Conclusion
Summary of Results
- Vulnerable Populations
Adults aged 65 and over accounted for more than 60% of all influenza-related deaths, confirming them as the most vulnerable group. Deaths in this group showed the highest variation and greatest concentration across states. A strong correlation was found between elderly population size and mortality, reinforcing population as a key risk driver. Although imputed values allowed limited analysis of the 0–4 age group, low death numbers suggest they are not disproportionately affected or critical to resource planning.
- Seasonality
The analysis confirmed a consistent seasonal pattern in influenza-related deaths across all U.S. regions. Deaths typically rise in November, peak in January, and decline by April. While all regions follow this timeline, the South consistently experiences a higher mortality rate. However, the Midwest, Northeast, and West show near-identical trends. Yearly totals revealed considerable variation in flu severity — suggesting that while the timing is predictable, the intensity fluctuates.
- Risk & Forecasting
Mapping showed that high-risk populations are not confined to the South, with large elderly populations in California, Florida, New York, and Pennsylvania. States were grouped by risk using elderly population thresholds:
Low (<500K), Medium (500K–1.25M), and High (>1.25M).
This risk classification aligned with death patterns and was validated through correlation analysis. Forecasted deaths for 2018 showed pronounced peaks in high-risk states and minimal fluctuation in low-risk states, supporting a risk-based planning model.
Recommendations
To support strategic medical staffing during the influenza season, it is recommended that personnel resources be allocated proportionally based on each state’s assigned risk category.
High-risk states — which have both large elderly populations and elevated flu-related mortality — should receive a greater share of available staff and be prioritised during peak months. Medium-risk states should be monitored closely and resourced appropriately based on seasonal forecasts. Low-risk states, with smaller elderly populations and minimal seasonal fluctuation, may require fewer flexible resources and can be supported through baseline staffing levels.
This proportional, risk-based approach will help ensure that staff are deployed where and when they are most needed, improving system responsiveness and reducing pressure on healthcare services during periods of high demand.
Reflection
Successes
The project delivered actionable insights into the spatial distribution of vulnerable populations, enabling the creation of a risk profile to guide regional staff allocations. It also enhanced understanding of influenza seasonality, supporting the strategic timing of deployments to meet anticipated healthcare demands.
Challenges
The available data lacked sufficient detail to account for all known risk factors in the analysis. In particular, the suppression of records for children under 5 limited risk assessment for this group, illustrating how data privacy laws can constrain public health analyses when key demographics are excluded.
Moving Forward
To evaluate the proposed deployment strategy, its impact should be monitored during the upcoming influenza season. Tracking performance indicators, such as staffing efficiency, response times, and patient outcomes across risk tiers, will help evaluate resource allocation and guide improvements.
Incorporating data on chronic health conditions and vaccination rates among seniors could further strengthen the analysis by offering a more comprehensive view of factors driving influenza outcomes. This would support more targeted and effective planning in future seasons.
David Griesel 