Table 2.

Sensitivity analysis for the 1918 scenario, using excess mortality rates derived from historical data from Copenhagen [35].

Mortality per 100,000 population, estimated no. of prevented deaths, and years of life lost prevented per vaccine dose, by age group and relative risk.

Influenza pandemic scenario, age group	Deaths		Years of life lost
	Rate per 100,000 population	Prevented per 100,000 vaccine doses, range	Rate per 10,000 population	Prevented per 10,000 vaccine doses, range	Relative riska
1918
<45 years	563	394–507	3209	2246–2888	28–68
45–64 years	210	147–189	992	694–893	8.8–21
⩾65 years	150	26–80	192	33–102	Reference
All ages	434	294–383	2344	1628–2102	…
1957
<45 years	6.4	4.5–5.8	36	25–32	0.2–0.4
45–64 years	45	32–41	131	92–118	0.6–1.6
⩾65 years	315	54–167	347	59–184	Reference
All ages	53	17–33	96	44–70	…
1968
<45 years	4.2	2.9–3.8	23	16–21	0.2–0.6
45–64 years	41	29–37	110	77–99	1.1–2.8
⩾65 years	151	26–80	167	28–89	Reference
All ages	30	11–21	60	32–48	…

NOTE. Scenarios are based on the 1918 experience in New York City and the 1957 and 1968 pandemic experience in the United States [11, 23, 25]. Vaccination was assumed to prevent 70%–90% of deaths in persons <65 years old and 17%–53% pf deaths in persons ⩾65 years old [14]. Results for age groups with the highest vaccine benefits are shown in boldface. See also the sensitivity analysis in table 2.

^aThe relative risk was calculated as the ratio of the prevented no. of years of life lost per vaccine dose in younger age groups to that in seniors.