Phenofhy

Calculating Summary Stats ​

This tutorial shows examples for calculate.py, which provides derived summaries and prevalence metrics for phenotypes. Use it to compute prevalence, generate summary tables for numeric and categorical traits, and quantify medication usage patterns.

Prevalence with calculate.prevalence ​

Compute counts and prevalence for categorical traits.

python
from phenofhy.calculate import prevalence

prev = prevalence(
    df,
    traits=[
        "derived.sex",
        "questionnaire.smoke_status_2_1",
        "questionnaire.alcohol_curr_1_1",
    ],
    denominator="nonmissing",
)

Add multiple denominators in a wide output:

python
prev = prevalence(
    df,
    traits=["derived.sex", "questionnaire.smoke_status_2_1"],
    denominators=["all", "nonmissing"],
    wide_output=True,
)

Summary tables with calculate.summary ​

Use summary() to compute numeric and categorical summaries, optionally stratified.

python
from phenofhy.calculate import summary

result = summary(
    df,
    stratify="derived.sex",
    categorical_traits=["questionnaire.smoke_status_2_1"],
    round_decimals=2,
)

numeric_df = result["numeric"]
categorical_df = result["categorical"]

Medication prevalence ​

For medication phenotypes, use calculate.medication_prevalence.

python
from phenofhy.calculate import medication_prevalence

per_med, grouped = medication_prevalence(
    df,
    codings=codings_df,
    medication_phenotypes=medication_phenotypes,
    denominator="all",
)