Epidemiology Data Presentation for Health Officials

Presenting epidemiological data to health officials is a specialized communication challenge. Your audience ranges from epidemiologists who understand confidence intervals to elected officials who need a clear headline and a recommendation. The same data has to serve both, and the presentation strategy has to bridge that gap without sacrificing scientific integrity.

The Core Challenge

Epidemiology deals in uncertainty. Rate estimates have confidence intervals. Surveillance data has reporting lags and ascertainment biases. Causal claims require careful language. Presenting this appropriately to a non-technical audience risks appearing wishy-washy. Oversimplifying it risks misleading decision-makers.

The resolution is to layer the presentation: clear headlines and actionable takeaways for the full audience, with technical detail in appendices for those who need it.

Choosing the Right Data Visualizations

Data visualization choices matter enormously for epidemiological data.

Epidemic Curves (Epi Curves)

For infectious disease outbreaks, case counts over time are typically displayed as epi curves — bar charts with time on the x-axis and case counts on the y-axis, often by date of symptom onset (not date of report, which introduces reporting lag bias).

Key elements for a clear epi curve:

• X-axis unit appropriate to the epidemic pace (days for fast-moving outbreaks, weeks or months for slower ones)
• Bars colored by case classification (confirmed, probable, suspected)
• Vertical lines marking key interventions or events
• Clear annotation of any reporting gaps or data quality issues

Maps and Geographic Distribution

For spatially distributed disease burden, choropleth maps (area shading by rate) and dot maps (one dot per case) communicate very different information. Choropleth maps are appropriate for population-normalized rates; dot maps for visualizing clustering in outbreak investigations.

Important caveats to communicate:

• Rates vs. counts (a large, sparsely populated county with a high rate may have fewer cases than a dense urban county with a low rate)
• Denominator uncertainty (population estimates in intercensal years carry error)
• Small-number instability (a county with 3 cases and 1,000 population has a "rate" that is statistically meaningless)

Time Trends

Line graphs showing incidence or mortality rates over time. Include:

• Rate on y-axis (per 100,000 or appropriate denominator)
• Confidence intervals as shaded bands
• Vertical lines marking program changes, policy changes, or major events
• Comparison line for state or national trend for context

Demographic Breakdowns

Bar charts or heatmaps showing rates by age group, sex, race/ethnicity, geography, or other relevant dimensions. Disparity ratios (the rate in the highest-burden group divided by the rate in the lowest-burden group) quantify health equity differences in a single number.

Slide Structure for a Surveillance Report Presentation

Slide 1 — Headline Summary

Three to five bullet points stating the most important findings in plain language. The secretary of health or the governor may only see this slide. It needs to be complete and accurate.

Example format:

• "[Disease] incidence increased 18% from 2024 to 2025, driven primarily by the 18–34 age group"
• "Disparity between highest- and lowest-income zip codes widened from 3.2x to 4.1x"
• "Hospitalization rate remains below the threshold associated with healthcare system strain"
• "Vaccination coverage in the highest-incidence counties averages 42% — well below the 70% threshold for herd protection"

Slide 2 — Surveillance System Overview

Brief description of data sources, case definitions, and known limitations. Even for a primarily policy audience, this context is essential for interpreting the numbers appropriately. One slide, not a methods section.

Slides 3–5 — Epidemiological Findings

Core surveillance data: trends over time, geographic distribution, demographic breakdown. Each visualization should have:

• A title that states the finding, not just the topic ("Incidence highest in rural counties, lowest in metro areas" not "Incidence by county type")
• Data source and date range cited
• A callout box or annotation highlighting the most important single thing to notice on the chart

Slide 6 — Population at Highest Risk

A profile of the communities or individuals bearing the greatest burden. Relevant for resource allocation and targeted intervention decisions.

Slide 7 — Modeled Projections (If Available)

If epidemic modeling is being presented, be explicit about model assumptions, scenario structure, and uncertainty ranges. Show multiple scenarios (low/medium/high) rather than point estimates. Officials who have been burned by overconfident model projections that didn't pan out are appropriately skeptical — earn their trust by showing the full uncertainty range.

Include a brief explanation of what the model does and does not account for. "This model projects spread under current conditions and does not account for potential policy interventions" is an essential caveat.

Slide 8 — Comparison to Benchmarks and Prior Periods

How does the current situation compare to:

• The same period last year
• The peak of the last comparable event or season
• State or national averages
• Targets or thresholds set by public health goals (Healthy People 2030, jurisdictional targets)

Benchmarks transform isolated numbers into interpretable context.

Slide 9 — Implications for Action

What do these data suggest about resource needs, policy changes, or program adjustments? This is where the epidemiologist's role ends and the policy recommendation begins — but your data should clearly point toward specific implications. Be willing to state what the data suggest, while being honest about uncertainty.

Slide 10 — Monitoring and Next Steps

What data will be collected next, when will the next update occur, and what thresholds or signals will trigger a reassessment or escalation? Decision-makers need to know the feedback loop.

Language Guidelines

Use precise language for uncertainty: "Rates increased significantly" is ambiguous. "Rates increased 23% [95% CI: 18–28%], exceeding our threshold for concerning trend" is precise.

Distinguish association from causation: "Areas with lower vaccination coverage show higher case rates" is an association. "Low vaccination coverage caused the outbreak" is a causal claim requiring more evidence.

Avoid jargon without definition: Incidence, prevalence, attack rate, Rt, and CFR are second nature to epidemiologists and opaque to everyone else in the room. Define each term on first use.

Build the deck with a clean structure in slide-deck.io, and maintain a full technical appendix separately for staff-level audiences who need the complete analysis.