Yen, Catherine MD, MPH*†; Armero Guardado, Julio A. MD‡; Alberto, Patricia MD‡; Rodriguez Araujo, David S. MD‡; Mena, Carlos MD§; Cuellar, Elizabeth MD¶; Nolasco, Jenny Brenda MD∥; De Oliveira, Lucia Helena RN, MSc**; Pastor, Desiree MD†; Tate, Jacqueline E. PhD*; Parashar, Umesh D. MB BS, MPH*; Patel, Manish M. MSc, MD*
From the *National Center for Immunization and Respiratory Diseases, Centers for Disease Control and Prevention, Atlanta, GA; †Epidemic Intelligence Service, Scientific Education and Professional Development Program Office (proposed), Centers for Disease Control and Prevention, Atlanta, GA; ‡Ministerio de Salud Pública y Asistencia Social, San Salvador, El Salvador; §Epidemiología, Hospital de Niños Benjamín Bloom, San Salvador, El Salvador; ¶Epidemiología, Hospital San Juan de Dios, Santa Ana, El Salvador; ∥Epidemiología, Hospital San Juan de Dios, San Miguel, El Salvador; **Pan American Health Organization, Washington, DC; and ††Pan American Health Organization, Bolivia.
Accepted for publication September 28, 2010.
The findings and conclusions of this report are those of the authors and do not necessarily represent the views of the Centers for Disease Control and Prevention.
Address for correspondence: Catherine Yen, MD, MPH, National Center for Immunization and Respiratory Diseases, Centers for Disease Control and Prevention, 1600 Clifton Rd NE, MS A-47, Atlanta, GA 30333. E-mail: email@example.com.
Rotavirus is the leading cause of diarrheal disease among children aged <5 years in El Salvador, resulting in an estimated 23,080 clinic visits, 2783 hospitalizations, and 295 deaths each year.1 After the World Health Organization recommendation for inclusion of rotavirus vaccines into national immunization programs in American and European countries,2 the Ministry of Health of El Salvador added rotavirus vaccine to its national childhood immunization program in October 2006. A recent postlicensure study from El Salvador showed that the monovalent rotavirus vaccine used by the country conferred 76% protection against rotavirus hospitalizations,3 similar to the efficacy for this vaccine in a pivotal clinical trial from Latin America.4 Although these studies assessed direct vaccine protection against rotavirus gastroenteritis, assessing the total public health benefits of vaccination in terms of impact on the national burden of childhood diarrhea from all causes is important for the Ministry of Health to better gauge the benefits of their investments. In addition, early data from the United States and Australia indicate that rotavirus vaccination may also indirectly benefit unvaccinated children in the community (ie, herd immunity), resulting in an overall impact greater than the direct benefits observed in the clinical trials and effectiveness studies.5–8
In this study, we compared annual rates of all-cause diarrhea and rotavirus-specific hospitalizations at sentinel hospitals and annual rates of the national burden of health care utilization for diarrhea from all causes among children <5 years of age in El Salvador before and after rotavirus vaccine introduction. We found that vaccination had a substantial public health impact on severe rotavirus disease as well as overall diarrhea events. We also identified important age-related changes in diarrheal incidence in El Salvador that highlights the need for ongoing rotavirus surveillance after vaccine introduction.
El Salvador is a low-middle income country in Central America with an estimated 7 million inhabitants and an annual birth cohort of approximately 160,000. The 21,000 km2 country is divided into 14 departments with 262 municipalities. The Ministry of Health of El Salvador purchases all childhood vaccines and had introduced the monovalent rotavirus vaccine into the routine childhood immunization program in October 2006. Rotavirus vaccination is recommended for administration in 2 doses at 2 and 4 months of age concurrently with other routine immunizations as designated by the Expanded Program on Immunization.
Vaccine Coverage Data
Because no rotavirus vaccine coverage surveys were available for the examined study period, we obtained vaccine coverage data from a case-control vaccine effectiveness study previously conducted at sentinel surveillance hospitals.3 We used card-confirmed vaccination rates for community controls from this study to estimate first and second dose rotavirus vaccine coverage for the catchment population of 7 sentinel hospitals. Because the sentinel surveillance sites were estimated to capture ∼50% of the country's hospitalizations, we assumed that the rotavirus vaccination rates of controls from this study would reasonably reflect coverage countrywide.
Sentinel Surveillance for Laboratory-confirmed Rotavirus Hospitalizations
We obtained diarrhea- and rotavirus-specific hospitalization data for January 2006 through December 2009 from a sentinel surveillance system established by the Ministry of Health. This sentinel rotavirus surveillance system comprises 7 hospitals located in the departments of San Salvador, Santa Ana, San Miguel, La Libertad, and La Paz. At each hospital, the Ministry conducted active surveillance for cases of acute diarrhea (≥3 stools in 24 hours) among children <5 years of age in both emergency and inpatient wards. Fecal samples were obtained from these children and tested for rotavirus at the national laboratory using a commercial enzyme immunoassay.
National Surveillance for Diarrhea-related Healthcare Visits
In El Salvador, the public health facilities of the Ministry of Health provide health care to approximately 85% of the population. With regard to diarrhea, these facilities provide weekly reports on diarrhea-related healthcare visits to the Ministry of Health. This system has provided stable reporting on total healthcare visits (inpatient and outpatient) related to diarrhea from all causes since 2005. From this reporting system, we obtained healthcare visit data related to diarrhea from all causes among children <5 years of age for January 2005 through December 2009.
Sentinel Surveillance Data
We examined rotavirus-specific hospitalizations from the 7 sentinel hospitals and calculated the rate of rotavirus hospitalizations for each surveillance year. We further compared the rate of these hospitalizations in 2008 and 2009, with the baseline rate from 2006. We excluded 2007 data from the comparison of pre- and postvaccine years as we considered it a transitional year during which rotavirus vaccine was still being introduced. To calculate these hospitalization rates, we used pooled population estimates for the departments in which the sentinel hospitals were located. Although exact data on the hospital catchment population were not available for the sentinel surveillance sites, these were the largest hospitals in each department, and approximately 60% of all children <5 years of age in El Salvador reside in these departments. We stratified our analyses by age group (<1 year, 1 to <2 years, 2 to <3 years, 3 to <4 years, and 4 to <5 years) and calculated reductions in rotavirus-specific hospitalization rates in 2008 and 2009 compared with that of 2006. Because vaccination began in late 2006, very few children older than 1 year in 2008 and older than 2 years in 2009 were vaccinated. Thus, we hypothesized that a significant decline in rotavirus hospitalizations in these unvaccinated age groups would provide evidence of indirect benefits from vaccination.
National Surveillance Data
We calculated annual rates of total (inpatient and outpatient) diarrhea-related healthcare events by using projected population estimates for 2003 to 2009 provided by the Ministry of Economy as the denominator data. We compared rates in 2008 and 2009, after rotavirus vaccination introduction, with the available baseline healthcare event rates using the mean number of events during 2005 and 2006. As with rotavirus hospitalizations from the sentinel sites, we excluded 2007 data from the comparison of pre- and postvaccine years because we considered it a transitional year. Rotavirus disease is strongly seasonal in El Salvador with >95% of laboratory-confirmed disease occurring during January through June (ie, rotavirus season).1 Therefore, we stratified the analysis of trends in national diarrhea events before and after vaccine introduction by rotavirus and nonrotavirus season. We hypothesized that a prominent decline during rotavirus season compared with the nonrotavirus season would improve the strength of association between rotavirus vaccination and any observed reduction of national diarrhea-related events. Age-specific data were not available for national healthcare visits.
Percent decline in rates and 95% confidence limits were computed using Excel and SAS statistical software, version 9.2. P values < 0.05, as calculated using the χ2 statistic, were deemed statistically significant.
Before the 2008 rotavirus season, rotavirus vaccine coverage for infants <1 year was estimated as 76% for the first dose. Among children aged 1 to <2 years, vaccine coverage was 84% for the first dose. Children ≥2 years were yet to be vaccinated before 2008. By the beginning of the 2009 rotavirus season, coverage was 78% for the first dose among infants <1 year. Among children aged 1 to <2 years and 2 to <3 years, vaccine coverage increased to 89% and 84%, respectively, for the first dose.
Sentinel Surveillance for Laboratory-confirmed Rotavirus Hospitalizations
Between January 2006 and December 2009, the 7 sentinel surveillance hospitals reported a total of 8287 diarrhea-related hospitalizations among children <5 years of age. In each year, there were distinct winter-spring peaks in diarrhea-related hospitalizations corresponding with peak numbers of positive rotavirus tests (Fig. 1). Most (95%) of the laboratory-confirmed rotavirus hospitalizations occurred during the rotavirus season from January through June. During the 2008 and 2009 rotavirus seasons, the winter seasonal peaks in diarrhea admissions were greatly blunted, and the number of rotavirus-confirmed hospitalizations decreased concurrently.
During 2006, the sentinel hospitals reported 2131 diarrhea hospitalizations among children <5 years, 1095 (51%) of which were laboratory confirmed as rotavirus (Table 1). During 2008, when the overall vaccine coverage in children <5 years was 32%, the rate of rotavirus hospitalizations declined by 81% (95% confidence interval [CI]: 78%–84%; P < 0.0001) compared with 2006. During 2009, with an overall vaccine coverage of 50% among children aged <5 years, rotavirus hospitalizations declined by 69% (95% CI: 65%–73%; P < 0.0001) compared with 2006 (Table 1). In 2008, the greatest declines of 84% to 86% in rotavirus hospitalizations compared with 2006 were observed in children <1 and 1 to <2 years of age, corresponding to age groups that were vaccinated (Table 1). However, sizeable reductions (41%–68%) were also observed among children ≥2 years, who were unvaccinated. In 2009, while significant reductions in rotavirus hospitalizations persisted among those aged ≤2 years, these effects were not present among older age groups who were unvaccinated. In fact, among children 3 to <4 years of age, the rate of rotavirus hospitalizations doubled from 2006. However, because the majority of rotavirus hospitalizations occurred among children <2 years of age, the overall decline in rotavirus hospitalizations among children <5 years remained significant.
National Surveillance for Diarrhea-related Healthcare Visits
The observed patterns of national all-cause diarrhea-related healthcare visits in El Salvador mimicked the pattern of rotavirus hospitalizations from the sentinel surveillance hospitals (Fig. 2A, B). National diarrhea-related healthcare visits in El Salvador decreased considerably during postvaccine introduction years 2008 and 2009 when compared with the prevaccine period 2005–2006 (Fig. 2B). Annual diarrhea-related healthcare visits decreased by 60,816 in 2008 and by 46,028 in 2009 when compared with a mean number of 164,307 visits in 2005 and 2006. Looking at rotavirus seasons (January through June), during 2005 to 2006, public health facilities reported a mean number of 116,702 diarrhea-related healthcare visits among children <5 years (Table 2). Compared with this prevaccine baseline, the number of diarrhea-related healthcare visits declined by 55,898 during the 2008 rotavirus season and by 41,006 during the 2009 rotavirus season. Diarrhea-related healthcare visit rates decreased from a mean rate of 14,402 per 100,000 children aged <5 years during prevaccine rotavirus seasons 2005 and 2006 to 7515 per 100,000 children during the 2008 rotavirus season (48% decline [95% CI: 47%–48%; P < 0.0001]) and 9428 per 100,000 children during the 2009 rotavirus season (35% decline [95% CI: 34%–35%; P < 0.0001]) (Table 2). In contrast, less sizeable declines in the number and rate of diarrhea-related healthcare visits were observed during July through December, when bacterial etiologies for diarrhea are more prevalent.
A recent study in El Salvador indicated that the monovalent rotavirus vaccine was highly effective against severe rotavirus gastroenteritis requiring hospitalization among children <2 years of age.3 Our study extends these findings by demonstrating sizeable declines after vaccine introduction in both rotavirus-specific hospitalizations and national diarrhea-related healthcare visits among children <5 years of age. The following findings support that the introduction of rotavirus vaccination is associated with the observed trends. First, the greatest reductions in rotavirus hospitalizations were observed among children in vaccine-eligible age groups (ie, children <1 year in 2008 and children <2 years in 2009). Second, declines in diarrhea-related healthcare visits during the postvaccine introduction years of 2008 and 2009 were greatest during the months when severe rotavirus disease (ie, rotavirus disease requiring hospitalizations) is most prevalent. Finally, rotavirus-specific data from sentinel hospital surveillance showed trends very similar to the overall national pattern of diarrhea healthcare visits. Before the implementation of rotavirus vaccination, the Ministry of Health estimated that rotavirus caused 295 deaths and 2783 admissions annually among children aged <5 years in El Salvador.1 If the reductions observed during 2008 and 2009 persist, we estimate that vaccination of a birth cohort in El Salvador could prevent approximately 1 death for every 531 infants vaccinated and 1 hospitalization for every 66 infants vaccinated before 5 years of age.
During 2008, significant declines in rotavirus hospitalizations were observed not only among children age-eligible for rotavirus vaccination, but also among children age-ineligible for rotavirus vaccination. Similar findings have been noted in other countries that have implemented rotavirus vaccination such as the United States and Australia, raising the possibility of indirect protection of unvaccinated children (ie, herd immunity) by vaccination of young infants.5–9 It was interesting to note that, during 2009, these indirect benefits disappeared, with an increase in the number of rotavirus hospitalizations among unvaccinated children >2 years of age. Given the decline in rotavirus activity observed during 2008, it is possible that more unvaccinated, susceptible children unexposed to rotavirus in the previous season had accumulated, resulting in greater rotavirus activity during the 2009 season. The relative increases in rotavirus hospitalizations observed during 2009 compared with 2008 may also be attributable, in part, to natural, secular variation in rotavirus activity.10 In addition, waning of immunity induced by vaccination may have resulted in a greater number of rotavirus infections among children aged <2 years, as has been suggested by a previous case-control studies.3,11 However, despite this increase in disease among the older age groups during 2009, we were reassured of the overall public health effect of vaccination by the persistent reduction of rotavirus hospitalizations among children aged <2 years in whom 85% of severe rotavirus disease occurs.
This study has several limitations. First, our findings from the sentinel hospital surveillance system may not be generalizable to all children aged <5 years in El Salvador, given the limited number of participating hospitals and their locations. Nevertheless, the 5 departments in which the 7 sentinel hospitals are located encompass approximately 60% of the <5 population and an estimated 48% of diarrhea-related hospitalizations among children <5 years in El Salvador. Second, we based vaccine coverage estimates on a limited number of children enrolled as controls in a previous case-control study; therefore, we cannot be sure if this estimate accurately represents the level of vaccine coverage at a national level. Third, misclassification of vaccination status in a case-control study is possible, although confirmation of vaccination history through vaccination card or registry review helped to minimize this.3 Finally, the observed trends in national diarrhea-related healthcare visits may have been affected by trends in other pathogens, such as norovirus or bacterial enteric diseases. However, given that the sizeable declines in visits occurred specifically during rotavirus seasons and were consistent with those observed in the sentinel surveillance data, we are reassured that these national trends were specific to rotavirus disease.
In summary, significant and sustained reductions in rotavirus hospitalizations and diarrhea-related healthcare visits among children aged <5 years in El Salvador after implementation of rotavirus vaccination demonstrate both direct and indirect benefits of vaccination and the immense public health benefits of vaccination. Given the year-to-year variability in trends by age, continued surveillance is necessary to monitor the epidemiology of rotavirus disease and long-term impact of vaccination.
1.Guardado JA, Clara WAW, Turcios RM, et al. Rotavirus in El Salvador: an outbreak, surveillance and estimates of disease burden, 2000–2002. Pediatr Infect Dis J. 2004;23(10 suppl):S156–S160.
2.World Health Organization. Rotavirus vaccines. Wkly Epidemiol Rec. 2007;82:285–295.
3.de Palma O, Cruz L, Ramos H, et al. Effectiveness of rotavirus vaccination against childhood diarrhoea in El Salvador: case-control study. BMJ. 2010;340:c2825.
4.Ruiz-Palacios GM, Perez-Schael I, Velazquez FR, et al. Safety and efficacy of an attenuated vaccine against severe rotavirus gastroenteritis. N Engl J Med. 2006;354:11–22.
5.Clark HF, Lawley D, Mallette LA, et al. Decline in cases of rotavirus gastroenteritis presenting to the children's hospital of Philadelphia after introduction of a pentavalent rotavirus vaccine. Clin Vaccine Immunol. 2009;16:382–386.
6.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.
7.Curns AT, Steiner CA, Barrett M, et al. Reduction in acute gastroenteritis hospitalizations among US children after introduction of rotavirus vaccine: analysis of hospital discharge data from 18 US states. J Infect Dis. 2010;201:1617–1624.
8.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.
9.Tate JE, Panozzo CA, Payne DC, et al. Decline and change in seasonality of US rotavirus activity after the introduction of rotavirus vaccine. Pediatrics. 2009;124:465–471.
10.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.
11.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.
© 2011 Lippincott Williams & Wilkins, Inc.