Example: Counfounding Bias - HIV and Needlestick

Lead Editor(s): Jeff Martin, MD

Treatment after HIV Needlestick Exposure

An example of counding can be found in this 1997 study in the New England Journal of Medicine (Li).

This is a case-control study in health care workers looking at the effectiveness of AZT in preventing HIV seroconversion after a needlestick accident.

Failure of Randomized Clinical Trial

Attempts to address this question with a RCT did not work because

Observational Study Parameters

So, the exposure in question is the use of AZT and the Cases were health care workers who had acquired HIV after a needlestick and

Cruse Analysis of Study Data

In the crude analysis, the OR was 0.61, with a confdence interval of 0.26 to 1.4. 0330_3HIV1.JPG

Could confounding be present?

Confounding - Look at Severity of Exposure

As seen in the figure below, we should be concerned about the severity of exposure.

HIV and Treatment - Stratified Anaysis

When one controls for severity of exposure, below, Now, we have sufficient evidence to reject the null hypothesis and conclude that AZT is effective in preventing HIV seroconversion.


Li, R. W., & Wong, J. B. (1997). Postexposure treatment of HIV. N Engl J Med, 337(7), 499-500; author reply 501.