A power analysis framework to aid the design of robust semi-field vector control experiments

Area of Science:

• Vector control research
• Experimental design methodology
• Statistical power analysis

Background:

• Semi-field experiments are vital for evaluating new vector control tools (VCTs) before large-scale field trials.
• Robust experimental design is essential for unbiased and informative results in semi-field studies.
• Power analysis is a key component for ensuring adequate statistical power in experimental designs.

Purpose of the Study:

• To develop and demonstrate a methodology for determining optimal semi-field experimental designs.
• To guide researchers in selecting the appropriate number of chambers, sampling frequency, and mosquito numbers for VCT evaluations.
• To achieve sufficient statistical power for assessing single or combined VCT impacts.

Main Methods:

• Simulated data analysis using a generalized linear mixed-effects model.
• Estimation of statistical power for various experimental designs.
• Comparison of short-term (24h) vs. long-term (3-month) experiments.
• Evaluation of single vs. combined intervention applications.

Main Results:

• The number of chambers and inter-chamber variance were dominant factors influencing statistical power.
• Increased chambers, sampling frequency, and mosquito numbers generally improved power.
• High variance between chambers significantly reduced power, emphasizing the need for standardized conditions.
• Combined interventions required more chambers than single interventions; experiment duration had minimal impact on key design aspects.

Conclusions:

• Optimizing semi-field experiment design requires balancing choices with available resources.
• The provided power analysis framework and tutorial support robust experimental design.
• This methodology facilitates the development of new and effective vector control tools (VCTs).