ABPM Public Health cheat sheet

Incidence proportion (risk) = new cases during period / population at risk at start of period.

Incidence rate (density) = new cases / total person-time at risk; person-time = persons × time observed.

Prevalence ≈ incidence × average duration of disease (for steady-state, low-prevalence conditions).

Point prevalence = cases at a single time / population at that time; period prevalence uses any time within a window.

Relative risk = incidence in exposed / incidence in unexposed; used in cohort studies.

Odds ratio = (a/c)/(b/d) = ad/bc from a 2×2 table; approximates RR when disease is rare.

Attributable risk (risk difference) = incidence exposed − incidence unexposed; population attributable fraction = Pe(RR−1)/(1+Pe(RR−1)).

Sensitivity = TP/(TP+FN); specificity = TN/(TN+FP); PPV = TP/(TP+FP); NPV = TN/(TN+FN).

Bias types: selection (differential entry), information (recall/measurement), confounding (third variable associated with both exposure and outcome).

Standard error of the mean = s/√n; 95% CI for a mean ≈ mean ± 1.96·SE (large samples).

Type I error (α) = rejecting a true null; type II error (β) = failing to reject a false null; power = 1−β.

Sample size for a proportion: n ≈ Z²·p(1−p)/d²; larger n needed for smaller margins of error.

Chi-square tests independence of two categorical variables; expected cell counts <5 require Fisher's exact test.

t-test compares means of two groups or one mean to a constant; ANOVA compares means across 3+ groups.

Pearson r measures linear association (−1 to +1); r² is the proportion of variance explained.

Linear regression: Y = β0 + β1X + ε; β1 is change in mean Y per one-unit change in X.

Positive predictive value rises as prevalence rises; sensitivity and specificity are prevalence-independent.

Confounding in regression is addressed by including the covariate; interaction is tested with a product term.

Donabedian model of quality: structure → process → outcomes.

Institute for Healthcare Improvement (IHI) Triple Aim: improve population health, enhance patient experience, reduce per-capita cost.

Medicare Part A covers inpatient hospital; Part B covers outpatient and physician services; Part D covers prescription drugs.

Medicaid is jointly funded by federal and state governments and administered by states.

Managed care models: HMO (gatekeeping, closed network), PPO (open network, no gatekeeping), POS (hybrid).

PDSA cycle: Plan → Do → Study → Act for continuous quality improvement.

Cost-effectiveness ratio = (cost intervention − cost comparator)/(effect intervention − effect comparator), expressed as cost per unit of health gain.

Health services pyramid: primary care → secondary care → tertiary care; primary care is first-contact, continuous, coordinated care.

Exposure pathway: source → environmental medium → route of exposure (inhalation, ingestion, dermal) → target organ.

Dose-response assessment identifies the relationship between dose and magnitude of adverse effect.

Risk characterization integrates hazard identification, dose-response, and exposure assessment.

LD50 is the dose lethal to 50% of test animals; lower LD50 = greater acute toxicity.

Reference dose (RfD) = no-observed-adverse-effect level (or benchmark) divided by uncertainty factors.

Primary air pollutants under the Clean Air Act: particulate matter, ozone, CO, NO2, SO2, lead.

Waterborne pathogens of concern include Giardia, Cryptosporidium, and Legionella; chlorine residual indicates disinfection protection.

Biomagnification: persistent lipophilic compounds (e.g., methylmercury, PCBs) increase in concentration up the food chain.

Indoor air hazards: radon (second leading cause of lung cancer), environmental tobacco smoke, CO from combustion appliances.

Social determinants of health: economic stability, education, social/community context, health care access, neighborhood/built environment.

Health belief model constructs: perceived susceptibility, severity, benefits, barriers, cues to action, self-efficacy.

Transtheoretical (stages of change) model: precontemplation, contemplation, preparation, action, maintenance.

Social cognitive theory: behavior is shaped by reciprocal determinism among person, behavior, and environment; self-efficacy is central.

Ecological model levels: individual, interpersonal, community, organizational, policy/enabling environment.

Diffusion of innovations: innovators, early adopters, early majority, late majority, laggards; adoption follows an S-curve.

Precede-Proceed model phases: social assessment, epidemiological assessment, educational/ecological diagnosis, administrative/policy diagnosis, implementation, evaluation.

Relative risk is often expressed as the inverse of the odds ratio for a protective factor; effect sizes inform intervention targeting.

Cultural competence requires awareness of one's own biases, knowledge of other cultures, and skills for cross-cultural communication.