The statistical linkage key (SLK-581) is a common tool for record linkage in Australia, due to its ability to provide some privacy protection. However, newer privacy-preserving approaches may provide greater privacy protection, while allowing high-quality linkage.
Objective:
To evaluate the standard SLK-581, encrypted SLK-581 and a newer privacy-preserving approach using Bloom filters, in terms of both privacy and linkage quality.
Method:
Linkage quality was compared by conducting linkages on Australian health datasets using these three techniques and examining results. Privacy was compared qualitatively in relation to a series of scenarios where privacy breaches may occur.
Results:
The Bloom filter technique offered greater privacy protection and linkage quality compared to the SLK-based method commonly used in Australia.
Conclusion:
The adoption of new privacy-preserving methods would allow both greater confidence in research results, while significantly improving privacy protection.
Australian Bureau of Statistics (2013) National Early Childhood Education and Care Collection: Data Collection Guide. Cat. 4240.0.55.002. Canberra:Australian Bureau of Statistics.
2.
Australian Government (2012) Privacy Amendment (Enhancing Privacy Protection) Bill 2012, Explanatory Memorandum. Canberra: Australian Government.
3.
Australian Institute of Health and Welfare (AIHW) (2009) Enhancing the Alcohol and Other Drug Treatment Services National Minimum Data Set. Canberra: AIHW.
4.
Australian Institute of Health and Welfare (AIHW) (2013) Disability Services Minimum Data Set: Data Guide. Canberra: AIHW.
BassJGarfieldC (2002) Statistical linkage keys: how effective are they?Symposium on Health Data Linkage, Sydney 2002. Available at: http://www.publichealth.gov.au/symposium.html(accessed 26 April 2016).
7.
BertinoELinDJiangW (2008) A survey of quantification of privacy preserving data mining algorithms. Privacy-preserving Data Mining34: 183–205.
8.
BoydJHFerranteAMO’KeefeCM. (2012) Data linkage infrastructure for cross-jurisdictional health-related research in Australia. BMC Health Services Research12: 480.
9.
BoyleDIRRafaelN (2011) BioGrid Australia and GRHANITE: privacy-protecting subject matching. Studies in Health Technology and Informatics168: 24–34.
10.
ChristenP (2012) Data Matching. New York: Springer.
11.
ChristenPGoiserK (2007) Quality and complexity measures for data linkage and deduplication. In: GuilletFHamiltonH (eds) Quality Measures in Data Mining Studies in Computational Intelligence. New York:Springer, pp. 127–151.
12.
Community Services Ministers Advisory Council (2004) Statistical Data Linkage in Community Services Data Collections. Canberra: Community Services Ministers Advisory Council.
13.
El EmamKDankarFKVaillancourtR. (2009) Evaluating the risk of re-identification of patients from hospital prescription records. The Canadian Journal of Hospital Pharmacy62: 307.
14.
GrahamC (2012) Anonymisation: managing data protection risk code of practice. Cheshire, UK: Information Commissioner’s Office. Available at: https://ico.org.uk/media/1061/anonymisation-code.pdf(accessed 26 April 2016).
15.
HetzelD (2006) Data linkage research—can we reap benefits for society without compromising public confidence?Australian Health Consumer2: 27–28.
JaroMA (1989) Advances in record-linkage methodology as applied to matching the 1985 census of Tampa, Florida. Journal of the American Statistical Association89: 414–420.
18.
KarmelR (2005) Transitions between Aged Care Services. Data Linkage Series no. 2. Canberra: AIHW.
19.
KarmelRAndersonPGibsonD. (2010) Empirical aspects of record linkage across multiple datasets using statistical linkage keys: the experience of the PIAC cohort study. BMC Health Services Research10: 41.
20.
KarmelRRosmanD (2008) Linkage of health and aged care service events: comparing linkage and event selection methods. BMC Health Services Research8: 149.
21.
KelmanCBassAHolmanD. (2002) Research use of linked health data: a best practice protocol. Australian and New Zealand Journal of Public Health26: 5.
22.
KuzuMKantarciogluMDurhamE. (2011) A constraint satisfaction cryptanalysis of Bloom filters in private record linkage. Privacy Enhancing Technologies pp. 226–245. Berlin Heidelberg: Springer.
23.
LawrenceGDinhITaylorL (2008) The centre for health record linkage: a new resource for health services research and evaluation. Health Information Management Journal37: 60–62.
24.
LovettRFisherJAl-YamanF. (2008) A review of Australian health privacy regulation regarding the use and disclosure of identified data to conduct data linkage. Australian and New Zealand Journal of Public Health32: 282–285.
25.
MitchellRJCameronCMMcClureRJ. (2015) Data linkage capabilities in Australia: practical issues identified by a population health research network ‘proof of concept project’. Australian and New Zealand Journal of Public Health39: 319–325.
26.
O’KeefeCMConnollyC. (2011) Regulation and perception concerning the use of health data for research in Australia. Electronic Journal of Health Informatics6: 16.
RandallSMFerranteAMBoydJH. (2013) Privacy-preserving record linkage on large real world datasets. Journal of Biomedical Informatics50: 205–212.
30.
RandallSMFerranteAMBoydJH. (2014) Privacy-preserving record linkage on large real world datasets. Journal of Biomedical Informatics50: 205–212.
31.
RosmanDGarfieldCFullerS. (2002) Measuring data and link quality in a dynamic multi-set linkage system. Sydney (NSW): Symposium on Health Data Linkage. Available at: http://www.publichealth.gov.au/symposiumpdf/rosman_a.pdf(accessed 26 April 2016).
32.
RyanTHolmesBGibsonD. (1999) A National Minimum Data Set for Home and Community Care. Canberra: Australian Institute of Health and Welfare.
33.
SchnellRBachtelerTReiherJ (2009) Privacy-preserving record linkage using Bloom filters. BMC Medical Informatics and Decision Making9(1): 41.
34.
TaylorLKIrvineKIannottiR. (2014) Optimal strategy for linkage of datasets containing a statistical linkage key and datasets with full personal identifiers. BMC Medical Informatics and Decision Making14: 85.
