Clinical Trial Statistical Power Model
00
Determine the probability of detecting a treatment effect (statistical power) given uncertainty in drug efficacy and patient adherence.
NOTEAbout This Model
## Determine the probability of detecting a treatment effect (statistical power) given uncertainty in drug efficacy and patient adherence. This model estimates **Statistical Power** based on uncertain inputs: Effect Size, Dropout Rate. Run the simulation to see the distribution of possible outcomes.
VARIABLEDropout Rate
beta(3.5, 16.5)
Clinical trials typically see 5-30% patient attrition depending on study duration and burden.
dropout_rate
VARIABLEEffect Size
triangular(0.1, 0.3, 0.5)
Cohen's d effect size for clinical significance typically ranges from small (0.2) to medium (0.5).
effect_size
FORMULAEffective N
enrolled_sample_size * (1 - dropout_rate)
Effective sample size after accounting for dropouts.
effective_n
FORMULANcp Delta
effect_size * sqrt(effective_n / 2)
Non-centrality parameter (delta) for the power calculation based on effect size.
ncp_delta
FORMULAStatistical Power
normal_cdf(ncp_delta - 1.96)
Approximate statistical power using a normal approximation for a two-tailed test.
normal_cdf
statistical_power
OUTPUTStatistical Power
statistical_power
Probability of detecting a real treatment effect if one exists. Below 80% risks missing real effects (underpowered). Above 90% is robust but expensive. Target 80-90% for most trials.
CONSTANTEnrolled Sample Size
100
Initial number of participants enrolled per treatment arm before dropouts.
enrolled_sample_size
What is Carlo?
Carlo is a visual tool for Monte Carlo simulation. Model uncertainty by dragging probability distributions, connecting them visually, and running thousands of scenarios instantly.