Diarrhea is a leading cause of childhood morbidity and mortality.1–4 Rotavirus, a pathogen associated with severe gastroenteritis, was the leading cause of global childhood diarrheal deaths in 2015, despite the licensure and increased use of rotavirus vaccines since 2006.5,6
Two oral rotavirus vaccines commonly used today [Rotarix (RV1, GlaxoSmithKline Biologicals, Wavre, Belgium) and RotaTeq (RV5, Merck & Co., Kenilworth, New Jersey, USA)] were licensed in 2006, followed by a global recommendation of use in 2009.5,7,8 Rotavirus vaccination has provided >80% protection against severe rotavirus illness in several middle- and high-income countries; however, lower efficacy and effectiveness have been observed in low- and middle-income countries.9 Hypotheses for attenuated effectiveness include that underlying health conditions (including environmental enteric dysfunction, helminth infection and poor nutritional status) may reduce oral vaccine immune response, or that natural and vaccine-derived immunity is insufficient to combat the diversity of strains and high rotavirus incidence in endemic lower-income settings.10–14 Water and sanitation improvements may diminish barriers to rotavirus vaccine efficacy;15,16 indeed, evidence suggests improved water, sanitation and hygiene conditions can improve seroconversion of oral rotavirus vaccination.17 We therefore hypothesized that water and sanitation conditions might modify the impact of rotavirus vaccination on childhood diarrhea rates.
Peru added RV1, administered in 2 doses at ages 2 and 4 months, to its national vaccine schedule in 2009.18–20 Before vaccine introduction, rotavirus was the leading cause of severe diarrhea in Peruvian children under 5 years of age, responsible for 64,000 annual outpatient visits, with 63% of children experiencing at least 1 episode of rotavirus diarrhea by age 5.21 The decade of vaccine introduction was characterized by rapid economic development and poverty declines, with improvements in water and sanitation access, although changes varied geographically.22,23
Rotavirus diarrhea significantly decreased in a peri-urban community in Lima in the 2 years after vaccine introduction.24 In Loreto, the rotavirus vaccine successfully prevented rotavirus diarrhea in infants, but protection was not sustained for ages 1–2 years.18 Heterogeneity in vaccine impact may exist across settings within Peru, although this has yet to be rigorously evaluated across geographic regions.
Our research utilizes a highly spatially and temporally resolved dataset of clinic visits for diarrhea in Peru from 2005 to 2015, as well as extensive data on piped water, sewerage and poverty. We examine long-term trends to understand the impact of routine rotavirus vaccination in Peru, and to explore how sociodemographic factors modify this impact.
MATERIALS AND METHODS
Geographic and Temporal Scope
Peru is comprised of 25 regions, which encompass 195 provinces, which are further sub-divided into districts. Each district belongs to a single province, allowing for aggregation of district data to the provincial level. Analyses were conducted for the 195 provinces of Peru covering January 2005–December 2015. Data from districts assigned to a new (196th) province in 2014 were analyzed with their former province.
Data Sources and Definitions
The Peruvian Ministry of Health (MINSA, Ministerio de Salud), in collaboration with the National Epidemiology Network, collects obligatory weekly surveillance reports on diarrhea visits from all public inpatient and outpatient clinics in Peru. Private clinics belonging to the National Epidemiology Network also send weekly reports. Acute diarrhea cases refer to patients presenting to a clinic with an increased frequency of bowel movements (≥3 in 24 hours), or in fluidity or volume of stool compared to usual, with onset within the past 2 weeks. Case reports are aggregated by age group (<1 year, 1–4 years and ≥5 years old), and assigned to the patient’s district of residence. We use the term “childhood diarrhea rate” to describe the rate of clinic visits for diarrhea in children under 5 years old.
MINSA provided district-level counts of the first and second doses of rotavirus vaccine given to infants in Peru from 2008 to 2015 and census-derived estimates of the annual infant (<1 year) population. Coverage increased quickly after the national vaccine introduction in 2009: by 2010, most infants were receiving both doses of RV1. Based on vaccine administration data, we classified 2005–2009 as the “pre-(rotavirus) vaccine era” and 2010–2015 as the “post-vaccine era”, or “rotavirus vaccine era”. We considered eras rather than estimates of the percentage of infants vaccinated due to difficulties in obtaining stable provincial-level RV1 coverage estimates.
Water, Sanitation and Population
The National Institute of Statistics and Informatics (INEI, Instituto Nacional de Estadística e Informática) conducted national censuses in 2007 and 2017. Participation was obligatory. El Sistema de Focalización de Hogares (SISFOH, the Household Focalization System) conducted an interim census from February 2012 to September 2013. Participation was voluntary and enumerated respondents were limited to those residing in their households for at least 6 months. SISFOH included approximately 24 million residents compared with 27 million in the 2007 census.25
Population by age was enumerated for each district during the 2007 and 2017 censuses. Province-level child population estimates used for determining the diarrhea rate were imputed for non-census years assuming a linear population change between censuses.
SISFOH and the national census collect data on the main source of household drinking water, and on household sanitation. The percentages of households with piped water access (households for which the primary drinking water source was water piped inside or outside of the home, but within the building area) and sewerage access (households for which the method of excreta disposal is a toilet connected to a piped sewerage system) were calculated for each province in years that census data were available (additional definitions in Text, Supplemental Digital Content 1; http://links.lww.com/INF/D928). Data from SISFOH were used as annual estimates for 2012. Data for interim years were imputed for each province from available data, assuming a linear change in percentage access between data points. The continuous percentages of province households with piped water, as well as the percentages with sewerage access, were split into quartiles, based on the annual estimates for 2005–2015. This categorization allowed us to maintain adequate data in each group in the pre- and post-vaccine eras, and does not impose a linear relationship between coverage and the childhood diarrhea rate. Each quartile included 25% of the observations of the annual province-level percentages of access to piped drinking water/sewerage. For example, the first (lowest) quartile of piped water access included observations in which 0%–40% of households in the province had access to piped water that year. The definitions of the quartiles are displayed in Table, Supplemental Digital Content 2; http://links.lww.com/INF/D929.
INEI calculated the percentage of households in each district falling below the poverty line for the years 2007, 2009 and 2013. Households were considered below the poverty line if monthly per capita expenditure was below the amount needed to acquire goods and services adequate to satisfy basic needs.25–27 Provincial-level estimates of the percentage of households in poverty were calculated using population-weighted district-level data. Data for interim years were imputed assuming a linear change between reporting years. The continuous percentage of households in poverty was split into quartiles using the same methodology as water/sewerage access estimates.
The El Niño Southern Oscillation, or simply “El Niño”, is a global pattern of climate variability associated with unusual warming of the Pacific Ocean near the equator occurring approximately every 2–7 years.28 El Niño events are associated with increased temperature and changes in precipitation patterns in Peru. The US National Oceanic and Atmospheric Administration reports data on the Oceanic Niño Index (ONI), calculated using a standard 3-month mean of sea surface temperature anomalies in the Niño 3.4 region of the Pacific Ocean.29 El Niño periods were defined using the ONI, with values 0.5–0.9, 1.0–1.4 and ≥1.5 corresponding to weak, moderate and strong El Niño events, respectively.30 We compared 3-month periods with a weak El Niño or no El Niño to 3-month periods with moderate (October–December 2009, January–March 2010 and April–June 2015) or strong (July–September 2015 and October–December 2015) El Niño events.
Data analysis was performed using R 3.5.1 (R Foundation for Statistical Computing, Vienna, Austria) and SAS 9.4 (SAS Institute, Inc., Cary, NC). We compared the childhood diarrhea rate in the pre-vaccine and post-vaccine eras, controlling for long-term trend and El Niño seasons. We also accounted for: (1) access to piped water, (2) access to sewerage and (3) poverty levels in separate models. We considered each of these 3 variables as main effects and also assessed the interaction between each of these and the rotavirus vaccine era, to determine if any of these factors modified the association between ongoing rotavirus vaccination and the childhood diarrhea rate.
Specifically, we analyzed the quarterly (3-month) count of childhood clinic visits for diarrhea by province using negative binomial generalized estimating equations, after finding over-dispersion of the outcome in a Poisson model. We aggregated data in 3-month intervals to assess longer-term trends, to align our data with El Niño seasons and to limit the influence of potential autocorrelation between weekly cases. We ran 3 models to separately consider quartiles of piped water access, sewerage access and poverty, as these risk factors were highly correlated with one another. Each of these 3 interaction models included a binary term for the rotavirus vaccine era; 1 risk factor of interest (categorical quartile of piped water access, sewerage access, or poverty); the interaction between the first 2 variables, to assess whether the risk factors modified the rotavirus vaccine/diarrhea relationship; a continuous variable for year to control for secular trend; and an indicator for moderate/strong El Niño events. We also ran 3 main effects models without the interaction terms. We accounted for clustering at the province level with an exchangeable correlation structure. Models included a population offset for the natural log of person-weeks.
Some districts (<1%) did not participate in diarrhea surveillance for the entire study period. The child populations of districts that did not start contributing diarrhea surveillance data until after January 2005, or stopped reporting prior to December 2015, were not counted toward the provincial population during the weeks the district was not reporting, that is, we did not consider children from non-reporting districts to be at risk when not reporting.
Emory University’s Institutional Review Board reviewed the study protocol and determined the research was exempt from requiring human subjects research approval, given that the data were provided in an aggregated format and did not include personal identifiers.
The analysis ultimately included data from 194 Peruvian provinces. Ocros province was excluded due to inconsistent diarrhea surveillance reporting. Data were analyzed from 1838 districts, with <1% of districts (18 districts) having any missing diarrhea reports.
Childhood Diarrhea Rate
Nationally, there were 28.8 [95% confidence interval (CI): 28.7–28.9) annual clinic visits for diarrhea per 100 children <5 years old in 2005, and the rate generally decreased throughout the study period, both in periods of the pre-vaccine era (eg, 2006–2008) and post-vaccine era (Fig. 1). In 2009, the year during which the rotavirus vaccine was added to the national immunization schedule, there was an annual rate of 25.6 (95% CI: 25.5–25.7) clinic visits for diarrhea per 100 children. The lowest annual rate was in 2014 (18.9 diarrhea clinic visits per 100 children, 95% CI: 18.8–19.0). There was substantial variability in the childhood diarrhea rate between provinces (Figure, Supplemental Digital Content 3; http://links.lww.com/INF/D930).
Rotavirus Vaccine Coverage
Rotavirus vaccination began in 40% of provinces in 2008. By 2009, all but 3 provinces were administering rotavirus vaccines. By 2010, the majority of Peruvian infants (estimated 75%) were receiving both doses of RV1 (Figure, Supplemental Digital Content 4; http://links.lww.com/INF/D931). Coverage increased until 2012, then remained relatively stable through 2015.
Piped Water, Sewerage and Poverty
Access to piped water and sewerage generally increased over time, and poverty generally decreased, although estimates varied substantially between provinces (Table, Supplemental Digital Content 2; http://links.lww.com/INF/D929).
The study year (2005–2015) was significantly associated with the childhood diarrhea rates across all models, controlling for the other variables. There was an annual reduction of approximately 3% in the childhood diarrhea rate throughout the study period (2005–2015; Tables 1–2 and Table, Supplemental Digital Content 5; http://links.lww.com/INF/D932).
Controlling for secular trends and El Niño events, the childhood diarrhea rate was 6.7% lower in the post-vaccine era compared with the pre-vaccine era [incidence rate ratio (IRR): 0.93, 95% CI: 0.90–0.97]; in models considering piped water or sewerage, the association varied by quartile of water/sewerage access. Specifically, in the interaction model considering access to piped water (Table 1), we did not observe a lower childhood diarrhea rate in the post-vaccine era in the first (lowest) quartile of piped water access (ie, for provinces in which <40% of households had piped water). Compared with the pre-vaccine era, there were lower childhood diarrhea rates in the post-vaccine era for provinces in quartiles 2–4 of piped water access (Table 1 and Fig. 2). The interaction model considering sewerage was similar: we did not observe lower childhood diarrhea rates in the post-vaccine era in the lowest quartile of sewerage access, but the rate was lower in the post-vaccine era in sewerage quartiles 2–4 (Table 2 and Fig. 2).
In the main effect models, we did not observe an overall association between access to piped water (Table 1) or access to sewerage (Table 2) and the rate of clinic visits for childhood diarrhea; however, the associations differed between the pre- and post-vaccine eras. Higher piped water access was associated with a lower childhood diarrhea rate in the rotavirus vaccine era only (Table 1, interaction model). In that era, the incidence rate of childhood diarrhea clinic visits ranged from 9% to 18% lower in the second–fourth quartiles of piped water access, compared with the lowest piped water access quartile. Higher sewerage access also tended to be associated with a lower diarrhea rate in the rotavirus vaccine era (Table 2, interaction model). No association between poverty quartile and the childhood diarrhea rate was observed (Table, Supplemental Digital Content 5; http://links.lww.com/INF/D932).
Across models, the childhood diarrhea rates were 6% higher (IRR: 1.06, 95% CI: 1.04–1.09) during 3-month periods with moderate or strong El Niño events (Tables 1 and 2 and Table, Supplemental Digital Content 5; http://links.lww.com/INF/D932).
The implementation of a national rotavirus vaccine program in Peru was associated with a significantly lower rate of childhood clinic visits for diarrhea, controlling for long-term trend; however, benefits of vaccination were not evident in provinces in the lowest group of access to piped water or sewerage.
The variable for the study year was significantly associated with childhood diarrhea rates across models, even controlling for the other factors such as access to piped water and sewerage. The overall decreasing trend in childhood clinic visits for diarrhea may reflect health/demographic changes in Peru beyond ones we measured, such as parental education, malnutrition and hygiene education.
In other middle-income countries in Latin America, rotavirus vaccination reduced annual acute gastroenteritis cases by 17%–55% in children under 2 years old.9 While our analysis estimated lower overall reductions in childhood diarrhea, we analyzed data for all children under 5 years old, as data were aggregated in a way that precluded an analysis specifically among children <2 years. National vaccination for infants began in 2009, thus vaccination was unlikely to reach high levels among children 2–4 years of age until the last years of our study, although older children may have benefited from indirect effects of vaccination. Rotavirus infections are more common in children 0–2 years of age (compared with 3–4 years)20 and RV1 may have a larger impact in this younger age group.
There are several potential explanations for the greater reductions in the rate of clinic visits for childhood diarrhea observed from the pre-vaccine to post-vaccine era in provinces with higher access to piped water and sewerage. First, households without piped water may have worse drinking water quality and reduced water quantity for hygiene. Children living in these conditions, and without access to sewerage, could be predisposed to environmental enteric dysfunction, which may diminish oral vaccine impact.11,31 Second, the leading etiologies of diarrhea cases may differ based on piped water and sewerage coverage. Compared with bacterial diarrheagenic pathogens, person-to-person transmission may be a more important transmission route for rotavirus, with water having a lesser role.32,33 There could be more bacterial diarrheal infections in areas with low piped drinking water and sewerage access, where people rely on surface water and other potentially unsafe drinking water sources; thus, reducing viral diarrhea may not have had an appreciable impact on the childhood diarrhea rate. Third, areas with higher piped water and sewerage access may have better vaccine coverage/delivery. Despite unstable provincial-level estimates of the percentage of infants receiving rotavirus vaccines, we assessed whether province-level access to piped water or sewerage was associated with the percentage of infants receiving a second dose of RV1, and found that piped water access and rotavirus vaccine coverage were weakly correlated (Figure, Supplemental Digital Content 6; http://links.lww.com/INF/D933) and sewerage access and rotavirus vaccine coverage were moderately positively correlated (Figure, Supplemental Digital Content 7; http://links.lww.com/INF/D934).
Access to improved drinking water (especially piped water) and improved sanitation are generally associated with lower diarrhea rates.34 We did not observe a protective effect of piped water or sewerage access on the rate of childhood diarrhea clinic visits in the pre-vaccine era. Interpolated province-level estimates may not adequately capture effects of piped water and sewerage beyond the secular childhood diarrhea trend. Areas with higher access to piped water/sewerage may also have better clinic access, allowing a higher rate of care-seeking in clinics when children experience a diarrhea episode. It is also possible the water/sewerage improvements did not translate to reductions in diarrhea cases; indeed, several large recent trials have failed to find an association between improved water/sanitation conditions and child diarrhea.35–40 Piped water may not be microbiologically safe; furthermore, service may be intermittent, potentially leading to unsafe storage methods and/or supplementation with unimproved water sources.41 Likewise, access to sewerage may be insufficient to limit children’s exposure to human/animal feces.
In the post-vaccine era, higher piped drinking water access was associated with a lower diarrhea rate, and the childhood diarrhea rate tended to be lower in areas with higher access to sewerage. This difference in the impact of piped water and sewerage between the pre-vaccine and post-vaccine eras may be explained by shifting dominant diarrhea etiologies. Water and sanitation interventions may have been insufficient to interrupt transmission of diarrheal disease pathogens in the pre-vaccine era, but had a greater impact on the residual diarrhea burden after the introduction of RV1. This is in line with the hypothesis that water/sanitation interventions may have varying effects in settings where pre-intervention conditions, predominant exposure transmission routes and underlying etiologies of diarrheal disease differ.35
Childhood diarrhea rates were higher during moderate and strong El Niño seasons, in agreement with other assessments of El Niño and diarrhea in Peru,42–48 although this association was not previously described on a national scale.
We were unable to evaluate provincial-level percentages of infants who received both RV1 doses. The counts of second vaccine doses given to infants exceeded the estimated infant population in many provinces (Figure, Supplemental Digital Content 4; http://links.lww.com/INF/D931), perhaps resulting from an underestimated infant population, or overestimation of vaccine administration. Accurately capturing provincial-level infant populations is challenging with a decadal census, especially given the rapidly-changing population pyramid and high rates of internal and international migration in Peru.49
While having a national surveillance system for clinic visits for childhood diarrhea allowed us to conduct a spatially- and temporally-detailed analysis of childhood diarrhea trends including both urban and rural areas, there were limitations of using clinic data. Data on the total number of all-cause clinic visits throughout the study were unavailable, and clinic utilization likely varied across provinces and over time. It is possible that diarrhea access and reporting improved over the course of the study, in which case the reductions in enumerated clinic visits for diarrhea that we report may be less than reductions in childhood diarrhea cases warranting clinical care. There may have been worse access to clinics in areas with higher poverty, which may explain why the overall rate of clinic visits for childhood diarrhea was not higher in provinces with higher poverty. Children presenting to clinics with diarrhea are likely those with more severe illness than non-reported community cases, and potentially less severe than cases requiring hospitalization, so our results are not directly comparable to studies with different outcomes (such as diarrhea hospitalizations). The etiology of the diarrhea cases was unavailable, thus we were unable to estimate the impact of RV1 specifically on rotavirus diarrhea.
Provincial-level estimates of access to piped water, sewerage and poverty were estimated using a linear interpolation of data collected at 3 time points for each measure, which may not capture abrupt changes. High correlation between piped water access, sewerage and poverty led us to examine these factors in isolation and limited our ability to differentiate between their possible effects.
This analysis utilized an ecologic study design, which is well-suited to study large-scale impacts of population-level interventions.50 It is one of few studies to consider other risk factors for diarrheal disease in a national evaluation of rotavirus vaccination, and underscores the importance of considering modifying factors in such analyses. Our results suggest that water and sanitation conditions may operate synergistically with rotavirus vaccination to reduce childhood clinic visits for diarrhea in Peru. Therefore, implementation of rotavirus vaccination with lower provision of piped water (<40% of households) or sewerage (<17% of households) may reduce the health impact of vaccination. Additionally, improving access to piped water and sewerage may be important to address the residual burden of diarrheal diseases in the rotavirus vaccination era.
This study utilized data collected by the Peruvian Ministry of Health and the Peruvian National Institute of Statistics and Informatics (INEI). We acknowledge their contributions and thank their staff, as well as the national census enumerators, for their data collection efforts. We thank Gaspar Moran, Nancy Hidalgo and Hector Benavides Rullier of INEI for their guidance in data collection and interpretation. We thank Angela Rozo at Emory University, and Diego Fano and Vilma Tapia from Universidad Peruana Cayetano Heredia for assistance in project management and translation, and Mitchel Klein and Howard Chang (Emory University) for epidemiologic and statistical guidance.
1. Wang H, Bhutta ZA, Coates MM, et al.Global, regional, national, and selected subnational levels of stillbirths, neonatal, infant, and under-5 mortality, 1980–2015: a systematic analysis for the Global Burden of Disease Study 2015. Lancet. 2016;388:1725–1774.
2. Kassebaum NJ, Arora M, Barber RM, et al.Global, regional, and national disability-adjusted life-years (DALYs) for 315 diseases and injuries and healthy life expectancy (HALE), 1990–2015: a systematic analysis for the Global Burden of Disease Study 2015. Lancet. 2016;388:1603–1658.
3. Troeger C, Colombara DV, Rao PC, et al.Global disability-adjusted life-year estimates of long-term health burden and undernutrition attributable to diarrhoeal diseases in children younger than 5 years. Lancet Glob Health. 2018;6:e255–e269.
4. Institute for Health Metrics and Evaluation (IHME). Findings from the Global Burden of Disease Study 2017. 2018.Seattle, WA, United States: IHME;
5. Troeger C, Forouzanfar M, Rao PC, et al.Estimates of global, regional, and national morbidity, mortality, and aetiologies of diarrhoeal diseases: a systematic analysis for the Global Burden of Disease Study 2015. Lancet Infect Dis. 2017;17:909–948.
6. Rotavirus Immunization Coverage Estimates by Country. 2018. World Health Organization; Available at: http://apps.who.int/gho/data/node.main.ROTACn
. Accessed February 14, 2019.
7. World Health Organization. Meeting of the immunization Strategic Advisory Group of Experts, April 2009 – conclusions and recommendations. Wkly Epidemiol Rec. 2009;84:220–236.
8. de Oliveira LH, Danovaro-Holliday MC, Matus CR, et al.Rotavirus vaccine introduction in the Americas: progress and lessons learned. Expert Rev Vaccines. 2008;7:345–353.
9. Patel MM, Glass R, Desai R, et al.Fulfilling the promise of rotavirus vaccines: how far have we come since licensure? Lancet Infect Dis. 2012;12:561–570.
10. Lopman BA, Pitzer VE, Sarkar R, et al.Understanding reduced rotavirus vaccine efficacy in low socio-economic settings. PLoS One. 2012;7:e41720.
11. Lagos R, Fasano A, Wasserman SS, et al.Effect of small bowel bacterial overgrowth on the immunogenicity of single-dose live oral cholera vaccine CVD 103-HgR. J Infect Dis. 1999;180:1709–1712.
12. Cooper PJ, Chico ME, Losonsky G, et al.Albendazole treatment of children with ascariasis enhances the vibriocidal antibody response to the live attenuated oral cholera vaccine CVD 103-HgR. J Infect Dis. 2000;182:1199–1206.
13. Ahmed T, Arifuzzaman M, Lebens M, et al.CD4+ T-cell responses to an oral inactivated cholera vaccine in young children in a cholera endemic country and the enhancing effect of zinc supplementation. Vaccine. 2009;28:422–429.
14. Santos N, Hoshino YGlobal distribution of rotavirus serotypes/genotypes and its implication for the development and implementation of an effective rotavirus vaccine. Rev Med Virol. 2005;15:29–56.
15. Prüss-Ustün A, Bartram J, Clasen T, et al.Burden of disease from inadequate water, sanitation
and hygiene in low- and middle-income settings: a retrospective analysis of data from 145 countries. Trop Med Int Health. 2014;19:894–905.
16. Acosta AM, Chavez CB, Flores JT, et alThe MAL-ED Network Investigators. The MAL-ED study: a multinational and multidisciplinary approach to understand the relationship between enteric pathogens, malnutrition, gut physiology, physical growth, cognitive development, and immune responses in infants and children up to 2 years of age in resource-poor environments. Clin Infect Dis. 2014;59(suppl 4):S193–S206.
17. Church JA, Rukobo S, Govha M, et al.The impact of improved water, sanitation
, and hygiene on oral rotavirus vaccine immunogenicity in Zimbabwean infants: substudy of a Cluster-randomized Trial. Clin Infect Dis. 2019;69:2074–2081.
18. Mohan VR, Karthikeyan R, Babji S, et alEtiology, Risk Factors, and Interactions of Enteric Infections and Malnutrition and the Consequences for Child Health and Development (MAL-ED) Network Investigators. Rotavirus infection and disease in a multisite birth cohort: results from the MAL-ED Study. J Infect Dis. 2017;216:305–316.
19. Schweitzer A, Pessler F, Akmatov MKImpact of rotavirus vaccination on coverage and timing of pentavalent vaccination – Experience from 2 Latin American countries. Hum Vaccines Immunother. 2016;12:1250–1256.
20. Linhares AC, Stupka JA, Ciapponi A, et al.Burden and typing of rotavirus group A in Latin America and the Caribbean: systematic review and meta-analysis. Rev Med Virol. 2011;21:89–109.
21. Ehrenkranz P, Lanata CF, Penny ME, et al.Rotavirus diarrhea disease burden in Peru: the need for a rotavirus vaccine and its potential cost savings. Rev Panam Salud Publica. 2001;10:240–248.
22. The World Bank. Country Overview: Peru. 2018. Washington, DC: The World Bank; Available at: http://www.worldbank.org/en/country/peru/overview
. Accessed February 16, 2019.
23. Eagin B, Graham JPA study of water and sanitation
access trends in Peru: where do inequities persist? J Water Sanit Hyg Dev. 2014;4:499–508.
24. Chang MR, Velapatiño G, Campos M, et al.Rotavirus seasonal distribution and prevalence before and after the introduction of rotavirus vaccine in a peri-urban community of Lima, Peru. Am J Trop Med Hyg. 2015;92:986–988.
25. Instituto Nacional de Estadística e Informática. Mapa de Pobreza Provincial y Distrital 2013. 2015. Lima, Peru: Instituto Nacional de Estadística e Informática; Available at: https://www.inei.gob.pe/media/MenuRecursivo/publicaciones_digitales/Est/Lib1261/Libro.pdf
. Accessed July 28, 2018.
26. Instituto Nacional de Estadística e Informática. Mapa de Pobreza Provincial y Distrital 2007: Metodología, Anexos Metodológicos. 2009. Lima, Peru: Dirección Técnica de Demografía e Indicadores Sociales del Instituto Nacional de Estadística e Informática; Available at: https://www.inei.gob.pe/media/MenuRecursivo/publicaciones_digitales/Est/Lib0911/index.htm
. Accessed February 3, 2018.
27. Instituto Nacional de Estadística e Informática. Mapa de Pobreza Provincial y Distrital 2009: El Enfoque de La Pobreza Monetaria. 2010. Lima, Peru: Dirección Técnica de Demografía e Indicadores Sociales del Instituto Nacional de Estadística e Informática; Available at: https://www.inei.gob.pe/media/MenuRecursivo/publicaciones_digitales/Est/Lib0952/Libro.pdf
. Accessed February 3, 2018.
28. Guimarães Nobre G, Muis S, Veldkamp TIE, et al.Achieving the reduction of disaster risk by better predicting impacts of El Niño and La Niña. Prog Disaster Sci. 2019;2:100022.
29. National Weather Service Climate Prediction Center: Cold & Warm Episodes by Season. National Oceanic and Atmospheric Administration, National Weather Service, National Centers for Environmental Prediction Climate Prediction Center. 2018. Available at: https://origin.cpc.ncep.noaa.gov/products/analysis_monitoring/ensostuff/ONI_v5.php
. Accessed November 19, 2018.
30. Moore SM, Azman AS, Zaitchik BF, et al.El Niño and the shifting geography of cholera in Africa. Proc Natl Acad Sci USA. 2017;114:4436–4441.
31. Kosek MN, Mduma E, Kosek PS, et al.Plasma tryptophan and the kynurenine-tryptophan ratio are associated with the acquisition of statural growth deficits and oral vaccine underperformance in populations with environmental enteropathy. Am J Trop Med Hyg. 2016;95:928–937.
32. Sengupta PRotavirus: the challenges ahead. Indian J Community Med. 2009;34:279–282.
33. Glass RI, Bresee JS, Turcios R, et al.Rotavirus vaccines: targeting the developing world. J Infect Dis. 2005;192(suppl 1):S160–S166.
34. Wolf J, Prüss-Ustün A, Cumming O, et al.Assessing the impact of drinking water and sanitation
on diarrhoeal disease in low- and middle-income settings: systematic review and meta-regression. Trop Med Int Health. 2014;19:928–942.
35. Cumming O, Arnold BF, Ban R, et al.The implications of three major new trials for the effect of water, sanitation
and hygiene on childhood diarrhea and stunting: a consensus statement. BMC Med. 2019;17:173.
36. Patil SR, Arnold BF, Salvatore AL, et al.The effect of India’s total sanitation
campaign on defecation behaviors and child health in rural Madhya Pradesh: a cluster randomized controlled trial. PLoS Med. 2014;11:e1001709.
37. Null C, Stewart CP, Pickering AJ, et al.Effects of water quality, sanitation
, handwashing, and nutritional interventions on diarrhoea and child growth in rural Kenya: a cluster-randomised controlled trial. Lancet Glob Health. 2018;6:e316–e329.
38. Clasen T, Boisson S, Routray P, et al.Effectiveness of a rural sanitation
programme on diarrhoea, soil-transmitted helminth infection, and child malnutrition in Odisha, India: a cluster-randomised trial. Lancet Glob Health. 2014;2:e645–e653.
39. Rogawski McQuade ET, Platts-Mills JA, Gratz J, et al.Impact of water quality, sanitation
, handwashing, and nutritional interventions on enteric infections in rural Zimbabwe: the Sanitation
Hygiene Infant Nutrition Efficacy (SHINE) Trial. J Infect Dis. 2019;221:1379–1386.
40. Pickering AJ, Djebbari H, Lopez C, et al.Effect of a community-led sanitation
intervention on child diarrhoea and child growth in rural Mali: a cluster-randomised controlled trial. Lancet Glob Health. 2015;3:e701–e711.
41. Shaheed A, Orgill J, Ratana C, et al.Water quality risks of ‘improved’ water sources: evidence from Cambodia. Trop Med Int Health. 2014;19:186–194.
42. Lama JR, Seas CR, León-Barúa R, et al.Environmental temperature, cholera, and acute diarrhoea in adults in Lima, Peru. J Health Popul Nutr. 2004;22:399–403.
43. Bennett A, Epstein LD, Gilman RH, et al.Effects of the 1997-1998 El Niño episode on community rates of diarrhea. Am J Public Health. 2012;102:e63–e69.
44. Speelmon EC, Checkley W, Gilman RH, et al.Cholera incidence and El Niño-related higher ambient temperature. JAMA. 2000;283:3072–3074.
45. Ramírez IJCholera resurgence in Piura, Peru: examining climate associations during the 1997–1998 El Niño. GeoJournal. 2015;80:129–143.
46. Checkley W, Epstein LD, Gilman RH, et al.Effect of El Niño and ambient temperature on hospital admissions for diarrhoeal diseases in Peruvian children. Lancet. 2000;355:442–450.
47. Salazar-Lindo E, Pinell-Salles P, Maruy A, et al.El Niño and diarrhoea and dehydration in Lima, Peru. Lancet. 1997;350:1597–1598.
48. Ramírez IJ, Grady SCEl Niño, climate, and cholera associations in Piura, Peru, 1991–2001: a wavelet analysis. EcoHealth. 2016;13:83–99.
49. Morley SChanges in rural poverty in Perú 2004–2012. Lat Am Econ Rev. 2017;26:1.
50. Hertz-Picciotto I. Environmental Epidemiology. In: Modern Epidemiology. 2008:3rd ed. Philadelphia: Wolters Kluwer Health, Lippincott Williams & Wilkins; 598–619.