Introduction

Diarrhea is among the most common morbidities affecting children under 5 years old with an estimated 960 million cases worldwide in 2015. Diarrhea mortality rate has declined 39% between 2005 and 2015, but it still is the fourth leading cause of death in this age group, responsible for killing half a million children [1••]. More than 70% of these deaths occur in children 2 years of age and younger [2]. These figures are estimates of deaths in which diarrhea was the primary cause identified. However, as suggested by Ahmed et al. in a verbal autopsy study in Bangladesh, diarrhea may contribute to children mortality as a secondary cause which may make its impact even larger [3]. Low- and middle-income countries disproportionately bear the burden of diarrheal diseases. It is estimated that in 2015, diarrhea caused over 71 million DALYs of which 45 million were among children under 5 years old and 99% of these occurred in low- and middle-income countries [1••]. The African, South Asian, and South East Asian regions have the highest burden of diarrhea in the world [4] (Fig. 1).

Fig. 1
figure 1

Percentage of deaths attributable to diarrheal diseases among children under 5 years old in 2015. Modified with permission. Source: WHO and Maternal and Child Epidemiology Estimation Group (MCEE) provisional estimates 2015, as presented in UNICEF’s report: One is too many: Ending child deaths from pneumonia and diarrhea. ISBN: 978-92-806-4859-1

Children that survive diarrhea are confronted with long-term consequences that include malnutrition and developmental delays. A recent episode of diarrhea was associated with underweight, stunting, and wasting in a Burkina Faso study among children under 5 years [5]. A vicious cycle ensues where diarrheal diseases predispose children to malnutrition, and this in turn makes them more vulnerable to severe diarrhea and diarrhea-related mortality [6, 7]. In a recent birth cohort study by Haworth EJN et al., Nepali children exposed to diarrhea at an early age were 2.5 times more likely to have a disability 10 years later than children without early diarrhea [8]. Diarrhea early in life may have an independent effect on cognitive function that is not necessarily mediated by the effects on nutritional status [9]. Thus, diarrhea hinders children developing to their full potential and may affect future productivity and quality of life. In addition to these effects on the individual, diarrhea imposes a large burden on household and health system economics in low-income countries further perpetuating poverty [10]. As reported by Ngabo et al., the hospitalization of a child with diarrhea can cost as much as the family’s income for the entire month [11].

Acute diarrhea is caused by more than 20 different viral, bacterial, and parasitic pathogens. However, recent studies have provided insight about the entero-pathogens responsible for the highest burden and their differences in age and geographic distribution. According to the Global Enteric Multicenter Study (GEMS) conducted in Africa and Asia among children under 5 years old seeking medical care, the most common pathogens causing moderate to severe diarrhea were rotavirus, Cryptosporidium, enterotoxigenic E. coli producing heat stable toxin, and Shigella [12•]. In this study, heat-stable toxin producing enterotoxigenic E. coli and enteropathogenic E. coli were associated with increased mortality in children younger than 12 months, and Cryptosporidium was associated with increased mortality in children between 12 and 23 months old [12•]. In contrast, the Multisite Birth Cohort Study (MAL-ED), a community-based birth cohort that conducted intensive surveillance of diarrhea in South America, Africa, and Asia, showed that norovirus, rotavirus, Campylobacter, astrovirus, and Cryptosporidium cause the highest burden on children younger than 12 months. In the same study, Campylobacter, norovirus, rotavirus, astrovirus, and Shigella caused most of the cases of diarrhea among children 12 to 23 months of age [13••]. Overall, rotavirus, Shigella, adenovirus, and Cryptosporidium were associated with more severe diarrhea [13••]. Smaller studies in China, Senegal, and Turkey showed similar results and confirmed the notion that interventions directed to control a limited number of pathogens are likely to have a large impact on diarrhea incidence and mortality [14,15,16]. In addition, the emergence of multidrug-resistant bacterial enteropathogens in low-income countries underscores the need for effective prevention strategies for diarrheal diseases [17]. A recent study from Bihar in India showed high levels of resistance to penicillins, cephalosporins, and quinolones, including extended spectrum beta-lactamase producing enterotoxigenic E. coli, enteropathogenic E. coli, and enteroaggregative E. coli [18].

Lack of access to clean water and hygiene, poor sanitation, malnutrition, and non-exclusive breastfeeding are important risk factor for diarrhea in children. Pathogen-specific characteristics like virulence and infectivity also play an important role in diarrhea epidemiology [19]. Several interventions focusing on specific risk factors and pathogens have produced an estimated reduction of diarrhea incidence of 10.4% and diarrheal deaths of 20.8% in recent years [1••]. Most importantly, some of these approaches have shown that child mortality associated with diarrhea is preventable. In this manuscript, we will review interventions used to prevent diarrhea among children and their impact or potential to cause significant declines in disease burden.

Vaccines

The development of vaccines directed against enteropathogens causing the highest burden of diarrhea will likely achieve a large reduction in morbidity and mortality. A few of these vaccines are already commercially available and several others are under investigation (extensively reviewed in 20••, 21••). Several rotavirus and Vibrio cholerae vaccines are internationally available and endorsed by the World Health Organization for disease control. Norovirus and Shigella vaccine candidates are in advanced clinical development stages [20••, 21••].

Rotavirus Vaccines

In low-income countries, 80% of children are infected with rotavirus at least once before reaching 1 year of age [22]. Rotavirus diarrhea is associated with moderate to severe disease and death. In the MAL-ED cohort, almost 5% of diarrhea episodes in the community among children younger than 2 years old were attributed to rotavirus infection [13••]. However, between 10 and 27% of moderate to severe diarrhea episodes were attributed to rotavirus among children of the same age in the GEMS study, which included children with more severe diarrhea [12•]. In 2013, an estimated 37% of the diarrhea-associated deaths were caused by rotavirus [23]. Thus, vaccines that prevent rotavirus moderate to severe diarrhea and mortality have a large impact. Several rotavirus vaccines are currently offered commercially, but only two, Rotarix (GlaxoSmithKline Biologicals) and Rotateq (Merck & Company), have been extensively studied and are pre-qualified by the World Health Organization (WHO) [20] (Table 1). The effectiveness of these vaccines is similar and varies according to the region and local risk for diarrhea [24, 30, 31].

Table 1 Available rotavirus vaccines used in humans

The introduction of Rotarix and Rotateq vaccines is associated with significant reductions in rotavirus hospitalizations and diarrhea mortality. A systematic review of observational studies on the impact of rotavirus vaccine program implementation showed an overall 38% reduction in hospitalizations and 42% reduction in mortality due to acute gastroenteritis among children under 5 years of age. In the same age group, the reduction in rotavirus related hospitalizations was reduced 80% [32]. In Botswana, vaccine coverages between 76 and 90% in children up to 2 years old achieved reductions in gastroenteritis-related hospitalizations and deaths of 23 and 22%, respectively [33]. Armah et al. reported a reduction of hospital admissions between 28 and 48% among the first 3 years of vaccine introduction in Ghana [34]. Shah et al. estimated that 134,714 hospitalizations and 20,986 deaths were prevented in 2016 after the introduction of rotavirus vaccination in 33 African countries [35]. In addition, rotavirus vaccination is associated with significant cost prevention for the healthcare system in low- and high-income countries [36, 37]. Given the evidence on effectiveness and cost savings, the WHO recommends introducing rotavirus vaccine in all national vaccination programs since 2013 [22].

Vibrio cholerae Vaccine

V. cholerae causes severe acute diarrhea leading to rapid dehydration. Cholera has a regional distribution closely related to poverty and lack of access to clean water and sanitation. Cholera is known to cause large outbreaks, but also occurs as an endemic illness mainly in countries of Africa and South Asia. The estimated numbers of annual cholera cases and associated deaths are 2.9 million and 95,000, respectively [38].

WHO emphasizes that WASH interventions are essential for cholera control, but it also recommends cholera vaccination as a complement in high-risk endemic areas and in outbreaks [39]. Three oral cholera vaccines are available and qualified by WHO: Dukoral (Valneva), Shanchol (Shantha Biotechnics), and Euvichol (Eubiologics) [40] (Table 2). Dukoral is recommended for ages 2 years and up and has an efficacy between 78 and 85% against cholera and may provide short-term cross-protection against enterotoxigenic E. coli [44]. The other two vaccines, Shanchol and Euvichol, are recommended for ages 1 year and up and have an efficacy between 40 and 87%. Except for Vaxchora, all cholera vaccines are given in two dose regimens with 2–6 weeks in between administrations. However, a recent systematic review and meta-analysis suggested that a single dose of oral cholera vaccine provides protection against cholera in 87% of subjects at 2 months and 33% at 3 years [45]. These data support single dose use in outbreaks where urgency and logistics for deployment may prevent the use of 2 dose regimens. Based on cost, production capacity, and implementation potential, Shanchol and Euvichol were chosen to create an oral cholera vaccine global stockpile in 2013. This stockpile has delivered more than 7 million doses of oral cholera vaccine in 14 countries for humanitarian emergencies and outbreaks [44].

Table 2 Available V. cholerae vaccines used in humans

WASH

WHO estimates that 88% of diarrheal diseases worldwide can be attributed to unsafe water, inadequate sanitation, and poor hygiene. In 2015, 844 million people lacked access to basic drinking water services (drinking water free of pathogens or chemicals within 30-min round trip time), and 159 million people were still collecting drinking water from surface sources. For the same period, 2.3 billion lacked basic sanitation, and 892 million people were still practicing open defecation. Almost half of the people living in low- and middle-income countries completely lacked handwashing facilities [46]. Thus, significant numbers of the world’s population have no access to water, sanitation, and hygiene (WASH), and their countries are facing challenges in implementation of education and infrastructure.

A recent Cochrane database review by Clasen et al. evaluated the impact of interventions to improve water quality on diarrhea. As compared with interventions to improve water safety at the source, interventions at the point of use showed significant benefits in diarrhea prevention. Household chlorination, flocculation, and solar disinfection decreased the risk of diarrhea by 23, 31, and 38%, respectively. Filtration, especially with Biosand systems, showed the highest impact in reducing diarrhea risk by 52% [47•]. Filtration may be especially effective in preventing diarrhea by specific pathogen resistant to other interventions like Cryptosporidium [48].

Promotion of handwashing with soap and water decreases the risk of diarrhea by 25% [49, 50]. However, rates of handwashing before eating and after using the toilet are low and inconsistent in low- and middle-income countries. A study in 51 countries evaluating availability of handwashing materials as proxy for handwashing behaviors showed that between 0 and 96% of the households had soap and water in their handwashing facilities. The availability of materials was significantly more common in wealthy countries and urban settings [51]. In a study by Pengpid and Peltzer, among adolescents from 9 African countries, only 62% reported washing hands consistently before eating, 58% after using the toilet, and 35% using soap [52]. A similar study in three countries of the Pacific Micronesia showed handwashing rates before eating and after toileting of 65 and 80% [53]. Educational interventions aimed at increasing handwashing before eating, after using the toilet, and using soap achieve improvements in behaviors when infrastructure is available and soap is provided [54]. Studies to support the effectiveness of educational interventions to increase handwashing among children are scarce and their results on diarrhea risk reduction are inconsistent [55]. Gover et al. reported on a different approach to changing handwashing behavior based on modifying the environment rather than focusing on education alone. Their study in Bangladesh showed that construction of handwashing facilities in schools including environmental “nudges” conducive to handwashing was as effective as high-intensity education to maintain the behavior among children 5 months after the intervention [56].

A recent meta-analysis evaluating WASH interventions and risk of diarrhea included only one study regarding the use of latrines and failed to show any effect [49]. In contrast, an older meta-analysis showed that 11 of the 13 studies evaluating different interventions to improve excreta disposal showed an effect on diarrhea risk. However, the heterogeneity between studies precluded the calculation of pooled effects [57]. Other studies have failed to show a difference in diarrhea risk in communities where improved sanitation interventions were implemented compared to those without them [58, 59]. The variety of interventions aimed at improving excreta disposal and the inability to isolate their effects from other WASH interventions may explain the variability of the results.

The calculations of cost-benefit of combined sanitation and safe water interventions estimated an average ratio of dollar returned/dollar invested of 4.3 and a cost for universal access to water and sanitation of 35 billion dollars. In addition to the benefits in diarrhea prevention, improvements in safe water, hygiene, and sanitation access decrease soil-transmitted helminth infections and trachoma and improves nutrition in children adding to the cost-benefit estimations [60,61,62]. Data on combination of the most effective WASH intervention and the integration of these in existing health promotion programs is lacking. Implementation of interventions may find economic, political, and cultural barriers.

Breastfeeding

Breastfeeding has been recognized as a cost-effective strategy to reduce infant diarrhea worldwide. Several studies have documented the impact on reduction of diarrhea rates of early initiation of breastfeeding, exclusive breastfeeding, and duration of breastfeeding. However, rates of exclusive breastfeeding among children younger than 6 months in low- and middle-income countries are 37%, and only 63% receive any breast milk between ages 6 and 23 months [63•]. A prospective study in Vietnam among 1049 6-month-old infants found that exclusive breastfeeding at 6 weeks of age significantly reduced the odds of inpatient admission for diarrheal illness (OR 0.37, 95% CI 0.15 to 0.88) [64]. Khan et al. reported that infants that interrupted exclusive breastfeeding before 6 months of age were between 2 and 5 times more likely to get diarrhea than infants that continued breastfeeding. In this study, the authors estimated that 27% of diarrhea episodes could have been prevented if exclusive breastfeeding would not have been interrupted [65]. A study in 9 sub-Saharan African countries with high diarrhea-associated mortality demonstrated that infants who initiated breastfeeding within the first hour of life and those who continued exclusive breastfeeding until 6 months of age were 20 and 50% less like to have diarrhea respectively after adjusting for demographic and socio-economic factors. These effects were maintained even after adjusting for differences in sanitation [66]. A review of Brazilian studies published in Portuguese and English languages showed that interventions aimed at promoting good breastfeeding practices (i.e., lactation centers) resulted in reductions of diarrhea rates and hospital admissions. Similarly to other studies, there were higher diarrhea rates and nutritional deficits among prematurely weaned infants [67]. Acharya et al. demonstrated that the diarrhea-protective effects of breastfeeding were annulled when other liquids or food was introduced among infants in Nepal. In this study, partial breastfeeding was associated with 3.5 times more diarrhea compared with exclusive breastfeeding [68]. In a meta-analysis of observational studies, Shen et al. were unable to demonstrate an independent association between breastfeeding and rotavirus infection. However, the studies were highly heterogeneous and showed significant differences in diarrhea risk and definitions of rotavirus diarrhea [69].

Some of the protective effects of breastfeeding are thought to be related to maternal antibodies present in breast milk. Human milk oligosaccharides influence infant microbiome, functioning as prebiotics to support growth of specific bacteria [70]. Breast milk contains dendritic cells, T cells, and plasma cells with gut-related phenotypes specific for maternal gut flora [63•]. Exclusive breastfeeding prevents the ingestion of contaminated or improperly preserved beverages and food. Also, by avoiding the use of feeding bottles, it reduces the risk of cross-contamination of milk with unsafe water or improperly washed hands.

Victora et al. used an extensive compilation of systematic reviews to evaluate outcomes associated with breastfeeding. The authors estimated that scaling up exclusive breastfeeding in children under 6 months to 90% or more could prevent up to half of all diarrhea episodes and 72% of diarrhea-related admissions [63•]. Breastfeeding rates over 90% are very ambitious considering that very few countries worldwide have a prevalence of breastfeeding over 85%. Implementation of high rates of breastfeeding will be challenging due to high variability in resources and local feeding culture even within countries. Barriers to exclusive breastfeeding include type and place of delivery, perceptions about breast milk, complications of breastfeeding, maternal employment or student status, and breastfeeding practice support in the household, community, and healthcare facilities [71, 72]. Thus, educating mothers and the community on the importance and practice of breastfeeding and strengthening the health system to promote and support this practice may increase rates and have a broader impact on maternal and child health. Promotional interventions increase rates of breastfeeding. A review of 110 studies of breastfeeding interventions described a 43% increase in rates of exclusive breastfeeding at day 1 (RR 1.43, 95% CI 1.19–1.42), 30% at 1 month (RR 1.30, 95% CI 1.19–1.42), 90% at 6 months (RR 1.90, 95% CI 1.54–2.34), along with a decrease in rates of no breastfeeding. Individual and group counseling provided in healthcare facilities had the greatest impact on exclusive breastfeeding practices [73].

Nutrition

Several studies have linked malnutrition to an increased risk of diarrheal disease [74, 75]. Malnutrition is also a consequence of repeated episodes of diarrhea. In addition to the enteric function and microbiota changes, inflammation, nutrient malabsorption induced by diarrhea [76•], and the resistance from caregivers to continue feeding during diarrhea episodes increase the likelihood of malnutrition. A study on child-feeding practices including 390 caregivers in a peri-urban community of Lima in Peru showed that 73% of caregivers discontinued normal feeding during episodes of diarrhea and 40% withheld vegetables and fruits [77]. Continued feeding during episodes of diarrhea is associated with faster recovery of intestinal function and does not prolong duration of diarrhea. There is an opportunity for educational interventions targeting feeding practices during episodes of diarrhea.

Zinc is currently included in the WHO essential medicine list for diarrhea treatment and has been shown to be an effective component of diarrhea control. Zinc deficiency has been proposed to increase susceptibility to diarrhea due to this element’s role in regulation of intestinal fluid transportation and mucosal integrity as well as its role in the immune system [78]. A randomized clinical trial in 379 children younger than 5 years old in Iran showed lower frequency of diarrhea (4.5 vs 5.3 stools per day, P = 0.004) in the group receiving zinc compared to the group receiving standard of care. This study also showed shorter duration of hospital stay in the group receiving zinc supplementation [79]. In Cairo, Eskander et al. studied 80 children aged 2 to 30 months admitted with diarrhea. Zinc deficiency (levels less than 60 μg/dl) was present in 56% of cases of diarrhea, and low levels of zinc correlated with dehydration and duration of hospital stay [80]. Similarly, a study on 354 children with diarrhea in India showed that median serum zinc levels were significantly lower in children with dehydration than those without dehydration (69.64 vs 82.86, p < 0.001) [81]. A Cochrane systematic review which included 33 trials and 10,841 children showed that zinc supplementation could be of benefit in areas where the prevalence of zinc deficiency or malnutrition is high. However, the data did not support routine administration of zinc in children less than 6 months of age.

Children in low-resource countries are at increased risk of vitamin A deficiency, which affects approximately 190 million under 5 years old [82]. The prevalence of Vitamin A deficiency remains high in South Asia and sub-Saharan Africa [83]. Vitamin A supplementation has been studied in children with the intent to improve immune function and mortality associated with diarrhea as well as other infectious diseases. A Cochrane database review found that Vitamin A supplementation to children under 5 years of age was associated with a 15% reduction in diarrhea risk (RR 0.85, 95% CI 0.82 to 0.87) and 12% reduction in diarrhea mortality ((RR 0.88, 95% CI 0.79 to 0.98) [84].

Iron deficiency anemia is the most common nutrient deficiency worldwide, and iron supplementation and fortification is widely recommended in infants at risk of iron deficiency anemia. It is well documented that infants with iron deficiency anemia are at risk of poor cognitive, motor, and physiologic development. At the same time, several studies have reported an increased risk of infant diarrhea with iron supplementation. Iron-containing supplements can increase pathogen abundance, increase gut inflammation, and increase risk of diarrhea in infants receiving supplementation [85, 86•]. This leads to a complex relationship between nutrition, iron deficiency anemia, and risk of infection. Long-term studies on the risks and benefits of iron supplementation are needed.

There is little data to support routine use of probiotics in infant diarrhea. A recent review of evidence noted decreased duration of diarrhea by 1 day with no changes in stool output, with the most significant evidence for prevention of antibiotic associated diarrhea, (decrease from 23 to 9.6%, RR 0.48, 0.26 to 0.89) and no evidence to support that probiotics use decreases risk of infection [87].

Conclusions

Childhood diarrhea remains an urgent health priority and a leading cause of morbidity and mortality worldwide. While there have been significant advances in this field, lower income countries and lower income populations within countries remain disproportionately affected. Vaccine development and implementation, particularly of rotavirus, exclusive breastfeeding, and widespread efforts in improvement of sanitation and education have been shown to be successful strategies in prevention of childhood diarrhea. Further research should be directed at reductions of inequalities in implementing these practices and should evaluate the effectiveness of combining interventions for the prevention of diarrhea.