Estimating Treatment Effects Using Observational Data
- Ralph B. D’Agostino, Jr, PhD;
- Ralph B. D’Agostino, Sr, PhD
- Author Affiliations: Department of Biostatistical Sciences, Wake Forest University School of Medicine, Winston Salem, NC (Dr D’Agostino, Jr); and Department of Mathematics and Statistics, Boston University, and Harvard Clinical Research Institute, Boston, Mass (Dr D’Agostino, Sr).
- Corresponding Author: Ralph B. D’@Agostino, Sr, PhD, Department of Mathematics and Statistics, Boston University, 111 Cummington St, Boston, MA 02215 (ralph{at}bu.edu).
Since this article does not have an abstract, we have provided the first 150 words of the full text.
- KEYWORDS:
- ANALYSIS OF VARIANCE
- CLINICAL TRIALS
- DATA INTERPRETATION, STATISTICAL
- RANDOMIZED TRIALS
- RESEARCH DESIGN
- SELECTION BIAS
The randomized clinical trial (RCT) is the ideal method for measuring treatment effects. Participants in clinical trials are randomly assigned to a treatment or control group. Randomization reduces biases by making treatment and control groups “equal with respect to all features,” except the treatment assignment. When randomization is performed correctly, differences in efficacy found by statistical comparisons can be attributed to the difference between the treatment and control.1
However, the RCT does not necessarily provide the final answer to treatment effectiveness, as there are many restrictions that limit generalizability. For example, RCTs are often restricted to patients with limited disease, comorbidity, and concomitant medications. Thus, RCTs generally demonstrate efficacy rather than effectiveness, where efficacy is the treatment effect under the restricted conditions of the RCT and effectiveness is the treatment effect under the conditions of usual practice.1
Observational, nonrandomized studies have a role when RCTs are not available, …
