* Total Points: 0
Back to Lessons Running the Numbers 0 pts Module 6 · Lesson 3
Introduction

Running the Numbers

Drive the four levers yourself and watch what sample size actually buys

Nobody Develops Intuition from Formulas

You now know the four levers: sample size, effect size, alpha, variability. What you don't have yet is a feel for the exchange rates — how much power one more patient buys, how brutally the cost climbs as the effect shrinks, why "we'll just enroll 50 and see" almost never works.

This lesson gives you a live power calculator. You'll set up studies, break them, and fix them. By the end, when someone proposes detecting a 5% difference with 60 patients, you'll feel the wrongness before you compute it.

The calculator uses the standard normal-approximation formulas — the same math inside most power software. It's built for intuition and journal-club sanity checks. For an actual protocol, confirm with dedicated software or a statistician (they'll add refinements like continuity corrections that nudge n slightly upward).

The Recipe

Every two-arm sample size calculation needs the same four ingredients.

1. The comparison, in absolute terms

For a binary outcome: the control event rate and the treatment event rate you're designing to detect (e.g., SSI of 20% vs 10%). For a continuous outcome: the difference in means and the outcome's standard deviation (e.g., 2 days shorter stay, SD 5 days). This pair encodes both the effect size and the variability levers.

2. Alpha

Almost always 0.05, two-sided. Stricter alpha (0.01) buys fewer false alarms at the cost of more patients.

3. Target power

80% is the conventional floor; 90% is increasingly expected for definitive trials. The step from 80% to 90% costs about a third more patients — you'll verify this yourself in a minute.

4. Reality adjustments

The formula's n is the number of analyzable patients. Real studies lose people to dropout, crossover, and missing data — so inflate: enrolling n / (1 − dropout rate) protects the analysis (e.g., need 200, expect 20% attrition → enroll 250). Section 4 covers the other adjustments that show up in real protocols.

That's the whole recipe. When a methods section says "assuming X% vs Y%, alpha 0.05 and 80% power, we required N patients per arm," you now know each ingredient — and after Lesson 2, you know which one to interrogate.

The Calculator

Two-arm comparison, two-sided test. Drag the sliders and watch power respond.

Experiment 1 — the price of precision: Set SSI rates to 20% vs 10% and find the n that first reaches 80% power (about 197/arm). Now nudge the treatment rate to 15% — a 5-point difference instead of 10 — and find n for 80% again (about 903). Feel that? Nearly 5× the patients for half the effect.
Experiment 2 — the 90% surcharge: At 20% vs 10%, compare the "n for 80%" and "n for 90%" readouts. The extra ~34% of patients is the standing price of dropping your miss rate from 20% to 10%.
Experiment 3 — the flat tail: Set rates to 30% vs 15% and slide n from 20 up to 120: power climbs from ~21% to 80%. Now slide from 240 up to 1000: you're spending 760 patients per arm to buy the last two points of power. Sample size has diminishing returns — the first patients are the valuable ones.
Experiment 4 — continuous outcomes: Switch to Continuous mode. Detecting a 2-day LOS difference with SD 5 needs ~99/arm. Now double the SD to 10 (a messier population) and watch the same clinical question quadruple in price. Variability is a lever — cleaner outcomes and tighter populations are free power.

What Real Protocols Add

The formula is the floor. Four adjustments separate a textbook calculation from a fundable one.

Dropout inflation

Divide, don't multiply: needing 200 analyzable patients with 20% expected attrition means enrolling 200/0.8 = 250, not 200 × 1.2 = 240. The difference grows with the dropout rate, and using the wrong version is a surprisingly common protocol error.

Survival outcomes: events, not patients

For time-to-event analyses (Module 4), power comes from the number of events — deaths, amputations, occlusions — not the number of enrolled patients. A 500-patient trial with 30 events is a 30-event trial. This is why low-event-rate trials run for years and why "we'll just follow them longer" is a legitimate power strategy.

Same logic underlies logistic regression's events-per-variable rule from Module 5 — in rare-outcome studies, events are the real currency everywhere.

Unequal allocation

2:1 randomization (often used when patients or surgeons prefer the new treatment) costs power: for the same total n, 2:1 needs about 12% more patients than 1:1 to reach the same power. Beyond 3:1 the penalty climbs steeply. Balanced arms are the efficient default; deviations need a reason.

Multiple endpoints and interim looks

Each additional primary endpoint or interim analysis spends alpha (more chances for a false alarm), so honest designs split the alpha or use formal spending rules — and that stricter effective alpha raises the required n. A power statement that names one endpoint while the abstract celebrates a different one is a Module 1 red flag with a Module 6 explanation.

Reading order for any power paragraph: ingredients (rates, delta, alpha, power) → arithmetic (does the n check out?) → adjustments (dropout? events? allocation?) → and finally, did the analysis actually use the endpoint the study was powered for?

Exercise: Run Your Own Numbers

Use the calculator below to answer — these are real design problems, not memory questions.

Question 1 of 8

Lesson Complete!

0
Total Points Earned
Exercise (0/8 correct) +0 pts
Lesson Completed +100 pts

Running the Numbers

Module 6 - Lesson 3 complete

Key Takeaways

  • Four ingredients: absolute rates (or mean difference + SD), alpha, target power, reality adjustments. Every power paragraph is these four.
  • Exchange rates: halving delta ≈ 4× the patients; 80% → 90% power ≈ +34%; alpha 0.05 → 0.01 ≈ +50%.
  • Diminishing returns: power climbs steeply at small n and crawls near the top — the first patients are the valuable ones.
  • Dropout: divide by (1 − rate), don't multiply by (1 + rate).
  • Survival studies run on events, not enrollment — follow-up duration is a power lever.
  • This calculator is for intuition and journal club; protocols deserve dedicated software and ideally a statistician.

Next lesson: The Underpowered Study Problem — what to do (and what never to claim) when the sample size is fixed by reality, plus the post-hoc power fallacy that reviewers keep requesting anyway.