- Email: [email protected]
Effectiveness of rotavirus pentavalent vaccine under a universal immunization programme in Israel, 2011–2015: a case–control study
Clinical Microbiology and Infection 24 (2018) 53e59
Contents lists available at ScienceDirect
Clinical Microbiology and Infection journal homepage: www.clinicalmicrobiologyandinfection.com
Original article
Effectiveness of rotavirus pentavalent vaccine under a universal immunization programme in Israel, 2011e2015: a caseecontrol study K. Muhsen 1, *, E. Anis 2, U. Rubinstein 3, E. Kassem 4, S. Goren 1, L.M. Shulman 1, 5, M. Ephros 6, 7, D. Cohen 1 1)
Department of Epidemiology and Preventive Medicine, School of Public Health, Sackler Faculty of Medicine, Tel Aviv University Ramat Aviv, Tel Aviv, Israel Division of Epidemiology, Ministry of Health, Jerusalem, Israel Department of Pediatrics, Laniado Medical Center, Netanya, Israel 4) Department of Pediatrics, Hillel Yaffe Medical Center, Hadera, Israel 5) Central Virology Laboratory, Ministry of Health, Tel Hashomer, Israel 6) Department of Pediatrics, Carmel Medical Center, Haifa, Israel 7) Faculty of Medicine, TechnioneIsrael Institute of Technology, Haifa, Israel 2) 3)
a r t i c l e i n f o
a b s t r a c t
Article history: Received 19 January 2017 Received in revised form 14 April 2017 Accepted 18 April 2017 Available online 22 April 2017
Objectives: The use of rotavirus pentavalent vaccine (RotaTeq®) as a sole vaccine within rotavirus universal immunization programmes remains limited. We examined the effectiveness of RotaTeq in preventing rotavirus gastroenteritis (RVGE) hospitalization in Israel, after the introduction of universal immunization against the disease. Methods: A test-negative caseecontrol study included age-eligible children for universal RotaTeq immunization (aged 2e59 months, born in 2011e2015). Cases (n ¼ 98) were patients who tested positive for rotavirus by immunochromatography; those who tested negative (n ¼ 628) comprised the control group. Information on rotavirus immunization history was obtained through linkage with a national immunization registry. Vaccination status was compared between cases and controls, adjusted odds ratios (aORs) were obtained from logistic regression models, and vaccine effectiveness calculated as (1 aOR)*100. Results: Immunization with RotaTeq was less frequent in RVGE cases (73.5%) than in controls (90.1%), p < 0.001; this association persisted after controlling for potential confounders. Effectiveness of the complete vaccine series was estimated at 77% (95% confidence interval (CI): 49e90) in children aged 6 e59 months, and 86% (95% CI: 65e94) in children aged 6e23 months; whereas for the incomplete series, the respective estimates were 72% (95% CI: 28e89) and 75% (95% CI: 30e91). Vaccine effectiveness was estimated at 79% (95% CI: 45e92) against G1P[8]-associated RVGE hospitalizations and 69% (95% CI: 11 e89) against other genotype-RVGE hospitalizations. Conclusions: High effectiveness of RotaTeq as the sole rotavirus vaccine in a universal immunization programme was demonstrated in a high-income country. Although partial vaccination conferred protection, completing the vaccine series is warranted to maximize the benefit. K. Muhsen, Clin Microbiol Infect 2018;24:53 © 2017 European Society of Clinical Microbiology and Infectious Diseases. Published by Elsevier Ltd. All rights reserved.
Editor: Dr C. Pulcini Keywords: Rotavirus Vaccine Gastroenteritis Effectiveness Post-marketing
Introduction
* Corresponding author. Khitam Muhsen, Department of Epidemiology and Preventive Medicine, School of Public Health, Sackler Faculty of Medicine, Tel Aviv University, Ramat Aviv, 6139001 Tel Aviv, Israel. E-mail address: [email protected] (K. Muhsen).
Rotavirus is the main cause of severe diarrhoea in children [1] and a leading cause of child mortality in developing countries [2]. Two rotavirus vaccines became available globally in recent years, both administered orally [3,4]. One is a monovalent vaccine containing attenuated human G1P[8] rotavirus strain (Rotarix®, GlaxoSmithKline Biologicals, Brentford, UK), administered in two doses
http://dx.doi.org/10.1016/j.cmi.2017.04.018 1198-743X/© 2017 European Society of Clinical Microbiology and Infectious Diseases. Published by Elsevier Ltd. All rights reserved.
54
K. Muhsen et al. / Clinical Microbiology and Infection 24 (2018) 53e59
at ages 2 and 4 months [4]. The other is a pentavalent live bovineehuman reassortant vaccine (RotaTeq®, Merck and Co., Inc., Whitehouse Station, NJ, USA) containing strains representing G1, G2, G3, G4 and P[8] genotypes, and administered in three doses, at ages 2, 4 and 6 months [3]. Although different disease severity scoring systems were used, both vaccines showed high efficacy (85e95%) in preventing severe rotavirus gastroenteritis (RVGE) in clinical trials conducted in Europe and the Americas [3,4]. Rotavirus vaccine efficacy was low (40e61%) in developing countries [5]. The World Health Organization (WHO) recommends including rotavirus vaccine in national immunization programmes worldwide [6]. By September 2016, a decade after rotavirus vaccines became available, 81 countries had introduced rotavirus immunization into national immunization programmes and seven countries had introduced rotavirus immunization in certain regions only [7]. Presently, most countries use Rotarix [7]. Significant and sustained reduction in the burden of RVGE was observed in countries that introduced rotavirus immunization, including Israel [8e14]. These studies employed mostly ecological design, comparing the incidence of RVGE in the entire target population, both vaccinated and unvaccinated groups, before and after the introduction of universal rotavirus immunization. The impact of universal rotavirus immunization on disease burden in the population is a function of vaccine coverage, vaccine effectiveness and possibly herd immunity. Numerous postmarking evaluations have assessed rotavirus vaccine effectiveness, mostly using a testnegative caseecontrol study design [15e19]. In this efficient and cost-saving design, rotavirus immunization history is compared between RVGE cases and control diarrhoea patients who test negative for rotavirus [20]. Due to the limited introduction of rotavirus vaccines in high-income countries, and since most of them introduced Rotarix, updated evidence of the net effectiveness of RotaTeq as a sole vaccine under universal vaccination policy in high-income settings, outside the USA, remains scarce [17,21]. In Israel, both rotavirus vaccines were licensed and became available by the end of 2007 in the private market, with partial refund through the health maintenance organizations. In December 2010, RotaTeq was introduced in the national immunization programme, free of charge for all children, and since that time it has been the only vaccine used. An Israeli study that comprised 70% Bedouin children reported 63% vaccine effectiveness for RotaTeq [21]. Bedouins are a small sub-group of the Arab minority in Israel, and typically live in low socioeconomic and suboptimal environmental conditions, thus generalizability from that study remains limited. We examined the effectiveness of RotaTeq in a universal immunization programme, in preventing RVGE hospitalization during 2011e2015 in Israel, as a model of a high-income country.
RotaTeq is administered at ages 2, 4 and 6 months; the estimated vaccine coverage is >80%. The sampling frame included children who were hospitalized due to diarrhoea (three or more watery stools during a 24 h period) [13,14,23e25]. During the study period, research staff in each hospital were instructed to systematically collect stool samples from all children hospitalized for diarrhoea and to submit the samples for rotavirus antigen testing at the local laboratory in each hospital. The current test-negative caseecontrol study was restricted to birth cohorts of the years 2011e2015 who were eligible for RotaTeq under universal immunization and hospitalized during 2011e2015. We set a lowest age limit at admission as 2 months, since this is the recommended age for the first RotaTeq dose. Children who were not tested for rotavirus or without information on rotavirus immunization were excluded from the analysis. Parents were interviewed concerning demographic and clinical information on the number of stools passed, vomiting and fever. Information on rotavirus vaccination history was obtained through linkage with the national immunization registry at the Ministry of Health. If parents expressed refusal to vaccinate their child, that information would be documented in the child's medical record, and was collected through the linkage with the immunization registry. Information on immunization was collected for children hospitalized until October 2015. Vaccination status was determined up to the admission date. Children were classified as unvaccinated or vaccinated with at least one dose, and according to the number of doses. Socioeconomic status (SES) was classified on a scale of 1e10, according to the socioeconomic rank of the town of residence, as defined by the Israel Central Bureau of Statistics [26]. The ranks were grouped into three categories of SES: 1e3 (low), 4e5 (intermediate) and 6e10 (high). Laboratory methods Stool samples were collected from patients within the first 48 h of hospitalization. Specimens were tested for rotavirus antigen by immunochromatography (Rotavirus Dipsticks, Hylabs Rehovot & Novamed, Jerusalem, Israel), according to manufacturers' instructions. In practice, 87% of the analysed stools were tested within 2 days of admission, 95% within 3 days. Cases were children with a confirmed rotavirus-positive test in their stool, and controls were patients who tested negative for rotavirus. After performing the rotavirus test, the remaining faecal material was transferred to the Central Virology Laboratory in cool conditions. Specimens were kept at 70 C until genotyping was performed [14].
Methods
Statistical methods
Study population and design
c2-Test and Fisher's exact test, as appropriate, were used to examine differences between vaccinated and unvaccinated children in sociodemographic characteristics, season/year of admission and birth year. Similarly, differences between RVGE cases and testnegative controls were examined. Logistic regression models were fitted to obtain adjusted odds ratio (aOR) and 95% confidence interval (CI) for immunization status before admission. The models were adjusted for vaccination status, age, year and season of admission and SES. Vaccine effectiveness was calculated as (1 aOR), and expressed as a percentage [27]. The primary analyses were limited to children 6e59 months of age, assuming that those children were all eligible for the full vaccine series. Additional analyses were performed with
A hospital-based active surveillance study was conducted between November 2007 and February 2016 in northern Israel, following a generic protocol of the WHO [22]. Details of the study design have been published elsewhere [13,14,23,24]. The study included children 0e59 months of age residing in the catchment area of three hospitals: Hillel Yaffe in Hadera, Carmel in Haifa, and Laniado in Netanya. Both Jewish and Arab residents live in the study area. According to the Universal National Health Insurance Law, all Israelis have access to medical care. The national immunization programme is conducted at maternal and child health clinics.
K. Muhsen et al. / Clinical Microbiology and Infection 24 (2018) 53e59
stratification by age group, including an exploratory analysis limited to infants two to five months of age, who were eligible to receive at least one dose. Data were analysed using SPSS version 23. Ethical considerations The study protocol was approved by the Institutional Review Boards of all participating hospitals and by the Ministry of Health. Parents of the participating children signed informed consent forms. Results Overall, 1437 of the children hospitalized in the participating hospitals with gastroenteritis during the study period were born during 2011e2015; of these, 1272 were aged 2e59 months and eligible for rotavirus immunization in the framework of the universal vaccination programme. Rotavirus was tested among 831 (65.3%) children (Supplementary Fig. 1). Compared to children not tested for rotavirus, a greater proportion of children tested passed 6 stools/day and vomited: 56.7% vs 38.7% (p < 0.001) and 78.2% vs 72.6% (p ¼ 0.024), respectively. Among both groups, 51.3% had fever (38 C) at admission. Among children tested for rotavirus, data on rotavirus immunization status was available for 726. Rotavirus was detected among 98 (13.5%), and the rest tested negative and served as controls. Children with missing data for rotavirus immunization status and rotavirus stool test included higher proportions of toddlers aged 24e59 months, children from high-SES communities and children admitted in 2014e2015, compared to children with complete information (Table 1). Among children aged 2e59 months, 88 (12.1%) did not receive any dose of RotaTeq; for 26 (29.5%) of them this was due to parental refusal. The proportion of immunized children with at least one dose differed by age, being highest at ages 6e23 months. This percentage was lower in patients born in 2015 than during 2011e2014 (65.2% vs 88.6%, p ¼ 0.004), and in children admitted in 2011 than in those admitted during 2012e2015 (Table 2). The proportion of vaccinated children did not differ according to hospital of admission (p ¼ 0.8). RVGE cases were older, and included more Jewish children and residents of intermediate- to high-SES towns than did controls.
55
Cases were admitted significantly more often during the winter, the typical rotavirus season in Israel; this in addition to differences between cases and controls in the year of admission (Table 3). Cases and controls did not differ according to hospital of admission (p ¼ 0.3). For all age groups, the proportion of children immunized with at least one RotaTeq dose, as well as with the complete vaccine series, was lower among RVGE patients than controls (Table 4). In a multivariate analysis, immunization with RotaTeq conferred significant protection against RVGE hospitalization in children aged 6e59 months, with similar effectiveness for any dose (76%; 95% CI: 48e89%) and for the complete series (77%; 95% CI: 49e90%) (Table 5). A lower point estimate of vaccine effectiveness of the complete RotaTeq series was found in children aged 6e11 months (82%; 95% CI: 36e95%) vs children aged 12e23 months (96%; 95% CI: 77e99%), although CIs were overlapping. The effectiveness for both age groups combined (6e23 months) was 86% (95%; CI: 65e94%) (Table 5). Socioeconomic status of residence and population group were highly correlated (lower SES in Arab patients: Spearman's r ¼ 0.77; p < 0.001); therefore, the two variables were not included in the same model. Similar results were obtained from models that included the variable population group (Supplementary material 2). The number of children aged 24e59 months was small to conduct a separate analysis. In the 2e5-month age group, the proportion of immunized RVGE cases was lower than among controls (Table 4), and RotaTeq conferred protection in this age group also (Table 5). Faecal samples were tested for rotavirus genotype in 41/98 (41.8%) of RVGE patients. Among these, 20 (48.8%), eight (19.5%), seven (17.1%), two (4.9%) and one were G1P[8], G3P[8], G9P[8], G2P [4] and G12P[8], respectively. In three additional samples, either the G- or P-type was non-typeable. Among 20 RVGE cases of G1P[8] genotype, 13 (65.0%) were immunized with RotaTeq compared to 90.1% in control children (age-adjusted OR: 0.21; 95% CI: 0.08e0.55; p ¼ 0.002), whereas in cases with other genotypes, 16/ 21 (76.2%) were immunized (age-adjusted OR: 0.31; 95% CI: 0.11e0.89; p ¼ 0.029). This yields vaccine effectiveness of 79% (95% CI: 45e92%) and 69% (95% CI: 11e89%) against hospitalizations for RVGE caused by these genotypes, respectively.
Table 1 Characteristics of children with available data on rotavirus and vaccination status, compared to those with missing data on these variables Characteristics Age (months) 2e5 6e11 12e23 24e59 Sex, males Population group: Jews Year of admission 2011 2012 2013 2014 2015 Season of admission DeceFeb (Winter) MareMay (Spring) JuneAug (Summer) SepeNov (Autumn) SES of place of residence Low: 1e3 Intermediate: 4e5 High: 6e10 SES, socioeconomic status.
Available data (n ¼ 726)
Missing data (n ¼ 546)
184 (25.3%) 242 (33.3%) 228 (31.4%) 72 (9.9%) 386 (53.2%) 459 (63.2%)
94 (17.2%) 173 (31.7%) 189 (34.6%) 90 (16.5%) 302 (55.3%) 329 (60.6%)
39 (5.4%) 161 (22.2%) 165 (22.7%) 235 (32.4%) 126 (17.4%)
13 (2.4%) 105 (19.2%) 111 (20.3%) 197 (36.1%) 120 (22.0%)
194 169 165 198
130 125 141 150
p value <0.001
0.4 0.3 0.009
0.17 (26.7%) (23.3%) (22.7%) (27.3%)
227 (31.5%) 378 (52.4%) 116 (16.1%)
(23.8%) (22.9%) (25.8%) (27.5%)
167 (31.3%) 249 (46.6%) 118 (22.1%)
0.018
56
K. Muhsen et al. / Clinical Microbiology and Infection 24 (2018) 53e59
Table 2 Vaccination status of children included in the analysis, according to various characteristics Characteristics Age (months) 2e5 6e11 12e23 24e59 Sex Males Females Population group Jews Arabs SES of place of residence Low: 1e3 Intermediate: 4e5 High: 6e10 Birth year 2011 2012 2013 2014 2015 Year of admission 2011 2012 2013 2014 2015
Total
Vaccinated with at least one RotaTeq dose
184 242 228 72
136 (73.9%) 222 (91.7%) 217 (95.2%) 63 (87.5%)
386 340
338 (87.6%) 300 (88.2%)
459 267
407 (88.7%) 231 (86.5%)
227 378 116
198 (87.2%) 337 (89.2%) 100 (86.2%)
218 203 171 111 23
191 (87.6%) 181 (89.2%) 158 (92.4%) 93 (83.8%) 15 (65.2%)
39 161 165 235 126
29 (74.4%) 145 (90.1%) 146 (88.5%) 213 (90.6%) 105 (83.3%)
p value <0.001a
0.7a
0.3a
0.6a
0.002a 0.10b
0.021a 0.8b
SES, socioeconomic status. a c2-Test. b p for trend: by c2 linear by linear association.
Table 3 Rotavirus gastroenteritis cases and rotavirus test-negative controls by sociodemographic factors Factors Age (months) 2e5 6e11 12e23 24e59 Sex (males) Population group: Jews SES of place of residence Low: 1e3 Intermediate: 4e5 High: 6e10 Birth year 2011 2012 2013 2014 2015 Year of admission 2011 2012 2013 2014 2015 Month/season of admission DeceFeb (Winter) MareMay (Spring) JuneAug (Summer) SepeNov (Autumn)
RVGE cases (N ¼ 98)
Rotavirus-test negative controls (N ¼ 628)
16 (16.3%) 34 (34.7%) 39 (39.8%) 9 (9.2%) 54 (55.1%) 74 (75.5%)
168 (26.8%) 208 (33.1%) 189 (30.1%) 63 (10.0%) 332 (52.9%) 385 (61.3%)
19 (19.6%) 57 (58.8%) 21 (21.6%)
208 (33.3%) 321 (51.4%) 95 (15.2%)
29 (29.6%) 32 (32.7%) 21 (21.4%) 14 (14.3%) 2 (2.0%)
189 (30.1%) 171 (27.2%) 150 (23.9%) 97 (15.4%) 21 (3.3%)
7 (7.1%) 10 (10.2%) 36 (36.7%) 21 (21.4%) 24 (24.5%)
32 (5.9%) 151 (24.0%) 129 (20.5%) 214 (34.1%) 102 (16.2%)
59 (60.2%) 22 (22.4%) 6 (6.1%) 11 (11.2%)
135 147 159 187
p value 0.098a 0.07b
0.6 0.007 0.018a 0.006b
0.8
<0.001a 0.2b
<0.001
RVGE, rotavirus gastroenteritis; SES, socioeconomic status. a c2-Test. b p value for trend by c2 linear by linear association.
(21.5%) (23.4%) (25.3%) (29.8%)
K. Muhsen et al. / Clinical Microbiology and Infection 24 (2018) 53e59 Table 4 Differences between RVGE cases and rotavirus test-negative controls in immunization status Age/RotaTeq dose(s)
RVGE cases
Rotavirus-test negative controls
Age 2e59 months 1 dose 1 or 2 doses 3 doses Age 6e59 months 1 dose 1 or 2 doses 3 doses Age 2e5 months 1 dose Age 6e11 months 1 dose 1 or 2 doses 3 doses Age 12e23 months 1 dose 1 or 2 doses 3 doses Age 6e23 months 1 dose 1 or 2 doses 3 doses
n ¼ 98 72 (73.5%) 26 (26.5%) 46 (46.9%) n ¼ 82 64 (78.0%) 18 (22.0%) 46 (56.1%) n ¼ 16 8 (50.0%) n ¼ 34 26 (76.5%) 14 (41.2%) 12 (35.3%) n ¼ 39 31 (79.5%) 4 (10.3%) 27 (69.2%) n ¼ 73 57 (78.1%) 18 (24.7%) 39 (53.4%)
n ¼ 628 566 (90.1%) 203 (32.3%) 363 (57.8%) n ¼ 460 438 (95.2%) 75 (16.3%) 363 (78.9%) n ¼ 168 128 (76.2%) n ¼ 208 196 (94.2%) 54 (26.0%) 142 (68.3%) n ¼ 189 186 (98.4%) 15 (7.9%) 171 (90.5%) n ¼ 397 382 (96.2%) 69 (17.4%) 313 (78.8%)
p value <0.001 <0.001a
<0.001 <0.001a
0.023 <0.001 <0.001a
<0.001 <0.001a
<0.001 <0.001a
RVGE, rotavirus gastroenteritis. a p value for trend by c2 linear by linear association.
Discussion We report that immunization with RotaTeq was less frequent in patients with RVGE than in controls. The association remained significant after adjustment for age, year of and season of admission, year of birth, and SES of residence. The effectiveness of the complete vaccine series was 77% (95% CI: 49e90%) in children aged
57
6e59 months; 82% (95% CI: 36e95%) in children aged 6e11 months, and 96% (95% CI: 77e99%) in children aged 12e23 months; overall vaccine effectiveness was 86% (95% CI: 65e94%) for the two age groups (6e23 months) combined. We previously showed that children aged <2 years comprised the main risk group for RVGE hospitalization in the preimmunization period [23], and here we demonstrate that universal immunization with RotaTeq confers protection for this age group. Generally, a vaccine induces immunity to susceptible individuals, and subsequently reduces transmission of the infectious agent in the population, thus leading to decreased incidence of disease. Eventually, the impact of the vaccine is observed in the entire population and depends on vaccine effectiveness and coverage, and likely the induction of herd immunity. We have already demonstrated prominent reductions in RVGE hospitalizations and clinic visits in children aged <5 years in Israel following universal immunization with RotaTeq [12e14], and consequent reductions in costs to the healthcare system and families. The estimated effectiveness reported in this study is comparable with reports from the USA (85e87%) [28] and Australia (89%) [11]; whereas in Finland, where vaccine coverage exceeds 90%, effectiveness was estimated at 94% [17]. The relatively low effectiveness of RotaTeq reported for Bedouin children in Israel [21] warrants particular attention. We found that infants aged 2e5 months, with partial vaccination, had lower likelihood of RVGE hospitalization than their unvaccinated peers. Collectively, ours and other reports [15,19,21,28] show that rotavirus immunization confers protection against severe RVGE during the first 2 years of life. Diarrhoeal diseases in infancy are associated with diminished physical growth [29] and even with worse cognitive function [30] in children living in low-to-middle-income countries. Therefore, assessing whether rotavirus immunization
Table 5 Association between immunization with RotaTeq and rotavirus gastroenteritis hospitalizations Age/RotaTeq dose(s) Age 2e59 months 1 dose 1 or 2 doses 3 doses Age 6e59 months 1 dose 1 or 2 doses 3 doses Age 2e5 months 1 dose Age 6e11 months 1 dose 1 or 2 doses 3 doses Age 12e23 months 1 dose 1 or 2 doses 3 doses Age 6e23 months 1 dose 1 or 2 doses 3 doses
Unadjusted OR (95% CI)
Adjusted ORa (95% CI)
Vaccine effectivenessb (95% CI)
0.30 (0.18e0.51) 0.31 (0.17e0.56) 0.30 (0.17e0.52)
0.29 (0.16e0.53) 0.29 (0.14e0.59) 0.29 (0.15e0.57)
71% (47e84) 71% (41e86) 71% (43e85)
0.18 (0.09e0.35) 0.29 (0.13e0.66) 0.16 (0.08e0.31)
0.24 (0.11e0.52) 0.28 (0.11e0.72) 0.23 (0.10e0.51)
76% (48e89) 72% (28e89) 77% (49e90)
0.31 (0.11e0.89)
0.12 (0.03e0.57)
c
88% (43e97)
0.20 (0.07e0.53) 0.39 (0.13e1.13) 0.13 (0.04e0.37)
0.22 (0.07e0.70) 0.57 (0.14e2.40) 0.18 (0.05e0.64)
d
78% (30e93) 43% (e140 to 86) 82% (36e95)
0.06 (0.02e0.25) 0.10 (0.02e0.56) 0.06 (0.02e0.24)
0.04 (0.01e0.24) 0.05 (0.01e0.44) 0.04 (0.01e0.23)
f
0.14 (0.07e0.30) 0.25 (0.10e0.59) 0.12 (0.05e0.26)
0.16 (0.07e0.40) 0.25 (0.09e0.70) 0.14 (0.06e0.35)
e e
f f
96% (76e99) 95% (56e99) 96% (77e99) 84% (60e93) 75% (30e91) 86% (65e94)
OR, odds ratio; CI, confidence interval. a Adjusted for age, year of admission, season of admission, year of birth (2011e2014 vs 2015) and socioeconomic status (SES) of place of residence. b Vaccine effectiveness was calculated as (1 adjusted OR)*100. c Since there were only 16 rotavirus gastroenteritis cases aged 2e5 months, we included only two variables in the models: rotavirus immunization and age as a continuous variable. d The variables age and SES of place of residence were excluded from the final model since they were not statistically significant in the model (p ¼ 0.12 and p ¼ 0.6, respectively) due to the small number of cases (n ¼ 34); this did not much alter the results. e The variable SES of place of residence was excluded from the final model since it was not statistically significant in the model (p ¼ 0.5) and due to the small number of cases (n ¼ 34); this did not much alter the results. f The variable age was excluded from the final model since it was not statistically significant in the model (p ¼ 0.3) and due to the small number of cases (n ¼ 39); this did not much alter the results.
58
K. Muhsen et al. / Clinical Microbiology and Infection 24 (2018) 53e59
might carry gains in children's growth and development is a subject for future research. The point estimate of RotaTeq effectiveness against G1P[8]associated RVGE hospitalizations was higher than for other genotype-RVGE hospitalizations: 79% (95% CI: 45e92%) vs 69% (95% CI: 11e89%). Nonetheless, due to the small number of samples genotyped and the overlapping confidence intervals, differential effectiveness by genotype cannot be concluded. Rotavirus G1P[8] genotype was the most frequent strain circulating before and after the introduction of universal immunization with RotaTeq in Israel [14]. We also reported a rise in G3P[8] and G9P[8] between the two periods [14]. Though this trend may be unrelated to vaccine pressure, continued monitoring of genotype distribution in RVGE cases and genotype-specific effectiveness is needed. Our study has limitations. The relatively small sample of hospitalized children with RVGE yielded wide confidence intervals and precluded comparisons of RotaTeq effectiveness among age- and genotype-specific groups, and according to waning immunity. Although we aimed to test rotavirus in all children hospitalized with gastroenteritis, rotavirus was tested in only 65% of eligible patients. Rotavirus testing was more frequent in young children and in children with more severe illness. Hence, the generalizability of our findings might be limited to more severe RVGE cases. Data on SES were only available according to residence, which raises the possibility of residual confounding. Nonetheless, SES of residence was associated only with RVGE, not with vaccination status, thus implying that SES is a not a strong confounder in our population. This situation is likely due to high accessibility to vaccines in the national immunization programme, which are free of charge, coupled with a wide geographic spread of maternal and child health clinics. This is a strength of our study, in addition to the selection of cases and controls from the same source population, in the framework of a systematic surveillance during five years of universal immunization with RotaTeq, and the access of information on immunization status from a national registry. In summary, we demonstrated high effectiveness of RotaTeq as the sole rotavirus vaccine within a universal immunization programme in a high-income country. Although partial vaccination conferred significant protection, efforts to complete three vaccine doses are warranted to achieve maximum benefit to children's health and society.
Acknowledgements We would like to thank the research staff at the participating hospitals (M. Yunes, F. Abu Rakia, S. Primer, T. Zim, P. ShakedMishan, A. Rimer), at the Central Virology Laboratory (Ms I. Zilberstient) and at Tel Aviv University (Ms A. Bialik and Mr M. Brik).
Appendix A. Supplementary data Supplementary data related to this article can be found at http:// dx.doi.org/10.1016/j.cmi.2017.04.018.
Transparency declaration The work was supported by the World Health Organization, Department of Immunization, Vaccination and Biologicals (V27181-190) in the first year and by the Israel National Institute for Health Policy and Research (grant 2011/154) (DC-PI). The authors report no conflicts of interest.
References [1] Kotloff KL, Nataro JP, Blackwelder WC, Nasrin D, Farag TH, Panchalingam S, et al. Burden and aetiology of diarrhoeal disease in infants and young children in developing countries (the Global Enteric Multicenter Study, GEMS): a prospective, caseecontrol study. Lancet 2013;382:209e22. [2] Tate JE, Burton AH, Boschi-Pinto C, Parashar UD, World Health OrganizationCoordinated Global Rotavirus Surveillance Network. Global, regional, and national estimates of rotavirus mortality in children <5 years of age, 2000e2013. Clin Infect Dis 2016;62(Suppl. 2):S96e105. [3] Vesikari T, Matson DO, Dennehy P, Van Damme P, Santosham M, Rodriguez Z, et al. Safety and efficacy of a pentavalent humanebovine (wc3) reassortant rotavirus vaccine. N Engl J Med 2006;354:23e33. [4] Ruiz-Palacios GM, Perez-Schael I, Velazquez FR, Abate H, Breuer T, Clemens SC, et al. Safety and efficacy of an attenuated vaccine against severe rotavirus gastroenteritis. N Engl J Med 2006;354:11e22. [5] Armah GE, Sow SO, Breiman RF, Dallas MJ, Tapia MD, Feikin DR, et al. Efficacy of pentavalent rotavirus vaccine against severe rotavirus gastroenteritis in infants in developing countries in sub-Saharan Africa: a randomised, doubleblind, placebo-controlled trial. Lancet 2010;376:606e14. [6] Anonymous. Rotavirus vaccines. WHO position paper e January 2013. Wkly Epidemiol Rec 2013;88:49e64. [7] International Vaccine Access Center (IVAC). Vaccine information management system (VIMS) global vaccine introduction report, September 2016. Baltimore: Johns Hopkins Bloomberg School of Public Health; 2016. Available at: http:// www.jhsph.edu/research/centers-and-institutes/ivac/vims/ [accessed 22 December 2016]. [8] Sabbe M, Berger N, Blommaert A, Ogunjimi B, Grammens T, Callens M, et al. Sustained low rotavirus activity and hospitalisation rates in the postvaccination era in Belgium, 2007 to 2014. Euro Surveill 2016;21. [9] Paulke-Korinek M, Kollaritsch H, Aberle SW, Zwazl I, Schmidle-Loss B, csei A, et al. Sustained low hospitalization rates after four years of rotavirus Ve mass vaccination in Austria. Vaccine 2013;31:2686e91. [10] Tate JE, Mutuc JD, Panozzo CA, Payne DC, Cortese MM, Cortes JE, et al. Sustained decline in rotavirus detections in the United States following the introduction of rotavirus vaccine in 2006. Pediatr Infect Dis J 2011;30: S30e4. [11] Buttery JP, Lambert SB, Grimwood K, Nissen MD, Field EJ, Macartney KK, et al. Reduction in rotavirus-associated acute gastroenteritis following introduction of rotavirus vaccine into Australia’s national childhood vaccine schedule. Pediatr Infect Dis J 2011;30:S25e9. [12] Muhsen K, Chodick G, Goren S, Anis E, Ziv-Baran T, Shalev V, et al. Change in incidence of clinic visits for all-cause and rotavirus gastroenteritis in young children following the introduction of universal rotavirus vaccination in Israel. Euro Surveill 2015;20. [13] Muhsen K, Rubenstein U, Kassem E, Goren S, Schachter Y, Kremer A, et al. A significant and consistent reduction in rotavirus gastroenteritis hospitalization of children under 5 years of age, following the introduction of universal rotavirus immunization in Israel. Hum Vaccin Immunother 2015;11:2475e82. [14] Muhsen K, Kassem E, Rubenstein U, Goren S, Ephros M, Cohen D, et al. Incidence of rotavirus gastroenteritis hospitalizations and genotypes, before and five years after introducing universal immunization in Israel. Vaccine 2016;34:5916e22. [15] Bar-Zeev N, Jere KC, Bennett A, Pollock L, Tate JE, Nakagomi O, et al. Population impact and effectiveness of monovalent rotavirus vaccination in urban Malawian children 3 years after vaccine introduction: ecological and caseecontrol analyses. Clin Infect Dis 2016;62(Suppl. 2):S213e9. pez B, Benoit SR, Xuya M, [16] Gastanaduy PA, Contreras-Roldan I, Bernart C, Lo et al. Effectiveness of monovalent and pentavalent rotavirus vaccines in Guatemala. Clin Infect Dis 2016;62(Suppl. 2):S121e6. [17] Hemming-Harlo M, Vesikari T, Uhari M, Renko M, Salminen M, TorcelPagnon L, et al. Sustained high effectiveness of RotaTeq on hospitalizations attributable to rotavirus-associated gastroenteritis during 4 years in Finland. J Pediatr Infect Dis Soc 2016 Oct 19. M, Heylen E, [18] Braeckman T, Van Herck K, Meyer N, Pirçon JY, Soriano-Gabarro et al. Effectiveness of rotavirus vaccination in prevention of hospital admissions for rotavirus gastroenteritis among young children in Belgium: caseecontrol study. BMJ 2012;345. [19] Tate JE, Ngabo F, Donnen P, Gatera M, Uwimana J, Rugambwa C, et al. Effectiveness of pentavalent rotavirus vaccine under conditions of routine use in Rwanda. Clin Infect Dis 2016;62:S208e12. [20] Tate JE, Patel MM, Cortese MM, Payne DC, Lopman BA, Yen C, et al. Use of patients with diarrhea who test negative for rotavirus as controls to estimate rotavirus vaccine effectiveness through caseecontrol studies. Clin Infect Dis 2016;62:S106e14. [21] Leshem E, Givon-Lavi N, Tate JE, Greenberg D, Parashar UD, Dagan R. Realworld effectiveness of pentavalent rotavirus vaccine among Bedouin and Jewish children in southern Israel. Clin Infect Dis 2016;62:S155e60. [22] World Health Organization, Vaccines and Biologicals. Generic protocols for (i) hospital-based surveillance to estimate the burden of rotavirus gastroenteritis in children and (ii) a community-based survey on utilization of health care services for gastroenteritis in children. Geneva: WHO; 2002. [23] Muhsen K, Shulman L, Rubinstein U, Kasem E, Kremer A, Goren S, et al. Incidence, characteristics, and economic burden of rotavirus gastroenteritis
K. Muhsen et al. / Clinical Microbiology and Infection 24 (2018) 53e59 associated with hospitalization of Israeli children <5 years of age, 2007e2008. J Infect Dis 2009;200(Suppl. 1):S254e63. [24] Muhsen K, Shulman L, Kasem E, Rubinstein U, Shachter J, Kremer A, et al. Effectiveness of rotavirus vaccines for prevention of rotavirus gastroenteritisassociated hospitalizations in Israel: a caseecontrol study. Hum Vaccin 2010;6. [25] Baqui AH, Black RE, Yunus M, Hoque AR, Chowdhury HR, Sack RB. Methodological issues in diarrhoeal diseases epidemiology: definition of diarrhoeal episodes. Int J Epidemiol 1991;20:1057e63. [26] Government of Israel, Central Bureau of Statistics. Characterization and classification of geographic units by the socio-economic level of the population 2008. 2013. Available at: http://www.cbs.gov.il/webpub/pub/text_page_eng. html?publ¼100&CYear¼2008&CMonth¼1 [accessed 2 June 2017].
59
[27] Weinberg GA, Szilagyi PG. Vaccine epidemiology: efficacy, effectiveness, and the translational research roadmap. J Infect Dis 2010;201:1607e10. [28] Staat MA, Payne DC, Donauer S, Weinberg GA, Edwards KM, Szilagyi PG, et al. Effectiveness of pentavalent rotavirus vaccine against severe disease. Pediatrics 2011;128:E267e75. [29] Mondal D, Minak J, Alam M, Liu Y, Dai J, Korpe P, et al. Contribution of enteric infection, altered intestinal barrier function, and maternal malnutrition to infant malnutrition in Bangladesh. Clin Infect Dis 2012;54:185e92. [30] Berkman DS, Lescano AG, Gilman RH, Lopez SL, Black MM. Effects of stunting, diarrhoeal disease, and parasitic infection during infancy on cognition in late childhood: a follow-up study. Lancet 2002;359:564e71.
Contents lists available at ScienceDirect
Clinical Microbiology and Infection journal homepage: www.clinicalmicrobiologyandinfection.com
Original article
Effectiveness of rotavirus pentavalent vaccine under a universal immunization programme in Israel, 2011e2015: a caseecontrol study K. Muhsen 1, *, E. Anis 2, U. Rubinstein 3, E. Kassem 4, S. Goren 1, L.M. Shulman 1, 5, M. Ephros 6, 7, D. Cohen 1 1)
Department of Epidemiology and Preventive Medicine, School of Public Health, Sackler Faculty of Medicine, Tel Aviv University Ramat Aviv, Tel Aviv, Israel Division of Epidemiology, Ministry of Health, Jerusalem, Israel Department of Pediatrics, Laniado Medical Center, Netanya, Israel 4) Department of Pediatrics, Hillel Yaffe Medical Center, Hadera, Israel 5) Central Virology Laboratory, Ministry of Health, Tel Hashomer, Israel 6) Department of Pediatrics, Carmel Medical Center, Haifa, Israel 7) Faculty of Medicine, TechnioneIsrael Institute of Technology, Haifa, Israel 2) 3)
a r t i c l e i n f o
a b s t r a c t
Article history: Received 19 January 2017 Received in revised form 14 April 2017 Accepted 18 April 2017 Available online 22 April 2017
Objectives: The use of rotavirus pentavalent vaccine (RotaTeq®) as a sole vaccine within rotavirus universal immunization programmes remains limited. We examined the effectiveness of RotaTeq in preventing rotavirus gastroenteritis (RVGE) hospitalization in Israel, after the introduction of universal immunization against the disease. Methods: A test-negative caseecontrol study included age-eligible children for universal RotaTeq immunization (aged 2e59 months, born in 2011e2015). Cases (n ¼ 98) were patients who tested positive for rotavirus by immunochromatography; those who tested negative (n ¼ 628) comprised the control group. Information on rotavirus immunization history was obtained through linkage with a national immunization registry. Vaccination status was compared between cases and controls, adjusted odds ratios (aORs) were obtained from logistic regression models, and vaccine effectiveness calculated as (1 aOR)*100. Results: Immunization with RotaTeq was less frequent in RVGE cases (73.5%) than in controls (90.1%), p < 0.001; this association persisted after controlling for potential confounders. Effectiveness of the complete vaccine series was estimated at 77% (95% confidence interval (CI): 49e90) in children aged 6 e59 months, and 86% (95% CI: 65e94) in children aged 6e23 months; whereas for the incomplete series, the respective estimates were 72% (95% CI: 28e89) and 75% (95% CI: 30e91). Vaccine effectiveness was estimated at 79% (95% CI: 45e92) against G1P[8]-associated RVGE hospitalizations and 69% (95% CI: 11 e89) against other genotype-RVGE hospitalizations. Conclusions: High effectiveness of RotaTeq as the sole rotavirus vaccine in a universal immunization programme was demonstrated in a high-income country. Although partial vaccination conferred protection, completing the vaccine series is warranted to maximize the benefit. K. Muhsen, Clin Microbiol Infect 2018;24:53 © 2017 European Society of Clinical Microbiology and Infectious Diseases. Published by Elsevier Ltd. All rights reserved.
Editor: Dr C. Pulcini Keywords: Rotavirus Vaccine Gastroenteritis Effectiveness Post-marketing
Introduction
* Corresponding author. Khitam Muhsen, Department of Epidemiology and Preventive Medicine, School of Public Health, Sackler Faculty of Medicine, Tel Aviv University, Ramat Aviv, 6139001 Tel Aviv, Israel. E-mail address: [email protected] (K. Muhsen).
Rotavirus is the main cause of severe diarrhoea in children [1] and a leading cause of child mortality in developing countries [2]. Two rotavirus vaccines became available globally in recent years, both administered orally [3,4]. One is a monovalent vaccine containing attenuated human G1P[8] rotavirus strain (Rotarix®, GlaxoSmithKline Biologicals, Brentford, UK), administered in two doses
http://dx.doi.org/10.1016/j.cmi.2017.04.018 1198-743X/© 2017 European Society of Clinical Microbiology and Infectious Diseases. Published by Elsevier Ltd. All rights reserved.
54
K. Muhsen et al. / Clinical Microbiology and Infection 24 (2018) 53e59
at ages 2 and 4 months [4]. The other is a pentavalent live bovineehuman reassortant vaccine (RotaTeq®, Merck and Co., Inc., Whitehouse Station, NJ, USA) containing strains representing G1, G2, G3, G4 and P[8] genotypes, and administered in three doses, at ages 2, 4 and 6 months [3]. Although different disease severity scoring systems were used, both vaccines showed high efficacy (85e95%) in preventing severe rotavirus gastroenteritis (RVGE) in clinical trials conducted in Europe and the Americas [3,4]. Rotavirus vaccine efficacy was low (40e61%) in developing countries [5]. The World Health Organization (WHO) recommends including rotavirus vaccine in national immunization programmes worldwide [6]. By September 2016, a decade after rotavirus vaccines became available, 81 countries had introduced rotavirus immunization into national immunization programmes and seven countries had introduced rotavirus immunization in certain regions only [7]. Presently, most countries use Rotarix [7]. Significant and sustained reduction in the burden of RVGE was observed in countries that introduced rotavirus immunization, including Israel [8e14]. These studies employed mostly ecological design, comparing the incidence of RVGE in the entire target population, both vaccinated and unvaccinated groups, before and after the introduction of universal rotavirus immunization. The impact of universal rotavirus immunization on disease burden in the population is a function of vaccine coverage, vaccine effectiveness and possibly herd immunity. Numerous postmarking evaluations have assessed rotavirus vaccine effectiveness, mostly using a testnegative caseecontrol study design [15e19]. In this efficient and cost-saving design, rotavirus immunization history is compared between RVGE cases and control diarrhoea patients who test negative for rotavirus [20]. Due to the limited introduction of rotavirus vaccines in high-income countries, and since most of them introduced Rotarix, updated evidence of the net effectiveness of RotaTeq as a sole vaccine under universal vaccination policy in high-income settings, outside the USA, remains scarce [17,21]. In Israel, both rotavirus vaccines were licensed and became available by the end of 2007 in the private market, with partial refund through the health maintenance organizations. In December 2010, RotaTeq was introduced in the national immunization programme, free of charge for all children, and since that time it has been the only vaccine used. An Israeli study that comprised 70% Bedouin children reported 63% vaccine effectiveness for RotaTeq [21]. Bedouins are a small sub-group of the Arab minority in Israel, and typically live in low socioeconomic and suboptimal environmental conditions, thus generalizability from that study remains limited. We examined the effectiveness of RotaTeq in a universal immunization programme, in preventing RVGE hospitalization during 2011e2015 in Israel, as a model of a high-income country.
RotaTeq is administered at ages 2, 4 and 6 months; the estimated vaccine coverage is >80%. The sampling frame included children who were hospitalized due to diarrhoea (three or more watery stools during a 24 h period) [13,14,23e25]. During the study period, research staff in each hospital were instructed to systematically collect stool samples from all children hospitalized for diarrhoea and to submit the samples for rotavirus antigen testing at the local laboratory in each hospital. The current test-negative caseecontrol study was restricted to birth cohorts of the years 2011e2015 who were eligible for RotaTeq under universal immunization and hospitalized during 2011e2015. We set a lowest age limit at admission as 2 months, since this is the recommended age for the first RotaTeq dose. Children who were not tested for rotavirus or without information on rotavirus immunization were excluded from the analysis. Parents were interviewed concerning demographic and clinical information on the number of stools passed, vomiting and fever. Information on rotavirus vaccination history was obtained through linkage with the national immunization registry at the Ministry of Health. If parents expressed refusal to vaccinate their child, that information would be documented in the child's medical record, and was collected through the linkage with the immunization registry. Information on immunization was collected for children hospitalized until October 2015. Vaccination status was determined up to the admission date. Children were classified as unvaccinated or vaccinated with at least one dose, and according to the number of doses. Socioeconomic status (SES) was classified on a scale of 1e10, according to the socioeconomic rank of the town of residence, as defined by the Israel Central Bureau of Statistics [26]. The ranks were grouped into three categories of SES: 1e3 (low), 4e5 (intermediate) and 6e10 (high). Laboratory methods Stool samples were collected from patients within the first 48 h of hospitalization. Specimens were tested for rotavirus antigen by immunochromatography (Rotavirus Dipsticks, Hylabs Rehovot & Novamed, Jerusalem, Israel), according to manufacturers' instructions. In practice, 87% of the analysed stools were tested within 2 days of admission, 95% within 3 days. Cases were children with a confirmed rotavirus-positive test in their stool, and controls were patients who tested negative for rotavirus. After performing the rotavirus test, the remaining faecal material was transferred to the Central Virology Laboratory in cool conditions. Specimens were kept at 70 C until genotyping was performed [14].
Methods
Statistical methods
Study population and design
c2-Test and Fisher's exact test, as appropriate, were used to examine differences between vaccinated and unvaccinated children in sociodemographic characteristics, season/year of admission and birth year. Similarly, differences between RVGE cases and testnegative controls were examined. Logistic regression models were fitted to obtain adjusted odds ratio (aOR) and 95% confidence interval (CI) for immunization status before admission. The models were adjusted for vaccination status, age, year and season of admission and SES. Vaccine effectiveness was calculated as (1 aOR), and expressed as a percentage [27]. The primary analyses were limited to children 6e59 months of age, assuming that those children were all eligible for the full vaccine series. Additional analyses were performed with
A hospital-based active surveillance study was conducted between November 2007 and February 2016 in northern Israel, following a generic protocol of the WHO [22]. Details of the study design have been published elsewhere [13,14,23,24]. The study included children 0e59 months of age residing in the catchment area of three hospitals: Hillel Yaffe in Hadera, Carmel in Haifa, and Laniado in Netanya. Both Jewish and Arab residents live in the study area. According to the Universal National Health Insurance Law, all Israelis have access to medical care. The national immunization programme is conducted at maternal and child health clinics.
K. Muhsen et al. / Clinical Microbiology and Infection 24 (2018) 53e59
stratification by age group, including an exploratory analysis limited to infants two to five months of age, who were eligible to receive at least one dose. Data were analysed using SPSS version 23. Ethical considerations The study protocol was approved by the Institutional Review Boards of all participating hospitals and by the Ministry of Health. Parents of the participating children signed informed consent forms. Results Overall, 1437 of the children hospitalized in the participating hospitals with gastroenteritis during the study period were born during 2011e2015; of these, 1272 were aged 2e59 months and eligible for rotavirus immunization in the framework of the universal vaccination programme. Rotavirus was tested among 831 (65.3%) children (Supplementary Fig. 1). Compared to children not tested for rotavirus, a greater proportion of children tested passed 6 stools/day and vomited: 56.7% vs 38.7% (p < 0.001) and 78.2% vs 72.6% (p ¼ 0.024), respectively. Among both groups, 51.3% had fever (38 C) at admission. Among children tested for rotavirus, data on rotavirus immunization status was available for 726. Rotavirus was detected among 98 (13.5%), and the rest tested negative and served as controls. Children with missing data for rotavirus immunization status and rotavirus stool test included higher proportions of toddlers aged 24e59 months, children from high-SES communities and children admitted in 2014e2015, compared to children with complete information (Table 1). Among children aged 2e59 months, 88 (12.1%) did not receive any dose of RotaTeq; for 26 (29.5%) of them this was due to parental refusal. The proportion of immunized children with at least one dose differed by age, being highest at ages 6e23 months. This percentage was lower in patients born in 2015 than during 2011e2014 (65.2% vs 88.6%, p ¼ 0.004), and in children admitted in 2011 than in those admitted during 2012e2015 (Table 2). The proportion of vaccinated children did not differ according to hospital of admission (p ¼ 0.8). RVGE cases were older, and included more Jewish children and residents of intermediate- to high-SES towns than did controls.
55
Cases were admitted significantly more often during the winter, the typical rotavirus season in Israel; this in addition to differences between cases and controls in the year of admission (Table 3). Cases and controls did not differ according to hospital of admission (p ¼ 0.3). For all age groups, the proportion of children immunized with at least one RotaTeq dose, as well as with the complete vaccine series, was lower among RVGE patients than controls (Table 4). In a multivariate analysis, immunization with RotaTeq conferred significant protection against RVGE hospitalization in children aged 6e59 months, with similar effectiveness for any dose (76%; 95% CI: 48e89%) and for the complete series (77%; 95% CI: 49e90%) (Table 5). A lower point estimate of vaccine effectiveness of the complete RotaTeq series was found in children aged 6e11 months (82%; 95% CI: 36e95%) vs children aged 12e23 months (96%; 95% CI: 77e99%), although CIs were overlapping. The effectiveness for both age groups combined (6e23 months) was 86% (95%; CI: 65e94%) (Table 5). Socioeconomic status of residence and population group were highly correlated (lower SES in Arab patients: Spearman's r ¼ 0.77; p < 0.001); therefore, the two variables were not included in the same model. Similar results were obtained from models that included the variable population group (Supplementary material 2). The number of children aged 24e59 months was small to conduct a separate analysis. In the 2e5-month age group, the proportion of immunized RVGE cases was lower than among controls (Table 4), and RotaTeq conferred protection in this age group also (Table 5). Faecal samples were tested for rotavirus genotype in 41/98 (41.8%) of RVGE patients. Among these, 20 (48.8%), eight (19.5%), seven (17.1%), two (4.9%) and one were G1P[8], G3P[8], G9P[8], G2P [4] and G12P[8], respectively. In three additional samples, either the G- or P-type was non-typeable. Among 20 RVGE cases of G1P[8] genotype, 13 (65.0%) were immunized with RotaTeq compared to 90.1% in control children (age-adjusted OR: 0.21; 95% CI: 0.08e0.55; p ¼ 0.002), whereas in cases with other genotypes, 16/ 21 (76.2%) were immunized (age-adjusted OR: 0.31; 95% CI: 0.11e0.89; p ¼ 0.029). This yields vaccine effectiveness of 79% (95% CI: 45e92%) and 69% (95% CI: 11e89%) against hospitalizations for RVGE caused by these genotypes, respectively.
Table 1 Characteristics of children with available data on rotavirus and vaccination status, compared to those with missing data on these variables Characteristics Age (months) 2e5 6e11 12e23 24e59 Sex, males Population group: Jews Year of admission 2011 2012 2013 2014 2015 Season of admission DeceFeb (Winter) MareMay (Spring) JuneAug (Summer) SepeNov (Autumn) SES of place of residence Low: 1e3 Intermediate: 4e5 High: 6e10 SES, socioeconomic status.
Available data (n ¼ 726)
Missing data (n ¼ 546)
184 (25.3%) 242 (33.3%) 228 (31.4%) 72 (9.9%) 386 (53.2%) 459 (63.2%)
94 (17.2%) 173 (31.7%) 189 (34.6%) 90 (16.5%) 302 (55.3%) 329 (60.6%)
39 (5.4%) 161 (22.2%) 165 (22.7%) 235 (32.4%) 126 (17.4%)
13 (2.4%) 105 (19.2%) 111 (20.3%) 197 (36.1%) 120 (22.0%)
194 169 165 198
130 125 141 150
p value <0.001
0.4 0.3 0.009
0.17 (26.7%) (23.3%) (22.7%) (27.3%)
227 (31.5%) 378 (52.4%) 116 (16.1%)
(23.8%) (22.9%) (25.8%) (27.5%)
167 (31.3%) 249 (46.6%) 118 (22.1%)
0.018
56
K. Muhsen et al. / Clinical Microbiology and Infection 24 (2018) 53e59
Table 2 Vaccination status of children included in the analysis, according to various characteristics Characteristics Age (months) 2e5 6e11 12e23 24e59 Sex Males Females Population group Jews Arabs SES of place of residence Low: 1e3 Intermediate: 4e5 High: 6e10 Birth year 2011 2012 2013 2014 2015 Year of admission 2011 2012 2013 2014 2015
Total
Vaccinated with at least one RotaTeq dose
184 242 228 72
136 (73.9%) 222 (91.7%) 217 (95.2%) 63 (87.5%)
386 340
338 (87.6%) 300 (88.2%)
459 267
407 (88.7%) 231 (86.5%)
227 378 116
198 (87.2%) 337 (89.2%) 100 (86.2%)
218 203 171 111 23
191 (87.6%) 181 (89.2%) 158 (92.4%) 93 (83.8%) 15 (65.2%)
39 161 165 235 126
29 (74.4%) 145 (90.1%) 146 (88.5%) 213 (90.6%) 105 (83.3%)
p value <0.001a
0.7a
0.3a
0.6a
0.002a 0.10b
0.021a 0.8b
SES, socioeconomic status. a c2-Test. b p for trend: by c2 linear by linear association.
Table 3 Rotavirus gastroenteritis cases and rotavirus test-negative controls by sociodemographic factors Factors Age (months) 2e5 6e11 12e23 24e59 Sex (males) Population group: Jews SES of place of residence Low: 1e3 Intermediate: 4e5 High: 6e10 Birth year 2011 2012 2013 2014 2015 Year of admission 2011 2012 2013 2014 2015 Month/season of admission DeceFeb (Winter) MareMay (Spring) JuneAug (Summer) SepeNov (Autumn)
RVGE cases (N ¼ 98)
Rotavirus-test negative controls (N ¼ 628)
16 (16.3%) 34 (34.7%) 39 (39.8%) 9 (9.2%) 54 (55.1%) 74 (75.5%)
168 (26.8%) 208 (33.1%) 189 (30.1%) 63 (10.0%) 332 (52.9%) 385 (61.3%)
19 (19.6%) 57 (58.8%) 21 (21.6%)
208 (33.3%) 321 (51.4%) 95 (15.2%)
29 (29.6%) 32 (32.7%) 21 (21.4%) 14 (14.3%) 2 (2.0%)
189 (30.1%) 171 (27.2%) 150 (23.9%) 97 (15.4%) 21 (3.3%)
7 (7.1%) 10 (10.2%) 36 (36.7%) 21 (21.4%) 24 (24.5%)
32 (5.9%) 151 (24.0%) 129 (20.5%) 214 (34.1%) 102 (16.2%)
59 (60.2%) 22 (22.4%) 6 (6.1%) 11 (11.2%)
135 147 159 187
p value 0.098a 0.07b
0.6 0.007 0.018a 0.006b
0.8
<0.001a 0.2b
<0.001
RVGE, rotavirus gastroenteritis; SES, socioeconomic status. a c2-Test. b p value for trend by c2 linear by linear association.
(21.5%) (23.4%) (25.3%) (29.8%)
K. Muhsen et al. / Clinical Microbiology and Infection 24 (2018) 53e59 Table 4 Differences between RVGE cases and rotavirus test-negative controls in immunization status Age/RotaTeq dose(s)
RVGE cases
Rotavirus-test negative controls
Age 2e59 months 1 dose 1 or 2 doses 3 doses Age 6e59 months 1 dose 1 or 2 doses 3 doses Age 2e5 months 1 dose Age 6e11 months 1 dose 1 or 2 doses 3 doses Age 12e23 months 1 dose 1 or 2 doses 3 doses Age 6e23 months 1 dose 1 or 2 doses 3 doses
n ¼ 98 72 (73.5%) 26 (26.5%) 46 (46.9%) n ¼ 82 64 (78.0%) 18 (22.0%) 46 (56.1%) n ¼ 16 8 (50.0%) n ¼ 34 26 (76.5%) 14 (41.2%) 12 (35.3%) n ¼ 39 31 (79.5%) 4 (10.3%) 27 (69.2%) n ¼ 73 57 (78.1%) 18 (24.7%) 39 (53.4%)
n ¼ 628 566 (90.1%) 203 (32.3%) 363 (57.8%) n ¼ 460 438 (95.2%) 75 (16.3%) 363 (78.9%) n ¼ 168 128 (76.2%) n ¼ 208 196 (94.2%) 54 (26.0%) 142 (68.3%) n ¼ 189 186 (98.4%) 15 (7.9%) 171 (90.5%) n ¼ 397 382 (96.2%) 69 (17.4%) 313 (78.8%)
p value <0.001 <0.001a
<0.001 <0.001a
0.023 <0.001 <0.001a
<0.001 <0.001a
<0.001 <0.001a
RVGE, rotavirus gastroenteritis. a p value for trend by c2 linear by linear association.
Discussion We report that immunization with RotaTeq was less frequent in patients with RVGE than in controls. The association remained significant after adjustment for age, year of and season of admission, year of birth, and SES of residence. The effectiveness of the complete vaccine series was 77% (95% CI: 49e90%) in children aged
57
6e59 months; 82% (95% CI: 36e95%) in children aged 6e11 months, and 96% (95% CI: 77e99%) in children aged 12e23 months; overall vaccine effectiveness was 86% (95% CI: 65e94%) for the two age groups (6e23 months) combined. We previously showed that children aged <2 years comprised the main risk group for RVGE hospitalization in the preimmunization period [23], and here we demonstrate that universal immunization with RotaTeq confers protection for this age group. Generally, a vaccine induces immunity to susceptible individuals, and subsequently reduces transmission of the infectious agent in the population, thus leading to decreased incidence of disease. Eventually, the impact of the vaccine is observed in the entire population and depends on vaccine effectiveness and coverage, and likely the induction of herd immunity. We have already demonstrated prominent reductions in RVGE hospitalizations and clinic visits in children aged <5 years in Israel following universal immunization with RotaTeq [12e14], and consequent reductions in costs to the healthcare system and families. The estimated effectiveness reported in this study is comparable with reports from the USA (85e87%) [28] and Australia (89%) [11]; whereas in Finland, where vaccine coverage exceeds 90%, effectiveness was estimated at 94% [17]. The relatively low effectiveness of RotaTeq reported for Bedouin children in Israel [21] warrants particular attention. We found that infants aged 2e5 months, with partial vaccination, had lower likelihood of RVGE hospitalization than their unvaccinated peers. Collectively, ours and other reports [15,19,21,28] show that rotavirus immunization confers protection against severe RVGE during the first 2 years of life. Diarrhoeal diseases in infancy are associated with diminished physical growth [29] and even with worse cognitive function [30] in children living in low-to-middle-income countries. Therefore, assessing whether rotavirus immunization
Table 5 Association between immunization with RotaTeq and rotavirus gastroenteritis hospitalizations Age/RotaTeq dose(s) Age 2e59 months 1 dose 1 or 2 doses 3 doses Age 6e59 months 1 dose 1 or 2 doses 3 doses Age 2e5 months 1 dose Age 6e11 months 1 dose 1 or 2 doses 3 doses Age 12e23 months 1 dose 1 or 2 doses 3 doses Age 6e23 months 1 dose 1 or 2 doses 3 doses
Unadjusted OR (95% CI)
Adjusted ORa (95% CI)
Vaccine effectivenessb (95% CI)
0.30 (0.18e0.51) 0.31 (0.17e0.56) 0.30 (0.17e0.52)
0.29 (0.16e0.53) 0.29 (0.14e0.59) 0.29 (0.15e0.57)
71% (47e84) 71% (41e86) 71% (43e85)
0.18 (0.09e0.35) 0.29 (0.13e0.66) 0.16 (0.08e0.31)
0.24 (0.11e0.52) 0.28 (0.11e0.72) 0.23 (0.10e0.51)
76% (48e89) 72% (28e89) 77% (49e90)
0.31 (0.11e0.89)
0.12 (0.03e0.57)
c
88% (43e97)
0.20 (0.07e0.53) 0.39 (0.13e1.13) 0.13 (0.04e0.37)
0.22 (0.07e0.70) 0.57 (0.14e2.40) 0.18 (0.05e0.64)
d
78% (30e93) 43% (e140 to 86) 82% (36e95)
0.06 (0.02e0.25) 0.10 (0.02e0.56) 0.06 (0.02e0.24)
0.04 (0.01e0.24) 0.05 (0.01e0.44) 0.04 (0.01e0.23)
f
0.14 (0.07e0.30) 0.25 (0.10e0.59) 0.12 (0.05e0.26)
0.16 (0.07e0.40) 0.25 (0.09e0.70) 0.14 (0.06e0.35)
e e
f f
96% (76e99) 95% (56e99) 96% (77e99) 84% (60e93) 75% (30e91) 86% (65e94)
OR, odds ratio; CI, confidence interval. a Adjusted for age, year of admission, season of admission, year of birth (2011e2014 vs 2015) and socioeconomic status (SES) of place of residence. b Vaccine effectiveness was calculated as (1 adjusted OR)*100. c Since there were only 16 rotavirus gastroenteritis cases aged 2e5 months, we included only two variables in the models: rotavirus immunization and age as a continuous variable. d The variables age and SES of place of residence were excluded from the final model since they were not statistically significant in the model (p ¼ 0.12 and p ¼ 0.6, respectively) due to the small number of cases (n ¼ 34); this did not much alter the results. e The variable SES of place of residence was excluded from the final model since it was not statistically significant in the model (p ¼ 0.5) and due to the small number of cases (n ¼ 34); this did not much alter the results. f The variable age was excluded from the final model since it was not statistically significant in the model (p ¼ 0.3) and due to the small number of cases (n ¼ 39); this did not much alter the results.
58
K. Muhsen et al. / Clinical Microbiology and Infection 24 (2018) 53e59
might carry gains in children's growth and development is a subject for future research. The point estimate of RotaTeq effectiveness against G1P[8]associated RVGE hospitalizations was higher than for other genotype-RVGE hospitalizations: 79% (95% CI: 45e92%) vs 69% (95% CI: 11e89%). Nonetheless, due to the small number of samples genotyped and the overlapping confidence intervals, differential effectiveness by genotype cannot be concluded. Rotavirus G1P[8] genotype was the most frequent strain circulating before and after the introduction of universal immunization with RotaTeq in Israel [14]. We also reported a rise in G3P[8] and G9P[8] between the two periods [14]. Though this trend may be unrelated to vaccine pressure, continued monitoring of genotype distribution in RVGE cases and genotype-specific effectiveness is needed. Our study has limitations. The relatively small sample of hospitalized children with RVGE yielded wide confidence intervals and precluded comparisons of RotaTeq effectiveness among age- and genotype-specific groups, and according to waning immunity. Although we aimed to test rotavirus in all children hospitalized with gastroenteritis, rotavirus was tested in only 65% of eligible patients. Rotavirus testing was more frequent in young children and in children with more severe illness. Hence, the generalizability of our findings might be limited to more severe RVGE cases. Data on SES were only available according to residence, which raises the possibility of residual confounding. Nonetheless, SES of residence was associated only with RVGE, not with vaccination status, thus implying that SES is a not a strong confounder in our population. This situation is likely due to high accessibility to vaccines in the national immunization programme, which are free of charge, coupled with a wide geographic spread of maternal and child health clinics. This is a strength of our study, in addition to the selection of cases and controls from the same source population, in the framework of a systematic surveillance during five years of universal immunization with RotaTeq, and the access of information on immunization status from a national registry. In summary, we demonstrated high effectiveness of RotaTeq as the sole rotavirus vaccine within a universal immunization programme in a high-income country. Although partial vaccination conferred significant protection, efforts to complete three vaccine doses are warranted to achieve maximum benefit to children's health and society.
Acknowledgements We would like to thank the research staff at the participating hospitals (M. Yunes, F. Abu Rakia, S. Primer, T. Zim, P. ShakedMishan, A. Rimer), at the Central Virology Laboratory (Ms I. Zilberstient) and at Tel Aviv University (Ms A. Bialik and Mr M. Brik).
Appendix A. Supplementary data Supplementary data related to this article can be found at http:// dx.doi.org/10.1016/j.cmi.2017.04.018.
Transparency declaration The work was supported by the World Health Organization, Department of Immunization, Vaccination and Biologicals (V27181-190) in the first year and by the Israel National Institute for Health Policy and Research (grant 2011/154) (DC-PI). The authors report no conflicts of interest.
References [1] Kotloff KL, Nataro JP, Blackwelder WC, Nasrin D, Farag TH, Panchalingam S, et al. Burden and aetiology of diarrhoeal disease in infants and young children in developing countries (the Global Enteric Multicenter Study, GEMS): a prospective, caseecontrol study. Lancet 2013;382:209e22. [2] Tate JE, Burton AH, Boschi-Pinto C, Parashar UD, World Health OrganizationCoordinated Global Rotavirus Surveillance Network. Global, regional, and national estimates of rotavirus mortality in children <5 years of age, 2000e2013. Clin Infect Dis 2016;62(Suppl. 2):S96e105. [3] Vesikari T, Matson DO, Dennehy P, Van Damme P, Santosham M, Rodriguez Z, et al. Safety and efficacy of a pentavalent humanebovine (wc3) reassortant rotavirus vaccine. N Engl J Med 2006;354:23e33. [4] Ruiz-Palacios GM, Perez-Schael I, Velazquez FR, Abate H, Breuer T, Clemens SC, et al. Safety and efficacy of an attenuated vaccine against severe rotavirus gastroenteritis. N Engl J Med 2006;354:11e22. [5] Armah GE, Sow SO, Breiman RF, Dallas MJ, Tapia MD, Feikin DR, et al. Efficacy of pentavalent rotavirus vaccine against severe rotavirus gastroenteritis in infants in developing countries in sub-Saharan Africa: a randomised, doubleblind, placebo-controlled trial. Lancet 2010;376:606e14. [6] Anonymous. Rotavirus vaccines. WHO position paper e January 2013. Wkly Epidemiol Rec 2013;88:49e64. [7] International Vaccine Access Center (IVAC). Vaccine information management system (VIMS) global vaccine introduction report, September 2016. Baltimore: Johns Hopkins Bloomberg School of Public Health; 2016. Available at: http:// www.jhsph.edu/research/centers-and-institutes/ivac/vims/ [accessed 22 December 2016]. [8] Sabbe M, Berger N, Blommaert A, Ogunjimi B, Grammens T, Callens M, et al. Sustained low rotavirus activity and hospitalisation rates in the postvaccination era in Belgium, 2007 to 2014. Euro Surveill 2016;21. [9] Paulke-Korinek M, Kollaritsch H, Aberle SW, Zwazl I, Schmidle-Loss B, csei A, et al. Sustained low hospitalization rates after four years of rotavirus Ve mass vaccination in Austria. Vaccine 2013;31:2686e91. [10] Tate JE, Mutuc JD, Panozzo CA, Payne DC, Cortese MM, Cortes JE, et al. Sustained decline in rotavirus detections in the United States following the introduction of rotavirus vaccine in 2006. Pediatr Infect Dis J 2011;30: S30e4. [11] Buttery JP, Lambert SB, Grimwood K, Nissen MD, Field EJ, Macartney KK, et al. Reduction in rotavirus-associated acute gastroenteritis following introduction of rotavirus vaccine into Australia’s national childhood vaccine schedule. Pediatr Infect Dis J 2011;30:S25e9. [12] Muhsen K, Chodick G, Goren S, Anis E, Ziv-Baran T, Shalev V, et al. Change in incidence of clinic visits for all-cause and rotavirus gastroenteritis in young children following the introduction of universal rotavirus vaccination in Israel. Euro Surveill 2015;20. [13] Muhsen K, Rubenstein U, Kassem E, Goren S, Schachter Y, Kremer A, et al. A significant and consistent reduction in rotavirus gastroenteritis hospitalization of children under 5 years of age, following the introduction of universal rotavirus immunization in Israel. Hum Vaccin Immunother 2015;11:2475e82. [14] Muhsen K, Kassem E, Rubenstein U, Goren S, Ephros M, Cohen D, et al. Incidence of rotavirus gastroenteritis hospitalizations and genotypes, before and five years after introducing universal immunization in Israel. Vaccine 2016;34:5916e22. [15] Bar-Zeev N, Jere KC, Bennett A, Pollock L, Tate JE, Nakagomi O, et al. Population impact and effectiveness of monovalent rotavirus vaccination in urban Malawian children 3 years after vaccine introduction: ecological and caseecontrol analyses. Clin Infect Dis 2016;62(Suppl. 2):S213e9. pez B, Benoit SR, Xuya M, [16] Gastanaduy PA, Contreras-Roldan I, Bernart C, Lo et al. Effectiveness of monovalent and pentavalent rotavirus vaccines in Guatemala. Clin Infect Dis 2016;62(Suppl. 2):S121e6. [17] Hemming-Harlo M, Vesikari T, Uhari M, Renko M, Salminen M, TorcelPagnon L, et al. Sustained high effectiveness of RotaTeq on hospitalizations attributable to rotavirus-associated gastroenteritis during 4 years in Finland. J Pediatr Infect Dis Soc 2016 Oct 19. M, Heylen E, [18] Braeckman T, Van Herck K, Meyer N, Pirçon JY, Soriano-Gabarro et al. Effectiveness of rotavirus vaccination in prevention of hospital admissions for rotavirus gastroenteritis among young children in Belgium: caseecontrol study. BMJ 2012;345. [19] Tate JE, Ngabo F, Donnen P, Gatera M, Uwimana J, Rugambwa C, et al. Effectiveness of pentavalent rotavirus vaccine under conditions of routine use in Rwanda. Clin Infect Dis 2016;62:S208e12. [20] Tate JE, Patel MM, Cortese MM, Payne DC, Lopman BA, Yen C, et al. Use of patients with diarrhea who test negative for rotavirus as controls to estimate rotavirus vaccine effectiveness through caseecontrol studies. Clin Infect Dis 2016;62:S106e14. [21] Leshem E, Givon-Lavi N, Tate JE, Greenberg D, Parashar UD, Dagan R. Realworld effectiveness of pentavalent rotavirus vaccine among Bedouin and Jewish children in southern Israel. Clin Infect Dis 2016;62:S155e60. [22] World Health Organization, Vaccines and Biologicals. Generic protocols for (i) hospital-based surveillance to estimate the burden of rotavirus gastroenteritis in children and (ii) a community-based survey on utilization of health care services for gastroenteritis in children. Geneva: WHO; 2002. [23] Muhsen K, Shulman L, Rubinstein U, Kasem E, Kremer A, Goren S, et al. Incidence, characteristics, and economic burden of rotavirus gastroenteritis
K. Muhsen et al. / Clinical Microbiology and Infection 24 (2018) 53e59 associated with hospitalization of Israeli children <5 years of age, 2007e2008. J Infect Dis 2009;200(Suppl. 1):S254e63. [24] Muhsen K, Shulman L, Kasem E, Rubinstein U, Shachter J, Kremer A, et al. Effectiveness of rotavirus vaccines for prevention of rotavirus gastroenteritisassociated hospitalizations in Israel: a caseecontrol study. Hum Vaccin 2010;6. [25] Baqui AH, Black RE, Yunus M, Hoque AR, Chowdhury HR, Sack RB. Methodological issues in diarrhoeal diseases epidemiology: definition of diarrhoeal episodes. Int J Epidemiol 1991;20:1057e63. [26] Government of Israel, Central Bureau of Statistics. Characterization and classification of geographic units by the socio-economic level of the population 2008. 2013. Available at: http://www.cbs.gov.il/webpub/pub/text_page_eng. html?publ¼100&CYear¼2008&CMonth¼1 [accessed 2 June 2017].
59
[27] Weinberg GA, Szilagyi PG. Vaccine epidemiology: efficacy, effectiveness, and the translational research roadmap. J Infect Dis 2010;201:1607e10. [28] Staat MA, Payne DC, Donauer S, Weinberg GA, Edwards KM, Szilagyi PG, et al. Effectiveness of pentavalent rotavirus vaccine against severe disease. Pediatrics 2011;128:E267e75. [29] Mondal D, Minak J, Alam M, Liu Y, Dai J, Korpe P, et al. Contribution of enteric infection, altered intestinal barrier function, and maternal malnutrition to infant malnutrition in Bangladesh. Clin Infect Dis 2012;54:185e92. [30] Berkman DS, Lescano AG, Gilman RH, Lopez SL, Black MM. Effects of stunting, diarrhoeal disease, and parasitic infection during infancy on cognition in late childhood: a follow-up study. Lancet 2002;359:564e71.