Secondary Logo

Journal Logo

Vaccine Reports

The Burden of Acute Diarrheal Disease in Young Hospitalized Urban South African Children Five Years After Rotavirus Vaccine Introduction: A Retrospective Descriptive Study

Makgatho, Euphrasia MBChB, DCH*; Patel, Firuzan MSc, MED Child Health, Neurodevelopment*; Solomon, Fatima MFamMed†,‡; Groome, Michelle J. PhD†,‡; Lala, Sanjay G. PhD*,§; Vallabh, Preeteeben FCPaed*; Dangor, Ziyaad PhD*,†,‡

Author Information
The Pediatric Infectious Disease Journal: July 2019 - Volume 38 - Issue 7 - p 752-756
doi: 10.1097/INF.0000000000002304


Diarrhea is a leading cause of childhood mortality and morbidity worldwide.1 In 2015, diarrheal deaths accounted for 8.6% of under-5 mortality, and an estimated 957.5 million cases occurred in children less than 5 years of age globally1; additionally, the burden of diarrheal disease is disproportionately higher in low-income and low-middle–income countries.2–5 In South Africa, diarrheal disease is the second leading cause of premature death after HIV/AIDS.6 Children less than 2 years of age, those who are malnourished, and HIV infected are at a higher risk of morbidity and mortality due to diarrheal disease.1,2,4,7

Rotavirus infection is the leading cause of severe diarrhea in young children, accounting for approximately 29.3% of diarrheal deaths in children less than 5 years, globally.1 The rotavirus vaccine was introduced at 6 and 14 weeks into the South African expanded program on immunization in 2009.8 The coverage of the second dose of rotavirus vaccine has been above 85% in South Africa since 2011.9 Since the introduction of the rotavirus vaccine, studies have shown significant reductions in severe rotavirus-associated diarrhea, and more importantly, a 38% decline in all-cause diarrheal illness in both high-income and low-middle–income countries.10 Studies in South Africa have reported a 33%–57% reduction in all-cause diarrhea.11–13 Five years into the era of high rotavirus vaccine coverage (>85%), we determined whether the reductions in acute diarrheal disease burden were maintained in young children hospitalized with a diarrheal illness; additionally, we determined predictors of mortality in these children.


This was a retrospective review of a discharge summary database of children less than 5 years of age hospitalized with an acute diarrheal illness at the Chris Hani Baragwanath Academic Hospital (CHBAH) from January 2015 to December 2016. Until April 2014, when the primary-level Bheki Mlangeni District Hospital (BMDH) became operational, CHBAH was the only state funded pediatric hospital that served the population of Soweto and neighboring suburbs. The under-5 population served by CHBAH was estimated as 183,497 in 2015 and 183,764 in 2016 by Statistics South Africa (2013, personal communication). This population comprises mainly black Africans from low to middle income households.14

The standard-of-care for the management of children presenting with diarrhea in this hospital has remained unchanged over the study period. Children seen at primary healthcare clinics with diarrhea are managed according to the Integrated Management of Childhood Illness protocol, which includes the use of oral rehydration solution (ORS) and zinc supplementation. The national rate of use of ORS and Zinc is, however, only 51% and 37%, respectively, with only 28% of under-5s with diarrhea receiving both.15 Those children with dehydration and/or those who do not tolerate oral rehydration at the primary healthcare clinics are referred to CHBAH or BMDH for further management and/or hospitalization. BMDH is, however, a primary level hospital and there are only 32 pediatric beds and with only a third bed occupancy.16 In addition BMDH refers all complicated cases to CHBAH. In addition to the clinic and BMDH referrals, some children with diarrheal disease may present directly to CHBAH. The degree of dehydration, as assessed by the attending physician is the main criteria used to decide whether the child will be hospitalized; the hospital protocol describes the degree of dehydration as:1 mild dehydration—the child is alert with few clinical signs of dehydration: thirst, drinks eagerly and has a decreased urine output,2 moderate dehydration—the child has 2 of the following signs: sunken eyes, depressed anterior fontanel, absent tears, thirst/drinks eagerly, restless/irritable, a slow skin pinch but <2 seconds and sticky mucosa, and3 severe dehydration—the child has worsening of symptoms, reduced level of consciousness, markedly reduced urine output and may be in shock with barely palpable pulses, capillary refill time ≥3 seconds, cold mottled limbs. Children with moderate or severe dehydration are hospitalized.

The provision of antiretroviral therapy (ART) and the prevention of mother-to-child transmission program in South Africa has led to a reduction in HIV transmission rates from 3.6% in 2011 to 1.5% in 2016.17 Among women attending antenatal clinics, however, the prevalence of HIV has remained at approximately 30% and therefore almost one third of newborns in Soweto are HIV exposed (HE).17

Statistical Analysis

Data were extracted from a discharge summary database captured by the Respiratory and Meningeal Pathogens Research Unit. Children less than 5 years of age with any ICD 10 (A00–A09) discharge diagnosis of a diarrheal illness were included in the study. We excluded a discharge diagnosis of nosocomial diarrhea, chronic noninfective gastroenteritis and other specified noninfective gastroenteritis and/or colitis. Data were stratified by age, gender, HIV status and presence of a coexisting acute lower respiratory tract infection (aLRTI) (ICD10 codes: B25.0, B20.6, J13–J18, J21 and J22).

The HIV status of cases was confirmed using Respiratory and Meningeal Pathogens Research Unit database and National Health Laboratory Service TrakCare system (InterSystems, Cambridge, Massachusetts, USA). HIV exposure was defined as children born to HIV-infected women with either a negative HIV polymerase chain reaction result, or if the HIV polymerase chain reaction test result was unavailable.

The incidence was calculated as the number of diarrhea cases per 100,000 population estimates (Statistics South Africa, 2013, personal communication). Case fatality rates were reported as a proportion of the total number of diarrheal cases. Anthropometry software (World Health Organization AnthroPlus, version 1.0.4 WHO, Geneva, Switzerland) was used to calculate weight-for-age and length/height-for-age Z-scores.18 The Mann-Whitney U test was used to compare continuous variables, and the Chi-squared or Fisher exact test was used to compare categorical variables. A P value <0.05 was considered statistically significant. Variables that were found to have a P value <0.20 in a univariate analysis were then included into a multivariable logistic regression analysis, and adjusted odds ratios (aORs) and 95% CI were reported.

STATA version 13.0 (College Station, Texas USA) and STATISTICA version 13.2 (Dell Inc. Tulsa, USA) were employed for statistical analysis. Permission to conduct the study was obtained from the University of the Witwatersrand Human Research Ethics Committee (M170501).


Over the 2-year period, 15,942 children less than 5 years were admitted to the pediatric wards at CHBAH; of these, 2365 (14.8%) had diarrheal disease (Fig. 1). Approximately, 50.4% (n = 1193) of all diarrheal hospitalization occurred in infants (Table 1). More than one third of the study population were underweight for their age (826/2260 = 36.5%) and/or stunted (629/1801 = 34.9%). The median length of hospitalization was 2 days [interquartile range (IQR): 1–5] and the case fatality ratio was 2.9% (n = 68). The HIV status was known in 2015 (85.2%) of children; of these, 147 (7.4%) were HIV infected. The yearly incidence of diarrheal disease was 675.8 per 100,000 (95% CI: 638.8–714.3) in 2015 and 612.2 per 100,000 (95% CI: 577.0–648.9) in 2016. Most of the admissions (58.0%) occurred during the autumn and winter seasons (from May to August) (Fig. 2). Fewer admissions (n = 450; 19.0%) were observed during spring (September–November).

Demographic characteristics of children (n= 2365) hospitalized with diarrheal illness
Flow diagram of children hospitalized with diarrhea. ICD10 codes: A02: other salmonella infections; A03: shigellosis; A04.7: enterocolitis due to Clostridium difficile; A05: other bacterial food-borne intoxication, not elsewhere specified; A07.1: giardiasis (lamliasis); A07.2: cryptosporidiosis; A09: diarrhea and gastroenteritis of presumed infectious origin; A09.0, A09/B96.2: diarrhea and gastroenteritis due to Escherichia coli; K52.8: other specified noninfective gastroenteritis and colitis; K52.9, noninfective gastroenteritis and colitis unspecified; A09/Y95: nosocomial infection.
Seasonal pattern of diarrheal disease hospitalizations over the study period.

Children who died were younger (median age: 8.6 months; IQR: 3.6–15.1) than those who were discharged (12.2 months; IQR: 6.3–22.0; P = 0.002) (Table 2). Children less than 6 months of age were 6.9 (95% CI: 2.1–22.9) times more likely to die from diarrheal illness than older children (P < 0.002). Two thirds of children who demised were underweight (41/62; 66.1%) and/or stunted (13/19; 68.4%), but these anthropometric parameters were not significantly different when compared with survivors. Children diagnosed with a coexisting aLRTI had 3.7 (95% CI: 1.2–11.5; P = 0.021) fold greater odds of death. HIV-infected children had an increased odds of death (aOR: 9.1, 95% CI: 2.6–31.6; P = 0.001) (Table 2), but HE infants did not (P = 0.398).

Predictors of Mortality in Children Hospitalized With Diarrheal Illness

We further compared clinical characteristics between HE and HIV-unexposed (HU) infants less than 6 months of age because these infants were at the highest risk of diarrheal disease. Of 578 infants less than 6 months of age, 45 (7.8%) were HIV infected, 211 (36.5%) were HE, 293 (50.7%) were HU and 29 (5.0%) were unknown. The median weight-for-age and height-for-age Z-scores were lower in HE compared with HU infants; P < 0.001 and P = 0.046, respectively (Table, Supplemental Digital Content 1, HE infants, however, did not have a more prolonged hospital stay or more coexistent aLRTI compared with HU infants.

Coexisting aLRTI was diagnosed in 405 (17.1%) children with diarrheal illness (Table, Supplemental Digital Content 2, Children with aLRTI had a longer duration of hospitalization: median stay of 3 (IQR: 1–7) versus 1 (IQR: 1–4) day, P < 0.001. Children with aLRTI were significantly more underweight and stunted, P = 0.010 and 0.030, respectively. Overall, HIV coinfection was more common in children with an aLRTI (10.9% vs. 5.3%, P < 0.001).


Five years into an era of high-coverage rotavirus vaccination, acute diarrheal disease hospitalization is a major cause of hospital admissions (14.8% of total admissions) in children less than 5 years of age at this tertiary hospital. The incidence estimates for 2016 (612 per 100,000) are, however, 58% lower than the 2006–2008 prevaccine rates (1470 per 100,000) and 37% lower than reported in 2010 (970 per 100,000), the year after rotavirus vaccine introduction.13 Nonetheless, despite the declining trends, the incidence of diarrheal disease is 2-fold higher than in high-income countries.13,19 In our setting, the overall case fatality ratio from acute diarrheal cases (2.9%) was lower than the prerotavirus vaccine rate (3.5%), but higher than comparative global and South African estimates (2.0% and 2.2%, respectively) over the same period.2,7,20

Rotavirus vaccination, together with other interventions (eg, ORS, Zinc supplementation, improved sanitation, access to clean water, upscaling of prevention of mother-to-child transmission and ART), is likely to have reduced the burden of disease in our setting. Nonetheless, in children with diarrheal disease, those who are young, HIV infected and those with a coexisting aLRTI have an increased risk of death; these findings are similar to reports from other low-income and low-middle–income settings.5 In nonseverely malnourished Kenyan children with diarrheal disease, those who were younger or had abnormal respiratory signs were found to have a 1.7- and 3.6-fold higher risk of death, respectively.21

Approximately, one third of our patients were underweight and/or stunted, which is higher than the rates observed in other South African studies.22 In our study, however, we did not have access to the preillness weight, and therefore, some children may have been misclassified as being underweight. More importantly, however, one third were also stunted and two thirds of deaths occurred in children who were stunted, and these findings are consistent with reports describing increased mortality rates in malnourished children with diarrhea.23–25

This rate of HIV infection in our study was higher than the national HIV prevalence rate in children less than 5 years,26 and most likely represents a referral bias to a tertiary care hospital. Historically, HIV infection increases the risk of death from diarrhea 11-fold, and despite improved access to ART in our setting, we found a 9.1 increased odds of death in HIV-infected children.7,27 We were unable to correlate HIV viral loads and CD4+ lymphocyte counts in HIV-infected children with acute diarrheal disease—a study limitation—but this is an important area for further research. Of the children less than 6 months of age, over one third were HE, an observation consistent with antenatal HIV prevalence rates.17 An increased risk of mortality and morbidity is reported in HE compared with HU infants,28,29 and a number of reasons are postulated to account for the increased vulnerability in HE infants, including a lack of protective maternal antibody transfer to fetus, and an altered immune system in the infant.29,30 In our study, however, there were no significant differences in terms of length of stay (LOS), death, age at admission and coexisting aLRTI between the HE and HU infants. This observation could be attributed to the improved maternal access to early ART (>95% of pregnant women with HIV accessed antiretroviral treatment in 2016).31

Children with coexisting aLRTI were 3.7 (95% CI: 1.2–11.5) times more likely to die (Table 2 and Table, Supplemental Digital Content 2; These children were younger (P = 0.002) and had longer LOS (P < 0.001) than those without an aLRTI. Children who were HIV infected were also more likely to have an aLRTI. Several reports from low-income and middle-income countries report an increase in aLRTI incidence with or following a diarrheal illness,32,33 possibly because diarrheal disease and LRTI share common risk factors such as immunodeficiency, HIV infection, malnutrition and young age.

The vast majority of admissions occurred in the autumn–winter months with fewer admissions occurring in spring. Previous South African and international studies confirm winter peaks in diarrheal hospitalizations in children less than 5 years of age.13,19,34,35 These winter peaks were more pronounced in the prerotavirus vaccine era, whereas in the post vaccine era, the diarrheal disease season appears to start later with less pronounced peaks.13,19

This was a retrospective study and had several limitations. We were unable to determine the etiologic agents responsible for acute diarrheal disease because stool samples are not routinely collected. The percentage of underweight children may have been overestimated because preillness and postrecovery (ie, discharge) weights are not routinely recorded; nonetheless, the large proportion of stunting indicates that malnutrition is a significant problem in our study population. The duration and use of ART, HIV viral loads, level of maternal education, vaccination status and breast-feeding practices are not recorded in the database.

In conclusion, diarrheal disease is a significant cause of morbidity and mortality in our setting and measures aimed at preventing diarrheal illnesses should continue. Our study indicates higher mortality rates in children hospitalized with diarrhea who are less than 6 months of age, HIV infected and have coexisting aLRTI: decreasing the mortality rates in these high-risk children should be the focus of further studies.


1. Global Burden of Diseases Collaborators. 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.
2. Walker CLF, Rudan I, Liu L, et al. Global burden of childhood pneumonia and diarrhoea. Lancet. 2013;381:1405–1416.
3. WHO. Global Health Estimates for 2016: Disease burden by cause, Age, Sex, by country and by region, 2000–2016. 2018.Geneva, Switzerland: WHO.
4. UNICEF. One is Too Many: Ending Child Deaths from Pneumonia and Diarrhea. 2016.New York, NY: UNICEF.
5. Walker CLF, Perin J, Aryee MJ, et al. Diarrhea incidence in low- and middle-income countries in 1990 and 2010: a systematic review. BMC Public Health. 2012;12:220.
6. Mayosi BM, Benatar SR. Health and health care in South Africa–20 years after Mandela. N Engl J Med. 2014;371:1344–1353.
7. Groome MJ, Madhi SA. Five-year cohort study on the burden of hospitalisation for acute diarrhoeal disease in African HIV-infected and HIV-uninfected children: potential benefits of rotavirus vaccine. Vaccine. 2012;30(suppl 1):A173–A178.
8. The National Department of Health. Immunisation That Works The Vaccinator’s Manual. 2012:4th ed. Pretoria, South Africa: The National Department of Health; 27.
9. UNICEF. South Africa: WHO and UNICEF Estimates of Immunization Coverage: 2015 Revision. 2016:New York, NY: UNICEF; 1–17.
10. Burnett E, Jonesteller CL, Tate JE, et al. Global impact of rotavirus vaccination on childhood hospitalizations and mortality from diarrhea. J Infect Dis. 2017;215:1666–1672.
11. Groome MJ, Page N, Cortese MM, et al. Effectiveness of monovalent human rotavirus vaccine against admission to hospital for acute rotavirus diarrhoea in South African children: a case-control study. Lancet Infect Dis. 2014;14:1096–1104.
12. Msimang VM, Page N, Groome MJ, et al. Impact of rotavirus vaccine on childhood diarrheal hospitalization after introduction into the South African public immunization program. Pediatr Infect Dis J. 2013;32:1359–1364.
13. Groome MJ, Zell ER, Solomon F, et al. Temporal association of rotavirus vaccine introduction and reduction in all-cause childhood diarrheal hospitalizations in South Africa. Clin Infect Dis. 2016;62(suppl 2):S188–S195.
14. Statistics South Africa. Census 2011. 2012. Available at: Accessed March 10, 2019.
15. UNICEF. Estimates of child cause of death, Diarrhoea 2018;Available at: Accessed March 10, 2019.
16. Mashele S; Soweto Urban. Jabulani Hospital 40% empty, short of 166 staff. 2015. MAY 29, 2015. Available at: Accessed March 10, 2019.
17. South African National AIDS Council (SANAC). South Africa’s National Strategic Plan for HIV, TB and STIs 2017–2022. 2017. 6. Pretoria, South Africa: SANAC; Available at: Accessed: March 10, 2019
18. WHO. Global Database on Child Growth and Malnutrition. Available at: Accessed March 10, 2019.
19. Martinón-Torres F, Martinón-Torres N, Bouzón Alejandro M, et al. Acute gastroenteritis hospitalizations among children aged < 5 years before and after introduction of rotavirus vaccines: a hospital-based surveillance study in Galicia, Spain. Hum Vaccin Immunother. 2012;8:946–952.
20. Magasana V, Chirinda W. Massyn N, Peer N, English R, et al. Child under 5 years diarrhoea case fatality rate. In: District Health Barometer 2015/2016. 2016:Durban, South Africa: Health Systems Trust. 119–124.
21. Akech S, Ayieko P, Gathara D, et al.; Clinical Information Network authors. Risk factors for mortality and effect of correct fluid prescription in children with diarrhoea and dehydration without severe acute malnutrition admitted to Kenyan hospitals: an observational, association study. Lancet Child Adolesc Health. 2018;2:516–524.
22. Horwood C, Butler LM, Vermaak K, et al. Disease profile of children under 5 years attending primary health care clinics in a high HIV prevalence setting in South Africa. Trop Med Int Health. 2011;16:42–52.
23. Caulfield LE, de Onis M, Blössner M, et al. Undernutrition as an underlying cause of child deaths associated with diarrhea, pneumonia, malaria, and measles. Am J Clin Nutr. 2004;80:193–198.
24. Black RE, Victora CG, Walker SP, et al.; Maternal and Child Nutrition Study Group. Maternal and child undernutrition and overweight in low-income and middle-income countries. Lancet. 2013;382:427–451.
25. Rice AL, Sacco L, Hyder A, et al. Malnutrition as an underlying cause of childhood deaths associated with infectious diseases in developing countries. Bull World Health Organ. 2000;78:1207–1221.
26. Shisana O, Rehle T, Simbayi LC, et al. South African National HIV Prevalence, Incidence and Behaviour Survey, 2012. 2014.Cape Town, South Africa; HSRC Press.
27. Thea DM, St Louis ME, Atido U, et al. A prospective study of diarrhea and HIV-1 infection among 429 Zairian infants. N Engl J Med. 1993;329:1696–1702.
28. Marinda E, Humphrey JH, Iliff PJ, et al.; ZVITAMBO Study Group. Child mortality according to maternal and infant HIV status in Zimbabwe. Pediatr Infect Dis J. 2007;26:519–526.
29. Kurewa EN, Gumbo FZ, Munjoma MW, et al. Effect of maternal HIV status on infant mortality: evidence from a 9-month follow-up of mothers and their infants in Zimbabwe. J Perinatol. 2010;30:88–92.
30. Abu-Raya B, Kollmann TR, Marchant A, et al. The immune system of HIV-exposed uninfected infants. Front Immunol. 2016;7:383.
31. UNAIDS. UNAIDS Data 2017 Reference. 2017:Geneva, Switzerland: UNAIDS; 40.
32. Schmidt WP, Cairncross S, Barreto ML, et al. Recent diarrhoeal illness and risk of lower respiratory infections in children under the age of 5 years. Int J Epidemiol. 2009;38:766–772.
33. Walker CLF, Perin J, Katz J, et al. Diarrhea as a risk factor for acute lower respiratory tract infections among young children in low income settings. J Glob Health. 2013;3:010402.
34. Mapaseka SL, Dewar JB, van der Merwe L, et al. Prospective hospital-based surveillance to estimate rotavirus disease burden in the Gauteng and North West Province of South Africa during 2003-2005. J Infect Dis. 2010;202(suppl):S131–S138.
35. Steele AD, Peenze I, de Beer MC, et al. Anticipating rotavirus vaccines: epidemiology and surveillance of rotavirus in South Africa. Vaccine. 2003;21:354–360.

diarrhea; rotavirus vaccine

Supplemental Digital Content

Copyright © 2019 Wolters Kluwer Health, Inc. All rights reserved.