Share this article on:

Impact of Rotavirus Vaccine on Childhood Diarrheal Hospitalization After Introduction Into the South African Public Immunization Program

Msimang, Veerle M. Y. MSc*; Page, Nicola PhD*; Groome, Michelle J. MD, MSc†‡; Moyes, Jocelyn MD, MSc*; Cortese, Margaret M. MD§; Seheri, Mapaseka PhD; Kahn, Kathleen MD, PhD‖**††; Chagan, Meera MD, PhD‡‡; Madhi, Shabir A. MD, PhD*†‡; Cohen, Cheryl MD, MSc*

The Pediatric Infectious Disease Journal: December 2013 - Volume 32 - Issue 12 - p 1359–1364
doi: 10.1097/INF.0b013e3182a72fc0
Vaccine Reports

Background: Oral rotavirus vaccine was introduced into the South African routine immunization program in August 2009 administered at 6 and 14 weeks with no catch-up. We described the change in rotavirus-associated diarrheal hospitalizations among children <5 years at 3 sentinel sites from 2009 through 2011.

Methods: During 2009 through 2011, we compared the proportion of enrolled children aged <5 years hospitalized with acute gastroenteritis and testing rotavirus positive. We used hospital data to determine the change in diarrhea hospitalizations and estimated total numbers of rotavirus hospitalizations by adjusting for nonenrolled patients. Stool samples were tested for rotavirus using enzyme immunoassay.

Results: In 2009 (May–December), 46% (404/883) of samples among children <5 years tested rotavirus positive, decreasing to 33% (192/580) (P < 0.001) in 2010 and 29% (113/396) (P < 0.001) in 2011. Compared with May–December 2009, total diarrhea hospitalizations among children aged <5 years was one-third lower in May–December of 2010 and 2011. Among infants, adjusted rotavirus hospitalizations were 61% (n = 267) and 69% (n = 214) lower, respectively, in 2010 and 2011 when compared with 2009 (n = 689), and 45 and 50 percentage points greater than the reduction in rotavirus-negative cases. Among children <5 years, rotavirus hospitalizations were 54% and 58% lower in 2010 and 2011, compared with 2009 (40 and 44 percentage points greater than reduction in rotavirus-negative cases). Rotavirus reductions occurred in rural and urban settings.

Conclusion: Using published estimates of rotavirus hospitalization burden, we estimate that at least 13,000 to 20,000 hospitalizations in children <2 years were prevented in the 2 years after rotavirus vaccine introduction.

Supplemental Digital Content is available in the text.

From the *National Institute for Communicable Diseases of the National Health Laboratory Services; Department of Science and Technology/National Research Foundation: Vaccine Preventable Diseases; Medical Research Council: Respiratory and Meningeal Pathogens Research Unit, University of the Witwatersrand, Johannesburg, South Africa; §National Center for Immunization and Respiratory Diseases, Centers for Disease Control and Prevention, Atlanta, GA; MRC/Diarrhoeal Pathogens Research Unit, University of Limpopo Medunsa Campus, Pretoria; MRC/Wits Rural Public Health and Health Transitions Research Unit (Agincourt), Faculty of Health Sciences, University of the Witwatersrand, Johannesburg, South Africa; **Umeå Centre for Global Health Research, Umeå University, Umeå, Sweden; ††INDEPTH Network, Accra, Ghana; and ‡‡Department of Paediatrics, University of Kwazulu-Natal, Durban, South Africa.

Accepted for publication July 1, 2013.

This surveillance work was supported by an unconditional grant for the conduct of Sentinel Site Rotavirus Surveillance from GlaxoSmithKline (GSK).

N.P. reports receiving fees for speaking engagements by GSK and Merck (MSD) Pty (Ltd). S.A.M. has received honorarium from GSK and MERCK. All authors have submitted the PIDJ Form for Disclosure of Potential Conflict of Interest. Conflicts that the editors consider relevant to the content of the article have been disclosed in the Acknowledgment section.

The funding sources had no involvement in the research, writing or the decision to submit the article for publication. The findings and conclusions in this report are those of the authors. The authors have no other funding or conflicts of interest to disclose.

Address for correspondence: Veerle M. Y. Msimang, MSc, Veterinary Epidemiology and Public Health, National Institute for Communicable Diseases, 1 Modderfontein Road, Private Bag X4, Sandringham 3121, South Africa. E-mail:

Supplemental digital content is available for this article. Direct URL citations appear in the printed text and are provided in the HTML and PDF versions of this article on the journal’s website (

Rotavirus causes an estimated 453,000 deaths and 2 million hospitalizations annually worldwide,1,2 accounting for 30% to 50% of diarrheal hospitalizations in children <5 years.3–6 In children <5 years, diarrhea accounts for approximately 11% of deaths in South Africa,5,7,8 and among children <2 years, rotavirus infection is estimated to cause 17,644 to 25,630 hospitalizations annually.9

A randomized, placebo-controlled trial, evaluating a 2-dose versus 3-dose schedule, demonstrated that 2 doses of monovalent rotavirus vaccine administered at ages 10 and 14 weeks with oral polio vaccine was 72% efficacious (95% confidence interval: 40–88) in preventing severe rotavirus gastroenteritis in South African infants aged <1 year.10,11 Monovalent rotavirus vaccine was introduced into the South African Expanded Program on Immunization in August 2009 administered at 6 weeks (with oral polio vaccine) and 14 weeks of age. The decision to immunize at 6 weeks rather than 10 weeks of age (as was evaluated in the efficacy trial)10 was premised on enabling earlier exposure to rotavirus vaccine and maximizing the opportunity of vaccinating at the earliest Expanded Program on Immunization visit. There are few published data evaluating the effectiveness or efficacy of the vaccine schedule used in South Africa, or the impact of rotavirus vaccine following introduction in African countries.10,12,13

We describe trends in rotavirus hospitalizations at 3 hospitals during the 2 years (2010 and 2011) after introduction of rotavirus vaccine in South Africa in 2009.

Back to Top | Article Outline


Rotavirus Sentinel Surveillance

A prospective hospital-based surveillance system for diarrhea was established in April 2009 at 3 sentinel sites (Fig., Supplemental Digital Content 1, These included 2 hospitals in Gauteng Province: Chris Hani Baragwanath Hospital (CHBH), a large tertiary referral hospital (2625 beds) in urban Soweto, and Dr. George Mukhari hospital (DGMH) (1500 beds), 37 km North of Pretoria, serving a periurban population; and Matikwana and Mapulaneng (M/M) hospitals, a regional referral (250 beds) and a district level hospital (180 beds) in rural Bushbuckridge district, Mpumalanga Province, 402 km east of Johannesburg. All hospitals have both inpatient and outpatient facilities. In South Africa, children with diarrhea generally present to public clinics where care is provided free of charge and free hospital referral is based on healthcare worker assessment. Treatment of diarrhea focuses on rehydration and follows standard Integrated-Management-Childhood-Illness guidelines. The decision for admission to and discharge from hospital is made by the attending physician.

Children aged <5 years, hospitalized for acute diarrhea (defined as <7 days duration at the time of hospital admission, with ≥3 loose stools in a 24-hour period), were screened for enrolment during working hours, Monday through Friday. We collected clinical and demographic data and a stool sample within 48 hours of hospitalization from consenting patients. The monthly number of diarrhea hospitalizations among children aged <5 years (the age group available) was obtained from hospital registers.

Testing of stool samples was performed at the National Institute for Communicable Disease in Johannesburg for samples from CHBH and M/M and at the Medical Research Council/Diarrhoeal Pathogens Research Unit, University of Limpopo, for samples from DGMH. Samples were tested using the Prospect Rotavirus ELISA (Oxoid, Basingstoke, UK), except during 2009 when the National Institute for Communicable Disease used the GastroVir immunochromatographic strip (CorisBioconcept, Gembloux, Belgium) and confirmed equivocal results by genotyping reverse transcription polymerase chain reaction.14 Stool samples found insufficient for testing (11/939 in 2009 and 15/1179 in 2010) were excluded. Real-time HIV polymerase chain reaction testing was performed using the Roche COBASAmpliPrep/COBASTaqMan HIV-1 Qual test on dried blood spot specimens for all consenting participants.

Back to Top | Article Outline

Rotavirus Vaccine Uptake

South Africa introduced oral monovalent rotavirus vaccine (Rotarix, GlaxoSmithKline Biologicals, Rixensart, Belgium) in August 2009. Monthly numbers of first and second rotavirus vaccine doses administered were obtained from the relevant subdistricts and the annual birth cohort from the national department of health information system. We estimated the coverage among children aged <1 year from each surveillance site’s catchment area on April 1, 2010 and 2011, by counting the number of doses administered (from vaccine introduction through March 2010 and April 2010, through March 2011, respectively) and dividing by the annual birth cohort. Assuming the vaccine was available from August 1, 2009, and given at ages 6 weeks and 14 weeks without catch-up, the earliest cohort of children that may have received rotavirus vaccine at age 6 weeks were those born in mid-June 2009. These children would have been aged 9.5 months on April 1, 2010, and 21.5 months on April 1, 2011. For the cohort of children aged <1 year on April 1, 2010, approximately 50% would have been eligible to have received 2 doses of rotavirus vaccine. By April 2010, no child aged 12–23 months would have been eligible to have previously received rotavirus vaccine. However, by September 2010 and April 2011, 21% and 79%, respectively, of 12- to 23-month-old children would have been eligible to have received rotavirus vaccine.

Back to Top | Article Outline

Data Analysis

The total number of diarrhea hospitalizations and adjusted number of rotavirus and nonrotavirus diarrhea hospitalizations from May through December 2010 and 2011 were compared with those from May through December 2009 by age group and surveillance site. Although there was no known systematic change in surveillance practices, the proportion of total diarrhea hospitalizations enrolled in surveillance varied by year at the sites. Therefore, we did not estimate reductions based only on enrolled children. To estimate the total number of rotavirus hospitalizations for each calendar month, we used the number of rotavirus hospitalizations enrolled in the surveillance system and adjusted this by adding the additional number expected if all eligible children hospitalized for diarrhea had been enrolled that month. For each month, the additional number expected in each age group was obtained by using the information from those enrolled (age distribution, percentage positive in each age group) and applying these to the number not enrolled. The adjusted number of nonrotavirus (rotavirus negative) diarrheal hospitalizations was the total diarrheal hospitalizations minus the adjusted number of rotavirus hospitalizations. The proportion of children tested that were rotavirus positive in 2010 and 2011 was compared with that for 2009, and the proportion of enrolled rotavirus-positive and rotavirus-negative children in each age group was compared over time using a χ2 test.

Before rotavirus vaccine introduction, the rotavirus season in Gauteng generally started in April and peaked in May/June with a second small peak around September.15 To analyze the timing of the rotavirus season, data were included from the start of the surveillance (April 15, 2009, at the earliest site) through December 31, 2011. For each year, the start of the rotavirus season was defined as the first of 2 consecutive weeks during which the detection ratio for rotavirus was ≥20% among enrolled children at all sites combined and the end of the season as the first week that was followed by 2 consecutive weeks with a detection rate of <20% (based on annual median rotavirus detection ratios between 14% and 21% found at DGMH during 2005 to 2009).6

Back to Top | Article Outline


Vaccine Coverage and Surveillance Enrolment

Estimated rotavirus vaccine coverage among infants increased between 2010 and 2011 (Table 1). Mpumalanga province (where M/M site is situated) experienced vaccine shortage in April, May and June of 2010 resulting in lower 2011 coverage than the other areas.



An estimated 85% (4027/4725) of all screened diarrheal admissions were eligible for enrolment, and 70% (2821/4027) of eligible patients were enrolled. The main reasons for nonenrolment were the parent being unavailable or refusal to give consent for their child to take part in surveillance and have their stool tested for rotavirus. Adequate stool specimens were collected from 93% (2632/2821) of children enrolled. Of these, 48% (1266/2632) were enrolled from CHBH, 27% (716/2632) from DGMH and 25% (650/2632) from M/M.

Back to Top | Article Outline

Trends in Rotavirus and Total Diarrheal Hospitalizations

The total number of diarrheal hospitalizations among children aged <5 years was 32% lower during May–December 2010 and 33% lower in May–December 2011, compared with 2009. The adjusted number of rotavirus hospitalizations May–December 2010 and May–December 2011 among children aged <5 years was 54% (n = 403) and 58% (n = 371) lower, respectively, compared with 2009 (n = 877), (Table 2). In comparison, there was only a 14% reduction in rotavirus-negative hospitalizations in 2010 and 2011. Rotavirus detection ratios compared with 2009 were 0.72 (P < 0.001) in 2010 and 0.64(P < 0.001) in 2011.



Among infants, adjusted numbers of rotavirus hospitalizations were reduced substantially by 61% (422/689) in 2010 and 69% (475/689) in 2011 (versus 2009), whereas the reductions in rotavirus-negative hospitalizations were 16% and 19%, respectively. Rotavirus detection ratios compared with 2009 were 0.69 (P < 0.001) in 2010 and 0.50 (P < 0.001) in 2011.

Among 12- to 23-month-old children, hospitalizations for rotavirus diarrhea decreased by 35% (58/168) in 2010 and 21% (35/168) in 2011 compared with 2009, which was similar to the reduction in rotavirus-negative hospitalizations. Rotavirus detection ratios remained stable compared with 2009 (0.92, P = 0.55 in 2010 and 1.12, P = 0.47 in 2011) in the 12- to 23-month age group. Very few children aged 2–4 years were hospitalized with rotavirus disease pre- or postvaccine introduction (Table 2).

When examined by site, for children <5 years and infants, the reductions were generally of similar magnitude at Matikwana/Mapulaneng (M/M) (rural) and CHBH (urban), for total diarrhea hospitalizations, and rotavirus and rotavirus-negative hospitalizations. MM, however, had less reduction in rotavirus-negative hospitalizations among infants in 2011 compared with CHBH (Table, Supplemental Digital Content 2, At DGMH (periurban) in 2010, the reductions in rotavirus hospitalizations in age groups <5- and <1-year old was similar to the other sites in 2010 but less in 2011, and there was no reduction or a modest increase in rotavirus-negative hospitalizations. At each site, the proportion of infants tested that were rotavirus positive was lower in 2010 and 2011 compared with 2009.

Back to Top | Article Outline

Age Distribution of Enrolled Children by Rotavirus Test Results

From April 2009 through December 2011, the proportion of enrolled children testing rotavirus positive aged 2–11 months decreased from 73% to 55% whereas the proportion aged 12–23 months increased from 20% to 34%. No change occurred in the age distribution of the enrolled children testing negative for rotavirus from 2009 to 2011 (Fig., Supplemental Digital Content 3,

Back to Top | Article Outline

Timing of the Rotavirus Season

In 2009, rotavirus cases were detected from the start of the surveillance (ie, calendar week 16; detection rate 33%) and the season lasted until calendar week 40 (Fig. 1 and Fig., Supplemental Digital Content 4, The 2010 and 2011 season began in calendar weeks 20 and 21, respectively, and lasted until calendar weeks 36 and 40, respectively. The onset of the rotavirus season was later, and the period of time over which rotavirus was identified was at least 4 weeks shorter in both 2010 and 2011 compared with 2009.



Back to Top | Article Outline

HIV Prevalence and Diarrheal and Rotavirus Deaths

Data on HIV status were available for 16% (461/2821) of children enrolled into the surveillance program. Among rotavirus-positive patients from April 2009 to December 2011, the HIV prevalence was 13% (1/8) in 2009, 10% (4/39) in 2010 and 5% (2/37) in 2011 (P = 0.4). HIV prevalence among rotavirus-positive children was 11% (6/53) in infants and 4% (1/23) in 12- to 23-month age group.

Among children enrolled into the surveillance system April 15, 2009, through December 31, 2011, 92 diarrheal-related deaths (92/2821; 3% [26/916] in 2009, 3% [39/1131] in 2010 and 4% [27/714] in 2011) were recorded. Mortality varied by age group: 4% (69/1903) for <1 year, 3% (19/740) for 12–23 months and 3 children aged ≥2 years died (3/278). Of the 92 deaths, 11 children (12%) overall were positive for rotavirus (12% [3/26] in 2009, 13% [5/39] in 2010 and 11% [3/27] in 2011). Of the 11 rotavirus diarrheal deaths, 5 were coinfected with Escherichia coli O157:H7, 1 with Cryptosporidium spp. and 1 with E. coli O157:H7and Cryptosporidium spp. Nine of the rotavirus-positive deaths were <1 year, 1 was 1 year old and another was 2 years old. HIV results were only available for 7 of the rotavirus deaths of which 2 were HIV positive.

Back to Top | Article Outline


To our knowledge, this is the first multisite evaluation in a programmatic setting, of the impact of rotavirus vaccination in an African country. Our results indicate that introduction of rotavirus vaccine into the national immunization program in late 2009 was associated with a one-third reduction in total diarrhea hospitalizations in children <5 years during May–December of 2010 and 2011, compared with 2009. In this age group, we estimated a 54% and 58% reduction in rotavirus hospitalizations in May–December 2010 and 2011, which was 40 and 44 percentage points, respectively, greater than the reduction in rotavirus-negative hospitalizations. Among infants, the 61% and 69% reduction in rotavirus hospitalizations were 45 and 50 percentage points above the rotavirus-negative reduction. Importantly, the reductions in rotavirus hospitalizations among infants were found at each site, encompassing both rural and urban areas. Administrative data suggested 2-dose coverage among infants of 40–45% and 56–78% at the start of the 2010 and 2011 rotavirus seasons, respectively.

Although reductions in rotavirus hospitalizations among infants after vaccine introduction have been estimated in various ways, several countries in Latin America and Europe, United States and Australia have demonstrated notable reductions (~50%–90%) among infants in the first year after moderate or high vaccine uptake,16–21 approximating or exceeding that expected based on efficacy results from clinical trials. Some evaluations excluded the first postintroduction season with lower coverage, while we included that first season. Our reductions in South African infants also approximate that based on the clinical trial efficacy and overall coverage estimates. Annual reductions in overall diarrhea-associated hospitalizations among children eligible to receive the vaccines have been observed in the other regions,16,18,20–25 as we also demonstrated.16,18,22–29 Reductions in deaths from diarrhea in children <5 years after vaccine introduction have been demonstrated in Brazil and Mexico.23,27,30

Our data support a substantial reduction in burden among infants shortly after vaccine introduction, however, the data among children 12–23 months are more variable and based on smaller numbers. By date of birth, none of the children aged 12–23 months at the start of the first postvaccine season would have been eligible to have received rotavirus vaccine in early infancy, so disease reduction from direct protection was not expected. The site-specific data available among this age group did not provide clear evidence for an indirect effect, and there were too few rotavirus hospitalizations among older children at baseline to assess any change. An indirect effect among unvaccinated older age groups, because of infant rotavirus vaccination possibly interrupting transmission of rotavirus in the community, has been proposed in some Latin American19,24 and high-income countries (Belgium, Australia and the United States).17,18,26

Compared with 2009, the 2011 rotavirus hospitalizations among 12- to 23-month age group decreased at 2 sites (and of a magnitude greater than that observed for rotavirus-negative reduction at the largest site), but increased at 1 site and consequently did not inform on the sustainability of vaccine-induced immunity. This is pertinent, as data from a subset of children in the 2-dose arm of the clinical trial in South Africa followed through their second rotavirus season suggested waning of protection.12 The efficacy against severe rotavirus disease through the second follow-up season was 32% (95% confidence interval: −71–75) (mean age at end of follow-up 13.8 months),12 whereas the efficacy in the South African infants in the full 2-dose cohort was 72% (95% confidence interval: 40–88).10 Lower effectiveness in the second year of life has also been noted in postintroduction evaluations in low- and middle-income countries,19,24 although high effectiveness beyond infancy was maintained in some countries like the United States.19,23,26,28 Ongoing surveillance and vaccine effectiveness evaluations are underway in South Africa to address this issue. The overall impact of rotavirus vaccine in reducing the rotavirus disease burden will nevertheless be substantial in countries like South Africa, where most severe disease occurs in infancy and even if there is waning of protection in the second year of life.

Although data from South Africa may not be directly applicable to all settings in Africa, it does provide an indication of the potential benefits that a rotavirus vaccine can have in an African setting with high levels of poverty (25% of adults unemployed) and high under 5 mortality rate (51/1000 births in 2009), similar to many other countries in sub-Saharan Africa.29 In addition, these data are the first on vaccine impact in a high HIV prevalence setting (antenatal HIV prevalence 30.2% in 2010 and HIV prevalence among 0–4 years 3.3%, among 0–18 months 2.9% in 2008).29

There are limitations to our evaluation. Our surveillance program began in April/May 2009, so we used only May–December 2009 period for our baseline burden estimate. The numbers of enrolled children testing rotavirus positive at DGMH are similar to the median of the 2005 to 2008 prevaccine years at that site, suggesting that the 2009 season was a typical rotavirus prevaccine year and providing further reassurance that using 2009 data from the other 2 sites as the prevaccine estimate is a reasonable approach.13 Furthermore, although the absolute numbers of rotavirus hospitalizations in May–December would be lower than if we were able to capture the full calendar year, our percent reduction estimates in rotavirus hospitalizations should be similar. Administrative data on the number of diarrheal hospitalizations among children <5 years (but not finer age groups) were available at all 3 sites, but these data were of variable quality. For the age-specific estimates of rotavirus hospitalizations, we assumed that the age distribution among patients not enrolled into the surveillance program was similar to that among those enrolled. If this were not the case, our age-stratified estimates of total rotavirus hospitalizations could be biased. Nonetheless, at each site, the data available among children <5 years support a substantial reduction in rotavirus hospitalizations. The hospital administrative data suggest that a lower proportion of children hospitalized for diarrhea were enrolled into the surveillance program in 2011. It is likely that this is at least in part the result of reduced numbers of children being approached for enrolment. Additional possible contributing factors could be changes in patient presentation patterns with more patients being admitted at night and over weekends, shorter overnight stays such that some children were discharged before being approached by surveillance staff for enrollment, and increased refusals among those approached. If the rotavirus detection rate among nonenrolled children differed from those of enrolled children, this could have introduced bias into our study. Unfortunately, no data on the characteristics of eligible children who could not be enrolled were available. Finally, subdistrict administrative data were used for estimates of vaccine coverage, which are unlikely to be as accurate as vaccine coverage surveys as individual-level vaccination status data are not available. However, the coverage estimates among infants based on the administrative data are close to that expected based on the timing of vaccine introduction, and administrative data are more promptly available than coverage surveys.17,24,25

Data presented in this study are ecological and thus other factors that could affect trends in overall diarrhea and rotavirus hospitalizations should be considered. If there had been improved outpatient management, changes in hospital admission criteria, care-seeking behavior or reduced disease severity in 2010 and 2011 compared with 2009, these may have reduced the number needing hospitalizations. Data on outpatient management of patients with diarrhea are not available, but we were not aware of any such changes implemented at clinics and hospitals in the surveillance sites. During the study period there were improvements, however, in management of HIV-infected women and children, including Prevention of Mother to Child Transmission Programs and an increase in the number of HIV-infected children initiated on antiretroviral treatment. Improving the health status of HIV-infected children would be expected to reduce the number of children needing hospitalization for diarrheal illness.29 Our conservative estimates of rotavirus reductions (reductions above those for rotavirus-negative cases) help account for potential nonvaccine effects. Additionally, 2 findings from our surveillance program strongly support that rotavirus vaccine has impacted rotavirus disease. First, the 2010 and 2011 rotavirus seasons in South Africa, which began approximately 9 months and 22 months, respectively, after vaccine introduction, were delayed in onset and shorter in total duration compared with the prevaccine 2009 season. The change in timing has been detected in other countries after vaccine introduction.16,28,30,31 Second, after vaccine introduction, the age distribution of children testing rotavirus positive increased whereas the distribution for those testing rotavirus negative did not change. This, too, is an expected early finding after introduction of an effective rotavirus vaccine administered in infancy.32 Neither of these 2 findings would occur with interventions that reduce the overall burden of diarrhea hospitalizations but not specifically target rotavirus disease.

Using the estimated number of annual rotavirus hospitalizations in children <2 (17,644–25,630) from a South African research group,9 we can estimate the number of rotavirus hospitalizations prevented in children age <2 years using the percent reductions from our evaluation (Table 2). The most conservative estimate would be using the rotavirus reduction that is above the rotavirus-negative reduction for children <2 (56%–19% = 37% reduction in 2010; 60%–20% = 40% reduction in 2011), which yields 13,686–19,737 fewer rotavirus hospitalizations in the 2-year period among children<2 years. A high estimate would be using only the rotavirus reduction without subtracting the rotavirus-negative change, yielding 20,466 to 29,730 fewer rotavirus hospitalizations in the 2-year period in this age group.

In conclusion, our data support that rotavirus hospitalizations have decreased substantially among South African infants in the 2 consecutive seasons 2010 and 2011 after vaccine introduction in August 2009. Due to variability between the years, continued surveillance is required to fully ascertain the impact of the rotavirus vaccine. Particularly, it will be important to measure vaccine effectiveness in children aged 12–23 months to assess for any potential waning of immunity as well as ascertain that there is no shift of severe disease to older age groups because of transient protection conferred by vaccination. The reductions in rotavirus hospitalizations over the 2 postvaccine rotavirus seasons in South Africa and consistent reductions in findings published by other countries7,28,31,33 provide sufficient evidence that rotavirus vaccine implementation needs to be extended to more African countries to reduce the burden of childhood diarrhea.

Back to Top | Article Outline


We are grateful to the children and families who participated in this surveillance study and acknowledge the efforts of the surveillance, nursery and medical staff from the Chris Hani Baragwanath, Dr. George Mukhari, Mapulaneng and Matikwana hospitals. We appreciate the efforts of staff in laboratories from the National Institute for Communicable Disease of the National Health Laboratory Service in Johannesburg and at the MRC/Diarrhoeal Pathogens Research Unit, University of Limpopo, Medunsa Campus. We thank all the Expanded Programme of Immunisation and Department of Health staff and their Department of Health Information System/Department of Health Information Systems for providing data on rotavirus vaccine coverage.

Back to Top | Article Outline


1. Parashar UD, Hummelman EG, Bresee JS, et al. Global illness and deaths caused by rotavirus disease in children. Emerg Infect Dis. 2003;9:565–572
2. Tate JE, Burton AH, Boschi-Pinto C, et al.WHO-coordinated Global Rotavirus Surveillance Network. 2008 estimate of worldwide rotavirus-associated mortality in children younger than 5 years before the introduction of universal rotavirus vaccination programmes: a systematic review and meta-analysis. Lancet Infect Dis. 2012;12:136–141
3. Cortes J, Arvelo W, Lopez B, et al. Rotavirus disease burden among children <5 years of age - Santa Rosa, Guatemala, 2007–2009. Trop Med Int Health. 2012;17:254–259
4. Kamiya H, Nakano T, Kamiya H, et al.Rotavirus Epidemiology Study Group. Rotavirus-associated acute gastroenteritis hospitalizations among Japanese children aged <5 years: active rotavirus surveillance in Mie Prefecture, Japan. Jpn J Infect Dis. 2011;64:482–487
5. Steele AD, Peenze I, de Beer MC, et al. Anticipating rotavirus vaccines: epidemiology and surveillance of rotavirus in South Africa. Vaccine. 2003;21:354–360
6. Kane EM, Turcios RM, Arvay ML, et al. The epidemiology of rotavirus diarrhea in Latin America. Anticipating rotavirus vaccines. Rev Panam Salud Publica. 2004;16:371–377
7. Aguado T, Bhutta Z, De Oliveira LH, et al. Detailed review paper on rotavirus vaccines. In: WHO Strategic Advisory Group of Experts (SAGE) on Immunization, April 2009, World Health Organisation.
8. Bradshaw D, Groenewald P, Laubscher R, et al. Initial burden of disease estimates for South Africa, 2000. S Afr Med J. 2003;93:682–688
9. Mapaseka SL, Dewar JB, van der Merwe L, et al. Prospective hospital-based surveillance to estimate rotavirus disease burden in the Gauteng and North West Province of South Africa during 2003–2005. J Infect Dis. 2010;202(suppl):S131–S138
10. Madhi SA, Cunliffe NA, Steele D, et al. Effect of human rotavirus vaccine on severe diarrhea in African infants. N Engl J Med. 2010;362:289–298
11. . Rotavirus vaccines:an update. Wkly Epidemiol Rec. 2009;84:533–540
12. Madhi SA, Kirsten M, Louw C, et al. Efficacy and immunogenicity of two or three dose rotavirus-vaccine regimen in South African children over two consecutive rotavirus-seasons: a randomized, double-blind, placebo-controlled trial. Vaccine. 2012;30(suppl 1):A44–A51
13. Seheri LM, Page NA, Mawela MP, et al. Rotavirus vaccination within the South African Expanded Programme on Immunisation. Vaccine. 2012;30(suppl 3):C14–C20
14. Gentsch J, Gray J, Iturriza-Gómara M, et al. Manual of rotavirus detection and characterization. Geneva, Switzerland World Health Organization 2009 http://HYPERLINK
15. Page N.. The introduction of rotavirus vaccines into South Africa. SA Fam Pract. 2006;48:57–58
16. Cortes JE, Curns AT, Tate JE, et al. Rotavirus vaccine and health care utilization for diarrhea in U.S. children. N Engl J Med. 2011;365:1108–1117
17. Buttery JP, Lambert SB, Grimwood K, 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(suppl 1):S25–S29
18. Raes M, Strens D, Vergison A, et al. Reduction in pediatric rotavirus-related hospitalizations after universal rotavirus vaccination in Belgium. Pediatr Infect Dis J. 2011;30:e120–e125
19. Yen C, Armero Guardado JA, Alberto P, et al. Decline in rotavirus hospitalizations and health care visits for childhood diarrhea following rotavirus vaccination in El Salvador. Pediatr Infect Dis J. 2011;30(suppl 1):S6–S10
20. O’Ryan M, Lucero Y, Linhares AC. Rotarix(R): vaccine performance 6 years postlicensure. Expert Rev Vaccines. 2011;10:1645–1659
21. Patel M, Pedreira C, De Oliveira LH, et al. Association between pentavalent rotavirus vaccine and severe rotavirus diarrhea among children in Nicaragua. JAMA. 2009;301:2243–2251
22. Cortese MM, Tate JE, Simonsen L, et al. Reduction in gastroenteritis in United States children and correlation with early rotavirus vaccine uptake from national medical claims databases. Pediatr Infect Dis J. 2010;29:489–494
23. do Carmo GM, Yen C, Cortes J, et al. Decline in diarrhea mortality and admissions after routine childhood rotavirus immunization in Brazil: a time-series analysis. PLoS Med. 2011;8:e1001024
24. Molto Y, Cortes JE, De Oliveira LH, et al. Reduction of diarrhea-associated hospitalizations among children aged < 5 Years in Panama following the introduction of rotavirus vaccine. Pediatr Infect Dis J. 2011;30(suppl 1):S16–S20
25. Quintanar-Solares M, Yen C, Richardson V, et al. Impact of rotavirus vaccination on diarrhea-related hospitalizations among children < 5 years of age in Mexico. Pediatr Infect Dis J. 2011;30(suppl 1):S11–S15
26. Payne DC, Staat MA, Edwards KM, et al. Direct and indirect effects of rotavirus vaccination upon childhood hospitalisations in 3 US counties, 2006–2009. Clin Infect Dis. 2011;53:245–253
27. Richardson V, Hernandez-Pichardo J, Quintanar-Solares M, et al. Effect of rotavirus vaccination on death from childhood diarrhea in Mexico. N Engl J Med. 2010;362:299–305
28. Tate JE, Mutuc JD, Panozzo CA, 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(suppl 1):S30–S34
29. Day C, Gray A, Bugell EHealth and Related Indicators. South African Health Review. Health system Trust. 2011:120–248
30. Correia JB, Patel MM, Nakagomi O, et al. Effectiveness of monovalent rotavirus vaccine (Rotarix) against severe diarrhea caused by serotypically unrelated G2P[4] strains in Brazil. J Infect Dis. 2010;201:363–369
31. Yen C, Tate JE, Wenk JD, et al. Diarrhea-associated hospitalizations among US children over 2 rotavirus seasons after vaccine introduction. Pediatrics. 2011;127:e9–e15
32. Payne DC, Szilagyi PG, Staat MA, et al. Secular variation in United States rotavirus disease rates and serotypes: implications for assessing the rotavirus vaccination program. Pediatr Infect Dis J. 2009;28:948–953
33. Cortese M, Leblanc J, White KE, et al. Leveraging state immunization information systems to measure the effectiveness of rotavirus vaccine. Pediatrics. 2011;128:e1474–e1481

rotavirus vaccine; childhood diarrhea; hospitalization

Supplemental Digital Content

Back to Top | Article Outline
© 2013 by Lippincott Williams & Wilkins, Inc.