tests/test_risk_models/test_crc_pro_model.py

# pylint: disable=missing-docstring
"""Tests for the CRC-PRO colorectal cancer risk model.

Web calculator available at: https://riskcalc.org/ColorectalCancer/
"""

import pytest

from sentinel.risk_models import CRCProRiskModel
from sentinel.user_input import (
    AlcoholConsumption,
    Anthropometrics,
    AspirinUse,
    CancerType,
    Demographics,
    Ethnicity,
    FamilyMemberCancer,
    FamilyRelation,
    FamilySide,
    FemaleSpecific,
    HormoneUse,
    HormoneUseHistory,
    Lifestyle,
    NSAIDUse,
    PersonalMedicalHistory,
    RelationshipDegree,
    Sex,
    SmokingHistory,
    SmokingStatus,
    UserInput,
)

GROUND_TRUTH_CASES = [
    {
        "name": "low_risk",
        "input": UserInput(
            demographics=Demographics(
                age_years=55,
                sex=Sex.FEMALE,
                ethnicity=Ethnicity.ASIAN,
                anthropometrics=Anthropometrics(height_cm=152.4, weight_kg=45.4),
                education_level=3,
            ),
            lifestyle=Lifestyle(
                smoking=SmokingHistory(status=SmokingStatus.NEVER, pack_years=0.0),
                alcohol_consumption=AlcoholConsumption.MODERATE,
                multivitamin_use=True,
            ),
            personal_medical_history=PersonalMedicalHistory(
                chronic_conditions=[],
                nsaid_use=NSAIDUse.NEVER,
            ),
            family_history=[],
            female_specific=FemaleSpecific(
                hormone_use=HormoneUseHistory(estrogen_use=HormoneUse.NEVER),
            ),
        ),
        "expected": 1.0,
    },
    {
        "name": "medium_risk",
        "input": UserInput(
            demographics=Demographics(
                age_years=65,
                sex=Sex.MALE,
                ethnicity=Ethnicity.PACIFIC_ISLANDER,
                anthropometrics=Anthropometrics(height_cm=177.8, weight_kg=81.6),
                education_level=1,
            ),
            lifestyle=Lifestyle(
                smoking=SmokingHistory(status=SmokingStatus.NEVER, pack_years=0.0),
                alcohol_consumption=AlcoholConsumption.HEAVY,
                multivitamin_use=True,
                moderate_physical_activity_hours_per_day=0.0,
                red_meat_consumption_oz_per_day=1.0,
            ),
            personal_medical_history=PersonalMedicalHistory(
                chronic_conditions=[],
                aspirin_use=AspirinUse.NEVER,
            ),
            family_history=[
                FamilyMemberCancer(
                    relation=FamilyRelation.FATHER,
                    cancer_type=CancerType.COLORECTAL,
                    age_at_diagnosis=60,
                    degree=RelationshipDegree.FIRST,
                    side=FamilySide.PATERNAL,
                )
            ],
        ),
        "expected": 3.8,
    },
    {
        "name": "high_risk",
        "input": UserInput(
            demographics=Demographics(
                age_years=75,
                sex=Sex.MALE,
                ethnicity=Ethnicity.WHITE,
                anthropometrics=Anthropometrics(height_cm=177.8, weight_kg=158.8),
                education_level=5,
            ),
            lifestyle=Lifestyle(
                smoking=SmokingHistory(status=SmokingStatus.FORMER, pack_years=25.0),
                alcohol_consumption=AlcoholConsumption.HEAVY,
                multivitamin_use=True,
                moderate_physical_activity_hours_per_day=0.0,
                red_meat_consumption_oz_per_day=1.0,
            ),
            personal_medical_history=PersonalMedicalHistory(
                chronic_conditions=[],
                aspirin_use=AspirinUse.NEVER,
            ),
            family_history=[
                FamilyMemberCancer(
                    relation=FamilyRelation.MOTHER,
                    cancer_type=CancerType.COLORECTAL,
                    age_at_diagnosis=70,
                    degree=RelationshipDegree.FIRST,
                    side=FamilySide.MATERNAL,
                )
            ],
        ),
        "expected": 8.9,
    },
]


def _base_user(sex: Sex, **overrides):
    demographics = Demographics(
        age_years=50,
        sex=sex,
        ethnicity=Ethnicity.WHITE,
        anthropometrics=Anthropometrics(height_cm=170.0, weight_kg=75.0),
        education_level=4,
    )
    lifestyle = Lifestyle(
        smoking=SmokingHistory(status=SmokingStatus.FORMER, pack_years=10.0),
        alcohol_consumption=AlcoholConsumption.MODERATE,
        multivitamin_use=True,
    )
    personal_history = PersonalMedicalHistory(chronic_conditions=[])

    if sex == Sex.FEMALE:
        lifestyle.moderate_physical_activity_hours_per_day = None
        lifestyle.red_meat_consumption_oz_per_day = None
        personal_history.aspirin_use = None
        personal_history.nsaid_use = NSAIDUse.NEVER
        female_specific = FemaleSpecific(
            hormone_use=HormoneUseHistory(estrogen_use=HormoneUse.NEVER)
        )
    else:
        lifestyle.moderate_physical_activity_hours_per_day = 1.0
        lifestyle.red_meat_consumption_oz_per_day = 1.5
        personal_history.aspirin_use = AspirinUse.NEVER
        personal_history.nsaid_use = None
        female_specific = None

    family_history = overrides.get("family_history", [])

    return UserInput(
        demographics=demographics,
        lifestyle=lifestyle,
        personal_medical_history=personal_history,
        family_history=family_history,
        female_specific=female_specific,
    )


class TestCRCProRiskModel:
    def setup_method(self):
        self.model = CRCProRiskModel()

    @pytest.mark.parametrize("case", GROUND_TRUTH_CASES, ids=lambda case: case["name"])
    def test_ground_truth_validation(self, case):
        user = case["input"]
        score = self.model.compute_score(user)
        # scores return a string; ensure numeric and close to expected
        calculated = float(score)
        assert calculated == pytest.approx(case["expected"], abs=0.5)

    def test_metadata(self):
        assert self.model.name == "crc_pro"
        assert self.model.cancer_type() == "colorectal"
        assert "CRC-PRO" in self.model.description()
        assert "%" in self.model.interpretation()
        refs = self.model.references()
        assert isinstance(refs, list) and refs

    def test_validation_errors(self):
        """Test that model raises ValueError for invalid inputs."""
        # Test missing required field
        user_input = UserInput(
            demographics=Demographics(
                age_years=50,
                sex=Sex.FEMALE,
                ethnicity=Ethnicity.WHITE,
                anthropometrics=Anthropometrics(height_cm=165.0, weight_kg=65.0),
            ),
            lifestyle=Lifestyle(
                smoking=SmokingHistory(
                    status=SmokingStatus.NEVER, pack_years=None
                ),  # Missing pack_years
                multivitamin_use=True,
            ),
            personal_medical_history=PersonalMedicalHistory(
                nsaid_use=NSAIDUse.NEVER,
            ),
            family_history=[],
            female_specific=FemaleSpecific(
                hormone_use=HormoneUseHistory(estrogen_use=HormoneUse.NEVER),
            ),
        )

        with pytest.raises(ValueError, match=r"Invalid inputs for CRC-PRO:"):
            self.model.compute_score(user_input)

    def test_inapplicable_sex(self):
        """Test unsupported sex returns N/A."""
        user_input = UserInput(
            demographics=Demographics(
                age_years=50,
                sex=Sex.UNKNOWN,  # Unsupported sex
                ethnicity=Ethnicity.WHITE,
                anthropometrics=Anthropometrics(height_cm=175.0, weight_kg=70.0),
            ),
            lifestyle=Lifestyle(
                smoking=SmokingHistory(status=SmokingStatus.NEVER, pack_years=0.0),
                multivitamin_use=True,
            ),
            personal_medical_history=PersonalMedicalHistory(),
            family_history=[],
        )

        score = self.model.compute_score(user_input)
        assert "N/A" in score

    def test_age_out_of_range(self):
        """Test age outside validated range raises ValueError."""
        user = _base_user(Sex.MALE)
        user.demographics.age_years = 44  # Below minimum
        with pytest.raises(ValueError, match=r"Invalid inputs for CRC-PRO:"):
            self.model.compute_score(user)

    def test_missing_ethnicity(self):
        """Test missing ethnicity returns N/A."""
        user = _base_user(Sex.MALE)
        user.demographics.ethnicity = None
        result = self.model.compute_score(user)
        assert "Ethnicity" in result

    @pytest.mark.parametrize("sex", [Sex.MALE, Sex.FEMALE])
    def test_valid_score(self, sex):
        """Test that valid inputs produce numeric scores.

        Args:
            sex: Sex enum value to test.
        """
        user = _base_user(sex)
        score = self.model.compute_score(user)
        assert score not in ("N/A: Missing required data:")
        assert float(score) >= 0

    def test_family_history_detection(self):
        """Test that family history increases risk score."""
        user = _base_user(
            Sex.FEMALE,
            family_history=[
                FamilyMemberCancer(
                    relation=FamilyRelation.MOTHER,
                    cancer_type=CancerType.COLORECTAL,
                    age_at_diagnosis=60,
                    degree=RelationshipDegree.FIRST,
                    side=FamilySide.MATERNAL,
                ),
            ],
        )
        score = float(self.model.compute_score(user))

        user_without_family_history = _base_user(Sex.FEMALE)
        score_no_family = float(self.model.compute_score(user_without_family_history))

        assert score > score_no_family