One-Sided Versus Two-Sided Testing

8.11 Conclusions

I believe one-sided (benefit only) testing reflects a mindset of physicians and healthcare researchers who believe their intervention can produce no harm, a philosophy that has been shown repeatedly to be faulty, and dangerous to patients and their families. Investigators who agree to the one-sided (benefit only) approach to significance testing in a clinical experiment have closed parts of their minds to the possibility of harm entering into an avoidable flirtation with danger. In this sense, the one-sided test is not the disease, it is only a symptom.

We as physicians and healthcare workers feel strongly about the treatment programs we advocate. This is required in our profession. However, these strong feelings often betray us since our day-to-day experience does not provide an objective view of the treatments. We need every tool we can find to help us gain that objective vantage point. The use of a two-sided significance test is of utmost importance. A forceful, intelligent argument for ethics will argue not only for a two-sided test, but asymmetrical allocation of alpha. Prospective identification of alpha is again critical here, and community protection predominates all other concerns. A sensitive, perceptive, ethical approach to alpha allocation for sidedness can complicate experimental design, but complexity in the name of ethics is no vice.