Efficacy vs Effectiveness

Efficacy asks whether an intervention works under ideal, controlled conditions — an explanatory trial: selected patients, a strict protocol, high adherence, expert delivery. Effectiveness asks whether it works in routine practice — a pragmatic trial: unselected patients, ordinary clinicians, real-world adherence. Efficacy generally over-estimates the benefit you’ll actually see.

One intervention evaluated two ways. The left path is the tightly-controlled explanatory trial (efficacy); the right path is the messy, real-world pragmatic trial (effectiveness). The same drug can post a bigger number on the left than it ever delivers on the right.

Place a trial on the explanatory–pragmatic spectrum by its patients and its conditions. Narrow eligibility, run-in periods that weed out non-adherers, and per-protocol analysis push it toward efficacy. Broad entry, usual care, and intention-to-treat push it toward effectiveness.

Your ED population is comorbid, unselected, and variably adherent — nothing like a clean explanatory cohort. A drug that is efficacious in an ideal RCT may be far less effective in practice. Judging external validity guards against importing an inflated benefit to the bedside.

• Efficacy = ideal conditions (explanatory trial)
• Effectiveness = real world (pragmatic trial)
• Efficacy ≥ effectiveness, almost always

Explanatory vs pragmatic, same intervention — picture a new oral anticoagulant tested first in a tightly-run explanatory RCT (carefully selected patients, monitored adherence, expert centres) where it looks highly efficacious. The pragmatic trial of the same drug — unselected ED patients, routine clinics, ordinary adherence — typically shows a smaller real-world effect. The gap is not a contradiction: it is the price of generalisability. The pragmatic estimate is the one that predicts what your department will actually see.

• Treating efficacy and effectiveness as interchangeable — they answer different questions.
• Forgetting that trial adherence (monitored, incentivised) outstrips real-world adherence.
• Over-reading an explanatory trial’s effect as the benefit your patients will get — limited external validity.