The findings and conclusions in these presentations have not been formally disseminated by the Centers for Disease Control and Prevention and should not be construed to represent any agency determination or policy.

Wednesday, May 10, 2006
287

Suppression of Sexually Transmitted Disease (STD) Data to Protect Confidentiality: At What Cost?

Kristine T. Edwards1, Philip Christopher Delcher2, Jeff Stover2, Lori M. Newman3, and Sam Groseclose3. (1) Center for Food Security and Public Health, Iowa State University, 2150 College of Veterinary Medicine, Iowa State University, Ames, IA, USA, (2) Virginia Department of Health, Richmond, VA, (3) Division of STD Prevention, CDC, Atlanta, GA


Background:
It is important to balance value of state STD surveillance data dissemination to decision-makers and other interested parties with protection of individual confidentiality. Common methods intended to protect STD data confidentiality involve suppression rules such as “rule of 5” (deleting cells with numerator less than 5) and “Missouri Rule” (deleting cells using numerator-denominator algorithm). The history, effectiveness, rationale for use, and application of suppression rules among states are not well-documented.

Objective:
To describe state data suppression rules used and model relative impact of commonly used rules.

Method:
State data suppression rule usage was obtained from a national survey of STD incidence by county, race, and gender conducted by CDC's Office of Women's Health (OWH) . We modeled selected suppression rules using data supplied by Virginia to OWH for women with gonorrhea and described potential loss of useful information, complexity of application, and protection of confidentiality.

Result:
Twenty-one (43%) of 49 states did not suppress data, 4 (8%) applied “rule of 3”, 5 (10%) “rule of 5”, and 19 (39%), one of 10 different suppression rules. Applying “rule of 3”, “rule of 5”, and “Missouri rule” to Virginia data resulted in 20%, 30% and 0.01% data loss, respectively. Complexity of application was greatest using numerator-denominator rules and moderate for numerator-only rules. Plausible scenarios for breach of confidentiality were identified for all rules examined.

Conclusion:
There is no uniform application of suppression rules among state STD programs. Application of suppression rules can be time-consuming and complex, may result in significant data loss, and may still allow breach of confidentiality.

Implications:
Improved national guidelines for state release of STD data are needed to maximize public health data dissemination, minimize data re-release burden, and more uniformly protect individual confidentiality.