Jove
Visualize
Contact Us
JoVE
x logofacebook logolinkedin logoyoutube logo
ABOUT JoVE
OverviewLeadershipBlogJoVE Help Center
AUTHORS
Publishing ProcessEditorial BoardScope & PoliciesPeer ReviewFAQSubmit
LIBRARIANS
TestimonialsSubscriptionsAccessResourcesLibrary Advisory BoardFAQ
RESEARCH
JoVE JournalMethods CollectionsJoVE Encyclopedia of ExperimentsArchive
EDUCATION
JoVE CoreJoVE BusinessJoVE Science EducationJoVE Lab ManualFaculty Resource CenterFaculty Site
Terms & Conditions of Use
Privacy Policy
Policies

Related Experiment Videos

Probabilistic record linkage and a method to calculate the positive predictive value.

Tony Blakely1, Clare Salmond

  • 1Department of Public Health, Wellington School of Medicine, University of Otago, PO Box 7343, Wellington, New Zealand. tblakely@wnmeds.ac.nz

International Journal of Epidemiology
|January 24, 2003
PubMed
Summary

A new "duplicate method" accurately estimates the positive predictive value (PPV) of computerized record linkage for anonymous data. This approach is valuable for cohort studies needing reliable outcome ascertainment without personal identifiers.

Related Concept Videos

You might also read

Related Articles

Articles linked to this work by shared authors, journal, and citation graph.

Sort by
Same author

Effectiveness of COVID-19 vaccine booster doses in adults aged 50 years and over during the Omicron period in Victoria, Australia.

Communicable diseases intelligence (2018)·2026
Same author

Is morbidity expanding? An epidemiological framework for understanding morbidity expansion and multimorbidity.

International journal of epidemiology·2026
Same author

Lessons from COVID-19 for policy-facing epidemiologists needing to consider health, social, and economic impacts.

International journal of epidemiology·2026
Same author

The Health and Cost Impacts of Interventions Designed to Reduce Dietary Sodium Intake in Australia: A Modeling Study.

The Journal of nutrition·2025
Same author

Estimates of non-communicable disease expenditure by disease phase, sex, and age group for all OECD countries.

Population health metrics·2025
Same author

Housing as a social determinant of health: a contemporary framework.

The Lancet. Public health·2025

Area of Science:

  • Epidemiology
  • Biostatistics
  • Data Science

Background:

  • Computerized record linkage is crucial for cohort studies to determine study outcomes.
  • The accuracy of record linkage is typically assessed using sensitivity and positive predictive value (PPV).

Purpose of the Study:

  • To introduce and validate a novel 'duplicate method' for calculating the PPV of record linkage.
  • To provide a method for PPV estimation that does not require a validation subset of records with personal identifiers, making it suitable for anonymous data linkage.

Main Methods:

  • The 'duplicate method' assumes that the distribution of links between two files follows combinatorial probabilities.
  • This probabilistic assumption allows for the estimation of false positive links and subsequent PPV calculation.

Related Experiment Videos

  • The method was demonstrated using anonymous, probabilistic record linkage of New Zealand census and mortality data.

    • Main Results:

    • PPV estimates derived from the duplicate method aligned with theoretical expectations for probabilistic record linkage.
    • Sensitivity analyses confirmed the robustness of the PPV estimates.
    • The method proved effective for linkage projects utilizing anonymous datasets.

    Conclusions:

    • The 'duplicate method' offers a reliable approach to estimating the PPV of record linkage with anonymous data.
    • This method is particularly useful in epidemiological research where data privacy is a concern.
    • Further research is encouraged to validate and refine the accuracy of this novel linkage assessment technique.