Determinants of homonym and synonym rates of record linkage in disease registration.
1996
AbstractReliable record linkage is an essential component of the quality of population-based disease registration. Quality assessment of disease registries should, therefore, include quantitative approaches to describe the extent of record-linkage errors. The homonym and synonym rates have been proposed for this purpose. The homonym rate quantifies the proportion of distinct patients excluded from registration due to erroneous linkage with other patients. The synonym rate quantifies the proportion of unrecognized duplicate notifications on patients already registered in the registry. This paper provides an algebraic assessment of the determinants of both rates. It is shown how the homonym a…
Controlling false match rates in record linkage using extreme value theory
2011
AbstractCleansing data from synonyms and homonyms is a relevant task in fields where high quality of data is crucial, for example in disease registries and medical research networks. Record linkage provides methods for minimizing synonym and homonym errors thereby improving data quality. We focus our attention to the case of homonym errors (in the following denoted as ‘false matches’), in which records belonging to different entities are wrongly classified as equal. Synonym errors (‘false non-matches’) occur when a single entity maps to multiple records in the linkage result. They are not considered in this study because in our application domain they are not as crucial as false matches. Fa…
Effects of record linkage errors on disease registration
1998
Abstract:Reliable record linkage is a prerequisite for high-quality population-based disease registration. Rapid developments in computer processing have made record linkage both more efficient and more reliable in recent years. At the same time, concerns about confidentiality increasingly hinder record linkage in many disease registries. This paper provides basic algebraic models describing the effects of record linkage errors on monitoring disease incidence. Homonym errors, that is, erroneous linkage of records that pertain to distinct individuals, lead to underestimation of incidence in the registry population. The degree of underestimation strongly depends on the discriminating power of…
Assessing the risk of osteonecrosis of the jaw due to bisphosphonate therapy in the secondary prevention of osteoporotic fractures
2012
There is evidence that the use oral bisphosphonates can lead to osteronecrosis of the jaws (ONJ). Although the occurrence of ONJ appears rare among oral bisphosphonates (BPs) users, it is important to know that it exists and can be opportunely minimized. INTRODUCTION: The purpose of this study is to evaluate the association between BPs prescribed for the secondary prevention of osteoporotic fractures and the occurrence of ONJ. METHODS: An Italian record linkage claims database with a target population of around 18 million individuals (6 million over 55 years of age) constituted the data source. We conducted a nested case-control study within a cohort of individuals aged 55+ years old, who w…