Mortality among 5-17 year old children in Kenya
Bonface Ombaba Osano1,2,&, Fred Were1, Shanaaz Mathews2
1University of Nairobi, College of Health Sciences, Kenya, 2University of Cape Town, South Africa
Bonface Ombaba Osano, University of Nairobi, CHS, Department of Paediatrics & Child Health, KNH, Nairobi, Kenya
global mortality trends have changed over time and are expected to continue changing with a reduction in communicable diseases and an increase of non-communicable disease. Increased survival of children beyond five years may change mortality patterns for these children. There are few studies in Africa that explore the causes of mortality in children over five years. The objective of this study was to determine the mortality rate and clinical profiles of children aged 5-17 years who died in six Kenyan hospitals in 2013.
retrospective review of patients’ medical records to abstract data on diagnosis for those who died in year 2013. Data was analysed to provide descriptive statistics and explored differences in mortality rates between age groups and gender.
we retrieved 4,520 patient records. The in-hospital mortality rate was 3.5% (95%CI 3.0-4.1) with variations in deaths between the ages and gender. Among the deaths, 60% suffered from communicable diseases, maternal and nutritional causes; 41.3% suffered from non-communicable diseases. A further 11.9% succumbed to traumatic injuries. The predominant clinical diagnoses among patients who died were HIV/AIDS, respiratory tract infections and malaria.
infectious causes had the highest proportion of diagnoses among children aged 5-17 years who died.
Globally, mortality trends have changed over time and are expected
to continue changing with an increase in injuries, mental illness, drug
abuse and poisoning [1-3]. The burden of disease studies
have shown that communicable diseases are the leading cause of global mortality,
with a transition underway as non-communicable diseases are on the rise
There is great variation in causes of mortality between regions, age groups and
gender [2, 3]. Global burden of
disease studies show that mortality rates increase from 15 years for females
and 20 years for males [2, 4-6].
In developed countries, non-communicable diseases cause greater morbidity
and mortality in children than communicable diseases [1-3, 7].
Although the causes of mortality are shifting to non-communicable causes,
the burden of communicable disease remains significant in middle- and low-income
countries where communicable diseases still account for the majority of
mortality cases [1-3, 7]. When
causes of mortality are analysed in low income countries, reproductive health
issues (pregnancy and maternity related illnesses) dominate in females as
they become older, whereas injuries are a leading cause of mortality in
older males [1, 3].
The main causes of death in Sub-Saharan Africa are preventable diseases caused
by infectious agents [8-11]. Communicable diseases have
been the leading cause of mortality and still remain so in Sub-Saharan Africa,
unlike in developed countries where non-communicable causes are on the increase
These communicable diseases include infections like malaria, HIV/AIDS, pneumonia,
tuberculosis (TB) and diarrhoea, especially for younger children [2, 9, 15].
The communicable disease may be leading probably due to co-morbidities such as
malnutrition and HIV. In Kenya, for example, up to 11% of children below 5 years
are underweight and a higher fraction (26%) are stunted [16, 17].
These levels of malnutrition may be prevalent in the older children and thus
contribute to the mortality from infectious causes. Globally, complications from
pregnancy among teenagers contribute to mortality among adolescents .
This is mainly due to teenage pregnancies as a consequences of early marriages
in the context of developing countries such as in Africa where cultural practices
impact on health outcomes . The large proportion of girls under
18 years who are pregnant may contribute to the high maternal mortality in Sub-Saharan
Africa [16, 19].
Non-communicable diseases, mental illness and substance abuse also play a role
in adolescent mortality in Africa; however very little is known about their contribution
to mortality in developing countries as they are not well documented as in the
Injuries as a cause of mortality are expected to rise with increasing economic
prosperity resulting in epidemiological transition and a decrease in communicable
diseases, more so in developing countries [1
and unintentional injuries, such as assault, burns and road traffic accidents
(RTA) are a major cause of mortality among adolescents and young adults aged
15-24 years, and especially among males in Kenya [12
low-income countries [13
] and globally [2
Suicide is one of the leading causes of death among adolescents in the United
States of America and other high-income countries [2
In most regions of the world, suicide is more common in males than females [2
Outdoor injuries from violence are more common amongst males, while in females
indoor injuries, such as burns, are slightly more common [13
Cyclists and pedestrians are more likely to be killed than car occupants [23
Road traffic accidents are an emerging public health problem, especially from
motorcycles in Tanzania [25
]. In South Africa, Balme, et
al found poisoning to be a cause of childhood hospital admission. Accidental
poisoning may be more common among the younger age groups but intentional poisoning
(suicide attempts) as well as homicide may be more common with the older groups.
Pesticides and household cleaning products have also increased as causative agents
of poisoning [26
There is limited data available on children older than 15 years in Kenya, as
they have not been explored as a group on their own and are instead grouped with
adults or missed completely. In addition, there are few studies on burden of
disease based on Africa with a focus on children older than five [2
Civil registration of deaths is unreliable due to under-reporting and misclassification,
with ill-defined incorrect ICD-10 codes used in some instances [10
To inform appropriate prevention strategies it is necessary to know the pattern
of causes and magnitude of deaths across all ages [5
There is little information in Kenya on the exact causes of death among 5-17
years of age yet they have been shown to contribute to close to 10% of deaths.
In Kenya, close to 8% of deaths that occurred in 2003-2009 in a rural area of
the western region were in late adolescents (15-19 years) and young adults (20-24
]. Preventable infectious causes, such as malaria,
lower respiratory tract infections, TB, HIV, malaria and gastroenteritis, account
for a large proportion of mortality in Kenya as in the rest of Africa [2
Even though infectious diseases are the leading cause of mortality in Kenya,
the exact infectious causes have varied over time and by age [3
There seems to be a reduction of up to 60% in mortality from gastroenteritis
and HIV [12
With increased survival of children under five years of age, there are likely
to be an increase in children aged 6-18 in the near future and there is an urgent
need to understand their health issues better. The Kenyan Demographic and Health
Survey of 2008-2009 showed that children aged 5-19 years comprised 39% of the
Kenyan population; yet there is little information on the health of this age
]. This study aimed to determine the in-hospital
mortality rate, diagnoses at death of children aged 5-17 years admitted to six
hospitals in Kenya for the 2013 calendar year and to explore pattern of mortality
The study was carried out in five government hospitals spread across
different regions of Kenya and one private hospital in Nairobi.
The five government hospitals were purposively selected because they are
hospitals for the counties they serve and represent a wide geographical
area within the country. The private hospital was included to provide
a perspective of children from a higher socio-economic group. A retrospective
hospital based study using admission registers as the sampling
the 2013 calendar year was used to identify all children aged 5-17
years. The study included all children aged 5-17 as the study sample. Hospital
records were retrieved, reviewed and data abstracted from patient’s discharge
or death summaries for interest variables (age, gender and diagnosis at
death). The researcher or trained research assistants reviewed the whole
patient record to extract the required data where there were no summaries.
Three patients’ records from the private hospital were not available for
review due to medico-legal reasons and were excluded from this analysis.
In this study, age was categorised into groups (5-9, 10-14 and 15-17 years)
as a modification of the age categories used by Lozano ,
Patton , Wang , and the
Kenya Demographic and Heath Survey  with the upper
age cut-off of 17 years as the legal age cut-off for children. The mortality
proportion was calculated as all the deaths in the age group divided by
the age group total admissions, expressed as a percentage to describe the
mortality pattern. Pearson’s chi-square test and odds ratios were calculated
to explore differences in deaths between genders, age groups and HIV status
at 95% confidence interval. The patients’ diagnoses were categorised in
accordance with the Lozano  study classifications
and to allow for comparison.
Classification of diseases
The disease classification used was based on the study by Lozano, et al on global and regional mortality from 235 causes of death for 20 age groups in 1990 and 2010; a systematic analysis for the Global Burden of Disease (GBD) Study [3
] as shown in Table 1
. This classification was chosen to make comparison of findings with other international studies easier. However, many patients had multiple diagnoses with some appearing across disease classification categories and hence some patients fell in more than one disease category. Patients with injuries did not have full details on where it occurred, the cause, nature and type of injury and were therefore grouped together as injuries.
We obtained approval from the Human Research Ethics Committee at the Faculty of Health Sciences, University of Cape Town (UCT) and the Kenyatta National Hospital/University of Nairobi Ethics and Research Committee (KNH-UoN ERC). Hospital review boards and administration permission were obtained from all the hospitals where the study was conducted. Patient confidentiality was maintained at all times. Names of the patients/study participants were not collected by the study. Once the hospital number was noted, it was de-identified by giving it a code that could not be traced back to the patients. Only coded data was transported to conceal identities. In addition, the data was stored in files kept under lock and key with restricted access in secured offices. Electronic databases were encrypted and stored in password-protected, secure computers.
There were 160 deaths in the age group 5-17 years out of 4,520 total
admissions at the six hospitals in the study population. The in-hospital
mortality rate was 3.5% (95% CI, 3.0 – 4.1%) as shown in Table
2. The deaths of 5-17 year olds contributed 3.9% of all age groups
hospitals’ total deaths. The non-communicable causes had the highest category-specific
mortality proportion of 4.4% while both communicable diseases and
injuries were at 3.0%. Most of the deaths occurred in younger children aged
of age (40.8%), although the age specific mortality proportion
was highest in the age group 10-14 years. Those aged 15-17 years had the
with a 2.3% in-hospital mortality rate while those aged 10-14 years
had the worst outcome, with a 5.0% in-hospital mortality rate (Table
2). There were significantly (p<0.001) more male deaths (90 - 57%)
compared to female deaths (67-43%).
Diagnosis of patients who died
disease classification of children aged 5-17 years who died at the six hospitals
in 2013. Overall communicable causes were the main cause of death (60%). Table
shows the clinical diagnosis for the patients who died whereby infectious
diseases such as HIV, malaria and pneumonia were the leading cause of death.
There were four maternal deaths; three were from obstetric haemorrhage while
one was from severe eclampsia. Poisoning contributed to 42% of deaths from injuries,
while head injuries contributed 10.5%; and fractures and soft tissue injuries
contributed 21.3% of injury patient deaths. The leading diagnoses in those who
died were infectious (HIV, respiratory tract infections and malaria), haematological
or injuries. Table
gives the clinical diagnosis of the patients who died. Some patients
who died had multiple diagnoses across disease categories; hence disease categories
are cumulative and exceed the number of deaths.
We found that the proportion of deaths among 5-17 year olds in relation to total hospital deaths (3.9%) is lower than the age group’s contribution to overall hospital admissions (6.9%). This was lower than in a study in a rural area of the western region in Kenya where close to 8% of deaths were among adolescents aged 15-19 years and young adults aged 20-24 years . These studies are not directly comparable since the one in Western Kenya included young adults where mortality rates are expected to start rising based on the findings from global burden of disease studies [3, 21]. The Phillips-Howard study in Western Kenya drew on a sample aged 15-24 years and therefore a direct comparison is not possible. In addition, the western part of Kenya has a different disease pattern, such as higher rates of malaria than the regions included in this study . This study also found significantly more male than female deaths. This may be due to the fact that females were admitted mainly due to communicable diseases and maternity-related causes which may be appropriately managed at this level.
Communicable, maternal and nutritional causes were diagnosed in over 60% of patients who died with the leading ones being malaria (17.2% of deaths), HIV/AIDS (18.1% of deaths) and pneumonia (15.0% of deaths). The major diagnoses at death in this study are the common causes of death in prior studies in similar settings [2, 8-14, 29, 30].
These are all communicable and preventable disease but due to social conditions
and inequitable access to health care and services we still have large number
of children dying from preventable causes. Poor access and health seeking behaviour,
where patients come to formal healthcare facilities when it is very late when
illnesses have complicated and probably more difficult to cure, may contribute
to a high number of deaths from preventable and curable illnesses. Anaemia was
one of the leading diagnoses in patients who died. The anaemia may be considered
nutritional or a complication of the infectious causes such as malaria. There
were four maternal deaths and this may have been contributed to by the young
age of these mothers. Young maternal age is considered a risk factor for poor
pregnancy outcomes for both mother and baby [31, 32]. The proportion of 5-19 year old females who have started child bearing is higher than that found in prior studies and surveys.
In Kenya 17.7% of women aged 15-19 have begun
child bearing [16, 19]. This
figure increased slightly to 18% as per the 2013 KDHS .
In addition to the young child bearing age being a risk factor, reproductive
health services may not be adolescent friendly or easily accessible to these
children thus worsening the outcomes. The non-communicable diseases were
diagnosed in 41.3% of the overall deaths. This is in keeping with increases
of NCDs as a cause of death in developing countries probably due to improvements
in the socio-economic status [1-3, 12].
Better socio-economic status may shift morbidity and mortality from infectious
causes to non-communicable disease and injuries as seen in more developed
Injuries attributed to 11.9% of deaths. Males were more likely to die due to
injuries but this difference was not statistically significant. The lack
of a gender difference in injury deaths in this study was probably because
the injuries were analysed as grouped and were not specified by type or
cause. The limitations of incomplete documentation of exact nature of injury
(interpersonal violence, intentional or unintentional) could not allow for
separation to analyse this difference further. Some injury types, such as
interpersonal violence, may increase with age and are more common in males.
The changes in causes of mortality to non-communicable diseases that have been
noted in developed countries were not reflected in this study. Although
injuries and non-communicable diseases were documented, communicable diseases
were still the leading cause of death in Kenya for children aged 5-17 years
as in most other low income countries [2, 8-14, 29, 30].
This study depended on hospital records, which did not have full documentation for all patients. The gap in information was supplemented by reviewing the full patient records, inferring from the investigations done and the management plan for each of those patients. The diagnoses at death may have been multiple for each death and not the immediate cause of death could be specifically ascribed. Mortality rate was not calculated because of the unavailability of population data for each district and the catchment areas.
Children aged 5-17 years contribute 3.5% of all in hospital deaths. Infectious and maternal causes were the leading cause of hospital deaths (60%). The deaths were mainly from preventable and curable diseases, which should not be causing mortality with appropriate preventive and curative services. This calls for strengthening of prevention of the non-communicable and common infectious causes in this age group. This could be achieved through the school health programme because almost all of these children are in school and it is the easiest place to reach them.
What is known about this topic
- Communicable diseases are a leading cause of death in sub-Saharan Africa;
- Injuries and non-communicable causes of mortality are rising.
What this study adds
- Communicable causes are the leading causes of death in children aged 5-17 years in Kenya;
- The deaths were mainly from preventable and curable diseases, which should not be causing mortality with appropriate preventive and curative services.
The authors declare no competing interest.
Bonface O Osano conceptualised and designed the study, collected, analysed and interpreted the data. He drafted the article, did revisions and approved the version to be published; Fred Were participated in the design of the study, data analysis and interpretation. He revised and approved the version to be published; Shanaaz Mathews conceptualised and designed the study, analysed and interpreted the data. She revised and approved the version to be published. All authors have read and agreed to the final version of this manuscript.
The project described was supported by MEPI/PRIME-K grant from the US National Institutes of Health (NIH-R24TW008889). The content is solely the responsibility of the authors and does not necessarily represent the official views of the US National Institutes of Health. We acknowledge the support of the hospitals where the study was carried out and Prof Jawaya Shea (UCT).
1: classification of diseases as adopted from Lozano et al. (2010).
Global and regional mortality from 235 causes of death for 20 age groups
in 1990 and 2010; a systematic analysis for the GBD Study 
Table 2: deaths by age group and gender
diagnosis of patients who died (number, percentage and category-specific in-hospital
mortality rate for 5-17 year olds)
Table 4: clinical diagnosis for the 160 patients who died
- Mathers CD, Loncar D. Projections of Global Mortality and Burden of Disease from 2002 to 2030. PLoS Med. 2006 Nov; 3(11): e442. PubMed | Google Scholar
- Patton GC, Coffey C, Sawyer SM, Viner RM, Haller DM, Bose K et al. Global Patterns of Mortality in Young People: A systematic analysis of Population Health Data. The Lancet. 2009; 374(9693): 881-892. PubMed | Google Scholar
- Lozano R, Naghavi M, Foreman K, Lim S, Shibuya K, Aboyans V et al. Global and regional mortality from 235 causes of death for 20 age groups in 1990 and 2010: a systematic analysis for the Global Burden of Disease Study 2010. The Lancet. 2012; 380(9859): 2095-2128. PubMed | Google Scholar
- Adazu K, Lindblade KA, Rosen DH, Odhiambo PO, Kwach J, Van Eijk AM et al. Health and demographic surveillance in rural western Kenya: a platform for evaluating interventions to reduce morbidity and mortality from infectious diseases. The American Journal of Tropical Medicine and Hygiene. 2005; 73(6): 1151-1158. PubMed | Google Scholar
- Viner RM, Coffey C, Mathers C, Bloem P, Costello
A, Santelli J, Patton G. 50-year mortality trends in
children and young people: a study of 50 low-income, middle-income, and
The Lancet. 2011; 377(9772): 1162-1174.. PubMed | Google
- Santelli JS, Galea S. The Global Burden of Disease in 10-24-year-olds. The Lancet. 2011; 377(9783): 2058-2060. PubMed | Google Scholar
- Murray CJ, Lopez AD. Mortality by cause for eight regions of the World: Global Burden of Disease Study. The Lancet. 1997; 349(9061): 1269-1276. PubMed | Google Scholar
- Stein CE, Inoue M, Fat DM. The global mortality of infectious and parasitic diseases in children. Seminars in Pediatric Infectious Diseases. 2004; 15(3): 125-129. PubMed | Google Scholar
- Einterz EM, Bates M. Causes and circumstances of death in a district hospital in northern Cameroon, 1993-200 Rural Remote Health. 2011; 11(3): 1623. PubMed | Google Scholar
- Wang H, Dwyer-Lindgren L, Lofgren KT, Rajaratnam
JK, Marcus JR, Levin-Rector A et al. Age-specific and sex-specific mortality
in 187 countries, 1970-2010: a systematic analysis for the Global Burden
of Disease Study 20 The Lancet. 2012; 380(9859): 2071-2094. PubMed | Google
- Sawyer SM, Afifi RM, Bearinger LH, Blakemore S, Dick B, Ezeh AC et al. Adolescence: a foundation for future health. The Lancet. 2012; 379(9826): 1630-1640. PubMed | Google Scholar
- Phillips-Howard PA, Odhiambo FO, Hamel M, Adazu K, Ackers M, van Eijk AM et al. Mortality Trends from 2003 to 2009 among adolescents and young adults in rural Western Kenya using a Health and Demographic Surveillance System. PLOS ONE. 2012; 7(11): E47017. PubMed | Google Scholar
- Deen JL, Vos T, Huttly RA, Tulloch J. Injuries and Non-Communicable Diseases: emerging health problems of children in developing countries. Bulletin of the World Health Organisation. 1999; 77(6): 519-524. PubMed | Google Scholar
- Petit PL, van Ginneken JK. Analysis of hospital records in four African countries, 1975-1990, with emphasis on infectious diseases. Journal of Tropical Medicine & Hygiene. 1995; 98(4): 217-27. PubMed | Google Scholar
- Misganaw A, Haile Mariam D, Araya T, Ayele K. Patterns of mortality in public and private hospitals of Addis Ababa, Ethiopia. BMC Public Health. 2012 Nov 20; 12: 1007. PubMed | Google Scholar
- Kenya National Bureau of Statistics (KNBS) and ICF Macro. Kenya Demographic and Health Survey, 2008-09. KNBS and ICF Macro. Calverton, Maryland. 2010. Google Scholar
- Kenya National Bureau of Statistics (KNBS) & ICF Macro. Kenya Demographic and Health Survey, 2012-13. KNBS and ICF Macro, Calverton, Maryland. 2014. Google Scholar
- Mayor S. Pregnancy and childbirth are leading causes of death in teenage girls in developing countries. BMJ. 2004 May 15; 328(7449): 1152. PubMed | Google Scholar
- Nour NM. Health consequences of child marriage in Africa. Emerging Infectious Diseases. 2006; 12(11): 1644-1649. PubMed | Google Scholar
- Blum RW, Nelson-Mmari K. The health of young people in a global context. Journal of Adolescent Health. 2004; 35(5): 402-418. PubMed | Google Scholar
- Patton GC, Coffey C, Cappa C, Currie DR, Riley L, Gore F et al. Health of the world’s adolescents: a synthesis of internationally comparable data. Lancet. 2012 Apr 28; 379(9826): 1665-75. PubMed | Google Scholar
- Bell C, Clark D. Adolescent suicide. Pediatr Clin North Am. 1998 Apr; 45(2): 365-80. PubMed | Google Scholar
- Ameratunga S, Hijar M, Norton R. Road-traffic injuries: Confronting disparities to address a global-health problem. The Lancet. 2006; 367(9521): 1533-1540. PubMed | Google Scholar
- Saidi H, Oduor J. Trauma deaths outside the hospital: Uncovering the typology in Kenyan Capital. The Journal of Forensic and Legal Medicine. 2013; 20(6): 570-574.. PubMed | Google Scholar
- Chalya PL, Mabula JB, Ngayomela IH, Kanumba ES, Chandika AB, Giiti G, et al. Motorcycle injuries as an emerging public health problem in Mwanza City, North-Western Tanzania. Tanzania Journal of Health Research. 2010; 12(4): 214-221. PubMed | Google Scholar
- Balme KH, Roberts JC, Glasstone M, Curling L, Mann MD. The changing trends of childhood poisoning at a tertiary children's hospital in South Africa. South African Medical Journal. 2012; 102(3): 142-146. PubMed | Google Scholar
- Rudan I, Lawn J, Cousens S, Rowe AK, Boschi-Pinto
C, Tomaskovic L et al. Gaps in policy-relevant information on burden
of disease in children: a systematic review. The Lancet. 2005; 365(9476):
- Mung'ala VO, Snow RW. Death registration on the Kenyan Coast. East African Medical Journal. 1994; 71(11): 747-750. PubMed | Google Scholar
- Menge I, Esamai F, Van Reken D, Anabwani G. Paediatric morbidity and mortality at the Eldoret District Hospital, Kenya. The East African Medical Journal. 1995; 72(3): 165. PubMed | Google Scholar
- Campbell JD, Sow SO, Levine MM, Kotloff KL. The causes of hospital admission and death among children in Bamako, Mali. Journal of Tropical Pediatrics. 2004; 50(3): 158. PubMed | Google Scholar
- Kumar A, Singh T, Basu S, Pandey S, Bhargava V. Outcomes of pregnancy. Indian Journal of Pediatrics. 2007; 74(10): 927-931. PubMed | Google Scholar
- Talawar S, Venkatesh G. Outcome of Teenage Pregnancy. IOSR Journal of Dental and Medical Sciences (IOSR-JDMS). 2013; 6(6): 81-83. PubMed | Google Scholar