Secondary Logo

Journal Logo

Supplement

Reduction in Rotavirus-associated Acute Gastroenteritis Following Introduction of Rotavirus Vaccine Into Australia's National Childhood Vaccine Schedule

Buttery, Jim P. FRACP*†; Lambert, Stephen B. MB BS, PhD; Grimwood, Keith MD; Nissen, Michael D. FRACP, PhD‡§; Field, Emma J. MApp pi; Macartney, Kristine K. FRACP‖**; Akikusa, Jonathan D. FRACP*; Kelly, Julian J. FRACP*; Kirkwood, Carl D. PhD††‡‡

Author Information
The Pediatric Infectious Disease Journal: January 2011 - Volume 30 - Issue 1 - p S25-S29
doi: 10.1097/INF.0b013e3181fefdee
  • Free

Abstract

In Australia, prior to the introduction of vaccines, rotavirus-associated gastroenteritis was responsible for an estimated 10,000 hospital admissions annually, representing 50% of all gastroenteritis admissions.1–3 Rotaviral disease has been estimated to cause 22,000 emergency department (ED) presentations and 115,000 primary care consultations each year.2,3 Nosocomial acquisition was also common, with up to 14% of nongastroenteritis hospital admissions to general children's wards in the peak season developing rotavirus infection during their stay.4 The economic impact on the health care system for direct health care costs alone has been calculated at $30 million each year.3

For most of temperate Australia, where the majority of the population resides, disease peaks occur annually, typically in the Winter-Spring months between July and October.5 In Northern and Central Australia, especially in indigenous communities, large outbreaks have been characterized by less predictable seasonality, and overwhelming impact on communities of young children and the health care systems caring for them.6 Nationally, the age group typically affected is similar to elsewhere, with the peak age of symptomatic infection occurring between 6 and 24 months.7 Disease onset at a younger age is typical in Central Australian settings6,8 and in Indigenous children elsewhere.9 Similar to other developed countries, children in out-of-home day care have been demonstrated to have increased risk of disease.10

Despite this substantial morbidity, mortality is low, with fewer than 1 death per year attributed to rotavirus disease.2

In 2006, both the monovalent human rotavirus strain G1P[8] (RV1, Rotarix, GSK Biologicals, Rixensart, Belgium) and the pentavalent bovine-human reassortant (RV5, Rotateq, Merck, Philadelphia, PA) vaccines were licensed by the Australian regulator Therapeutic Goods Administration for infant administration using 2 and 3 dose regimens, respectively. Both vaccines were available for use from early 2006, and were funded federally as part of the National Immunization Program (NIP) free of charge to all Australian infants at the scheduled age of administration.

Rotavirus vaccines were introduced into the NIP from July 2007 to be administered as 2-dose (RV1) or 3-dose (RV5) regimens in conjunction with the routine 2, 4, and 6 month infant schedule. Prior to a cost-effectiveness recommendation from the Pharmaceutical Benefits Advisory Committee, RV1 had been introduced in the Northern Territory as a state-funded initiative, in October 2006.11 Due to concern about potential increased risk of intussusception observed following the withdrawn pentavalent rhesus-reassortant rotavirus vaccine (Rotashield) in the United States in children receiving their first dose over 3 months of age, the upper age limits from the prelicensure clinical trials and vaccine manufacturer product information were recommended to providers for both the first (RV5, 12.9 weeks; RV1, 14.9 weeks) and final (RV5, 32.9 weeks; RV1, 24.9 weeks) doses of each vaccine.

At the time of vaccine introduction, despite NIP inclusion, decisions and purchasing negotiations regarding which vaccine to implement were made by the 8 States and Territories. This provided a unique scenario where, within the 1 country, populations received either one vaccine or the other depending upon their state of residence. RV5 was used in Queensland, Victoria, and South Australia, whereas RV1 was administered in New South Wales, Western Australia, Northern Territory, Australian Capital Territory, and Tasmania. Approximately half Australia's estimated 290,000 annual birth cohorts received either vaccine, although this split changed in May 2009 when Western Australia moved to RV5 vaccine. Combined with the high coverage of infant immunizations achieved in the Australian NIP, this offered an opportunity to compare and contrast the performance of both vaccines in similar settings and populations. This implementation also occurred against a background of extensive rotavirus epidemiology over the preceding 15 years.5

This review describes the changes in rotavirus epidemiology following rotavirus vaccine introduction, including the impact upon hospital admissions, emergency visits, and pediatric short-stay units (SSU). We describe the impact in populations using both RV1 and RV5 vaccines, both in children eligible for NIP funded vaccine, and older children unlikely to have received rotavirus vaccination as they were beyond the upper age limits recommended for vaccination when the funded program commenced.

METHODS

To assess and compare the impact of the introduction of rotavirus vaccine in several Australian states during first 24 months, we have used data from several sources including published articles together with unpublished data from hospital admissions and rotavirus testing in Victoria.12–14 Immunization data were obtained from the Australian Childhood Immunization Register (ACIR).15 The ACIR records immunization data on all children under the age of 7 years enrolled in the universal Australian health insurance scheme, Medicare, and constitutes a nearly complete population register, as approximately 99% of the 290,000 children born each year are registered with Medicare by 12 months of age.15 The ACIR records notifications from providers of vaccinations received and dates of administration. Date of birth, State or Territory, postcode of residence, provider type and Indigenous status are also recorded.

For the states included in this review, New South Wales (NSW) used RV1, while RV5 was introduced into the schedule in 2 states analyzed, Victoria and Queensland. We assessed the impact of vaccination on laboratory confirmed rotavirus disease at 3 sites: Queensland Health Laboratories; Children's Hospital at Westmead (CHW), New South Wales, and Royal Children's Hospital (RCH) in Victoria.

In Queensland, a laboratory-confirmed diagnosis of rotavirus infection is required to be notified to Queensland Health Rotavirus notifications and laboratory testing data from Queensland Health laboratories were used to compare rotavirus testing and patterns of the proportions of positive samples before and after vaccine introduction.12 The only other Australian jurisdiction requiring notification of rotavirus infection was the Northern Territory. In NSW, routine rotavirus testing of children with acute gastroenteritis (AGE) admitted to CHW conducted as part of infection control practice was analyzed.13 In addition, laboratory-confirmed rotavirus cases from 2 public pathology providers and ED assignments were analyzed.14

In RCH, Melbourne, the major pediatric referral hospital in Victoria, rotavirus testing of AGE is discretionary as part of clinical management. Rotavirus positive and total gastroenteritis admissions to the SSU (for stays >8 hours anticipated to be less than 36 hours duration) and the inpatient wards were documented, along with the proportion of SSU activity due to gastroenteritis.

Hospital admissions due to all cause AGE in Queensland were defined as those in which the primary diagnosis field contained the following International Classification of Diseases 10 codes: A08.0 (Rotavirus enteritis); A08.2 (Adenoviral enteritis); A08.3 (Other viral enteritis); A08.4 (Viral intestinal infection, unspecified); A09 (Diarrhea and gastroenteritis of presumed infectious origin); A09.0 (Other and unspecified gastroenteritis and colitis of infectious origin); and A09.9 (Gastroenteritis and colitis of unspecified origin). In all centers, fecal testing for rotavirus was performed using Enzyme-linked immunosorbent assays.

Disease due to rotavirus and all-cause gastroenteritis were documented by age and year, allowing comparison of the cohorts born before and after rotavirus vaccination was included in the NIP (infants born after May 1, 2007). This allowed examination of disease presentations in an immunized cohort, and those unlikely to have been immunized as they were not age eligible for funded rotavirus vaccination.

In the first Queensland 12-month birth cohort eligible for RV5, 3-dose vaccine effectiveness was calculated for preventing hospitalization due to RV and non-RV AGE using the screening method.16

RESULTS

Vaccine Coverage

The ACIR estimated national vaccine coverage by December 2008 (18 months into commencement of the nationally funded program) was 87% for at least 1 dose of vaccine received by 4 months of age, and 84% for a full vaccine course (either 2 or 3 doses) by 13 months of age.17 The timeliness of rotavirus vaccination was excellent with 95% to 99% of doses given before upper age limits.18 Rotavirus coverage is lower for Indigenous children than non-Indigenous children with the difference greatest for the second dose of rotavirus vaccine at 11%.

Rotavirus Gastroenteritis and AGE Hospitalizations

Figure 1 shows the monthly distribution of rotavirus detected for each state in the period prevaccine (2001–2006) together with the 3 years postvaccine introduction (2007–2009). As depicted in Figure 2, in 2007, the first year post vaccine introduction, all states observed a slight decrease in rotavirus activity with a slight shift in disease peak. In 2008 and 2009, the rotavirus peak was significantly blunted in all locations. No difference was identified between states introducing either RV1 or RV5. These decreases were associated with little or no decline in the number of tests performed in this period (Figs. 1, 2).

FIGURE 1.
FIGURE 1.:
Rotavirus positive fecal samples 2001–2009. A, Westmead Children's Hospital, New South Wales. B, Royal Children's Hospital, Victoria. C, Queensland Health, Queensland.
FIGURE 2.
FIGURE 2.:
Laboratory tests performed, results, and proportion of specimens positive for rotavirus, Queensland Health Clinical and Statewide Services, January 2000 to July 2010.

In Queensland Health laboratories, the proportion of specimens tested that were positive for rotavirus fell immediately in all age-groups, when 2007 data were compared with mean values for 2000 to 2006.12 Between 2000 and 2006, during the peak month of rotavirus activity, the proportion of specimens positive for rotavirus ranged from a high of 58% to a low of 41% (Fig. 2). The equivalent figure for 2007 and 2008 was 24%, falling further to 17% in 2009.

Rotavirus positive specimens at RCH, Victoria are documented in Figure 1B. The age distribution of rotavirus hospitalizations at RCH is typical of the spectrum seen across Australia in the years 2003 to 2006 prevaccine introduction (Fig. 3). The majority of disease occurred in the 0 to 12 month period (53.1% of cases), with 24.2% of cases reported in infants 0 to 6 months of age, and 28.9% of cases in infants 7 to 12 months of age.

FIGURE 3.
FIGURE 3.:
Age distribution of rotavirus hospitalization—Royal Children's Hospital, Melbourne, Victoria.

A reduction in rotavirus hospitalizations was seen in all age groups, 0 to 12 months, 13 to 24 months, and 25 to 36 months by 68%, 68%, and 53% respectively in the 2007–2009 period. In NSW, a reduction of 75% was seen in 2008–2009 period compared with the 6-year period 2001–2006, with a 93% reduction in the <12 month age group.13 The Victorian data show an increase in children in 2009 in the 0 to 6 month age group, almost exclusively in children 0 to 3 months of age. A slight increase in the 0 to 5 month group was also reported in NSW, although the number of cases is small.

There was a statistically significant rate reduction in rotavirus-coded hospitalizations in Queensland for all those less than 20 years of age, when mean annual admission rates for 2007 and 2008 were compared with the mean values from 2000 to 2006.16 For those less than 5 years of age, there was also a significant reduction in nonrotavirus AGE coded hospitalizations. For those less than 5 years of age, the number of specific rotavirus-coded hospitalizations prevented was fewer than those prevented for non-RV AGE. This suggests that in younger age groups, misclassification of a rotavirus admission to a less-specific code is more common than correct classification as a rotavirus hospitalization.9 For a non-RV AGE code in any or the primary diagnostic field, vaccine efficacy was 62.3% and 63.9%, respectively. For RV, AGE-coded admissions, the corresponding figures are 89.3% and 93.9%.16

Age of Rotavirus Gastroenteritis Presentations: Evidence of Indirect Herd Effect

In Queensland, a reduction in rotavirus positive samples from children too old to have received RV5 under the funded NIP program since 2007 provides support for a possible indirect protective effect of RV5.12 At WCH, NSW, a decline in rotavirus hospitalizations was observed also in those aged 24 to 59 months, most of whom would not have been eligible for immunization, and those >60 months of age.13 Similarly, reductions in rotavirus admissions to RCH, Victoria were seen in children aged 13 to 24 months in 2008, and 25 to 36 months in 2009 (Fig. 3).

Impact on ED Presentations and SSU Admissions

In NSW, state-wide ED presentations during the Winter/Spring rotavirus season between 2001 to 2008, which were assigned a primary ED diagnosis of gastroenteritis or gastroenteritis symptoms were examined.14 2008 showed the lowest number and annualized rate of gastroenteritis presentations for the previous 8 years, in both the age group eligible for rotavirus vaccination (<15 months of age) and those too old to have received it under the NIP (15 months to 5 years).

Prior to the introduction of rotavirus vaccine into the NIP in 2007, gastroenteritis was the first or second most common cause of admission to the SSU at RCH, Victoria. Over the period 2005 to 2009, admissions for AGE to the SSU fell from a peak of 739 in 2006 to 166 by 2009. Figure 4 shows a plot of the total number of admissions to SSU and the proportion of admissions due to AGE by year between 2005 to 2009. Since 2007, there has been a fall in the total number of admissions to SSU commensurate with the drop in the number of admissions to SSU with AGE. At the same time, the proportion of SSU admissions related to AGE has fallen from just under 15% in 2005 to 5% in 2009.

FIGURE 4.
FIGURE 4.:
Short-stay unit admissions to RCH, Victoria 2005–2009, and the proportion due to acute gastroenteritis (AGE).

Nosocomial Rotavirus Gastroenteritis

At CHW, NSW, nosocomial rotavirus infections declined following vaccine introduction by an estimated 87% (the mean annual number of nosocomial cases in 2007 to 2009 was 4, compared with a mean of 31 annual nosocomial cases previously).13

DISCUSSION

From 2007, RV1 was used in 5 Australian States/Territories and RV5 was used in the other 3 States/Territories. This geographic split resulted in approximately one half of the Australian birth cohort receiving RV1 and one half receiving RV5, although this split changed in May 2009 when Western Australia moved to the RV5 vaccine.17 In the first 18 months of vaccine introduction, vaccine uptake was high, and timeliness of vaccine administration extremely high.

We describe findings from 3 states, 2 using RV5 and 1 RV1. These studies include state-wide public hospital laboratory testing and all hospitalization data from Queensland, and 2 studies based at the major pediatric hospital in each of the 2 most populous states, NSW and Victoria. All studies show not only reductions in rotavirus positive tests and hospital encounters, but also reductions in nonrotavirus coded episodes of gastroenteritis. Interpretation of this is somewhat complicated by different testing thresholds in different states, but suggests that even in settings where testing is routine, testing and coding practices underestimate the true rate of rotavirus presentations.12 This has been noted in other settings, including the United States.19

Reductions in childhood gastroenteritis encounters have been observed at all hospital levels, from emergency presentations and SSU admissions to all cause and rotavirus coded hospital admissions. Reductions in hospitalizations and rotavirus positive tests have also been observed in children too old to be eligible for NIP funded rotavirus vaccination, suggesting early evidence of a possible indirect protective effect of rotavirus vaccination. These results are supported by similar results from South Australia (RV5).20 Importantly, case control data from a rotavirus outbreak in Central Australia demonstrated 2-dose RV1 vaccine efficacy of 77.7% (95% confidence interval, 40.2%–91.7%) against hospitalization for gastroenteritis, with vaccine efficacy 84.5% (95% CI, 23.4%–96.9%) against confirmed cases of rotavirus infection.8 These indirect effects have also been reported from other countries who have introduced rotavirus vaccines into their immunization programs.21

In Australia, where rotavirus mortality was rare prior to vaccine introduction, the decision to implement infant rotavirus vaccination was based on the morbidity caused by rotavirus and the predicted cost-effectiveness of vaccination.3 These early data provide reassurance that vaccination has impacted directly and possibly indirectly on gastroenteritis morbidity, and its resultant health, societal and economic toll.

Australia offers a unique model of rotavirus vaccination. The state-based implementation of a single vaccine with high coverage and timeliness, but different states choosing between RV1 and RV5, may provide insights into the way these vaccines behave in similar populations. The constant pressure exerted on the circulating wild-type strains by the vaccine will continue to be monitored by the Australian Rotavirus Surveillance Program, to provide data on the impact of universal mass vaccination program. This “programmatic experiment” may provide further insights the 2 vaccines including genotype coverage, induction of herd immunity and safety. To date, the early results for rotavirus disease prevention are promising.

REFERENCES

1.Carlin JB, Chondros P, Masendycz P, et al. Rotavirus infection and rates of hospitalisation for acute gastroenteritis in young children in Australia, 1993–1996. Med J Aust. 1998;169:252–256.
2.Newall AT, MacIntyre R, Wang H, et al. Burden of severe rotavirus disease in Australia. J Paediatr Child Health. 2006;42:521–527.
3.Galati JC, Harsley S, Richmond P, et al. The burden of rotavirus-related illness among young children on the Australian health care system. Aust N Z J Public Health. 2006;30:416–421.
4.Ringenbergs ML, Davidson GP, Spence J, et al. Prospective study of nosocomial rotavirus infection in a pediatric hospital. Aust Paediatr J. 1989;25:156–160.
5.Kirkwood CD, Boniface K, Bogdanovic-Sakran N, et al. Rotavirus strain surveillance—an Australian perspective of strains causing disease in hospitalised children from 1997 to 2007. Vaccine. 2009;27(suppl 5):F102–F107.
6.Schultz R. Rotavirus gastroenteritis in the Northern Territory, 1995–2004. Med J Aust. 2006;185:354–356.
7.Buttery JP, Kirkwood C. Rotavirus vaccines in developed countries. Curr Opin Infect Dis. 2007;20:253–258.
8.Snelling TL, Schultz R, Graham J, et al. Rotavirus and the indigenous children of the Australian outback: monovalent vaccine effective in a high-burden setting. Clin Infect Dis. 2009;49:428–431.
9.Campbell SJ, Nissen MD, Lambert SB. Rotavirus epidemiology in Queensland during the pre-vaccine era. Commun Dis Intell. 2009;33:204–208.
10.Ferson MJ, Stringfellow S, McPhie K, et al. Longitudinal study of rotavirus infection in child-care centers. J Paediatr Child Health. 1997;33:157–160.
11.Mitchell AS, Isaacs D, Buttery J, et al. Funding of drugs: do vaccines warrant a different approach? Lancet Infect Dis. 2009;9:269–270; author reply 70–71.
12.Lambert SB, Faux CE, Hall L, et al. Early evidence for direct and indirect effects of the infant rotavirus vaccine program in Queensland. Med J Aust. 2009;191:157–160.
13.Macartney KK, Porwal M, Dalton D, et al. Decline in rotavirus hospitalisations following introduction of Australia's national rotavirus immunization program. J Paediatr Child Health. In press.
14.Belshaw DA, Muscatello DJ, Ferson MJ, et al. Rotavirus vaccination one year on. Commun Dis Intell. 2009;33:337–340.
15.Hull BP, Deeks SL, McIntyre PB. The Australian Childhood Immunization Register-A model for universal immunization registers? Vaccine. 2009;27:5054–5060.
16.Field EJ, Vally H, Grimwood K, et al. Pentavalent rotavirus vaccine and prevention of gastroenteritis hospitalizations in Australia. Pediatrics. 2010;126:e506–e512.
17.Macartney KK, Burgess MA. Rapid impact of rotavirus vaccination in the United States: implications for Australia. Med J Aust. 2009;191:131–132.
18.Rotavirus vaccine coverage and timeliness in Australia and the impact on timeliness of other recommended vaccines. In: PHAA 12th National Immunization Conference 2010; August 18, 2010; Adelaide, Australia.
19.Hsu VP, Staat MA, Roberts N, et al. Use of active surveillance to validate international classification of diseases code estimates of rotavirus hospitalizations in children. Pediatrics. 2005;115:78–82.
20.Marshall H, Clarke M, Davidson G, et al. Reduction in pediatric gastroenteritis admissions in south Australia in both vaccinated and unvaccinated children post introduction of rotavirus vaccine. European Society of Pediatric Infectious Diseases, Nice, France, May 6, 2010.
21.Tate JE, Curns AT, Cortese MM, et al. Burden of acute gastroenteritis hospitalizations and emergency department visits in US children that is potentially preventable by rotavirus vaccination: a probe study using the now-withdrawn rotashield vaccine. Pediatrics. 2009;123:744–749.
© 2011 Lippincott Williams & Wilkins, Inc.