Curriculum outcome trends.

Step 1 first-time pass-rate trend

Step 1 became pass/fail in January 2022, and the passing standard was raised from 194 to 196 at the same time. The "pass/fail effect" is unmistakable in the national data: the U.S. MD first-time pass rate fell from 95% to 91% in a single year, then kept drifting downward.

Source: USMLE Performance Data, published by the NBME (usmle.org → Performance Data). "First-time" means first attempt. Committees should read this as a regime change, not a curriculum failure: the entire nation moved together, and 2022–2024 is the comparable series — do not benchmark current cohorts against pre-2022 scored-era figures.

Step 2 CK: pass rates and score norms

Since Step 1 went pass/fail, Step 2 Clinical Knowledge is the score residency programs see. Its pass rate is far higher than Step 1's, while the national mean score has held steady.

Source: USMLE Performance Data (NBME). Note the repeat-taker collapse: failing once roughly doubles the odds of failing again. That makes the first-attempt pass rate a student-success variable, not just an institutional metric — committees own it either way.

What "distance from national average" means

Regulators and peer reviewers rarely judge a school against a fixed pass-rate number. They look at distance from the national average: how far below (or above) the national first-time rate a school's cohort sits. The concept generalizes to any committee benchmark:

• Distance is relative to the right reference. A 91% Step 1 pass rate was a red flag in 2021 (national: 95%) and near the mean in 2022 (national: 91%). The same number means different things in different years — always pair a local figure with that year's national figure.
• Direction matters more than position. A school two points below the national rate with a narrowing gap over three years is a different story than one two points below with a widening gap. Committees should report gap trajectories, not just gaps.
• First-time and eventual are different metrics. First-time pass rate reflects preparation; eventual pass rate (after retakes) reflects remediation capacity. Report both, labeled, and never let one stand in for the other in a leadership slide.

Using cohort trends for continuous improvement

Curriculum committees turn outcome trends into action. The discipline is the same whether the outcome is a Step pass rate, a shelf-exam mean, or an OSCE domain score:

• Build multi-year views. One cohort's dip is a data point; three cohorts' drift is a trend. The national series above shows why: year-to-year wobble of a point or two is normal, regime changes (2022) are not.
• Separate the cohort effect from the curriculum effect. Before attributing a change to a curricular intervention, check whether the cohort's entering credentials (MCAT, GPA — see the admissions funnel) moved too. A weaker entering class predicts weaker outcomes; that is measurement, not failure.
• Disaggregate thoughtfully. National-vs-local gaps can hide subgroup patterns. Disaggregate by meaningful groups when the cohort is large enough to do so without identifying individuals — and treat subgroup findings as investigation prompts, not conclusions.
• Close the loop in writing. A CQI cycle is: observe the trend, hypothesize the cause, intervene, re-measure, document. Committees that write down what they changed and what happened next build an institutional memory that survives turnover. Accreditation reviewers notice the difference — for that angle, see SILME.

Small cohorts need extra care

Most medical school classes are a few dozen to a couple hundred students. With small denominators, a handful of outcomes swing the percentage — a class of 60 where three students fail reports 95%; where five fail, 92%. That three-point "drop" is two people, not a trend. The benchmarking page covers small-n reporting discipline in detail: roll multi-year windows, report counts alongside rates, and resist the urge to react to a single cohort.

Sources

• USMLE Performance Data — NBME, at usmle.org (Performance Data pages): Step 1/2/3 first-time pass rates, score norms, passing-standard changes.
• NRMP 2024 Program Director Survey: 77% of program directors consider failed Step attempts when deciding whom to interview.
• NRMP Charting Outcomes in the Match, 2024 (4th edition): mean Step 2 CK of matched U.S. MD seniors, 250.4 (SD 13.2).

Figures as published through 2026-09-30. The NBME and NRMP publish new data on their own schedules; check the official sites for the latest editions.