Archive project · foundational work
US influenza season — staffing plan
Ahead of the 2018 US flu season, a healthcare organisation needed to decide how many extra medical staff to hire and how to spread them across the 50 states. This analysis turned CDC mortality and census data into a staffing plan.
Why it mattered
Converted CDC mortality and census data into a state-by-state hiring plan — so a fixed staffing budget lands where flu risk actually concentrates: the elderly, not the population at large.
91%
of flu deaths are 65+
51%
are 85+ alone
409k
flu deaths, 2009–17
50
states in the plan
01The questions
- Which age groups face the highest flu-related mortality risk?
- Which states need the most additional medical personnel?
- How should resources be allocated by vulnerability and death rate?
02What I found
- Flu deaths are overwhelmingly elderly — the 65+ groups account for 91% of all deaths across 2009–2017, and the 85+ band alone is 51%.
- Risk climbs steeply with age: 85+ (51%), 75–84 (27%), 65–74 (13%), then a sharp fall-off below 65.
- Under-5 counts don't appear in this dataset — CDC suppresses any state-year cell below 10 deaths, so those small numbers are hidden rather than truly zero; what's measurable here is dominated by the elderly.
- States that combine large elderly populations and high death counts are the clear staffing priority.
FIGHalf of all flu deaths are 85+
65+ — 91% of deathsUnder 65
share of 408,613 influenza deaths, 2009–2017 (CDC)
03Recommendations
- Concentrate additional staff in the states with the highest elderly flu mortality.
- Prioritise regions with the largest 65+ populations — especially the 85+ band that carries half the deaths.
- Match hiring to combined elderly-population-and-mortality, not raw state population.
04Tools & method
TableauExcelUS Census dataCDC mortality data (2009–2017)Data cleaning