Diarrheal Diseases

Robert E. Black

Christa L. Fischer Walker

Claudio F. Lanata

INTRODUCTION

Diarrheal diseases are an important global problem, causing high rates of morbidity and mortality, particularly in developing countries.¹ Diarrhea occurs most frequently in conditions of poor environmental sanitation and hygiene, inadequate water supply, poverty, and limited education. Over the last century, many countries experienced economic growth and overall environmental conditions improved dramatically, which in turn decreased the burden of diarrheal diseases in developed countries.² As middle-income and developing countries continue to undergo similar economic, social, and epidemiologic transitions, it is likely that the burden of diarrheal disease will continue to decline. Unfortunately, the benefits of economic and social transitions are slow to reach the poorest of the poor, among whom high rates of diarrhea morbidity and mortality remain problematic, especially among children younger than 5 years of age.

Although infectious diarrheal diseases rarely cause death in economically developed countries, diarrhea continues to cause substantial morbidity, and the associated costs of treatment and lost wages continue to be problematic.³, ⁴ In addition, high rates of diarrhea in some settings, such as hospitals and daycare centers, and outbreaks of foodborne diarrhea have resulted in increased concerns about this disease.⁵, ⁶

From the beginning of the 1980s, substantial global efforts were directed at reduction of diarrheal disease mortality.⁷ It was recognized that dehydration as a result of diarrhea played a substantial part in the case-fatality rate for this illness and that such dehydration could be prevented or treated with oral fluid and electrolyte replacement, along with continued feeding. This low-tech approach made effective therapy more widely available through national diarrheal disease control programs in developing countries. Similar clinical approaches have also improved the management of diarrhea in the United States, resulting in improved therapy and a reduction in diarrhea-related mortality. Now, with the introduction of zinc supplementation for the treatment of diarrhea, there is yet one more inexpensive and simple tool for preventing diarrhea mortality. In addition, increased efforts are being directed at prevention of diarrhea through research on the etiology, epidemiology, and transmission patterns of diarrheal diseases.

GENERAL EPIDEMIOLOGY DEFINITIONS

Diarrhea is a symptom complex characterized by stools of decreased consistency and increased number. Although it is possible to define diarrhea as the occurrence of these symptoms simply in comparison to that individual’s prior bowel pattern, epidemiologic studies have generally used a more generalizable definition.⁸ Most studies now define diarrhea as three or more loose or watery stools during a 24-hour period. At least 2 days free of diarrhea are usually required to define an episode as terminated. Dysentery is a diarrheal disease defined by the presence of blood in loose or liquid stools.

Although most diarrheal episodes resolve within a week, a small proportion continue for 2 weeks or more.⁹ Studies in many countries show that the distribution of episode durations is continuous, but skewed toward the longer durations. Thus any definition of “persistent” diarrhea is an arbitrary threshold but a useful indicator of severe disease for diarrhea control programs. The World Health Organization (WHO) operationally defines persistent diarrhea as an episode that lasts for at least 14 days.¹⁰, ¹¹ The term persistent diarrhea, as used by WHO, encompasses episodes that begin acutely and continue for longer
than their expected duration; it is not intended to include infrequent diarrheal disorders, such as hereditary syndromes, gluten-sensitive enteropathy, or other noninfectious conditions.

SOURCES OF DATA

Data to describe the epidemiology of diarrheal diseases come from three main sources: (1) prospective studies in households and health facilities, (2) passive or active surveillance systems, and (3) outbreak investigations. Cross-sectional surveys may also provide limited types of information.

Prospective studies in households in developing countries have generally used visits by health workers at an interval of no more than one week to collect data on symptoms and to obtain specimens of stool for etiologic testing. Such studies in more developed countries often use a combination of household visits and telephone contacts. Prospective studies in health facilities may be done in general outpatient clinics, hospital wards, or special populations; an example of the last is studies of nosocomial infections in newborn nurseries.

Passive surveillance is based on routine reports of a specified disease to public health officials by healthcare workers. In active surveillance, cases are identified by the researcher querying health workers or laboratories about a specific disease. Passive surveillance can be done on a wide scale and is relatively inexpensive, but its value may be limited by reporting that is incomplete, biased, and delayed. Active surveillance of selected diseases might be appropriate if more complete and timely information is needed. In general, developing countries do not have useful passive or active surveillance as a routine system for diarrheal diseases; disease surveillance systems are weak and rates of care-seeking for diarrheal disease are low. In the United States, surveillance systems exist for only a limited number of diarrheal disease pathogens that are reported through laboratory-based passive surveillance.

Investigation of outbreaks of diarrheal diseases can be very useful to rapidly develop information on the risk factors, transmission patterns, and control measures for enteric pathogens. Such investigations can lead to controls to contain that outbreak, as well as help develop measures to prevent future outbreaks.

In developing countries, cross-sectional surveys are commonly conducted to obtain information about health conditions and the use of health services. Although information on the presence of diarrhea in the respondent on the day of survey can be reported accurately, these data may be a biased indicator of the incidence of infectious diarrhea because the symptom of diarrhea is frequent, commonly mild, and may be affected by seasonal variations. Surveys of patients or their families may provide information on recent more clear and memorable events such as hospitalization or death from diarrheal diseases or medical care received for current episodes.

INCIDENCE

A summary of prospective, community-based studies in developing countries concluded that the median annual incidence of all diarrhea in children younger than 5 years of age was 3.4 episodes in 1990 and 2.9 episodes in 2010.¹² Diarrheal incidence has varied in the different settings in which it has been studied (Table 20-1). This variation could be due to methodologic differences, such as the definition of diarrhea, or surveillance techniques used in the study. In community-based studies, the incidence has been highest in studies that included a small number of children under surveillance and more frequent home visiting, suggesting that the other studies may have found lower rates because of underreporting.¹³ At the same
time, it is likely that actual differences in the incidence of diarrhea in different populations exist because of different environmental and host risk factors, as well as the relative frequency of various enteropathogens. Both methodologic- and setting-specific differences may also affect the distribution of diarrheal episode duration. Studies that have the most intensive and frequent surveillance are most likely to identify mild and short-duration episodes that might not otherwise be reported with less intensive case finding.

Table 20-1 Median Age-Adjusted Diarrhea Incidence Rates (Episodes per Child Year) for Children Under 5 Years of Age in Selected WHO Regions of the World

World Health Organization Region	Number of Prospective Studies Included in Estimate	Age-Adjusted Incidence Rate for Children 0-4 Years of Age
Sub-Saharan Africa	7	5.0
Latin America	14	5.0
Eastern Mediterranean	4	4.9
South-East Asia	6	3.1
Western Pacific	3	3.3
Global	33	3.5
Reproduced from Boschi-Pinto, C., C.F. Lanata, and R.E. Black, The Global Burden of Childhood Diarrhea in Maternal and Child Health Global Challenges, Programs, and Policies, J. Ehiri, Editor. 2009, p. 225-243 , with kind permission from Springer Science+Business Media B.V.

In developing countries, the incidence of diarrhea varies greatly with age. Generally the first 2 years of life have the highest incidence, followed by a decline with increasing age. Peak incidence is often at 6-11 months of age (Figure 20-1). The incidence in boys and girls is generally similar, or slightly higher in boys; however, in some countries boys may be taken to health facilities more often, giving the appearance of higher rates of diarrhea among that group.

In the United States, the Centers for Disease Control and Prevention (CDC) has estimated that there are 21-37 million episodes of diarrhea each year in children younger than 5 years of age; approximately 10% of these illnesses led to a visit to a physician.¹⁴ A prospective community-based study found an annual incidence rate of diarrhea in persons of all ages of 0.63 episodes per person-year of observation.¹⁵ The highest incidence occurred in infants, who had a rate of 1.43 episodes per person-year. Children, especially those attending daycare centers, have a higher incidence of diarrheal disease, due to a naïve immune system and person-to-person transmission within these settings. However, these rates may be declining, especially among children 6-23 months of age, as the widespread uptake of rotavirus vaccine decreases the incidence and severity of rotavirus diarrheal episodes.¹⁶

Figure 20-1 Annual age-specific incidence of diarrhea per 1000 person-years assessed by household surveillance. Reprinted from THE LANCET, Vol. 1, RE Black et al. Incidence and severity of Rotavirus and Escherichia coli Diarrhea in Rural Bangladesh. Page 142, © 1981, with permission from Elsevier.

Older children, adolescents, and adults typically have lower incidence rates of diarrhea than children younger than 5 years of age. In a systematic review of prospective and cross-sectional surveys in both developing and developed countries, incidence of diarrhea was less than 1 episode per person per year (Table 20-2).¹⁷ The elderly are often thought to be at particular risk of diarrheal diseases because of decreased immune function. This effect can be more pronounced in developed countries among adults in long-term care facilities with unsanitary conditions or increased risk of nosocomial transmission.

Another group of individuals from developed countries who are at increased risk of diarrhea are those who travel to developing countries.¹⁸ Numerous studies have demonstrated that approximately half of such travelers will develop diarrhea during a trip of approximately 2 weeks.

Table 20-2 Median Diarrhea Incidence Rates (Episodes per 100 Person-Years) by Age for Children > 5 Years of Age, Adolescents, and Adults in Selected Regions of the World

World Health	Number of Studies	Episodes per 100 Person-Years
Organization Region	Included in Estimate	5-14 Years	15-54 Years	≥ 55 Years
Americas	6	50.8	62.1	39.5
Eastern Mediterranean	3	88.4	88.4	66.4
Europe	3	56.4	37.2	78.7
South-East Asia	6	67.5	29.9	30.1
Western Pacific	5	26.5	44.3	50.4
Sub-Saharan Africa^*	NA	NA	NA	NA
^*No data were available from Sub-Saharan Africa. Walker, C.L. and R.E. Black (2010). Diarrhoea morbidity and mortality in older children, adolescents, and adults. Epidemiol Infect. 138(9): p. 1215-26. Reproduced with permission.

IMPACT OF DIARRHEA

It is well appreciated that diarrheal diseases are important causes of death in developing countries. In 2010, there were approximately 800,000 diarrheal-related deaths among children younger than 5 years of age.¹

This number represents a substantial reduction from the estimated 3.3 million diarrheal-related deaths per year in the1980s¹³ or from the 2.5 million diarrhea deaths per year in the 1990s.²⁰ The diarrheal mortality rate is highest in the first year of life.

The case-fatality rates in children in developing countries have been reported to range from 0.1% to 0.5% in settings as diverse as urban Central African Republic, rural Egypt, rural northern India, and rural Indonesia. Overall, the diarrheal case-fatality rate in children younger than 5 years of age in developing countries is estimated to be 0.2%. The case-fatality rate is highest in the youngest children. In rural India, the case-fatality rate for persistent diarrhea was reported to be 20 times higher than that for acute diarrhea.²¹

In more developed countries, the diarrhea mortality rate and the illness case-fatality rate are very low.³, ¹⁴ In developed countries, diarrhea deaths are rare among children; for example in the United States, fewer than 50 diarrhea deaths occur each year among children younger than 5 years of age.¹⁹

Diarrhea mortality among adults has not been widely studied in developing countries where vital registration data are not available to record the cause of death. In low- and middle-income countries, diarrhea mortality rates have been estimated to be in the range of 0.1-2.88/1000 person-years in African countries and 0.3-1.0/1000 person-years in Asian countries.¹⁷ Although diarrhea deaths among older children, adolescents, and adults are rare in developed countries, such disease remains an important cause of death among the elderly. In the United States in 2007, there were 6771 diarrhea deaths among older children, adolescents, and adults, with 82% of these deaths occurring in people 65 years of age and older.²²

In developing countries, the infectious diseases of childhood have an adverse effect on growth. Diarrheal diseases have the greatest effect of all the infectious diseases, possibly because of the concomitant reduction in appetite, altered feeding practices, and decreased absorption of nutrients, along with the very high prevalence of diarrhea in young children in these settings.²³ The magnitude of the effect of diarrheal diseases on growth seems to be modified by a number of factors. Continued breastfeeding and continued feeding during diarrhea can prevent weight faltering. In addition, children who consume a good diet will not only better withstand the illness, but also have the potential to grow more rapidly after the illness and thereby recover from any weight loss. Because most children in developed countries usually receive appropriate treatment and consume an adequate diet, the growth effects of diarrhea— especially given the low prevalence of the illness— is probably very small. However, in developing countries where children suffer from repeated episodes, the cumulative detrimental effect on growth can be substantial.²⁴

Diarrheal diseases impose an economic burden because of the costs of medical care, medications, and lost work. Because the illnesses can largely be managed by fluid and nutritional therapy, much of the medication use, such as antibiotics and so-called antidiarrheal drugs, are an unnecessary expense and are potentially hazardous. Current recommendations
for the treatment of diarrhea among children younger than 5 years of age include oral rehydation salts (ORS) and zinc supplementation, both of which are inexpensive and should be available for home management.²⁵

MICROBIAL ETIOLOGIES

Relative Importance of Enteropathogens

A large number of bacterial, viral, and parasitic agents have been associated with diarrhea in both developing and developed countries. Because the highest rates of diarrheal diseases and the most severe consequences generally occur in young children, most studies have focused on this age group. Older children and adults may become ill from the same enteropathogens, but the relative frequencies with which these organisms are encountered varies because of immunity acquired from prior infection or from differential exposure to the various pathogens.

Community-based studies are those in which household visits are made to identify cases of diarrhea and to collect fecal specimens for identification of enteropathogens. These studies give the best measure of the overall incidence of diarrheal disease because they are not based on severity or care-seeking and are not subject to bias in the ascertainment of cases. In a review of 61 studies (published 1990-2002) with comprehensive microbiology from developing countries, enterotoxigenic Escherichia coli (ETEC) constituted the largest proportion of episodes with a median of 14.1% (Table 20-3). The next most commonly found organism has been Giardia lamblia. Campylobacter species, enteropathogenic E. coli (EPEC), and rotavirus were identified in 7-8% of these diarrheal episodes. Other organisms were less frequently observed. The relative importance of these organisms was rather variable in the different studies.

Noroviruses are emerging as an important pathogens leading to diarrhea and vomiting. Because this kind of infection is typically epidemic and seasonal, it is often not included in yearlong studies prospective studies. In the United States, norovirus causes more than 21 million gastrointestinal illnesses each year, many of which will manifest as severe acute diarrhea disease.²⁶

Studies in health facilities—either outpatient clinics or hospital wards—involve a more selected group of patients who have sought care because of an illness of greater severity. A review of 98 studies done in outpatient health facilities in developing countries found that rotavirus was the most frequent enteropathogen, with a median incidence of 18% (Table 20-3). However, these studies demonstrated that bacterial pathogens predominated overall, accounting for approximately 40% of illnesses. Of these, the most common was C. jejuni. As illustrated in the range of percentage identification in Table 20-4, each of these bacterial pathogens can be very frequent in some settings. Among the parasites, C. parvum and G. lamblia were the most frequently identified. A review of 107 studies in hospitalized children found that rotavirus was the most common
pathogen, followed by EPEC and ETEC (Table 20-5). A number of recent studies in hospitalized children with diarrhea in East Asian countries have found that 40-60% had rotavirus identified.²⁷, ²⁸

Table 20-3 Percentage Identification of Selected Enteropathogens in Children with Diarrhea in Outpatient Clinic Studies in Developing Countries (98 studies)

Enteropathogen		Median (%)	Interquartile Range (%)
Campylobacter sp		12.6	4.0-16.7
Cryptosporidium parvum		2.5	0.8-4.0
Entameba histolytica		0.6	0.1-1.4
Enteropathogenic		9.1	4.5-19.4
	Escherichia coli
Enterotoxigenic		8.7	5.2-15.3
	Escherichia coli
Giardia lamblia		3.0	1.2-5.7
Rotavirus		18.0	13.3-30.7
Salmonella sp		3.2	1.3-6.7
Shigella sp		5.8	2.4-11.0

Table 20-4 Percentage Identification of Selected Enteropathogens in Children with Diarrhea in Community-Based Studies in Developing Countries (61 studies)

Enteropathogen		Median (%)	Interquartile Range (%)
Campylobacter sp		7.6	3.9-13.4
Cryptosporidium parvum		5.7	3.8-8.5
Entameba histolytica		3.5	0.8-5.8
Enteropathogenic		8.8	6.6-13.2
	Escherichia coli
Enterotoxigenic		14.1	6.7-22.6
	Escherichia coli
Giardia lamblia		10.2	7.4-18.8
Rotavirus		8.0	4.2-15.3
Salmonella sp		0.9	0.2-3.1
Shigella sp		4.6	2.2-7.6

Table 20-5 Percentage Identification of Selected Enteropathogens in Children with Diarrhea in Hospital Studies in Developing Countries (107 studies)

Enteropathogen	Median (%)	Interquartile Range (%)
Campylobacter sp	4.5	2.3-9.9
Cryptosporidium parvum	3.4	1.3-7.9
Entameba histolytica	0.7	0.2-3.7
Enteropathogenic	15.6	8.3-27.5
Escherichia coli
Enterotoxigenic	9.5	6.4-16.1
Escherichia coli
Giardia lamblia	1.6	0.5-5.5
Rotavirus	25.4	17.0-37.7
Salmonella sp	4.4	2.9-8.4
Shigella sp	5.6	2.8-10.4

In general, community-based studies identified an enteropathogen in approximately half of the episodes, and the health facility-based studies identified an enteropathogen in nearly 75% of the episodes— although these rates are difficult to estimate from the studies available, because most studies do not set out to test for all possible pathogens. The detection tests for many of the enteropathogens do not have optimal sensitivity, so these studies may have underestimated their incidence. Although it is likely that these enteropathogens individually account for only a small proportion of the episodes, collectively they may cause many of the episodes not associated with one of the more common enteropathogens. Other enteropathogens of interest are adenovirus and astrovirus,²⁹, ³⁰ as well as Bacteroides fragilis and Clostridium difficile,³¹ particularly among patients taking antibiotics.³²

Table 20-6 Median Percentage Identification of Campylobacter sp and Rotavirus from Cases of Diarrhea and Controls without Diarrhea in Community-Based Studies in Developing Countries

	Community-Based Studies			Health Facility-Based Studies
Enteropathogen	*(N)*	Diarrhea	Control	*(N)*	Diarrhea	Control
Campylobacter sp	9	8	7	29	7	2
Rotavirus	11	4	1	31	21	4
Reproduced from Black, R.E., et al., Incidence and etiology of infantile diarrhea and major routes of transmission in Huascar, Peru. Am J Epidemiol, 1989. 129(4): p. 785-99 . By permission of Oxford University Press.

The identification of an enteropathogen from feces during diarrhea does not necessarily mean that that organism is causing the illness. In fact, studies that have performed comprehensive microbiology analyses have often found two or more enteropathogens simultaneously, making it difficult to assign causality. These mixed infections could occur because the individual is exposed simultaneously or sequentially, or an individual may have an asymptomatic infection with one enteropathogen at the time of exposure to another disease-causing agent.

Asymptomatic enteric infections are common in developing country populations but less so in developed country settings. In community-based studies, routine assessment of enteropathogens allows for a comparison between times when the children have diarrhea and when they are healthy. Community-based studies often find a similar rate of identification of enteropathogens in children when they have diarrhea and when they do not. For example, in nine community-based studies in developing countries, the median identification of Campylobacter species was 8 % during diarrhea, compared to 7% when patients did not have diarrhea (Table 20-6). Some enteropathogens, such as rotavirus, may have a higher rate of identification during diarrhea episodes. In health facility-based studies, children who came to the same facility for a reason other than diarrhea may be used as controls. Often the relative prevalence of different enteropathogens is more distinct between diarrhea cases and controls in these studies.

The pathogenicity (i.e., the number of infections with diarrhea divided by the total number of infections) varies by enteropathogen and in some cases by age (Table 20-7). For example, the pathogenicity of rotavirus is lower in the first 6 months of life than in the second 6 months, presumably due to passive protection from maternally derived antibody in early infancy.³³ In contrast, some pathogens, such as Shigella species, have a higher pathogenicity early in childhood. This difference may arise because the initial infection induces some immunity, which protects more effectively against subsequent illness than against infection.

Table 20-7 Pathogenicity of Selected Enteropathogens by Age Group in Peruvian Infants in a Community-Based Study

		Age 0-5 Months			Age 6-11 Months
Enteropathogen		Infections with Diarrhea (N)	Total Infections	Pathogenicity	Infections with Diarrhea (N)	Total Infections	Pathogenicity
Campylobacter sp		73	177	0.41	65	188	0.35
Enterotoxigenic
	Escherichia coli	42	119	0.35	40	103	0.39
Rotavirus		18	33	0.55	23	28	0.82
Shigella sp		14	17	0.82	16	34	0.47
Reproduced from Black, R.E., et al., Incidence and etiology of infantile diarrhea and major routes of transmission in Huascar, Peru. Am J Epidemiol, 1989. 129(4): p. 785-99 . By permission of Oxford University Press.

Table 20-8 Percentage of Children <5 Years Experiencing Dehydration During Diarrheal Episodes by Enteropathogen in Two Community-Based Studies in Rural Bangladesh

Enteropathogen	Episodes (N)	Dehydration (N)	Dehydration (%)
Rotavirus	78	28	36
Vibrio cholerae	3	1	33
Enterotoxigenic Escherichia coli	322	17	5
Other	843	17	2
Total	1246	63	76
Data from Black, R.E., et al., Incidence and severity of rotavirus and Escherichia coli diarrhoea in rural Bangladesh. Implications for vaccine development. Lancet, 1981. 1(8212): p. 141-3 and Black, R.E., et al., Longitudinal studies of infectious diseases and physical growth of children in rural Bangladesh. II. Incidence of diarrhea and association with known pathogens. Am J Epidemiol, 1982. 115(3): p. 315-24 .

The virulence (i.e., the number of severe illnesses divided by the total number of illnesses) may also vary by enteropathogen. This can be illustrated by the propensity of the organism to cause an illness that leads to dehydration. In community-based studies in Bangladesh,³⁴, ³⁵ children with rotavirus diarrhea or cholera are most likely to develop dehydration (Table 20-8). Those persons with ETEC had a modestly increased rate of dehydration compared with all other types of diarrhea.

In the United States, the relative importance of the various enteropathogens differs from that in developing countries (Table 20-9). In studies done in health facilities, rotavirus is the most important enteropathogen associated with diarrhea, just as it is in developing countries, but this is expected to change owing to the widespread use of the rotavirus vaccine since 2006.³⁶ In general, the bacterial causes of diarrhea are less important in developed countries, although Campylobacter species and Salmonella species may be important in some settings. This pattern may be shifted closer to a developing country pattern in certain higher-risk populations. For example, residents of Indian reservations in the United States have shown higher rates of diarrhea than the general population, although good access to medical care has now reduced the diarrheal mortality to a very low level.³⁷, ³⁸ In such settings (Table 20-10), rotavirus is still the most important pathogen, but other organisms such as ETEC and Campylobacter species may play a more prominent role than they do in the general U.S. population.

In the United States, it is estimated that each year there are 1300 deaths, 56,000 hospitalizations, and 9.4 million illnesses caused by foodborne infections, counting both outbreaks and sporadic cases caused by 31 major enteric pathogens.³⁹ In 2006, the CDC reported more than 1200 outbreaks of foodborne disease.⁴⁰ In the 49% of cases for which a cause was identified, norovirus was the pathogen most often identified, being responsible for 54% of the outbreaks. Traditionally, Salmonella, Staphylococcus, and Clostridium perfringens have been considered the main responsible organisms, and they continue to be important. A number of other organisms that are commonly spread by the foodborne route (notably
Campylobacter and E. coli 0157:H7) have also caused substantial morbidity.

Only gold members can continue reading. Log In or Register to continue