Primary and Secondary Outcomes

Every study must declare a primary outcome — the single most important result that the study is designed to detect and that the sample size calculation is based on. This is the outcome that determines whether the study 'succeeded' or 'failed.' For example, in a trial of a new blood pressure drug, the primary outcome might be 'change in systolic blood pressure at 12 weeks.' The study is powered — meaning enough participants are enrolled — to detect a clinically meaningful difference in this specific outcome. Secondary outcomes are additional measurements that provide supporting information. In the same blood pressure trial, secondary outcomes might include diastolic blood pressure change, adverse events, medication adherence, and quality of life. These are reported but the study is not specifically powered to detect differences in these outcomes. Why does this distinction matter? Because testing multiple outcomes without a pre-specified primary outcome opens the door to cherry-picking — selectively reporting whichever outcome looks best. If you test 20 outcomes, on average one will appear statistically significant by chance alone (at the p < 0.05 level). Pre-specifying the primary outcome protects against this. Study registration (e.g., on ClinicalTrials.gov) requires declaring the primary outcome before enrollment begins. This transparency measure prevents researchers from changing the primary outcome after seeing the results — a practice called 'outcome switching' that is considered a form of reporting bias.