E-Value for Unmeasured Confounding

Problem

Return the E-value for unmeasured confounding, RR + sqrt(RR*(RR - 1)) with RR = max(risk_ratio, 1/risk_ratio). The E-value is the minimum strength of association an unmeasured confounder would need to explain the effect away. Return 0.0 when risk_ratio is not positive.

Hint

Fold protective effects to their reciprocal before applying the formula.

Test cases

case 1

in: [2]

expected: 3.414213562373095

case 2

in: [1]

expected: 1

case 3

in: [0.5]

expected: 3.414213562373095

case 4

in: [3]

expected: 5.449489742783178

case 5

in: [0]

expected: 0

Starter codePython

def e_value(risk_ratio):
    # Your code here
    pass

5 test cases · run your code against them in the editor