Method Validation Edition

Bland-Altman analysis for method agreement Limits of agreement with constant and non-constant precision, mean and median bias, replicate measurements, and confidence intervals on the limits — per EP09-A3.

See the agreement between methods at every concentration

A regression line tells you the average relationship between two methods. It doesn’t show you the individual differences, whether they’re constant or widen at higher concentrations, or whether any single patient result might be clinically misclassified. The difference plot shows all of that — which is why Bland-Altman has become the standard approach for visualising and quantifying method agreement across clinical laboratories, IVD development, and published research.

Analyse-it provides the full Bland-Altman implementation with three bias models, limits that adapt to the precision structure of your data, and confidence intervals on the limits themselves — not just on the bias estimate. The mountain plot alongside gives you the cumulative distribution of differences for a second perspective on agreement.

We use Analyse-it frequently for our verification and pre-verification work, in accordance with CLSI guidelines for in-vitro diagnostics. It’s saved time and effort compared to the hodge-podge of applications we used before, JMP, SAS, etc...

Brian Noland, Ph.D.
Principal Scientist, Product Development
Biosite / Inverness Medical Innovations

What's included

Difference plot with limits of agreement and allowable difference band

Differences plotted against the best estimate of the true value, with bias, limits of agreement, and allowable difference band overlaid. Specify allowable difference as absolute, percentage, or combination — such as “±4 mmol/L” or “10%, with a minimum of 5 mg/dL” — to assess whether agreement is within clinically acceptable limits. Plot as difference, relative difference, or ratio. Mountain plot alongside for the cumulative distribution of differences, also with allowable difference band.

Mean, median, and linear fit bias

Three bias models in one analysis. Mean bias for centred, symmetric differences. Median bias for skewed distributions. Linear fit when bias scales with concentration. Compare all three on the same data and choose the one that represents the relationship.

Constant and non-constant precision limits

Horizontal limits of agreement for constant precision across the range. V-shaped limits that widen proportionally when precision is non-constant. Nonparametric limits when differences are non-normally distributed.

Confidence intervals on the limits themselves

Confidence intervals on both the bias estimate and the upper and lower limits of agreement — reporting the uncertainty in the agreement boundary, not just the central estimate.

Partition the measuring range

Partition the data into separate measuring intervals, each with its own bias estimate, limits of agreement, and allowable difference assessment — for example, ±0.06 μg/L below the clinical threshold and ±6% above it. Use any fit within each interval per EP09-A3.

Replicate measurements with correct variance

Singlicate, duplicate, or any number of replicates. Within-subject variation estimated directly from the replicate structure and limits adjusted accordingly — not approximated by averaging.

Example analyses

See Bland-Altman results in detail — difference plots, limits of agreement, mountain plots, and allowable difference bands — using CLSI example datasets you can download and follow along with.

EP09-A3 — Appendix I
Bland-Altman with partitioned measuring range. Measuring range partitioned into 0 to 1.8 μg/L (allowable difference ±0.06 μg/L) and 1.8 to 100 μg/L (allowable difference ±6%).

EP21-A — Table 3
Sodium method comparison. 125 observations. Bland-Altman with mean bias, 95% limits of agreement with 90% CIs, mountain plot, and allowable difference ±4 mmol/L.

Part of the method comparison workflow

Bland-Altman is one of six methods in the method comparison analysis. For a regression-based bias estimate, see Passing-Bablok for a non-parametric approach or Deming and Weighted Deming for a parametric approach.

Software you can trust

Validated calculations you can defend at inspection Every calculation is performed by Analyse-it — no Excel formulas, no third-party functions. Results are validated against CLSI reference datasets, published datasets, and thousands of internal test cases before every release. See how we develop and validate Analyse-it →

Data stays in your facility Analyse-it runs entirely within Microsoft Excel on your PC. No cloud processing, no data transmission. Pre-submission data, patient-adjacent data, and in-process results stay within your facility under your own data governance controls.

Standard Excel workbooks anyone can open Every analysis is an ordinary .xlsx workbook. Share with colleagues, submit to regulatory affairs, archive for audit, open on any PC with Excel. No proprietary format, no licence required to view results. Colleagues and auditors see exactly what you see.

Results that can’t be accidentally broken Analysis output contains computed values, not formulas. Nothing to accidentally overwrite, no cell references to break, no formula errors to introduce. The results you reported are exactly what you’ll find when you reopen the workbook months or years later for an audit.

Technical details

CLSI protocols

EP09-A3: Measurement Procedure Comparison and Bias Estimation Using Patient Samples

Bland-Altman agreement

Mean or Median bias for constant precision (horizontal limits)
Linear regression fit bias new in v3.75 for non-constant precision (V-shaped limits)
Limits of agreement with confidence intervals

Study design

Singlicate, duplicate, and replicate measurements
Reduce or partition measuring interval new in v4.00

Plots

Difference plot with bias, limits of agreement and allowable difference
Plot difference, relative difference, or ratio
Histogram of differences
Mountain plot with allowable difference band new in v3.71

Method Validation Edition Bland-Altman analysis for method agreement Limits of agreement with constant and non-constant precision, mean and median bias, replicate measurements, and confidence intervals on the limits — per EP09-A3.