An unsupervised classification method for inferring original case locations from low-resolution disease maps
2006

Patient Privacy and Low-Resolution Disease Maps

Sample size: 550 publication 10 minutes Evidence: high

Author Information

Author(s): John S. Brownstein, Christopher A. Cassa, Isaac S. Kohane, Kenneth D. Mandl

Primary Institution: Children's Hospital Informatics Program at the Harvard-MIT Division of Health Sciences and Technology

Hypothesis

Can low-resolution disease maps effectively protect patient privacy from re-identification?

Conclusion

Lowering the resolution of disease maps does not adequately protect patient addresses from being re-identified.

Supporting Evidence

  • The method identified 26% of patient addresses from a low-resolution map.
  • For a higher-resolution map, 79% of patient addresses were correctly identified.
  • Addresses were predicted to be within 14 meters of the actual location for the publication quality map.

Takeaway

This study shows that even blurry maps can still reveal where patients live, which is not safe for their privacy.

Methodology

The study used a five-step process to reverse identify patient addresses from low-resolution maps.

Potential Biases

The method may not account for all variables affecting address re-identification.

Limitations

The study relied on simulated data rather than actual patient addresses.

Participant Demographics

The study focused on patient addresses in urban and suburban areas of Boston, MA.

Statistical Information

P-Value

p<0.05

Confidence Interval

95% CI, 27.4–30.4

Statistical Significance

p<0.05

Digital Object Identifier (DOI)

10.1186/1476-072X-5-56

Want to read the original?

Access the complete publication on the publisher's website

View Original Publication