Is it safe to pool the blank control data with the solvent control data?

Area of Science:

• Environmental toxicology
• Chemical risk assessment
• Statistical analysis in ecotoxicology

Background:

• Ecotoxicological studies frequently employ blank controls (BC) and solvent controls (SC) when test chemicals are dissolved in solvents.
• Current guidelines often suggest a conditional testing scheme: combine BC and SC data if not significantly different, otherwise use only SC for chemical effect testing.

Purpose of the Study:

• To investigate the statistical validity of the conditional testing scheme used in ecotoxicological studies.
• To evaluate the reliability of significance testing when BC and SC data are combined conditionally.

Main Methods:

• Simulating ecotoxicological data sets under conditions with and without significant differences between BC and SC.
• Applying the conditional testing scheme to the simulated data to assess its performance.

Main Results:

• The simulations demonstrated that the conditional testing scheme does not yield a statistically valid test.
• The proposed method for combining controls can lead to erroneous conclusions regarding chemical treatment effects.

Conclusions:

• The conditional testing scheme recommended in many ecotoxicological guidelines is statistically flawed.
• Alternative or revised statistical approaches are needed to ensure the validity of significance testing in ecotoxicological studies involving solvent controls.