Reduction of influenza A(H3N2)-associated symptoms by influenza vaccination in school aged-children during the 2014-2015 winter season dominated by mismatched H3N2 viruses

ABSTRACT

Background: Little is known about the vaccine effectiveness (VE) in attenuating the influenza-associated symptoms in children during the 2014–2015 influenza season in Beijing, China, in which there was a mismatch between the vaccine and circulating strain.

Methods: This study included 210 laboratory-confirmed influenza cases among children, who were enrolled in 2014, from November 1 to December 31. Standard demographic information and clinical symptoms were recorded. Influenza vaccination was confirmed via a vaccination registry. Univariate and multivariate analyses were used to estimate the odds of presenting with clinical symptoms among vaccinated and unvaccinated groups, adjusting for sex, age, area, BMI level, and chronic conditions.

Results: Among the 210 laboratory-confirmed influenza cases, 170 (81.0%) presented with fever ≥38°C. The other most common symptoms were cough (78.1%), sore throat (46.7%), nasal congestion/rhinorrhea (38.6%), headache (34.8%), fatigue (24.8%) and myalgia/arthralgia (16.2%). Approximately 9.0% (19/210) exhibited nausea/vomiting, and 2.4% (5/210) exhibited diarrhea/abdominal pain. Respiratory complications occurred in 5.7% (12/210) of the confirmed influenza cases. In 210 laboratory-confirmed cases, univariate and multivariate conducted after adjusted for the aforementioned characteristics suggested that the odds of fever ≥ 38°C were significantly reduced in vaccinated children (odds ratio [OR]: 0.42, 95% CI: 0.19–0.93; P = 0.033).

Conclusions: Influenza vaccination may reduce the clinical symptoms of laboratory-confirmed influenza cases potentially even in the mismatching season.

Introduction

Annual influenza epidemics can result in substantial morbidity and mortality globally.¹ Schools constitute enclosed societies with a high density of individuals, thus they are conducive to the spread of influenza among children via close contacts.^2,3 Compared with other populations, there is reportedly a higher incidence of influenza in school-age children.⁴ Therefore, school-age children play an important role in influenza transmission within schools, families, and communities.^3–7

Influenza vaccination is the most effective approach to preventing influenza.⁸ School-age children are a priority population for influenza vaccination, given their risk of infection and associated transmission.^1–3 School based influenza vaccination program can reduce adverse influenza-related outcomes.⁸

Annual assessments of influenza vaccine effectiveness (VE) are critical because effectiveness varies from season to season, and VE is influenced by the extent to which there is a match between the strains used in the vaccine and the actual circulating virus strains. During the 2014–2015 influenza season, low estimates of VE against influenza A(H3N2) viruses were reported in many countries, presumably because of a mismatch between the clade 3C.1 vaccine strain and the 3C.2a and 3C.3a clade circulating viruses.^9,10 The available influenza VE estimates during the 2014–2015 seasons are focused on the prevention of medically treated influenza illness attributed to H3N2 viruses, but data on preventive effectiveness against influenza-associated symptoms and complications are limited. In addition, it is unclear whether influenza vaccination conferred protection via symptom reduction in laboratory-confirmed influenza cases during the 2014–2015 influenza season, despite the fact that there was a strain mismatch between the vaccine and the circulating strains.

In Beijing, China, a school-based vaccination program has been operating since 2007, where by a free seasonal trivalent influenza vaccines (IIV3) are provided to elementary and high school students prior to the winter influenza season. A previous study suggests that the school-based vaccination program resulted in a modest degree of prevention of influenza illness caused by H3N2 viruses in children during school-based influenza outbreaks in the 2014–2015 influenza season in Beijing.¹¹ In this present study we attempted to evaluate the preventive value of influenza vaccination, with regard to reducing influenza-associated symptoms and complications in laboratory-confirmed influenza cases among children during school-based influenza outbreaks.

Results

A total of 3101students were included in this study, and of these there were 210 cases of laboratory-confirmed influenza, including 195 cases of A(H3N2) influenza and 15 of unsubtyped influenza A. During the 2014–2015 season, the respective influenza vaccination rates of the 3101 students and the 210 students with laboratory-confirmed influenza were 39.4% (1221/3101) and 43.3% (91/210). The distributions of age groups and areas differed between the vaccinated and unvaccinated groups in the 210 students with confirmed influenza (P < 0.001). Students who were vaccinated against influenza tended to be younger than those who were unvaccinated. Among the 210 confirmed influenza cases, 6.6% of vaccinated individuals were older than 15 years, compared with 15.1% of unvaccinated individuals. There were no significant differences in the other characteristics including sex, BMI level, and chronic disease condition between vaccinated and unvaccinated individuals (Table 1).Among the 210 students with laboratory-confirmed influenza, 170 (81.0%) presented with fever ≥38°C. The other most common symptoms were cough (78.1%), sore throat (46.7%), nasal congestion/rhinorrhea (38.6%), headache (34.8%), fatigue (24.8%) and myalgia/arthralgia (16.2%). Approximately 9.0% (19/210) exhibited nausea/vomiting, and 2.4% (5/210) exhibited diarrhea/abdominal pain. Respiratory complications occurred in 5.7% (12/210) of the confirmed influenza cases. Though the observation was not statistically significant, point estimates of multivariate logistic regression analyses indicated that vaccinated children were less likely to present with most of the symptoms and respiratory complications than unvaccinated children. In both univariate and multivariate analyses adjusted for the aforementioned characteristics vaccinated children were significantly less likely to exhibit fever ≥ 38°C than unvaccinated children (odds ratio [OR]: 0.42, 95% CI: 0.19–0.93; P = 0.033) (Table 2) .

Table 1.

Participant characteristics between vaccinated and unvaccinated laboratory-confirmed influenza cases^a during school influenza outbreaks in Beijing, China, November 1 to December 31, 2014.

	Vaccinated, N = 91	Unvaccinated, N = 119
Characteristics	n (%)	n (%)	P value
Gender			0.896
Male	49 (53.8)	63 (52.9)
Female	42 (46.2)	56 (47.1)
Age group, yr			<0.001
6–10	22 (24.2)	51 (42.9)
11–15	63 (69.2)	50 (42.0)
>15	6 (6.6)	18 (15.1)
Areas			<0.001
Urban	64 (70.3)	108 (90.8)
Rural	27 (29.7)	11 (9.2)
BMIb			0.584
Normal	57 (62.6)	79 (67.5)
Underweight	4 (4.4)	8 (6.8)
Overweight	16 (17.6)	18 (15.4)
Obesity	14 (15.4)	12 (10.3)
Chronic conditions			0.792
Yes	7 (7.9)	8 (6.9)
No	82 (92.1)	108 (93.1)

NOTE: Data are number (%) of participants, unless otherwise indicated. Due to a small number with missing values, not all column categories sum to the total number of vaccinated or unvaccinated students.

^a Cases: patients who tested positive for influenza virus in school influenza outbreaks included in the study, consisting of 195 of A(H3N2) influenza and 15 of unsubtyped influenza A

^bBMI = 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).

Table 2.

Comparisons of clinical symptoms between vaccinated and unvaccinated laboratory-confirmed influenza cases^a during school influenza outbreaks in Beijing, China, November 1 to December 31, 2014.

	Vaccinated N = 91	Unvaccinated N = 119	Total N = 210
Characteristics	n (%)	n (%)	n (%)	Unadjusted ORb (95%CI)	P value	Adjusted ORc (95%CI)	P value
Fever (≥38℃)	67 (73.6)	103 (86.6)	170 (81.0)	0.43 (0.22–0.88)	0.020	0.42 (0.19–0.93)	0.033
Cough	74 (81.3)	90 (75.6)	164 (78.1)	1.40 (0.72–2.75)	0.325	1.14 (0.54–2.42)	0.736
Sore throat	44 (48.4)	54 (45.4)	98 (46.7)	1.13 (0.65–1.95)	0.669	0.90 (0.48–1.66)	0.730
Nasal congestion/rhinorrhea	33 (36.3)	48 (40.3)	81 (38.6)	0.84 (0.48–1.48)	0.548	0.70 (0.38–1.28)	0.250
Nausea/vomiting	5 (5.5)	14 (11.8)	19 (9.0)	0.44 (0.15–1.26)	0.125	0.41 (0.13–1.24)	0.116
Diarrhea/abdominal pain	1 (1.1)	4 (3.4)	5 (2.4)	0.32 (0.04–2.91)	0.311	0.44 (0.04–4.31)	0.480
Headache	33 (36.3)	40 (33.6)	73 (34.8)	1.12 (0.63–1.99)	0.689	0.87 (0.47–1.60)	0.645
Fatigue	18 (19.8)	34 (28.6)	52 (24.8)	0.62 (0.32–1.18)	0.145	0.66 (0.34–1.30)	0.234
Myalgia/arthralgia	11 (12.1)	23 (19.3)	34 (16.2)	0.57 (0.26–1.25)	0.162	0.61 (0.27–1.39)	0.241
Chest pain	0 (0)	2 (1.7)	2 (1.0)	NA	NA	NA	NA
Chill	3 (3.3)	6 (5.0)	9 (4.3)	0.64 (0.16–2.64)	0.539	0.75 (0.17–3.32)	0.705
Respiratory complicationsd	3 (3.3)	9 (7.6)	12 (5.7)	0.42 (0.11–1.59)	0.199	0.42 (0.10–1.72)	0.227

NOTE: Data are number (%) of participants, unless otherwise indicated. Boldface indicates statistical significance. OR, odds ratio; CI, confidence interval. NA, not available (because of small sample size data could not be analyzed by unconditional logistic regression)

^aConfirmed influenza cases consisted of 195 of A(H3N2) influenza and 15 of unsubtyped influenza A.

^b Unadjusted OR was obtained for vaccination status by univariate unconditional logistic regression analysis.

^c Adjusted OR was obtained for vaccination status by multivariate unconditional logistic regression analysis controlling participant characteristics, including sex, age (modeled as a cubic spline), areas, BMI level, chronic conditions.

^d Respiratory complications: pneumonia, tracheitis, bronchitis and asthma.

Discussion

In the current study, the results suggested that fever may be alleviated by influenza vaccination in laboratory-confirmed influenza cases during school-based influenza outbreaks in the 2014–2015 Season in Beijing, China.

During the 2014–2015 influenza season, approximately 39.4% of students affected by school-based influenza outbreaks had been vaccinated against influenza, which was lower than the approximately 47% in the United States in the 2011–2012 influenza season.¹² However, the IIV3 coverage was only 10.8% among ambulatory patients aged 3–17 years in the 2014–2015 influenza season in Beijing.¹¹ In a study in Beijing it was reported that influenza consistently occurred in children, aged<15 years earlier than it occurred in adults, implying that school children may cause epidemic fluctuations.¹³ Reinforcing the potential value of school-based influenza vaccination programs, as a way to achieve high vaccination coverage is highly justified.¹⁴

Due to the mismatch between the H3N2 vaccine strain and the circulating strains during the 2014–2015 influenza season, the estimates of VE against medically attended influenza illness caused by H3N2viruses were very poor globally, such as 6% in the United States, −17% in Canada, 14.4% in European countries and −25% in China.^11,15–17 In a previous study, investigating school influenza outbreaks in China VE against laboratory-confirmed influenza in students was only 18% after adjustment for various characteristics.¹¹ However, in an influenza VE study investigating hospitalization in the US during the same season, VE for preventing H3N2-related hospitalization was 43%, which was better than estimates in patients consulting ambulatory care facilities.¹⁸

Remarkably, in the current study there was a significant reduction in the odds of developing high fever in vaccinated children infected with influenza. The study also suggested that influenza vaccination may have mitigated influenza-related symptoms in children who contracted influenza, though this observation was not statistically significant. Similarly, previous studies have also reported beneficial effects of influenza vaccination in patients who subsequently contract influenza. Influenza vaccination was effective, with regard to preventing clinical presentations, reducing illness severity, and reducing the odds of severe influenza outcomes in the event of influenza.^19,20

The current study had several limitations. First, because the data on students’ clinical symptoms were based on their parents’ or legal guardians’ recollections, we cannot rule out the possibility of recall bias. Second, the number of students with confirmed influenza was relatively small, as a consequence, the study may have lacked the power to detect statistically significant VE against some symptoms. Third, we did not attempt to collect information pertaining to aspects, such as students’ nutrition, hygiene habits, or preventive measures against influenza infection, which may have led to residual and unmeasured confounding.

In conclusion, in the present study, action influenza vaccination could reduce clinical symptoms in laboratory-confirmed influenza cases in a vaccine mismatched season despite low vaccine effectiveness against A(H3N2) influenza viruses.

Materials and methods

The surveillance systems of school influenza outbreaks

There are two syndrome surveillance systems are used to monitor influenza-like illness (ILI)and identify school outbreaks of influenza, (measured or self-reported temperature of ≥38°C with either cough or sore throat), and febrile illness of any etiology (involving a measured or self-reported temperature of ≥37.5°C). As described in a previous study,²¹ the respective definitions of an influenza outbreaks as determined by each surveillance system: 1) 10 or more epidemiological-linked ILIs identified in a school within 1 week; and 2) 10 or more febrile illnesses within a single school classroom within 2 days.

The staff of local Centers for Disease Prevention and Control (CDC) were requested to undertake field epidemiological investigations within a day of suspected outbreaks. Oral pharyngeal (OP) swabs were collected from at least 10 symptomatic cases from outbreaks school by local CDC staff, and transported to local CDC laboratories for the detection of influenza viruses via real-time reverse-transcription polymerase chain reaction (rRT-PCR) assay.

Participants

From November 1, 2014, approximately 14 days after the commencement of the school influenza vaccination campaign during the 2014–2015 season, to December 31, 2014, 70 influenza outbreaks were confirmed in elementary schools and high schools. Twenty-seven school outbreaks were excluded from the present study because vaccination was administered ≤14 days prior to the onset of an outbreak. A total of 3224 children in the classes affected by the 43 outbreaks were recruited, of which 123 children with incomplete or missing 2014–2015 vaccination records were excluded. Among the enrolled 3101 students, 210 students were laboratory-confirmed influenza cases.

Data collection

After obtaining informed consent from parents or legal guardians, participants’ individual information and clinical symptoms were provided by students and their parents or guardians via a questionnaire, including age, sex, height, weight, presence of chronic medical conditions, symptoms, and illness onset date. The Beijing Management System of Information on the Immunization Program was used to collect documented influenza vaccination records pertaining to 2014–2015 IIV3. School information was also collected, including the type of school (elementary, junior high, or high schools), areas (urban or rural), and the number of students who received IIV3.

Statistical analysis

Questionnaire and laboratory data were entered in duplicate using EpiData software, and statistical analysis was performed using SAS University Edition statistical software.

Categorical variables were described using frequencies and percentages. Pearson’s χ²-test was used to assess differences in characteristics between vaccinated and unvaccinated groups. Proportion of each influenza-associated clinical symptom was calculated in vaccinated and unvaccinated groups, within the laboratory-confirmed influenza cases. The clinical symptoms included were fever (temperature ≥38°C), cough, sore throat, nasal congestion/rhinorrhea, nausea/vomiting, diarrhea/abdominal pain, headache, fatigue, myalgia/arthralgia, chest pain chill, and respiratory complications (pneumonia, tracheitis, bronchitis, and asthma). Each specific symptom was used as an outcome. Associations between influenza vaccination in 2014–2015 season and the occurrence of each symptom in in laboratory-confirmed influenza cases were investigated using multivariate unconditional logistic regression analysis after controlling for sex, age (modeled as a cubic spline), BMI level, areas, and chronic conditions. All tests were two-sided, and P < 0.05 was deemed to indicate statistical significance.

The study was approved by institutional review board and human research ethics committee of Beijing CDC.

Funding Statement

This work was supported by Beijing Health System High Level Health Technology Talent Cultivation Plan (2013-3-098), Beijing Municipal Science and Technology Commission (D141100003114002), Capital’s Funds for Health Improvement and Research (2018-2-1013), Beijing Talents Fund (2014000021223ZK36).

Acknowledgments

We thank all the 70 schools participated in this study, and we also appreciate all of the support of the data collectors. Also, we thank the children and their parents, all health workers and teachers involved in this study for their sustained support.

Disclosure of potential conflicts of interest

No potential conflict of interest was reported by the authors.

Author’s contribution

This study was conceived by Wei Duan and Peng Yang. Wei Duan, Li Zhang and Shuangsheng Wu collected study data and Wei Duan also interpreted the data. All the co-authors in this study performed the study.