Confidence intervals and controlled clinical trials: incompatible tools for medical research

The probabilistic rationale for statistical design and analysis of clinical trials is random assignment. While arithmetic and mathematical formulations may be identical to those used with random samples, we should not indiscriminately borrow tools from survey sample methods. Specifically, the confidence interval should be used sparingly, if at all. Observations have an internal validity, within the clinical trial, with no basis for claims of quantitative external generalizability. Confidence intervals encourage an unnecessary dependence on statistical analysis when the careful design should allow the data to speak for itself. Confidence intervals encourage a statistical focus and statistical conclusions that ignore scientific context and misrepresent relationships among results from related research. The clinician is presented with information about population parameters when facing confidence intervals. These do not address questions about treatment and prognosis of an individual patient. Confidence intervals are particularly distracting when a clinical trial has failed to produce anticipated results. The clinical trial is the model research tool for clinical medical research, founded on randomization. The confidence interval is a statistical tool for parameter estimation based on population sampling concepts. These tools are incompatible.