What Is Blend Uniformity CV and How Do You Calculate It?

If you manufacture tablets, capsules, or any solid dosage form, one question matters above all: Is the active ingredient evenly distributed throughout the blend? The answer is quantified by the Blend Uniformity Coefficient of Variation (CV) — a simple but powerful statistic used in the pharmaceutical industry.

This article explains what blend uniformity CV means, how to calculate it, and why it is essential for quality control — all within the framework of USP <905> content uniformity thinking.

---

What It Is

Blend uniformity CV is the relative standard deviation (RSD) of assay results (e.g., % label claim) measured on multiple samples taken from a powder blend.

In the context of USP <905> (Content Uniformity), the goal is to ensure that each dosage unit contains the active ingredient within a narrow range of the labeled amount. Before tablets are compressed, you test the blend itself: if the blend is not uniform, the final tablets cannot be uniform.

The CV expresses the variability of those assay values as a percentage of the mean. A low CV (e.g., ≤ 5%) indicates a well-mixed blend; a high CV signals poor mixing, segregation risk, or sampling error.

---

How It Works / Formula

The blend uniformity CV is calculated in three steps:

  1. Sample the blend — take \( n \) samples (typically 10 or more) from different locations in the blender.
  2. Assay each sample — determine the active ingredient content as a percentage of label claim (or mg per sample).
  3. Compute the statistics:


\[
\text{Mean} (\bar{x}) = \frac{\sum x_i}{n}
\]

\[
\text{Standard deviation} (s) = \sqrt{\frac{\sum (x_i - \bar{x})^2}{n-1}}
\]

\[
\text{CV} (\%) = \frac{s}{\bar{x}} \times 100
\]

Where:
  • \( x_i \) = assay value of sample \( i \)
  • \( n \) = number of samples
  • \( \bar{x} \) = mean assay value
  • \( s \) = sample standard deviation (using \( n-1 \) denominator)


Interpretation guidance (typical industry practice):
  • CV ≤ 5% is generally considered acceptable for a well-mixed blend.
  • CV > 5% may indicate inadequate mixing, segregation, or a need to investigate the sampling method.


Note: USP <905> itself specifies acceptance criteria for finished dosage units (e.g., AV ≤ 15.0 for immediate-release products). The CV is a complementary tool used during blend characterization and process validation.



---

A Worked Illustrative Example

Example data (illustrative only):
Suppose you take 10 blend samples and obtain the following assay results (% label claim):

Sample | Assay (%)
  • 1 | 98.2
  • 2 | 101.5
  • 3 | 99.0
  • 4 | 100.8
  • 5 | 97.6
  • 6 | 102.1
  • 7 | 99.4
  • 8 | 100.2
  • 9 | 98.9
  • 10 | 101.0


Step 1 — Mean:
\[
\bar{x} = \frac{98.2 + 101.5 + 99.0 + 100.8 + 97.6 + 102.1 + 99.4 + 100.2 + 98.9 + 101.0}{10} = \frac{998.7}{10} = 99.87\%
\]

Step 2 — Standard deviation:
First, compute the squared deviations from the mean (shown for a few samples):
  • (98.2 − 99.87)² = 2.79
  • (101.5 − 99.87)² = 2.66
  • (99.0 − 99.87)² = 0.76

… and so on for all 10 samples.

Sum of squared deviations = 20.41
\[
s = \sqrt{\frac{20.41}{10-1}} = \sqrt{2.268} = 1.506\%
\]

Step 3 — CV:
\[
\text{CV} = \frac{1.506}{99.87} \times 100 = 1.51\%
\]

Conclusion: A CV of 1.51% is well below the 5% threshold, indicating excellent blend uniformity.

---

Common Pitfalls

  1. Sampling bias — Taking samples only from the top of the blender misses segregation at the bottom or sides. Always sample from multiple depths and locations.
  2. Sample size too small — With fewer than 10 samples, the CV estimate is unstable. Use at least 10, preferably more.
  3. Using population standard deviation (n) instead of sample (n−1) — This underestimates variability, especially at small n.
  4. Ignoring the mean — A low CV is meaningless if the mean is far from 100% label claim. Always report both.
  5. Confusing blend CV with finished product AV — Blend uniformity CV is a process indicator; USP <905> acceptance value (AV) applies to final dosage units.


---

Ready to Calculate?

Manual calculation is fine for a handful of samples, but for routine monitoring or validation batches, a dedicated tool saves time and prevents arithmetic errors. Use the free, instant Blend Uniformity CV calculator at 6SQ:
? https://www.6sq.com/tools/mixuni/

Enter your assay values, and the tool returns the mean, standard deviation, and CV — so you can focus on making your blend uniform, not on crunching numbers.
Invited:

0 replies, guests cannot view replies. For more features, please log in or register