NURS 8310 – Epidemiology and Population Health
Walden University DNP Program
Week 2 Study Notes: Data Sources, Collection Methods & Descriptive Epidemiology Primary Textbook Reference
Gordis, L. Epidemiology (6th ed., 2014 or latest). Elsevier. Focus: Chapters 2 (Epidemiologic Measures), 3 (Descriptive Epidemiology), and selected sections from Chapter 4 (Data Sources and Collection).
Core Learning Objectives for Week 2 Identify and evaluate major sources of epidemiologic data (vital statistics, surveys, registries, surveillance systems).
Compare passive vs. active surveillance, strengths/limitations of each.
Describe methods of data collection and common sources of error/bias in epidemiologic data.
Apply descriptive epidemiology principles using the person–place–time framework.
Perform and interpret basic descriptive analyses (crude/specific/age-adjusted rates, standardization methods).
Recognize how descriptive data generate hypotheses for analytic studies.
1. Major Sources of Epidemiologic DataSource Category
Examples
Strengths
Limitations
Use Case Examples
Vital Statistics
Birth/death certificates, fetal death reports
Complete (legal requirement), population-based, long-term trends
Underreporting (e.g., cause-of-death misclassification), limited detail
Mortality trends, life expectancy, infant mortality rates
National Surveys
NHANES, BRFSS, NHIS, MEPS
Nationally representative, rich behavioral/SDOH data, repeated cross-sections
Self-report bias, recall bias, non-response, cross-sectional (no temporality)
Prevalence of obesity, hypertension, smoking, health behaviors
Disease Registries
SEER (cancer), National ALS Registry, trauma registries
High-quality, detailed clinical/pathologic data, long-term follow-up
Incomplete coverage, reporting delay, expense
Cancer incidence/survival, rare disease tracking
Surveillance Systems
NNDSS (notifiable diseases), FluView, NVSS, YRBSS
Real-time or near real-time, outbreak detection, trend monitoring
Underreporting (passive), case definition variability
Notifiable diseases (e.g., measles, syphilis), influenza seasonality
Administrative Data
Hospital discharge (HCUP), claims data (Medicare/Medicaid), EHRs
Large volume, readily available, cost-effective
Coding errors, limited clinical detail, access restrictions
Readmission rates, procedure utilization, healthcare disparities
Specialized/Global
WHO Global Health Observatory, GBD Study, DHS (Demographic & Health Surveys)
International comparison, burden of disease estimates (DALYs)
Data quality varies by country, modeling assumptions
Global burden comparisons, low-resource setting epidemiology
Key Surveillance Types Passive surveillance: Providers report cases voluntarily → low cost, broad coverage, but underreporting common (e.g., most NNDSS diseases).
Active surveillance: Staff seek cases (chart review, calls) → higher sensitivity/completeness, resource-intensive (e.g., Active Bacterial Core surveillance [ABCs]).
Syndromic surveillance: Monitors pre-diagnostic indicators (chief complaints, pharmacy sales) → early outbreak detection (e.g., BioSense, ESSENCE).
2. Data Collection Methods & Sources of Error/BiasCommon Methods Direct observation (e.g., chart abstraction).
Self-report (surveys, interviews).
Laboratory/clinical measurement.
Administrative extraction (claims, EHR).
Major Sources of Error Measurement error: Misclassification (differential vs. non-differential).
Selection bias: Non-representative sample (e.g., volunteer bias in surveys).
Information bias: Recall bias (worse for cases), interviewer bias, social desirability bias.
Reporting bias: Underreporting in passive systems (stigma, mild cases).
Ecologic fallacy: Inferring individual-level conclusions from group data.
3. Descriptive Epidemiology – Core PrinciplesDescriptive epidemiology answers Who? Where? When? → generates hypotheses for analytic studies.Person–Place–Time Framework (expanded from Week 1)Person Age (most powerful predictor; age-specific rates essential).
Sex/gender (biological + social factors).
Race/ethnicity (health disparities, proxy for SDOH).
Socioeconomic status (education, income, occupation).
Behaviors (smoking, diet, physical activity).
Genetic factors (when available).
Example: Higher breast cancer incidence in women; higher prostate cancer mortality in Black men.
Place Geographic variation: local (neighborhood clusters), regional, national, global.
Urban vs. rural, climate, environmental exposures (e.g., air pollution, water quality).
Migration effects (healthy migrant effect).
Example: Higher melanoma in Australia (UV exposure); higher lead poisoning in older urban housing.
Time Secular trends (long-term): e.g., declining CVD mortality since 1960s (smoking reduction, better treatment).
Seasonal/cyclic: e.g., influenza winter peaks, rotavirus winter/spring.
Epidemic curves: point-source (common exposure, single peak), propagated (person-to-person spread, multiple waves), intermittent.
Period vs. cohort effects: e.g., birth cohort smoking patterns affect lung cancer rates decades later.
Epidemic Curve Interpretation Point-source: rapid rise/fall (e.g., foodborne outbreak).
Propagated: slower rise, successive waves (e.g., measles in unvaccinated community).
Use to generate hypotheses about mode of transmission.
4. Rate Calculation & StandardizationCrude Rates Crude mortality = (total deaths / mid-year population) × 1,000
Problem: Confounded by age structure (older populations have higher crude rates).
Specific Rates Age-specific mortality = (deaths in age group / population in age group) × 1,000
Age-Adjusted (Standardized) Rates Direct method: Apply age-specific rates to a standard population (e.g., 2000 U.S. standard million).
Formula: Σ (age-specific rate × standard population in age group) / total standard population
Indirect method: Uses standard rates applied to study population (SMR – standardized mortality ratio).
Purpose: Remove age confounding when comparing populations.
Example Calculation (Direct Method – Hypothetical)
Population A has higher crude mortality than Population B → but after age-adjustment, B is higher → Population B is older.5. High-Yield Concepts & DNP ApplicationsWhy descriptive before analytic? Descriptive data identify patterns → generate etiologic hypotheses → analytic studies test them.
DNP relevance: Use surveillance data to identify local health priorities.
Present age-adjusted rates to stakeholders to avoid misinterpretation.
Recognize limitations of self-report data when designing QI projects.
Apply person-place-time to target interventions (e.g., place-based obesity programs in food deserts).
Recommended Week 2 Activities Read Gordis Chapters 2–3 + selected sections of Chapter 4.
Practice calculating crude, specific, and adjusted rates using examples from Gordis or CDC data.
Explore CDC Wonder (https://wonder.cdc.gov) or County Health Rankings for real descriptive data.
Review an epidemic curve (e.g., COVID-19 U.S. waves) and label point-source vs. propagated features.
Prepare for discussion by choosing a health problem and describing its person-place-time pattern using public data.
These notes provide the foundation for understanding how raw data are transformed into meaningful descriptive patterns—the essential first step in any epidemiologic investigation. Mastery here is critical for Weeks 3–4 (measures of association and study designs). Good luck in Week 2 of NURS 8310!
