Nigerian Journal of Paediatrics

3Clinico demographic factors associated with diarrhoeal disease outcome in under five children A Nigerian tertiary hospital experience

Niger J Paediatr 2016; 43 (4): 246 – 251

ORIGINAL

Ndukwu C

Clinico - demographic factors

Onah S

Ebenebe J

associated with diarrhoeal disease

Osuorah D

outcome in under-five children:

A Nigerian tertiary hospital

experience

DOI:http://dx.doi.org/10.4314/njp.v43i4.3

Accepted: 4th August 2016

Abstract : Objectives: To deter-

period were aged below 5 years.

mine the proportion of under-5

Acute diarrheal disease accounted

Ndukwu C (

)

children presenting with diar-

for 32% and 38% respectively of

Onah S, Ebenebe J

rhoeal disease, and the clinico-

morbidity and mortality in these

Department of Paediatrics,

Nnamdi Azikiwe University

demographic variables associated

under-five children, with a case

Teaching Hospital, Nnewi.

with the outcome at the emer-

fatality rate of 18.3%. The mean

Email: ndukwu_ifey<@yahoo.com

gency paediatric ward (CHER) of

age at presentation was 11.96

the Nnamdi Azikiwe University

months with a male to female ratio

Osuorah D

Teaching Hospital, Nnewi, South-

of 1.4:1. Presentation with fast

Child Survival Unit,

East Nigeria.

breathing (Odds Ratio {OR} 2.6),

Medical Research Council UK,

Methods: Clinical and demo-

convulsion (OR 2.5), loss of con-

the Gambia unit, Gambia.

graphic details of patients with

sciousness (OR 4.3), increased

diarrhoeal disease as obtained

severity of dehydration, presence

from the Paediatrics emergency

of one or more comorbidities (OR

ward log book over an 18month

4.68) and a co-diagnosis of sepsis

period were analyzed using SPSS

(OR 3.23) significantly affected

and Microsoft excel software

the outcome. (p< 0.05 in all these).

packages. The clinical variables

Educational status of the mothers

(non-parametric) were tested for

also significantly affected the out-

degree of association with mortal-

come. (F=9.08, p=0.023)

ity using Spearman’s rank correla-

Conclusion:

Intensified

effort

tion. Chi square and Fischer exact

should be made to sensitize the

test were used to determine pres-

public about dangers of inappro-

ence of significant differences

priate therapy and late hospital

between certain variables. A p

presentation of children with diar-

value <0.05 was deemed signifi-

rhoeal disease. These will reduce

cant.

mortality-heralding complications.

Results: One thousand, one hun-

dred and sixty three out of 1,513

Keywords: diarrhoeal disease;

children admitted within the

mortality; sepsis

Introduction

ted through consumption of contaminated food and wa-

ter. Though there are effective interventions for diar-

Diarrhoea disease remains a leading cause of childhood

rhoea prevention like vaccines, promotion of early and

morbidity and mortality in developing countries like

exclusive breastfeeding, vitamin A supplementation,

Nigeria, and is presently described as one of the two

safe drinking water, improved sanitation; and safe effec-

biggest child killers globally.

1,2

It trails only acute respi-

tive treatment protocols such as the use of low-

ratory infections in the under-five mortality toll, with

osmolarity oral rehydration salts (ORS) and zinc, chil-

about 1.5 million under-five children dying each year

dren are still dying because these interventions are either

from diarrhoea and about 80% of these deaths occurring

not available or accessible to the vulnerable, especially

in the developing world.

in Africa and South Asia; India and Nigeria, topping the

charts.

2,3

In effect, diarrhoeal disease presently kills

When properly managed, the prognosis of diarrhoeal

more children than AIDS, malaria and measles com-

disease is excellent. Indeed, the mortality rate had im-

bined. In Nigeria, it accounts for 10% of morbidities

2,4

proved significantly over the last 40 years, following

amongst under- 5s.

global scaling up and widespread use of oral rehydration

therapy. Subsequently, with world attention diverted to

Diarrhoea has well defined risk factors and transmission

other global emergencies, there has been a decline over

routes with the inducing pathogens being often transmit-

the past two decades in the number of children accessing

247

recommended treatment for diarrhoeal disease in devel-

tre at the extreme of severe dehydration with signs of

oping countries, with an upsurge in the number of un-

cardiovascular compromise and impaired organ perfu-

sion are described as being in clinical shock. Available

treated or poorly treated children who present with a

spectrum of increasing severity of complications.

2,6,7

bed side investigations include Rapid diagnostic test for

Region specific continuous evaluation of the burden and

malaria, urinalysis, random blood sugar. Side lab inves-

outcome of diarrhoeal disease is thus needful with high-

tigations include retroviral screening, packed cell vol-

lighting of the factors that impact on prognosis, as a

ume and films for malaria parasite. Urgent main lab in-

prerequisite to consolidate strategies that will reduce the

vestigations include full blood count, serum electrolytes

case fatality of this highly preventable and treatable dis-

and chest X-ray. The diagnosis of sepsis in the CHER is

ease.

based on signs of systemic inflammatory response like

tachycardia, tachypnea, abnormally high or low tem-

perature, leucocytosis and presence of presumed or con-

Methods

firmed infection. Data was entered in both SPSS version

16 and Microsoft Excel computer software packages.

This was a retrospective study carried out in the Chil-

Analysis was also done using these software packages.

dren Emergency Room (CHER) of the Nnamdi Azikiwe

Clinical variables were tested for degree of association

University Teaching Hospital, Nnewi. This is a Federal

with mortality using Spearman’s rank correlation. Chi

Government owned tertiary and referral centre of excel-

square test was used to determine presence of significant

lence in Anambra state. This hospital receives referrals

differences between certain variables. A p value <0.05

from all around the state and some adjourning states like

was deemed significant. Relevant charts, tables and fig-

Delta and Enugu states. The CHER is manned by a full

ures were used to display frequency distributions of

retinue of staff running 2 shifts for doctors and 3 for

variables.

nurses. There are 2 consultant emergency paediatricians,

2 senior registrars during the morning shift and 1 senior

Study Results

registrar at night, a minimum of 1 resident doctor per

shift, 2 house officers during the day shift, 1 at night,

One thousand, five hundred and thirteen children were

and an average of 3 nurses in each of the nursing shifts.

admitted in the Children Emergency Room of the hospi-

The study population included all patients that were ad-

tal within the period. There were 872 males and 641

mitted in CHER from January 2013 to June 2014, with

females, with a male to female ratio of 1.4: 1. Children

acute diarrhoea. Those that were brought in dead and

under 5 years of age were 1163and constituted 77% of

those with diarrhoea that had lasted more than 2 weeks

all the patients admitted. Three hundred and seventy two

were excluded. Ethical approval to study morbidities

of these U-5 children (214 males and 158 females) pre-

and mortalities in CHER within the time frame was ob-

sented with acute diarrheal disease giving a prevalence

tained from the Research and Ethical Committee of the

of 32%, with a male to female ratio of 1.4: 1. The mean

hospital. Data contained in a structured proforma was

age of these children was 11.96 months + 8.7. See Table

obtained from the doctors’ paediatrics emergency ward

1 below.

log book. The CHER doctors’ log contains biodemo-

graphic information on admitted patients as well as rele-

Table 1: age and sex distribution of the patients

vant clinical details like symptoms, signs, diagnoses,

Age (months)

Male N(%)

Female N(%)

Total N(%)

investigation results, initial management and outcome in

1 – 11

125 (58.4)

97(61.4)

222(59.7)

the emergency room, as documented in the folders of

patients, retrieved after initial stabilization of the

12 – 35

80(37.4)

54(34.2)

134(36.0)

patients.

35 – 59

9(4.2)

7(4.4)

16(4.3)

Total

214(100)

158(100)

372(100)

Information extracted included the age, gender of pa-

Mean age is 11.96+ 8.7months, Median age is 10months

tients, birth order, parental occupation and highest edu-

cation level, presenting symptoms and duration of ill-

Other demographic characteristics: Averagely the

ness, important signs, bedside and urgent initial investi-

socioeconomic class was middle class with up to 34% of

gations, diagnoses and outcome. Criteria for diagnosis

the patients being from the lowest social strata. Other

of acute diarrhoea in the CHER is clinical, based on the

details are shown in table 2.

standard definition of ‘an increase in daily stool fluidity,

frequency and volume from what is considered normal

Table 2: Demographic characteristics of patients

for an individual. This is made after a detailed history

Characteristics

Frequency (n = 372)

Percentage (%)

and thorough physical examination. The aetiological

Birth order

322

86.6

diagnosis for diarrhoea was not included in the study.

13.4

Diagnoses of morbidities like malaria, sepsis and bron-

Maternal age(years)*

chopneumonia are initially made clinically, but quickly

<35

298

80.2

followed up by bed side, side lab and urgent main labo-

>35

19.8

Maternal education

ratory investigations. The grading of severity of dehy-

Nil formal

0.6

dration was based on the current World Health Organi-

Primary

16.2

zation’s clinically based criteria as no, some or severe

Secondary

191

51.3

dehydration. Routinely, children presenting in this cen-

Tertiary

118

31.8

*0.8%(3) of the mothers were teenagers aged between 18 and 19 years

248

Clinical characteristics of the patients : The major

ciated with lower educational level. (F=9.08, p=0.023)

presenting symptoms of the patients are illustrated in

See Figure 4. Bivariate analysis was equally significant

Figure 1. Most (73.9%) of the patients presented within

at 0.01 level.

the first week of their initial symptom, while 8.3% pre-

sented after 2 weeks of initial symptom, other than diar-

Fig 4: Relative percentage mortality per educational level

rhoea (acute diarrhoeal disease occurred in these ones

secondarily). There was clinical dehydration at presenta-

tion in 68% (257) of them, with up to 25% of these be-

ing in clinical shock. Figure 2 is a representation of the

hydration state of the patients at presentation.

Fig 1: Major presenting symptoms of the patients

Presenting complaints and outcome

Table 3 shows the cross-tabulation of presenting symp-

toms with the outcome of acute diarrhoea in the chil-

dren. Cconvulsions (OR 2.5, p<0.001), fast breathing

Fig 2: Levels of dehydration in the patients

(OR 2.6, p <0.001) and unconsciousness (OR 4.3,

p<0.001) were significantly associated with fatal out-

come in children with acute diarrheal disease. A signifi-

cant strength of association with mortality in the patients

was also noted with these symptoms using bivariate

analysis.

Table 3: Association of clinical symptoms with mortality

Symptom

Outcome

Odd’s ratio

p-value

(95% ci )

Mortality

Survived

Vomiting

54(18.4)

239(81.6)

1.0

0.885

Outcome of illness: Diarrhoea disease was responsible

n=293(100)

(0.6-1.8)

Fever

58(18.1)

263(81.9)

0.9

0.792

for 38% of the 178 under-five mortalities recorded

n=321(100)

(0.5-1.7)

within the study period. Sixty eight of the372 children

Convulsion

24(36.4)

42(63.6)

2.5

0.000

with diarrheal disease died, translating to a case fatality

N=66(100)

(1.7- 3.9)

Fast breathing

***

rate of 18%. Up to 72% of these diarrhoeal related

35(32.4)

73(67.6)

2.6

0.000

deaths (49 of 68 patients) occurred within the first 24

n=108(100)

(1.7-3.9)

Unconscious-

30(51.7)

28(48.3)

4.3

0.000

hours of admission with many of them dying within

ness

****

(2.9-6.3)

minutes or few hours of presentation. The outcomes of

n=58(100)

the admissions are represented in Figure 3 below.

* CI = confidence interval

Fig 3: Outcome of

Spearman’s correlation (0.217) significant at 0.005 level

***

diarrheal disease

Spearman’s correlation (0.234) significant at 0.005 level

****

amongst under-fives

Spearman’s correlation (0.372)significant at 0.005 level

Duration of illness and outcome : Effect of duration of

illness (time of onset of initial symptom) and duration of

diarrheal disease before presentation are represented on

table 4 below. The patients that presented earlier had a

significantlybetter outcome.

(F=10.69, p<0.05;X

=12.24, p<0.05)

Demographic characteristics and outcome: Forty five

(65%) of the children that died were infants while the

rest were from 12 to 36months old. However this was

not significant. (F=4.475, p=0.10). The gender of the

patients also did not significantly affect the outcome.

(X =2.69, p= 0.10)The only demographic characteristic

that had a significant relationship with the outcome was

maternal education with a higher relative mortality asso-

249

Table 4: Association between the duration of illness, diarrheal

Discussion

disease duration and outcome

Duration of

Outcome

Statistic

p-value

This study showed a high burden of diarrhoeal disease

illness

Mortality

Survived

morbidity and mortality amongst under-five year old

1-7 days

40(14.5)

235(85.5)

F=10.69

0.010

children that presented to the children emergency room

n=275(100%)

of the Nnamdi Azikiwe University Teaching Hospital

8-14 days

21(31.8)

45(68.2)

n=66(100%)

within the study period, being responsible for about a

15 to 30 days

4(23.5)

13(76.5)

third of morbidities and mortalities amongst them. This,

n=17(100%)

alongside the high case fatality, confirms as has been

>30 days

3(21.4)

11(78.6)

documented in several studies and surveys both globally

n=14(100%)

and in Nigeria in recent years, that diarrhoea remains a

Duration of diarrheal disease

major cause of death in under-5 children.

1-4,10-3

In the

X =12.24

1 day

12(11.9)

89(88.1)

0.007

children emergency room of another tertiary hospital in

n=101(100%)

South East Nigeria, diarrhoeal disease was the common-

2 to 3 days

23(19.0)

98(81.0)

est morbidity managed a year prior to this index study

n=121(100%)

and was responsible for 43% of all mortalities within the

4 to 7 days

18(16.5)

91(83.5)

period. A hospital based study in North central Nige-

n=109(100%)

8 to 14 days

15(36.6)

26(63.4)

ria, however, documented a far less prevalence (2.7%)

of diarrhoea among under-5 children. The lower preva-

n=38(100%)

lence recorded in the study may be attributed to the lar-

F= Fischer’s exact test

ger sampling frame which included all children admitted

X = Chi square test

in the Paediatric department of the tertiary Hospital over

a period of 2 years unlike our study which enrolled only

Presence of dehydration, other co-morbidities and

outcome : The level of dehydration of these children at

under-5 children admitted into the emergency unit of the

the time of presentation significantly affected the out-

Paediatric department. Their prevalence was even lower

come. (F=89.43, p=0.000). Of all the patients who pre-

than the overall Nigerian prevalence of 10%, obtained in

sented in clinical shock, 63.5% died while 14.3% and

a national survey in which the methodology involved

5.3% respectively, of all those that presented with severe

asking mothers about occurrence of diarrhoeal disease in

the two weeks preceding the survey. As diarrhoea tends

and some dehydration died. (See table 5) Presence of

one or more co-morbidities increased the odds of dying

to be seasonal which was noted in the same survey,

from diarrheal disease. (OR 4.68, p=0.000). Children

these results would be largely dependent on the time of

who had sepsis at presentation to CHER were about

the survey.

three times more likely to die from acute diarrhoea than

those without sepsis on presentation (OR 3.23, p=0.000)

The outcome of diarrhoeal disease in this study was

(See Table 5). Bivariate analysis also showed a signifi-

worse in the children that presented with dehydration.

cant strength of association.

This is not surprising as young children are known to be

at great risk of life – threatening dehydration because of

Table 5: Dehydration, co morbidities and outcome

their high daily fluid turnover and decreased ability to

conserve water like adults, thus being unlikely to toler-

Level of dehy-

Outcome

Statistic

p-value

ate increased loss of body water. The past gains of the

dration

Mortality

Survived

prevention of dehydration from diarrhoeal disease have

F =89.43

16(13.9)

97(86.1)

0.000

not been sustained, with studies documenting a de-

n=115(100)

creased awareness of the use and proper constitution of

Some

9(5.2)

164(94.8)

Oral Rehydration Salts (ORS).

2,12,15

n=171(100)

In a 2010 study in

Severe

3(14.3)

18(85.7)

South East Nigeria, while 76% of the mothers knew they

n=21(100)

should use ORS for a child with diarrhoea, only 14.2%

Shock

40(63.5)

23(36.5)

of them could correctly prepare ORS.

[15]

This decreased

n=63(100)

Co-morbidity

Odd’s ratio (CI)

awareness and knowledge will obviously negatively

One or more

60(26.2)

169(73.8)

4.68

0.000

impact on the disease outcome and may be partly why

n=229

(2.31-9.5)

diarrhoeal mortalities are soaring. Even though the

Sepsis

***

33(39.3)

51(60.7)

3.23

0.000

mothers in our study were not questioned on their

n=84

(2.15-4.87)

PEM

8(28.6)

20(71.4)

1.64

0.143

knowledge of oral rehydration, the proportion of our

n=28

(0.87-3.07)

patients presenting with moderate and severe dehydra-

Malaria

9(18.8)

39(81.2)

1.03

0.928

tion and shock (over 68%), with significantly increasing

n=48

(0.55-1.94)

mortality based on severity of dehydration is highly sug-

Bronchopneu-

7(30.4)

16(69.6)

0.119

monia

(0.90-3.36)

gestive of ignorance on the caregivers’ part. Other re-

n=23

searchers have equally documented a strong and positive

correlation of a higher degree of dehydration to fatal

outcome in childhood diarrhoea.

* Fischer’s exact test

**Spearman’s correlation (0.259) significant at 0.005 level

***Spearman’s correlation (0.294) significant at 0.005 level

Just as found in our study, children with diarrhoea who

present to hospital with secondary symptoms like fast

250

breathing and unconsciousness have been equally docu-

procurement of drugs prescribed by patent medicine

dealers are initially exhausted.

6,12

mented in other studies to have increased odds for mor-

This might explain

tality. Known diarrhoeal complications such as electro-

why more than 2 out of every 10 of these index children

lyte imbalance, hypoglycaemia and dehydration result in

studied presented with severe dehydration, with most of

respiratory symptoms, seizures, altered consciousness

these being in shock. The functionality of the primary

and ultimately death. Apart from these secondary

and secondary health care in the region, where most of

symptoms, other conditions which are known causes of

the diarrhoea cases should have been initially managed,

mortality in children actually co-existed with the diar-

is thus compromised as many of these children are hur-

rhoea disease in some of these children. This could ex-

riedly referred to the tertiary centre verbally, as experi-

plain the finding in this study of the presence of one or

enced in our centre, because of their critical condition.

more co morbidities being significantly associated with

mortality. Such findings have been documented previ-

ously.

3,12

Sepsis was the singular significant co-

morbidity that increased the likelihood of death from

Conclusion

diarrhoea disease in the study children. This is not sur-

prising as severe sepsis is a terminal event with a high

Diarrhoeal disease is still a major cause of under-five

fatal outcome in many infectious diseases, including

morbidity and mortality. The poor clinical state of many

diarrhoea in which clinical signs of severe sepsis with

of the index patients at presentation was associated with

septic shock could overlap with hypovolemic shock.

poor outcome of illness. Delayed presentation to hospi-

tal until there are attendant co-morbidities and complica-

The significantly worse outcome seen in those with

tions such as severe dehydration, convulsions, fast

longer duration of illness has been documented in other

breathing and loss of consciousness were significantly

studies.

12,17,18

This is expected because the longer lasting

associated with increase in the odds of death. Efforts to

a diarrhoea episode is, and the longer it takes before

educate caregivers about timely rehydration and early

effective and appropriate care is sought for a child with

hospital presentation could reduce these mortality-

diarrhoeal disease, the more likely it is for serious and

heralding complications.

unsalvageable physiologic aberrations to set in. The

poor clinical state at presentation of a significant per-

centage of the children in our study could be an indirect

Authors’ Contributions

reflection of the health seeking attitude and literacy level

CN and SO headed decision in diagnosis and manage-

of the population studied. This is obvious and justified

ment of most of the patients. CN entered and analyzed

by the fact that relatively, a higher percentage of chil-

the data. CN, SO and DO drafted the manuscript. CN,

dren of women with lower educational attainment died.

SO and JE revised the manuscript.

Poorly educated women are less likely to indulge in safe

Conflict of interest: None

hygienic practices and may not recognise danger signs

Funding: None

in ill children. It has been previously documented that

Nigerian children and children from other parts of Af-

rica are taken to health facilities when their illnesses are

considered severe with obvious failure of home and

Acknowledgements

other traditional therapy, contributing to late presenta-

tion to appropriate health facility.

6,12

With many of these

Drs Ofiaeli, Ngonadi, Nweke, Ameh, Azubuike and

parents influenced by cultural beliefs on the cause of and

Okafor, Paediatric residents in the emergency room at

treatment of diarrhoea, alternative options including

the time of the study for meticulous record keeping.

Reference

Ban Ki Moon Jonathan GE, Lake

Mohammod JC, Monira S, Tah-

National Population Commission

A, Osotimehin B, Stoltenberg J,

meed A, Pradip KB, Hasan A,

(NPC)[Nigeria] and ICF Interna-

Radi A et al. The Global Cam-

Shafiqul AS et al. Progression of

tional. Nigeria Demographic and

paign for the Health Millenium

severe sepsis to septic shock in

Health Survey, 2013. Abuja, Ni-

Development Goals- Report 2013.

under-five diarrheal children in an

geria and Rockville, Maryland,

Accelerating Progress in saving the

urban critical care ward in Bangla-

USA: WPC and ICF Interna-

lives of Women and Children.

desh: Identifiable risks, blood iso-

tional. 2014.

Ministry of Foreign Affairs, Nor-

lates and outcome. Bangladesh

Etea TD. Mother’s perception and

way. January 2013. http://

Crit Care J. 2014: 2(1); 10-15

treatment seeking behaviour for

www.norad.no/globalcampaign et

Centre for Disease Control and

childhood diarrhoea in Dendi

al. Accelerating progress in saving

Prevention. Global water, sanita-

District, West Shoa, Ethiopia.

the lives of women and children.

tion and hygiene. Global Diarrhoea

Glob J Med Pub Hlth 2014; 3 (3)

Global Campaign for the Health

Burden. US Department of Health

1-9

MDGs- Report 2013. United Na-

and Human Services.

Ene-Obong H, Iroegbu CU,

tions. 2013.

Uwaegbute AC. Perceived causes

UNICEF/WHO. Diarrhea: Why

and management of diarrhoea in

Children are Still Dying and What

young children by market women

can be Done. New York: UNICEF/

in Enugu state, Nigeria. J Hlth Pop

WHO; 2009

Nutr 2000; (2)97-102

251

World Health Organization. The

11. Anyanwu OU, Ezeanosike OB,

15. Adimora GN, Ikefuna AN,

Treatment of Diarrhoea: A manual

Ezeonu CT. Pattern and outcome

Ilechukwu G. Home management

for physicians and other senior

of admissions at the children emer-

of childhood diarrhoea: Need to

health workers. 2005 4 Revision

gency room at the Federal Teach-

intensify campaign. Niger J Clin

WHO/CDD/SER/80.2. 6-7

ing Hospital, Abakaliki. African J

Prac. 2011: 14(2) 237-41

National Institute for Health and

Med Hlth Sciences. 2014.13(1)6-

16. Stefano Guandalini and Carmen

Care Excellence. (NICE) Diar-

10 DOI: 10.4103/2384-

Cuffari. Diarrhoea disease. Med-

rhoea and vomiting in children:

5589.139435

scape 2014. Available from http://

Diarrhoea and vomiting caused by

12. Blessing I., Abhulimhen-Iyoha,

emedicine.medscape.com/

gastroenteritis: diagnosis, assess-

Okposio M . Risk factors for death

article/928598-overview. As-

ment and management in children

in under-five children presenting

sessed 01/05/2015.

younger than 5 years. NICE clini-

with acute diarrhoea in an urban

17. Lee WS, Boey CC. Chronic diar-

cal guidelines April 2009. As-

teaching hospital in Nigeria. Pa-

rhoea in infants and young chil-

sessed 21/6/15. http: //

cific J. Med Scie: 2013;12(1): 45-

dren: causes, clinical features and

publications.nice. org.uk/diarrhoea

outcome. Paediatr Child Health.

-and-vomiting-in-children-cg84/

13. UNICEF 2015. Monitoring the

1999 Jun;35(3):260-3.

notes-on-the-scope-of-the-

Situation of Children and Women.

18. Ndukwu CI, Onah SK. Pattern

guidance.

Diarrhoea Disease. Available from

and outcome of postneonatal

10. UNICEF 2015. Monitoring the

http://data.unicef.org/child-health/

pediatric emergencies in Nnamdi

Situation of Children and Women.

diarrhoeal-disease. Assessed on

Azikiwe University Teaching

Diarrhoea Disease. Available from

01/05/2015.

Hospital, Nnewi, South East Ni-

http://data.unicef.org/child-health/

14. Yilgwan CS, Okolo SN. Preva-

geria. Niger J Clin Prac 2015: 18

diarrhoeal-disease. Assessed on

lence of diarrhea disease and risk

(3) 348-53

01/05/2015.

factors in Jos University Teaching

Hospital, Nigeria. Ann Afr Med.

2012;11(4):217-21.