Prior to the introduction of rotavirus vaccine for all infants in the US in 2006, rotavirus gastroenteritis caused an estimated 205,000 – 272,000 emergency department (ED) visits annually among children under 5 years of age [1, 2]. Current US guidelines recommend either 2 doses of a monovalent rotavirus vaccine (RV1; Rotarix, GSK Biologicals) or 3 doses of a pentavalent rotavirus vaccine (RV5; Rotateq, Merck) for routine immunization of all US infants before the age of 8 months [1].

Since rotavirus vaccine introduction, there has been a dramatic decline in the burden of severe rotavirus disease in US children [3–8]. Moreover, rotavirus vaccination in infants offers indirect benefits by reducing rotavirus infection in older children and adults [9]. While the effect of vaccination on childhood hospitalizations for rotavirus has been well described, the effects of rotavirus vaccine on ED visits are less well documented. While rotavirus vaccination was approved primarily to reduce the risk of severe diarrhea (as measured by hospitalizations and deaths), an added benefit of vaccination would be reducing ED congestion through less diarrhea visits. Active surveillance through the New Vaccine Surveillance Network (NVSN) showed a 76% reduction in laboratory-confirmed rotavirus infection in ED visits in the first two seasons after vaccine introduction, but these data were limited to hospitals in three sentinel US counties [10]. A prior analysis using insurance databases showed a decrease in ED visits for gastroenteritis during the first four years of the vaccination program from 2008–2011 [5], but these data cannot be generalized to the entire US population.

With maturation of the rotavirus vaccine program, immunization coverage in children less than 3 years of age has increased from 44% in 2009 to 73% in 2013 [11]. We aim to provide a comprehensive evaluation of ED utilization among children <5 years of age for acute gastroenteritis through 2013 using robust and more nationally representative discharge data. We estimated baseline pre-vaccine burden and monitored trends in gastroenteritis- and rotavirus-coded ED visits among US children younger than 5 years in 10 US states over an 11-year period to account for expected year-to-year variation in rotavirus trends [12].

A total of 1,450,474 ED visits coded for gastroenteritis and 13,350 ED visits coded for rotavirus occurred in children <5 years of age in the 10 states from 2003–2013. Overall gastroenteritis-coded ED visits decreased significantly by 10.3% (±0.3%, p<.0001) in the six post-vaccine years (2008–2013) compared with the pre-vaccine years (2003–2006). ED visits were significantly reduced in 5 of the 6 post-vaccine years compared to the pre-vaccine period, with the largest decline of 20.4% (±0.5%, p<.0001) seen in 2012 (Table). Declines were similar among male and female children. By age group, there was a greater reduction in ED visits in age groups less than 2 years old (from 14.1%–20.6%, p<.0001), while visits remained unchanged or increased for children 2–5 years old. Hispanic children experienced a larger reduction in ED visits (23.1% ±1.0%) than White (13.7% ±0.5%) or Black (13.1% ±0.6%) children (p<.0001 for all comparisons).

The declines in ED visits were restricted to the months of the year from January to June during which rotavirus circulates in US Children, with a 21.2% (±0.4%, p<.0001) reduction in ED visits during these months overall in children <5 years of age in the post-vaccine period compared with pre-vaccine years. Similar to the annual reductions, the greatest reduction of 37.7% (±0.6%, p<.0001) during the rotavirus season was also seen in 2012. In contrast to the decline in the rotavirus-season months, a 9.5% (±0.6%, p<.0001) increase in ED visits was seen during the non-rotavirus season months from July to December.

Monthly gastroenteritis-coded ED visits (Figure 1A) decreased notably during rotavirus seasons in post-vaccine years relative to pre-vaccine years. The reductions in all-cause gastroenteritis ED visits mirrored the decline in ED visits assigned the rotavirus-specific code (Figure 1B), and, as expected, more prominent declines were seen in rotavirus-specific ED visits.

The largest and most consistent reductions in ED visits for gastroenteritis were in children <23 months of age (Figure 2).

Since the introduction of rotavirus vaccination in 2006, emergency department visits for all-cause acute gastroenteritis in children <5 years of age have declined through December 2013. The reduction in ED utilization has occurred primarily from January-June, the months when rotavirus infection is most common, and in children <2 years of age, the age group that is most likely to experience rotavirus infection [2]. These findings suggest strongly that the reduction in ED visits for gastroenteritis is likely attributable to rotavirus vaccination, as opposed to unmeasured contemporaneous factors or changes in coding practices.

Our results are consistent with and expand on results from prior studies showing a reduction in rotavirus detection, hospitalization, and resource utilization since vaccine introduction [3–6, 10, 17–20]. The most direct comparison can be made with an analysis of ED utilization using the 2001–2011 Truven Health MarketScan Commercial Claims and Encounters Database, an insurance claims database that is largely from large, self-insured companies, and excludes Medicaid recipients and the uninsured [5]. This analysis showed reductions in ED visits for gastroenteritis from −2 to 31% in the first five years following rotavirus vaccine introduction from 2007–2011, and a smaller decrease in -1 to 20% in outpatient (non-ED) visits for gastroenteritis in the same time period.

The change in ED visits for gastroenteritis (ranging from an increase of 1.8% to a decrease of 20.4% in individual years from 2008–2013) is not as dramatic as the change in hospitalizations (decreases from 31–55% in individual years from 2008–2012), using similar methods and a similar dataset [4]. There are at least two possible explanations for this finding. First, childhood gastroenteritis caused by rotavirus infection is more severe than that from other causes [21–23]; thus, the proportion of hospitalized diarrhea cases that are caused by rotavirus and thus are potentially vaccine preventable is greater than the proportion of diarrhea cases seen in ED settings. Second, the efficacy of rotavirus vaccines is greater against more severe rotavirus gastroenteritis [24–28], and thus vaccines would avert a greater proportion of rotavirus hospitalizations than ED visits.

Because laboratory testing for rotavirus is infrequently conducted among diarrhea patients seen in ED settings, our analysis focused on the less specific disease outcome of all-cause acute gastroenteritis. Despite the under-reporting for rotavirus in ED settings, it is reassuring that ED visits for rotavirus and all-cause gastroenteritis show matching reductions since vaccine introduction and matching post-vaccination biennial peaks, likely due to incomplete vaccine coverage.

Given the ecologic design our study, our examination of disease trends is potentially susceptible to bias from secular changes in the number of diarrhea-associated ED visits caused by pathogens other than rotavirus. In particular, secular trends associated with the emergence of new viral strains are well recognized for norovirus [29], which has emerged to become the leading cause of diarrhea hospitalization in US children following the decline in rotavirus after vaccine implementation [30]. In this regard, it is worth noting that the increase in ED visits seen in 2012–2013 temporally correspond to the emergence of the norovirus outbreak strain GII.4 Sydney [29].

In summary, our results contribute to the growing evidence of the impact of rotavirus vaccination in reducing the burden and severity of gastroenteritis and reducing utilization at multiple levels of the healthcare system in the United States. While rotavirus vaccination reduced ED visits more modestly than has been cited for hospitalizations, it still offers a valuable tool in reducing ED congestion, particularly in winter months. Increasing rotavirus vaccine coverage in eligible children remains a public health priority to sustain the declines in the burden of severe gastroenteritis in US children.