Relative incidence and individual-level severity of seasonal influenza A H3N2 compared with 2009 pandemic H1N1

BMC Infectious Diseases

Table 3 Estimated absolute risks of severe outcomes per infection

Outcome	Strain	Rounds	Infections^a	Estimated severe outcomes^b	Risk per 10,000 infections (Binomial 95% CI)^c
Excess all-cause deaths	A(H1N1)pdm09	1 to 2	655000	61	0.93 (0.66 , 1.31)
		2 to 3	520000	160	3.08 (2.09 , 4.55)
		3 to 4	971000	359	3.70 (2.81 , 4.88)
	A(H3N2)	1 to 2	158000	73	4.62 (2.27 , 9.30)
		2 to 3	520000	409	7.87 (5.35 , 11.6)
		3 to 4	67800	37	5.46 (1.89 , 14.9)
Excess respiratory deaths	A(H1N1)pdm09	1 to 2	655,000	57	0.87 (0.62 , 1.23)
		2 to 3	520,000	130	2.50 (1.70 , 3.70)
		3 to 4	971000	289	2.98 (2.27 , 3.93)
	A(H3N2)	1 to 2	158000	34	2.15 (1.06 , 4.33)
		2 to 3	520000	190	3.66 (2.48 , 5.40)
		3 to 4	67800	17	2.51 (0.87 , 6.83)
Excess respiratory hospitalisations	A(H1N1)pdm09	1 to 2	655000	3470	53.0 (37.7 , 74.8)
		2 to 3	520000	2480	47.7 (32.4 , 70.5)
		3 to 4	971000	4590	47.2 (36.0 , 62.4)
	A(H3N2)	1 to 2	158000	942	59.6 (29.3 , 120)
		2 to 3	520000	5300	102 (69.3 , 151)
		3 to 4	67800	476	70.2 (24.3 , 191)

^aEstimated cumulative number of seroconversions between study rounds, adjusted for the difference in the age distribution between study participants and the overall Hong Kong population. For each age group i in our study data (i = 1,2,3,4), which corresponds to age groups (2-18, 19-44,45-64,65+ ), we define Ai=Age group specific proportion of four fold rise in our study data x sample size of this study x Proportion of population in this age group
Then, number of infections can be calculated with the following formula
\( Hong\ Kong\ population\ X\frac{\sum_{i=1}^4\mathrm{Ai}}{our\ study\ sample\ size} \)
^bEstimated excess rate of adverse outcomes. Point estimates calculated as in [26]
^cBinomial confidence bounds reflecting uncertainty in the number of infections, rather than the number of adverse outcomes. See Discussion

ISSN: 1471-2334