Influenza vaccine effectiveness estimates against influenza A(H3N2) and A(H1N1) pdm09 among children during school-based outbreaks in the 2016–2017 season in Beijing, China

Figures & data

Figure 1. Flow chart of subject enrollment in the test-negative case–control design study for estimating influenza vaccine effectiveness among children during school-based outbreaks during the 2016–2017 influenza season in Beijing, China.

Figure 2. The time distribution of the eligible school influenza outbreaks included in the test-negative case–control study for estimating influenza vaccine effectiveness, 1 November 2016 Figure 1to 30 April 2017.

Table 1. Participant characteristics for estimates of influenza vaccine effectiveness for 2016–2017 trivalent influenza inactivated vaccine during school outbreaks of influenza in Beijing, 1 November 2016 to 30 April 2017.

Characteristics	Characteristics of Cases and Controls			Vaccination Status Among Cases and Controls
	Confirmed cases^a No. (Col.%)	Asymptomatic Controls^b No. (Col.%)	P value^d	Confirmed cases^a Vaccinated No./total (%)	P value^d	Asymptomatic Controls^b Vaccinated No./total (%)	P value^d
Overall	176(100)	628 (100)		30/176 (17.0)		229/628 (36.5)
Age group
6–10 years	140(79.5)	486 (77.4)	0.830	17/140 (12.1)	0.002	182/486 (37.4)	0.028
11–15 years	28 (15.9)	110 (17.5)		11/28 (39.3)		40/110 (36.4)
>15 years	8 (4.5)	32 (5.1)		2/8 (25.0)		7/32 (21.9)
Sex
Male	90 (51.1)	327 (52.1)	0.827	14/90 (15.6)	0.591	116/327 (35.5)	0.591
Female	86 (48.9)	301 (47.9)		16/86 (18.6)		113/301 (37.5)
Areas
Urban	110 (62.5)	382 (60.8)	0.687	20/110 (18.2)	0.605	129/382(33.8)	0.080
Rural	66 (37.5)	246 (39.2)		10/66 (15.2)		100/246 (40.7)
BMI^c
Normal	112 (63.6)	384 (61.1)	0.743	15/112 (13.4)	0.320	139/384 (36.2)	0.716
Underweight	17 (9.7)	58 (9.2)		5/17 (29.4)		22/58 (37.9)
Overweight	18 (10.2)	84 (13.4)		4/18 (22.2)		27/84 (32.1)
Obesity	29 (16.5)	102 (16.2)		6/29 (20.7)		41/102 (40.2)
Chronic conditions
Yes	5 (2.8)	30 (4.8)	0.266	2/5 (40.0)	0.202	10/30 (33.3)	0.715
No	171 (97.2)	598 (95.2)		28/171 (16.4)		219/598 (36.6)

NOTE: BMI: body mass index; NS: not statistically significant (P > 0.05).

^aConfirmed cases: patients who tested positive for influenza virus in an influenza outbreak included in the study.

^bAsymptomatic controls: classmates of cases in an influenza outbreak included in the study who were asymptomatic (i.e., without symptoms of fever, cough, or sore throat) during the period from the illness onset of the index case to the ending of the outbreak.

^cBMI = weight (kg)/height (m)^2; BMI of students were categorized into four levels (normal, underweight, overweight and obesity) according to the National Student Physical Health Standard (2014)

^dPearson χ2 test was used to assess differences between participant characteristic groups (rows) in the percentage vaccinated.

Table 2. Estimates of vaccine effectiveness (VE) by age group and type of school during outbreaks of influenza in schools in Beijing, 1 November 2016 to 30 April 2017.

	Confirmed cases^a	Asymptomatic controls^b		Unadjusted	Adjusted^c
	Vaccinated No./total (%)	Vaccinated No./total (/%)	P value	VE% (95% CI)	VE% (95% CI)
Overall	30/176 (17.0)	229/628 (36.5)	< 0.001	68 (50 to 80)	69 (51 to 81)
Age group
6–10 years	17/140 (12.1)	182/486 (37.4)	< 0.001	78 (59 to 88)	73 (53 to 85)
11–15 years	11/28 (39.3)	40/110 (36.4)	0.775	NA^d	NA^d
>15 years	2/8 (25.0)	7/32 (21.9)	0.850	NA^d	NA^d
Type of school
Primary school	21/148 (14.2)	194/520 (37.3)	< 0.001	72 (52 to 84)	78 (59 to 89)
Middle school	8/25 (32.0)	31/96 (32.3)	0.978	NA ^d	NA ^d
High school	1/3 (33.3)	4/12 (33.3)	1.000	NA^d	NA^d
Influenza subtype
Influenza A(H1N1)pdm09	8/36 (22.2)	229/628 (36.5)	0.083	57 (−4 to 83)	60 (−15 to 86)
Influenza A(H3N2)	20/136 (14.7)		< 0.001	74 (55 to 85)	73 (52 to 84)

NOTE. CI: confidence interval; VE, vaccine effectiveness. Vaccine effectiveness was estimated by comparing the vaccination coverage in cases and controls and calculated as 100 × (1 − odds ratio) using conditional logistic regression model.

^aConfirmed cases: patients who tested positive for influenza virus in an influenza outbreak included in the study.

^bAsymptomatic controls: classmates of cases in an influenza outbreak included in the study who were asymptomatic (i.e., without symptoms of fever, cough, or sore throat) during the period from the illness onset of the index case to the end of the outbreak.

^cAdjusted for age group, sex, location, BMI, chronic conditions by conditional logistic regression.

^dNot available because the models did not converge due to low sample size.

Table 3. Effect of previous vaccination on estimates of vaccine effectiveness (VE) in school outbreaks of influenza in Beijing, 1 November 2016 to 30 April 2017.

Vaccine(s) received	Vaccination status		VE (%) against any influenza virus		VE (%) against influenza A(H1N1)pdm09		VE (%) against influenza A(H3N2)
	Confirmed cases^a No. (Col.%)	Asymptomatic Controls^b No. (Col.%)	Unadjusted (95% CI)	Adjusted^c (95% CI)	Unadjusted (95% CI)	Adjusted^c (95% CI)	Unadjusted (95% CI)	Adjusted^c (95% CI)
Overall Unvaccinated	176 (100) 133 (75.6)	628 (100) 339 (54.0)
2015–2016 vaccination only	13 (7.4)	60 (9.6)	46 (−4 to 72)	46 (−5 to 72)	50 (−162 to 91)	56 (−89 to 90)	43 (−16 to 72)	39 (−28 to 71)
2016–2017 vaccination only	14 (8.0)	145 (23.1)	77 (58 to 87)	77 (58 to 87)	71 (−4 to 92)	70 (1 to 91)	80 (59 to 90)	79 (58 to 90)
Both 2016–2017 and 2015–2016 vaccinations	16 (9.1)	84 (13.4)	57 (19 to 77)	57 (17 to 78)	44 (−75 to 82)	46 (−44 to 80)	67 (27 to 85)	54 (−6 to 80)

NOTE. CI: confidence interval; VE, vaccine effectiveness. Vaccine effectiveness was estimated by comparing the vaccination coverage in cases and controls and calculated as 100 × (1 − odds ratio) using conditional logistic regression model.

^aConfirmed cases: patients who tested positive for influenza virus in an influenza outbreak included in the study.

^bAsymptomatic controls: classmates of cases in an influenza outbreak included in the study who were asymptomatic (i.e., without symptoms of cough, or sore throat) during the period from the illness onset of the index case to the ending of the outbreak.

^cAdjusted for age group, sex, areas, BMI, chronic conditions by conditional logistic regression.