Selection Bias: Types and Sources in Epidemiologic Studies

Graduate Depth 185 in the knowledge graph I know this Set as goal
Unlocks 5 downstream topics
selection-bias internal-validity study-design

Core Idea

Selection bias arises when the process selecting study participants is related to both exposure and outcome, distorting the true association. Common types include Berkson's bias (in case-control hospital studies), loss-to-follow-up bias (cohort studies), and healthy worker effect. Selection bias is a threat to internal validity and cannot be controlled in analysis once it occurs.

Explainer

Selection bias is about the gap between who you wanted to study and who you actually studied — and when that gap is systematically related to both the exposure and the outcome you're investigating. You already know from epidemiologic study designs that different designs have different sampling schemes: cohort studies enroll participants based on exposure status and follow them forward, while case-control studies sample based on outcome status. Each design has characteristic vulnerabilities to selection bias that flow directly from how participants enter the study.

The core mechanism is simple: selection bias occurs when the probability of being included in the study is not equal across all combinations of exposure and disease status. Write it this way: if P(selected | E=1, D=1) / P(selected | E=1, D=0) ≠ P(selected | E=0, D=1) / P(selected | E=0, D=0), then the observed odds ratio will not equal the true population odds ratio. The selection probabilities multiply through to distort the observed association. This is not a statistical problem solvable by larger samples or better analysis — the target population you want to make inferences about is not adequately represented in your sample.

Berkson's bias (or Berkson's fallacy) is the canonical selection bias in hospital-based case-control studies. When you recruit both cases and controls from hospitalized patients, you are sampling from people sick enough to be hospitalized — not from the general population. People with multiple conditions are disproportionately hospitalized. As a result, if your exposure (say, smoking) independently increases hospitalization risk, it will appear more commonly in both cases and controls than in the general population. If the control condition (say, appendicitis) is associated with smoking, you'll see artificially low odds ratios because smoking is inflated in the control group too. The bias flows entirely from the selection of controls from a non-representative hospital population.

Loss-to-follow-up bias strikes cohort studies when participants who drop out differ systematically from those who remain. If sicker participants are more likely to die or stop coming to clinic (informative censoring), and if their illness is related to the exposure, the remaining cohort is a healthier, non-representative subset. The classic example: workers in an occupational cohort who develop serious illness quit the workforce and are lost to follow-up; the remaining workers appear healthier than they truly are. This connects directly to the healthy worker effect — the observation that employed populations consistently show lower mortality than the general population in references tables, not because employment is protective but because severely ill people disproportionately don't work. Comparing a worker cohort to general population mortality tables therefore creates a systematic downward bias in estimated occupational hazard ratios.

The critical clinical and methodological lesson is that selection bias cannot be corrected analytically after the fact. Unlike confounding, which can sometimes be addressed by statistical adjustment if the confounders are measured, selection bias distorts the study sample itself — the data you have simply do not represent the target population in the relevant cells of the exposure-outcome table. The remedy must come at the design stage: community-based (rather than hospital-based) control recruitment, intensive follow-up to minimize loss, and explicit attention to what process selects participants into and out of the study and whether that process is related to exposure and outcome simultaneously.

Practice Questions 5 questions

Prerequisite Chain

Counting to 10Counting to 20Understanding ZeroThe Number ZeroCounting to FiveOne-to-One CorrespondenceCombining Small Groups Within 5Addition Within 10Addition Within 20Two-Digit Addition Without RegroupingTwo-Digit Addition with RegroupingAddition Within 100Repeated Addition as MultiplicationMultiplication Facts Within 100Division as Equal SharingDivision as Grouping (Measurement Division)Division: Grouping (Repeated Subtraction) ModelDivision: Fair Sharing ModelDivision as Equal SharingDivision as GroupingBasic Division FactsDivision Facts Within 100Two-Digit by One-Digit DivisionDivision with RemaindersRemainders and Quotients in DivisionDivision Word ProblemsIntroduction to Long DivisionFactors and MultiplesPrime and Composite NumbersEquivalent FractionsRelating Fractions and DecimalsDecimal Place ValueReading and Writing DecimalsComparing and Ordering DecimalsAdding and Subtracting DecimalsMultiplying DecimalsDividing DecimalsDividing FractionsMixed Number ArithmeticOrder of OperationsInteger Order of OperationsVariable ExpressionsCombining Like TermsOne-Step EquationsTwo-Step EquationsSolving Multi-Step EquationsEquations with Variables on Both SidesAngle Pairs: Complementary, Supplementary, and VerticalParallel Lines and TransversalsCorresponding AnglesAlternate Interior AnglesTriangle Angle Sum TheoremExterior Angle TheoremTriangle Inequality TheoremSimilar Triangles: AA SimilaritySimilar Triangles: SSS and SAS SimilarityProportions in Similar TrianglesRight Triangle Trigonometry IntroductionTrigonometric Ratios ReviewRadian MeasureConverting Between Degrees and RadiansThe Unit CircleGraphing Sine and CosineGraphing Tangent and Reciprocal Trigonometric FunctionsDerivatives of Trigonometric FunctionsAntiderivativesIterated Integrals and Fubini's TheoremDouble Integrals in Cartesian CoordinatesDouble Integrals over Rectangular RegionsDouble Integrals in Polar CoordinatesDouble Integrals: Definition and SetupIterated Integrals and Fubini's TheoremDouble Integrals over Rectangular RegionsDouble Integrals over General RegionsApplications of Double Integrals: Area, Mass, and MomentsTriple Integrals in Cartesian CoordinatesTriple Integrals in Cylindrical and Spherical CoordinatesChange of Variables and the Jacobian DeterminantApplications of Triple Integrals: Volume and MassVector Fields and Their RepresentationsLine Integrals of Vector FieldsGreen's TheoremSurface Integrals and Flux of Vector FieldsSurface Integrals and Flux of Vector FieldsDivergence Theorem: Flux and OutflowDivergence TheoremElectric FluxGauss's LawConductors in Electrostatic EquilibriumCapacitance and CapacitorsDielectricsDielectric Constant and Relative PermittivityElectric Field Inside Dielectric MaterialsDielectric Materials and PolarizationDielectric Susceptibility and PermittivityEnergy Density in Electric FieldsElectric Current and Current DensityElectrical Resistance and ResistivityOhm's Law and Circuit ElementsElectromotive Force (EMF) and BatteriesKirchhoff's Circuit Laws: Voltage and CurrentDC Circuit Network Analysis MethodsTransient Response in RC CircuitsRC CircuitsLC and RLC CircuitsAC Circuits: FundamentalsImpedance and ReactanceAC Power and ResonanceElectromagnetic WavesThe Electromagnetic SpectrumBlackbody Radiation and Planck's LawPhotoelectric EffectThe Photon: Light as QuantaCompton ScatteringWave-Particle Dualityde Broglie WavelengthHeisenberg Uncertainty PrincipleWavefunction and the Born RuleThe Schrödinger EquationState Vectors and WavefunctionsQuantum SuperpositionQuantum EntanglementBell Theorem and Bell InequalitiesPostulates of Quantum MechanicsScattering TheoryIntroduction to Scattering TheoryPartial Wave Analysis in ScatteringSpin Angular MomentumElectron Spin and Intrinsic Magnetic MomentStern-Gerlach Experiment: Spin Quantization and MeasurementElectron Diffraction and Matter Wave PropertiesDavisson-Germer Experiment: Crystal Diffraction of ElectronsElectron Diffraction and Matter Wave InterferenceWavefunctions and Probability Density InterpretationQuantum Superposition and Linear Combinations of StatesQuantum Operators and ObservablesCanonical Commutation Relations and UncertaintyHeisenberg Uncertainty Principle and Measurement LimitsTime-Independent Schrödinger Equation and EigenvaluesHydrogen Atom in Quantum MechanicsSpectral Lines and Energy TransitionsSelection Rules for Atomic TransitionsLS and jj Coupling Schemes in Multi-Electron AtomsPauli Exclusion Principle and Antisymmetric WavefunctionsElectron Configuration and the Aufbau PrincipleThe Periodic Table and Atomic Electronic StructureThe Periodic TableElectron ConfigurationPeriodic TrendsIonization EnergyIonic BondingLewis StructuresResonance Structures and Delocalized ElectronsResonance and Formal ChargeMolecular Polarity and Dipole MomentsIntermolecular ForcesStates of Matter and Phase Changes: Melting, Boiling, and SublimationGas Laws and the Ideal Gas EquationGas Stoichiometry and Volume-Volume CalculationsThermochemistry and EnthalpyHeat Capacity and CalorimetryEntropy and Molecular DisorderSpontaneity and ΔGEntropy and Gibbs Free EnergyChemical EquilibriumAcid-Base ChemistryOrganic Reaction Mechanisms and Arrow PushingElectrophilic Addition to AlkenesAromaticity and BenzeneDNA StructureCentral Dogma of Molecular BiologyThe Genetic CodeDNA MutationsDNA Repair MechanismsCell Cycle Checkpoints and Cancer PreventionMitotic Spindle Checkpoint and Chromosome SegregationKinetochore Structure and FunctionMitochondria: Structure and FunctionCellular Respiration OverviewBacterial Metabolism OverviewAntibiotic Resistance MechanismsInfectious Disease EpidemiologyFoundations of EpidemiologyMeasuring Disease Frequency: Incidence and PrevalenceEpidemiologic Study DesignsSelection Bias: Types and Sources in Epidemiologic Studies

Longest path: 186 steps · 927 total prerequisite topics

Prerequisites (1)

Leads To (2)