Epidemiology & Surveillance

Group Check-In & Collaboration (Page 2)

• Purpose: Reflect on content covered so far (including prework) and plan how to build confidence/proficiency with this week’s content.

• Steps:

  • 1 minute: Write down a question and what you can offer the group

  • 2 minutes: Share in pairs/triplets

  • 3 minutes: Share with the entire group

• Focus: Identify questions, gaps, and ways to support one another's learning.

Mini lecture: Describe the role of epidemiology (Page 3)

• Public health surveillance assesses and monitors population health.

• Without surveillance, health problems in the population cannot be identified.

• Surveillance is defined as an ongoing and systematic effort to collect, analyze and interpret health data.

• Surveillance can be used to:

  • Estimate magnitude of problems

  • Determine distribution of disease

  • Identify trends

  • Detect epidemics

  • Generate hypotheses

  • Evaluate control measures

  • Monitor changes in infectious agents

  • Detect changes in health practices

Surveillance: Core concepts (Page 3)

• Surveillance is ongoing, systematic data collection, analysis, and interpretation aimed at understanding population health.

• Functions include monitoring, trend detection, hypothesis generation, and evaluation of interventions.

Surveillance Types: Active vs Passive (Page 4)

• Active surveillance:

  • Identifying cases, often led by a public health agency

  • Resource-intensive and expensive

• Passive surveillance:

  • Provider-initiated (reportable)

  • Cheaper and less resource-intensive but often less complete/accurate (cases with no access may be unreported)

Sources of surveillance (Page 5)

• Data sources include: environmental monitoring systems, animals/vectors, labs, medical records, surveys, wastewater

• Group activity prompt (examples to find in groups):

  • Environmental surveillance example

  • Chronic disease surveillance example

  • Infectious disease surveillance example

  • Injury or substance use surveillance example

• Considerations for each source:

  • Who can access the data

  • Whether the system is passive or active

  • Geographical level of data (local, regional, national)

Measures of disease frequency (Page 6)

• Prevalence:

  • Definition: The number of existing cases of an outcome in a population at a given time

  • Formula (conceptual): $\text{Prevalence} = \frac{ \text{number of existing cases}}{ \text{total population}}$

• Cumulative incidence:

  • Definition: The number of new cases during a specified time period among the population at risk at the start

  • Important detail: Population at risk cannot already have the disease

  • Formula: $\text{Cumulative incidence} = \frac{ \text{new cases during period}}{ \text{population at risk at start}}$

• Incidence rate (incidence density):

  • Definition: The number of new cases per unit of person-time

  • Formula: $\text{Incidence rate} = \frac{ \text{new cases}}{ \text{person-time at risk}}$

• Case fatality (CFR):

  • Definition: The number of deaths from an outcome among the total number of cases

  • Formula: $\text{CFR} = \frac{ \text{deaths due to disease}}{ \text{number of cases of disease}}$

• Note: Full descriptions available in pre-work; review before class

Practice Calculations (Page 7)

• Purpose: Compute sample epidemiology measures as a group

• Activity: Worked Examples #1 and #2 (10 minutes)

Screening: Concept and purpose (Page 8)

• Screening is the process of using a test to determine whether someone has a health indicator or is likely to develop it.

• Not the same as a diagnosis; screening provides information about likelihood, not a definitive diagnosis.

When to screen & evaluation criteria (Page 9)

• Factors influencing screening decisions:

  • Prevalence or incidence of the condition

  • Time between biological onset and symptoms (window for detection)

  • Availability of treatment

  • Costs and invasiveness of screening

• How screening tests are evaluated:

  • Sensitivity: ability to identify those with the outcome; true positive rate

  • Specificity: ability to identify those without the outcome; true negative rate

  • Positive Predictive Value (PPV): probability that a person with a positive test actually has the outcome

  • Negative Predictive Value (NPV): probability that a person with a negative test does not have the outcome

Diagnostic test accuracy and the 2x2 framework (Page 10)

• Contingency table concepts:

  • True Positive (TP): screening positive and diagnosis positive

  • False Positive (FP): screening positive but diagnosis negative

  • True Negative (TN): screening negative and diagnosis negative

  • False Negative (FN): screening negative but diagnosis positive

• Verifying formulas:

  • PPV = $\frac{TP}{TP+FP}$

  • NPV = $\frac{TN}{TN+FN}$

  • Sensitivity = $\frac{TP}{TP+FN}$

  • Specificity = $\frac{TN}{TN+FP}$

• Note: The diagram typically shows a screening parameter leading to true/false positives/negatives and corresponding predictive values

Screening decision considerations (Page 11)

• Most screening tests yield a dichotomous (yes/no) result; cut-offs must balance false positives (FP) and false negatives (FN)

• Trade-off questions:

  • Do we prefer lower FP or FN?

  • Consequences of FP: unnecessary/invasive tests, resource use, potential harm from misdiagnosis

  • Consequences of FN: missed early treatment opportunity, potential ongoing transmission for infectious diseases, conditions easy to treat if identified early

• Other reasons not to screen:

  • Hidden costs or risks

  • Harm from false positives

  • Unwarranted reassurance from false negatives (FN)

  • Whether effective follow-up or treatment is available; risk communication needs

Practice Calculations (Page 12)

• Activity: Worked Examples #3 and #4 (10 minutes)

Quick Checks: Sample problems & answers (Pages 13-15)

• Quick Check 1 (Page 13): Test characteristics for cancer screening

  • Data: 1,000 people tested; 200 had cancer; test correctly identified 180 (TP = 180)

  • 800 without cancer; test correctly identified 720 as not having cancer (TN = 720)

  • Calculations:

  • Sensitivity = $\frac{TP}{TP+FN} = \frac{180}{200} = 0.90 = 90\%$

  • Specificity = $\frac{TN}{TN+FP} = \frac{720}{800} = 0.90 = 90\%$

  • Answer: A

• Quick Check 2 (Page 14): Case fatality rate during an outbreak

  • Data: 1,000 diagnosed; 75 died

  • CFR = $\frac{\text{deaths}}{\text{cases}} = \frac{75}{1000} = 0.075 = 7.5\%$

  • Answer: B

• Quick Check 3 (Page 15): Cohort follow-up with varying time

  • Scenario: 10,000 adults followed 5 years with dropout; need to account for varying follow-up time

  • Best measure when follow-up time varies: Incidence Rate

  • Answer: B

BRFSS Data Description Assignment (Page 16)

• Assignment (15 minutes, individual): HDR Homework - Week 3

• Questions:

  • What is BRFSS and why is it important?

  • Who gets surveyed in BRFSS and how is data collected?

  • How is your assigned health condition measured in BRFSS? (e.g., self-report question like “Have you ever been told by a doctor that you have…?”)

• Extra resources: https://www.cdc.gov/brfss/annualdata/annual2023.html

Closing & Week-long Deliverables (Page 17)

• End of week assignments:

  • Assignment: Find & Analyze an Epi Study (due 9/12)

  • Assignment: Description of Data Source (due 9/12)

Connections to Foundational Principles & Real-World Relevance

• Surveillance links to public health action: identification of health problems enables timely interventions and policy decisions.

• Distinguishing active vs passive surveillance helps allocate resources efficiently and understand data completeness.

• Measures of disease frequency provide different lenses on risk and burden: prevalence reflects burden at a point, while incidence concepts capture risk and speed of new cases.

• Screening concepts tie to preventive strategies: balancing early detection with harms and costs of false results, and recognizing that screening is a step in a broader diagnostic pathway.

• The 2x2 framework (TP, FP, TN, FN) underpins interpretation of test results and informs clinical/public health decisions.

• Real-world application examples: BRFSS data informs state and national health policy; epidemiology methods underpin outbreak response, chronic disease prevention, and health services planning.

Key Formulas (summary)

• Prevalence: See "Measures of disease frequency (Page 6)"

• Cumulative incidence: See "Measures of disease frequency (Page 6)"

• Incidence rate: See "Measures of disease frequency (Page 6)"

• Case fatality rate: See "Measures of disease frequency (Page 6)"

• Sensitivity: See "Diagnostic test accuracy and the 2x2 framework (Page 10)"

• Specificity: See "Diagnostic test accuracy and the 2x2 framework (Page 10)"

• PPV: See "Diagnostic test accuracy and the 2x2 framework (Page 10)"

• NPV: See "Diagnostic test accuracy and the 2x2 framework (Page 10)"