relative_risk(exposed_cases, exposed_total, control_cases, control_total)
This function computes the relative risk associated with a 2x2 contingency table (, section 2.2.3; , section 3.1.2). Instead of accepting a table as an argument, the individual numbers that are used to compute the relative risk are given as separate parameters. This is to avoid the ambiguity of which row or column of the contingency table corresponds to the "exposed" cases and which corresponds to the "control" cases. Unlike, say, the odds ratio, the relative risk is not invariant under an interchange of the rows or columns.
The R package epitools has the function riskratio, which accepts a table with the following layout
disease=0 disease=1
exposed=0 (ref) n00 n01
exposed=1 n10 n11
With a 2x2 table in the above format, the estimate of the CI is computed by riskratio when the argument method="wald" is given, or with the function riskratio.wald.
For example, in a test of the incidence of lung cancer among a sample of smokers and nonsmokers, the "exposed" category would correspond to "is a smoker" and the "disease" category would correspond to "has or had lung cancer".
To pass the same data to relative_risk, use
relative_risk(n11, n10 + n11, n01, n00 + n01)
The number of "cases" (i.e. occurrence of disease or other event of interest) among the sample of "exposed" individuals.
The total number of "exposed" individuals in the sample.
The number of "cases" among the sample of "control" or non-exposed individuals.
The total number of "control" individuals in the sample.
The object has the float attribute relative_risk, which is
rr = (exposed_cases/exposed_total) / (control_cases/control_total)
The object also has the method confidence_interval to compute the confidence interval of the relative risk for a given confidence level.
Compute the relative risk (also known as the risk ratio).
from scipy.stats.contingency import relative_risk
result = relative_risk(27, 122, 44, 487)
result.relative_risk
result.confidence_interval(confidence_level=0.95)
The following pages refer to to this document either explicitly or contain code examples using this.
scipy.stats._relative_risk:RelativeRiskResult
Hover to see nodes names; edges to Self not shown, Caped at 50 nodes.
Using a canvas is more power efficient and can get hundred of nodes ; but does not allow hyperlinks; , arrows or text (beyond on hover)
SVG is more flexible but power hungry; and does not scale well to 50 + nodes.
All aboves nodes referred to, (or are referred from) current nodes; Edges from Self to other have been omitted (or all nodes would be connected to the central node "self" which is not useful). Nodes are colored by the library they belong to, and scaled with the number of references pointing them