Practical introduction to record linkage for injury research

Clark, David E.

doi:10.1136/ip.2003.004580

Cited by 147 publications

(113 citation statements)

References 60 publications

Supporting

Mentioning

110

Contrasting

Order By: Relevance

Section: Data Matchingmentioning

confidence: 99%

“…It combines ideas from statistics, computer systems and operations research and has been greatly facilitated by high speed computing though it began earlier. A comprehensive overview is provided by Clark (2004) and there is an extensive literature review in ITF (2012).Its use in road safety research is relatively recent. It has been used to make estimates of underreporting of particular kinds of accidents or injuries, as in the Rhone region in France by Amoros et al (2007).…”

mentioning

confidence: 99%

See 1 more Smart Citation

The safety challenge of increased cycling

Short

Caulfield

2014

Transport Policy

View full text Add to dashboard Cite

Many countries have recently set out policy frameworks to support the growth of cycling. However, increased cycling can mean more collisions, injuries and even fatalities. This paper discusses the role of safety in cycling policy in the particular case of Ireland, which is one of the countries that has a government-endorsed policy to increase cycling. It examines available information on cycling, including police-reported accident data over the last fifteen years and more recent hospital accident data. Comparisons are made for injuries between the two sources and data matching and capture-recapture methods used to estimate injuries. The deficiencies in data are analysed and suggestions for improvements made. It is argued that gaps in available data have led to accident risks being poorly understood. The paper discusses how a convergence in cycling and safety policy can be achieved and suggests additional measures including a cycling safety target, increased communication on cycling safety, lower urban speeds and compulsory wearing of helmets for children. Background: Road Safety and Cycling in Ireland Road safetyIreland has had a highly successful decade in reducing road traffic fatalities and has jumped up the international safety league tables (International Transport Forum( IRTAD), 2011). Not previously known as a country with a particularly good road safety record, the achievements, especially in the latter half of the past decade have attracted a lot of attention from international bodies and from countries keen to understand the reasons for the improvements. With a fatality rate of less than 50 per million people, Ireland, in 2010, joined the elite group of around ten countries achieving this level (IRTAD, 2011). In terms of fatalities per vehicle kilometre, arguably a more correct measure, Ireland was rated fourth behind Iceland, Sweden and Great Britain, recording fewer than 5 deaths per billion vehicle kilometres. Cyclist fatalities have also fallen significantly and at 5 in 2010 were among the lowest (absolutely and per capita) in the OECD (IRTAD, 2011). CyclingBetween 1986 and 2002, cycling in Ireland declined dramatically as can be seen from Table 1. The number of people cycling to work almost halved and, because the number at work grew, the modal share by bike fell from 5.6% to below 2% (CSO 2011(a)). The decline in cycling to school or college was even more striking. Between 1986 and 2006, the number of children between 5 and 12 years of age cycling to school fell by 83%. The number of 13-18 year olds cycling to school fell even more, by 85% and the number of girls of this age cycling to school *Manuscript (without author's information) Click here to view linked References 2 fell by 98%. In 2011, there were as many people in this age group driving a car to school or college as there were cycling. (CSO, 2011 (a)).The decline halted between 2002 and 2006 and cycling started to grow again. Supported by policy, including a successful scheme to subsidise bicycle purchase through tax-free loans (...

show abstract

Section: Data Matchingmentioning

confidence: 99%

mentioning

confidence: 99%

The safety challenge of increased cycling

Short

Caulfield

2014

Transport Policy

View full text Add to dashboard Cite

show abstract

“…These conditional probabilities are conventionally known as m and u probabilities (Clark, 2004). This ranking of likelihood ratios and predetermined statistical error levels lead to two thresholds, a higher and lower, which determine membership of the sets A 1 , A 2 and A 3 .…”

Section: Classical and Bayesian Approachesmentioning

confidence: 99%

“…Using the odds form of Bayes' theorem and following Clark (2004), the posterior odds of a true match is the prior odds times the product of the likelihood ratios. This allows calculation of a direct relationship between the weights as defined above and the posterior probability (P) as:…”

Section: Decision Rules For Matches and Non-matchesmentioning

confidence: 99%

Record linkage for road traffic injuries in Ireland using police hospital and injury claims data

Short

Caulfield

2016

Journal of Safety Research

View full text Add to dashboard Cite

The study of non-fatal road traffic injuries is growing in importance. Since there are rarely comprehensive injury datasets, it is necessary to combine different sources to obtain better estimates on the extent and nature of the problem. Record linkage is one such technique. In this study, anonymised datasets from three separate sources of injury data in Ireland -hospitals, police and injury claims are linked using probabilistic and deterministic linkage techniques. A method is proposed that creates a 'best' set of linked records for analysis, useful when clerical review of undecided cases is not feasible. The linkage of police and hospital datasets shows results that are not dissimilar to those found in other countries, with significant police understatement especially of cyclist and motorcyclist injuries. The addition of the third dataset identifies a large number of additional injuries and demonstrates the error of using only the two main sources for injury data. The study also underlines the risk in relying on the LincolnPetersen capture-recapture estimator to provide an estimate of the total population concerned. The data show that road traffic injuries are significantly more numerous than either police or hospital sources indicate. It is also argued that no single measure can fully capture the range of impacts that a serious injury entails. Record Linkage for Road Traffic Injuries in Ireland using Record Linkage for Road Traffic Injuries in Ireland using Police Hospital and Injury Claims Data ABSTRACTThe study of non-fatal road traffic injuries is growing in importance. Since there are rarely comprehensive injury datasets, it is necessary to combine different sources to obtain better estimates on the extent and nature of the problem. Record linkage is one such technique. In this study, anonymised datasets from three separate sources of injury data in Ireland -hospitals, police and injury claims are linked using probabilistic and deterministic linkage techniques. A method is proposed that creates a 'best' set of linked records for analysis, useful when clerical review of undecided cases is not feasible. The linkage of police and hospital datasets shows results that are not dissimilar to those found in other countries, with significant police understatement especially of cyclist and motorcyclist injuries. The addition of the third dataset identifies a large number of additional injuries and demonstrates the error of using only the two main sources for injury data. The study also underlines the risk in relying on the Lincoln-Petersen capture-recapture estimator to provide an estimate of the total population concerned. The data show that road traffic injuries are significantly more numerous than either police or hospital sources indicate. It is also argued that no single measure can fully capture the range of impacts that a serious injury entails.

show abstract

“…First, as already mentioned, real-world data is dirty, and matching therefore is reliant upon the attributes available in common in all databases to be matched. In real-world applications, data quality can be one of the biggest obstacles to successful data matching [16]. Second, even when efficient blocking techniques are applied, the number of candidate record pairs can be very large, and the resulting computational requirements (run-times and memory foot-prints) can become challenging even for today's powerful computing platforms.…”

Section: Record Pair Classificationmentioning

confidence: 99%

Geocode Matching and Privacy Preservation

Christen

2009

Privacy, Security, and Trust in KDD

View full text Add to dashboard Cite

Abstract. Geocoding is the process of matching addresses to geographic locations, such as latitudes and longitudes, or local census areas. In many applications, addresses are the key to geo-spatial data analysis and mining. Privacy and confidentiality are of paramount importance when data from, for example, cancer registries or crime databases is geocoded. Various approaches to privacy-preserving data matching, also called record linkage or entity resolution, have been developed in recent times. However, most of these approaches have not considered the specific privacy issues involved in geocode matching. This paper provides a brief introduction to privacy-preserving data and geocode matching, and using several real-world scenarios the issues involved in privacy and confidentiality for data and geocode matching are illustrated. The challenges of making privacy-preserving matching practical for real-world applications are highlighted, and potential directions for future research are discussed.

show abstract

Practical introduction to record linkage for injury research

Abstract: doi:10.1136/ip.2003.00458

Cited by 147 publications

References 60 publications

The safety challenge of increased cycling

The safety challenge of increased cycling

Record linkage for road traffic injuries in Ireland using police hospital and injury claims data

Geocode Matching and Privacy Preservation

Contact Info

Product

Resources

About