THE DETERMINANTS AND EXTENT OF HOME ACCIDENTS IN

CHILDREN UNDER FIVE YEARS IN KISUMU DISTRICT, KENYA

BY:

DR. ROSEMARY AKINYI OBARA

(MBCh.B, NAIROBI, KENYA)

A THESIS SUBMITTED IN PART FULFILMENT

FOR THE AWARD OF A MASTER'S DEGREE IN PUBLIC HEALTH FROM

THE DEPARTMENT OF COMMUNITY HEALTH,

UNIVERSITY OF NAIROBI.

1998

University of NAIROBI Library

MEDICAL LIBRARY
UNIVERSITY OF NAIROBI
P. O. Box 19676
NAIROBI
 DECLARATION

I hereby declare that this thesis is my own original work and has not been presented for a degree

in any other university.

Signed:_________________________ D ate:___________________

Dr. Rosemary Akinyi Obara.

1
 APPROVAL

This thesis has been submitted for examination with our approval as supervisors.

1. Dr. E.N. Ngugi, Ph.D.,

Lecturer, Department o f Community Health,

University o f Nairobi.

Signed:_ _Date: !Z 3 > ] / i 1

2. Mr. L. Nyabola, [Link]., MS., M.S., Dip. Epid.

Lecturer, Department of Community Health,

University o f Nairobi.

Signed:_____ ___________________ Date: 2 - 7 ////< 7 ^

3. Mr. Andre Dzikus, BSc., MSc.,

Human Settlements Officer,

United Nations Centre For Human Settlements (UNCHS)

(HABITAT) United Nations, Gigiri, Nairobi

Signed: Date:

4. Prof. J. K. W ang’ombe, M.A., Ph.D.,

Chairman, Department o f Community Health,

11
 DEDICATION

This work is dedicated to:

My father, Mr. Gideon Obara and my mother, Mrs. Josephine A. Obara for their love,

encouragement and support;

My husband Dr. Stephen Olus Okeyo, my children Gregory, Valery, Antony, Theodore

and all the members of my family for their support and patience during this study.

in
 ACKNOWLEDGEMENTS

I wish to most sincerely extend my gratitude and appreciation to the following:

1. The Inter-faculty collaboration Programme (IFCP), especially Dr. Violet Kimani, the
coordinator for providing me with the Hinds enabling me to carry out the study;

2. My internal supervisors Dr. E. N. Ngugi, Mr. Nyabola and Dr. J. Kiamba for their
commitment and constant supervision and guidance during the study;

3. Mr. Njeru for his guidance on the sampling procedure;

4. My external supervisors, Mr. A. Dzikus and Dr. J. Alwar who gave me great advice and
material assistance in the field in the technical aspects of data collection;

5. The Chairman and all the lecturers of the Department o f Community Health, University
ofNairobi;

6. The MOH, Kisumu for assistance in identifying the research assistants and for allowing
them time off their schedules to participate in this study;

7. The District Officer Winam, together with the locational chiefs and sub-chiefs, for
assistance in identifying the study clusters and also for availing the village elders
(mlangos) to guide us during the study.

8. Mr. Paul Okullu, for assisting with facilities to print during the study, and providing me
with means o f transport during the entire data collection period;

9. All the research assistants for working tirelessly and sometimes in very difficult
conditions in order to get all the data collected promptly;

10. Mr. Kihoro for his guidance with SPSSPC+ computer programme during data analysis;

11. To my colleagues for their support, encouragement and co-operation;

12. Last but not least, to the people o f Kisumu for their co-operation and understanding
during the data collection.

IV
 TABLE OF CONTENTS

PAGE
Declaration..............................................................................................................................................i
Approval.................................................................................................................................................ii
Dedication.............................................................................................................................................iii
Acknowledgements..............................................................................................................................iv
Table o f Contents.................................................................................................................................. v
Abbreviations...................................................................................................................................... vii
List o f T ab les....................................................................................................................................viii
List of Figures...................................................................................................................................... ix

EXECUTIVE SUMMARY ................................................................................................................ x

CHAPTER 1 INTRODUCTION.................................................................................................... 1

CHAPTER 2 LITERATURE REVIEW........................................................................................ 4

CHAPTER 3 RESEARCH PROBLEM...................................................................................... 19

3.1 JUSTIFICATION OF STUDY.................................................................. 21
3.2 STUDY OBJECTIVES...............................................................................23
3.3 STUDY HYPOTHESIS..............................................................................24

CHAPTER 4 STUDY AREA.......................................................................................................25

4.1 KISUMU DISTRICT - ADMINISTRATIVE UNITS AND
BOUNDARIES...................................................................................................25

CHAPTER 5 METHODOLOGY OF THE STUDY................................................................. 30

5.1 STUDY DESIGN........................................................................................ 30
5.2 STUDY POPULATION............................................................................30
5.2.1 SAMPLE SIZE................................................................................ 31
5.2.2 SAMPLING PROCEDURE..........................................................31
5.3 INCLUSION/EXCLUSION CRITERIA...................................................34
5.4 ETHICAL CONSIDERATIONS............................................................... 34
5.5 DEFINITIONS.............................................................................................35
5.6 LOGISTICS.................................................................................................35
5.6.1 PREPARATION FOR DATA COLLECTION........................... 35
5.6.2 DATA COLLECTION TECHNIQUES........................................ 36
5.7 CONSTRAINTS ENCOUNTERED DURING THE STU DY..............37
5.8 APPROACHES TO DATA ANALYSIS..................................................38

v
CHAPTER 6 RESULTS .........................................................................................................41
6.1 SOCIO-DEMOGRAPHIC CHARACTERISTICS OF THE STUDY
SA M PLE.....................................................................................................41
6.2 ENVIRONMENTAL CHARACTERISTICS OF CHILD'S HOME
ACCIDEN TS.............................................................................................. 51
6.3 HOME ACCIDENTS............................................................................... 55
6.3.1 PREVALENCE OF INJURIES DUE TO
HOME ACCIDEN TS.................................................................... 55
6.3.2 SEVERITY OF HOME ACCIDEN TS........................................ 55
6.3.3 HOME ACCIDENTS BY CHILD'S SOCIO­
DEMOGRAPHIC CHARACTERISTICS....................... 58
6.3.4 ENVIRONMENTAL HEALTH CIRCUMSTANCES............... 64
6.3.5 COURSE OF ACTION FOLLOWING ACCIDENTS.................72
6.3.6 OUTCOME OF HOME ACCIDEN TS....................................... 76

CHAPTER 7 DISCUSSION, CONCLUSION, AND RECOMMENDATIONS................ 78

7.1 DISCUSSION .......................................................................................... 78
7.1.1 BACKGROUND CHARACTERISTICS OF
THE STUDY SA M PLE............................................................... 78
7.1.2 THE EXTENT AND DETERMINANTS
OF HOME ACCIDENTS .............................................................84
7.1.3 COURSE AND TIME OF ACTION
FOLLOWING ACCIDENTS ..................................................... 94
7.1.4 OUTCOME OF HOME ACCIDEN TS..................................... 96
7.2 CONCLUSION.........................................................................................97
7.3 RECOMMENDATIONS...................................................................... 100

REFERENCES................................................................................................. 102

LIST OF ANNEXES

Annex I Questionnaire.................................................................................. 105

Annex II Map o f Kisumu Municipality........................................................115

Annex ID Background Information on K isum u........................................... 116

vi
 ABBREVIATIONS

CRC Child resistant closure.

INSERM (French) Institut National de la Sante et de la

Recherche Medicale.

..... (The French National Institute of Health

and Medical Research).

MCH Maternal and Child Health.

TBSA Total body surface area.

BSA Body surface area.

WHO World Health Organization.

RTA Road Traffic Accidents.

NEISS National Electronic Injury Surveillance

System (America).

n Sample size

SD Standard Deviation

V ll
LIST OF TABLES PAGE

TABLE 4.1: ADMINISTRATIVE SECTIONS OF WINAM DIVISION IN

KISUMU MUNICIPALITY AND NUMBER OF HOUSEHOLDS
WITHIN THE SUBLOCATIONS...........................................................................27
TABLE 5.1: DISTRIBUTION OF CLUSTERS BY SUBLOCATION ................................ 33
TABLE 6.1: AGE/SEX DISTRIBUTION OF CHILDREN.................................................... 42
TABLE 6.2: DISTRIBUTION OF PARENTS BY AGE..........................................................43
TABLE 6.3: SUMMARY STATISTICS ON AGE FOR THE PARENTS........................... 43
TABLE 6.4: EDUCATION OF PA REN TS.............................................................................. 44
TABLE 6.5: OCCUPATION OF PARENTS............................................................................ 45
TABLE 6.6: MARITAL STATUS OF CHILD’S M OTHER.................................................. 46
TABLE 6.7: RESIDENTIAL STATUS OF PA REN TS...........................................................46
TABLE 6.8: DISTRIBUTION OF CARETAKERS BY A G E ................................................ 48
TABLE 6.9: DISTRIBUTION OF CARETAKERS BY EDUCATION LEVEL................. 49
TABLE 6.10: AREAS OF RESIDENCE...................................................................................... 51
TABLE 6.11: SOURCE OF COOKING FU E L.......................................................................... 53
TABLE 6.12: PRESENCE OR ABSCENCE OF RISK FACTORS IN THE HOME............ 54
TABLE 6.13: INJURIES DUE TO HOME ACCIDENTS.........................................................55
TABLE 6.14: ACCIDENT RATES IN THE 0-5 YEARS......................................................... 58
TABLE 6.15: ACCIDENTS IN THE VARIOUS AGE-GROUPS...........................................59
TABLE 6.16: ACCIDENTS BY S E X ..........................................................................................60
TABLE 6.17: ACCIDENT BY AGE OF CARETAKER.......................................................... 62
TABLE 6.18: ACCIDENT OCCURRENCE RATES IN THE DIFFERENT CARETAKER
CATEGORIES CONTROLLING FOR A G E ...................................................... 63
TABLE 6.19: ENVIRONMENTAL SAFETY AND ACCIDENT OCCURRENCE
RATES...................................................................................................................... 66
TABLE 6.20: ACCIDENT TYPE BY PLACE OF ACCIDENT..............................................67
TABLE 6.21: ACCIDENT OCCURRENCE BY THE LIGHTING SYSTEM .......................69
TABLE 6.22: ACCIDENT BY SOURCE OF COOKING FUEL.............................................71
TABLE 6.23: DISTRIBUTION OF ACCIDENTS BY TYPE OF COOKING FUEL.......... 72
TABLE 6.24: SUMMARY OF COURSE OF ACTION TAKEN ON THE SPECIFIC TYPES
OF ACCIDENTS....................................................................................................73
TABLE 6.25: TIME TAKEN TO ACT FOLLOWING SPECIFIC ACCIDENT
OCCURRENCE......................................................................................................75
TABLE 6.26: POISONING: TIME OF ACTION....................................................................... 75
TABLE 6.27: OUTCOME OF ACCIDENTS............................................................................. 76
LIST OF FIGURES PAGE

FIGURE 1 - TYPE OF HOUSES......................................................................................................52

FIGURE 2 - TYPES OF ARTIFICIAL LIGHTING..................................................................... 53

FIGURE 3 - SEVERITY OF ACCIDENTS.................................................................................... 56

FIGURE 4 - ACCIDENTS BY AGE OF CHILDREN.................................................................. 59

FIGURE 5 - ACCIDENT OCCURRENCE BY EDUCATION OF MOTHER.......................... 61

FIGURE 6 - ACCIDENT OCCURRENCE AND OCCUPATION OF MOTHER....................62

FIGURE 7 - ACCIDENTS BY AREA OF RESIDENCE.............................................................64

FIGURE 8 - ACCIDENTS IN RELATION TO TYPE OF HOUSE............................................ 65

FIGURE 9 - RATE OF ACCIDENTS BY TYPE OF DOMESTIC LIGHTING.......................69

IX
EXECUTIVE SUMMARY

A cross sectional survey was carried out in Kisumu Municipality, Kenya to measure the

extent and determinants of Home Accidents in 1465 children under the age o f five years who

were obtained from a total o f 774 households. The objectives o f the study were:

a) To determine the prevalence o f Home Accidents in children under five years, and the

extent o f the injuries resulting from these accidents with emphasis on the three most

common ones; namely bums, poisoning and falls.

b) To determine and compare the socio-demographic and environmental determinants of

accidents in these children.

c) To determine the types o f injury, severity o f injury, the immediate intervention and the

outcome status o f the victim following the occurrence at home, o f an accident.

d) To identify and recommend preventive measures that are feasible within the household

and its environment aimed at reducing the 0-5 years morbidity and mortality due to

accidents.

The sample was obtained randomly using multistage sampling procedure and the data was

collected over a period of one month from the sampled clusters (villages and town estates).

The child under five was the sampling unit. The main tool o f the study was a semi structured

questionnaire administered by trained enumerators, fluent in the local dialect (Dholuo),

Kiswahili and English. Data analysis was done using the SPSSPC computer program.

The study sample was found to have a male:female ratio o f 1:1.04 and a population

distribution not different from the national one. Judging from the parents education and

x
occupation status, as well as the housing, they were mainly of low to middle socio-economic

standing. Mothers were leading (80%) as the caretakers o f children. The main caretaker

other than parents was the maid (11%), followed by the grandparents (4%). The maids were

o f primary level o f education and mainly aged 13-18 years.

This was a predominantly urban (63%) population, 22% were in the peri-urban and slum

areas while 15% were in the rural area.

Observation o f risk factors, revealed that 59% o f households used unprotected or open fire

places; 58.1% homes had children under five years playing without routinely being

supervised; 29.4% homes had custody o f medicines and potentially harmful chemicals

within easy reach o f the children; Sharp instruments were within easy reach o f the

children in 36.1% of the homes; litter was scattered around the compound in 31% of

homes; 36.6% o f the households were assessed as crowded and only 15% had homes

being used as work place such as workshops. These risks pose unknown danger and

contribute to the occurrence o f accidents such as bums and scalds, poisoning, falls, and

others.

A total o f 456 children were reported to have suffered accidents within one year giving a

prevalence rate of 32.1%. The most commonly occurring accident was due to falls (42%),

followed by bums (31%), and poisoning (10%). The other accidents such as cuts, drowning

and suffocation altogether formed 17%. About half o f the injuries were o f moderate severity

while 32% were mild and 18% severe. The severe injuries were contributed to mainly by

bums/scalds and falls in equal proportion and totalling 80%. The other 20% were due to

xi
poisoning and other accidents.

O f all the accidents, mothers thought that 55% were preventable, 34%, not preventable

while 11% did not know.

The most important factors significantly associated with accidents occurrence were: the age

o f a child; gender or sex was not found to predict one's risk o f getting an accident ( 33.6%

and 30.8% for boys and girls respectively).

The none educated mothers, unemployed mothers and the house wives were associated with

lowest accident rates (29%) because they took direct care o f their children as compared to

the mothers with secondary or higher education who were more likely to leave their children

in other people’s hands when they went to work in formal or professional jobs (35%).

The leading accident rates were found in those staying in semi-permanent houses, and in

those staying in peri-urban and slum areas (42%) as compared to the urban (32%) and the

rural with 22% (p=0.0000). Accidents occurring within the house structure predominated

(51%), whereas those occurring in the compound were 49%. The kitchen was found to be

the place of accidents for the majority o f accidents mainly bums (50%) and poisoning

(39%). It was found that the accident rates were slightly higher in those homes with

observed risks as open fire places and where children played without supervision although

this was not statistically significant. Most of the accidents were found in households using

hurricane lamps (55%) and the open flame lamp (30%). The highest accident occurrence

X ll
rates were observed in those using firewood (35%), then those using gas (33%), charcoal

(32.5%), lastly kerosene (28%) and electricity (28%).

The course o f action following accidents was found to be appropriate as 88% o f them

were acted upon within two hours ( or 30 minutes in the case o f poisoning). O f all the

accidents, 25.5% resulted in admissions to health care institutions; 38.5% treated as out­

patients; 31.5% were given home remedy; 2.5% were given other treatments such as

prayers; and 2% were not acted upon. Most of the admissions were due to bums and

scalds (46%), then falls formed 21%; and lastly poisoning (15%).
b

It was noted that o f the accidents that led to admissions and outpatient treatments,

bums/scalds and falls were the leading; those that were treated as out patients, bums were

still leading together with poisoning. This shows the seriousness o f these accidents as

compared to falls which are leading in the treatments given at home and those that were

not treated at all. It then follows that poisoning, bums and scalds have the most

implications on the health system in general. These were all statistically significant

findings.

The outcome o f injuries due to home accidents was satisfactory. Many (85.7%) resulted in

recovery; while 12.2% had residual damages and chronic problems and 2% resulted in

death.

Bums and scalds and falls contributed the most to residual damages and scarring. These

meant a certain amount o f chronic problems and therefore additional burden after the

xm
initial recovery or partial recovery.

The deaths occurred due to poisoning (67%) and falls (11%) and others (22%). The

highest fatality rate was found in the category o f poisoning (13%); followed by others

(accidents other than bums and scalds, poisoning and falls) (5%); and falls (0.5%). There

were no deaths reported with bums.

From this research it is recommended that more commitment and research is still needed in

Kenya in the area o f accidents prevention and particularly on the specific contributory

factors. In order to curb childhood accident mortality and morbidity therefore, children

should be given more supervision which is age specific and appropriate as well as be

encouraged to play out o f the house in deliberately made safer environment and

specifically not be allowed in the kitchen. There should also be laws that ensure protection

of fireplaces, safe keeping o f medicines and harmful substances, which can be monitored

by health workers at the grassroots level, advocacy for repeated education o f caretakers

such as maids and accident prevention programs with active parental participation.

xiv
CHAPTER ONE

INTRODUCTION
INTRODUCTION

1.1 GENERAL INTRODUCTION

There are a number o f factors that give accidents a prominent position among public

health problems. Analysis shows that there are -few countries where accidents do not

appear among the five leading causes of death. In the american region in particular,

accidents are among the five leading causes o f death in all countries, whatever their level

o f development1.

The consequences for health are tremendous and greatly underestimated by the public

and even by the top-level decision-makers. Indeed, accidents are a major cause of

demands on the health system, whether at the primary health care level or at the hospital

level. On average in the industrialized countries, and also in many developing countries,

one hospital bed in ten is occupied by an accident victim1.

Accidents as a public health problem have been ignored world-wide as far as allocation of

funds is concerned. This can be seen in the USA (United States o f America) where

although accidents are responsible for the loss of more years o f productive life (YPL) than

cardiovascular diseases and cancers together, federal expenditure on accident research is

only one-tenth o f the research expenditure on cancer and less than one-fifth o f research

expenditure on diseases o f the circulatory system2. Contrary to many preconceived ideas,

accidents and their sequelae can be prevented if approached practically.

Accidents are responsible for significant mortality among young people. Moreover there

1
is excess male mortality from accidents, a characteristic that appears right from the first

year o f life2. Accidents also often cause disability.

In the public welfare hospitals in Paris for example, road accidents, occupational

accidents, domestic and sports accidents and suicides or other forms o f physical violence

are responsible for over 10% o f all admissions o f children, and injuries represent the third

most common cause o f admission, immediately after infectious diseases and

malformations3.

Despite the amount o f theoretical knowledge about the circumstances in which the

accidents occur as well as on the precautions that should be taken in order to reduce the

risk knowledge is still lacking on how much should be done towards accidents. There is

also lack o f prioritisation o f the real problems. The result is that nothing has been done,

especially in the developing countries, by way o f deliberate plans to reduce home

accidents in children.

WHO (World Health Organisation) has recognized this need and indeed is participating

actively in promoting specific research on accidents as well as the formation o f programs

such as the injury prevention program. The Child/Adolescent component of the WHO's

Injury Prevention Program has two main objectives;

• To promote epidemiological research on accidents in young people, particularly where

such research is negligible, in order to collect sufficient information on the nature and

the extent o f the problem to establish prevention policies.

2
• To evaluate these policies periodically and make the evaluations available to the

member states o f WHO so as to provide them with useful information on various

national experiences and thus assist them in formulating their own policy and defining

their priorities.

The socio-economic cost o f the problem must also be taken into account when

determining priorities. Accidents are expensive, certainly much more expensive than their

prevention, even though the cost o f prevention is greatly underestimated. But a large

number o f preventive and educational activities have been developed in various parts of

the world without ever being evaluated, and it is to be feared that much money has been

spent in programmes o f doubtful efficacy4.

3
 CHAPTER TWO

LITERATURE REVIEW
LITERATURE REVIEW

2.1. HOME ACCIDENTS IN GENERAL

The problem o f accidents in children and adolescents has been documented by WHO in

fifty countries including some developing countries unfortunately Kenya was not

included. In Egypt, among boys 5-14 years old, accidents accounted for 7.6% o f all deaths

from 1975-19795:

In a WHO study, accidents were found to be the leading cause o f death in different age

groups with different frequencies in various developing countries: In the 1-4 years age-

group, it was found to be the leading cause of death in two developing countries; In the 5-

9 years age-group, it was leading in seven out o f eight countries; In 10-14 years in 6 out of

9 countries6.

Rivara FP reviewed the recent literature on childhood injuries using a conceptual

framework for injury prevention and loss reduction. The framework applies the

epidemiologic factors o f host, agent, and environment to the study o f the events leading to

an injury and the injury itself. The role o f energy transfer as the etiologic agent for injury

was examined. He found that there is a need for further research involving the

characteristics o f the host, identification and modification o f injury vectors, and changes

in both the physical and sociocultural environments7.

In an attempt to test the assumption that health education directed at parents and children

can reduce childhood accidents, a controlled study was carried out in Ely , one area of

4
Cardiff, using conventional health education techniques, the campaign was carried out in

June and July 1981 and monitored by the numbers o f injured children attending the local

Accident and Emergency Department. Comparison o f accident numbers in Ely between

1980 and 1981 and between Ely and the whole o f Cardiff in 1981 showed no significant

change. A slight increase in trivial injuries suggested an increased willingness to attend

hospital. There was no change in the age distribution o f victims. The benefits of health

education and alternative methods o f accident prevention are discussed and the need for

further research is emphasized8.

The following are the results of a controlled clinical trial conducted to evaluate the

implementation o f a health education program intended to reduce the risk o f childhood

household injuries. The study population was randomly assigned into two

demographically comparable groups. Only the experimental group mothers received an

educational intervention consisting o f a tutorial, home safety-proofing assignments, and

follow-up. The homes o f the two groups were later assessed for hazards during an

unannounced visit by an interviewer who did not know to which group each home

belonged. A home safety score mean for the two groups was almost identical. The

program stimulated heightened interest and stated intent to improve, but did not result in

actual reduction o f household hazards. Active health education, as used and evaluated in

this study, appears to have limited effectiveness when applied to home safety. Approaches

such as "passive" measures may offer greater potential for household injury reduction.

Injuries claim the lives of more children each year than the next six leading paediatric

disorders combined, and produce injuries that require medical attention for one in three

5
children. In the pre-school age group, 91 per cent o f these accidents and over one-half the

resultant fatalities occur in the home9.

There was a case control study done where parents o f 171 children coming to the

Yale-New Haven Hospital Primary Care Centre for their 6-month check-up were

randomised into an intervention group and a control group. Parents in the intervention

group received a three-part individualised course in child safety that required active

parental participation. Parts 1, 2, and 3 were given at the 6-month, 9-month, and

12-month well-child visits, respectively. Parents in the control group received routine

safety education as provided at well-child visits. A total o f 109 home visits (55 for the

intervention group and 54 for the control group) were made out o f the families who

completed the educational phase of the study. Safety knowledge, number o f hazards in the

home, and reported accidents were assessed by a "blinded" community health worker

approximately 1 month after the 12-month well-child visit-. Parental safety knowledge was

assessed based upon pictorial hazard recognition. O f 13 possible hazards, the mean

number o f hazards recognised by the intervention group parents was 9.4 versus 8.4 by the

control group parents. A hazard score was determined for each family based on nine

possible hazards observed at the home visit. The mean hazard score for the intervention

group was 2.4 versus 3.0 for the control group. Parentally reported accidents and

accidents reported in hospital records were similar for both groups. Results o f this study

suggest that age-appropriate safety education that is repetitive and individualised and that

requires active parental participation results in an increase in parental knowledge and an

improvement in certain safety practices10.

6
 j-xx- * . the causes of serious domestic scald injuries in Birmingham were
In a different study, u*
Seventy-eight people sustained such scalds, 11 adults and 67
studied for one year. j ^ r

children Only 51 of the patients received satisfactory first aid. Most accidents could have

been avoided and the severity o f the injury would have been less if appropriate first aid

had been given11.

Another survey which w as a prospective one, was carried out over a three month period to

assess the incidence o f accidental poisoning in children. The survey also examined the

ole o f child resistant closures (CRC) and other form o f packaging in the prevention of

poisoning Two hundred and six children with suspected poisoning attended the casualty

departments of the th re e Dublin children's hospitals surveyed. Ninety three percent o f the

children were less than flve years old- M edications were thought to have been ingested by

65%, household or g ard en in g products by 34% and plants by 1%. Most children had no

symptoms and received no treatment except emesis or oral fluid and demulcents. Twenty

five percent required adm issio n to hospital. There wer.e no fatalities during the study

period Only 8 5% o f the medications involved had been in containers with CRC's.

Twenty percent were d i blister Packs or striPs- A substantial number o f poisonings

occurred at a time w he*1 medicines were not in their normal place although in most cases

they were still in the o rig in a l container. The author proposed a two-phase approach to

poisoning prevention, r* * ore widespread use o f CRC's and public education on safe storage

of medicines12.

More than 153 000 A it^ e ric a n s die annually as a result o f accidental injuries-nearly half of

7
them from motor vehicle accidents, the rest from falls, bums, poisoning, and other causes.

Injuries are the leading cause o f death between ages 1 and 44, and account for

approximately 55 percent o f fatalities for those aged 15 to 24. Further, it is estimated that

in 1979, 70 million people suffered non-fatal accidental injuries requiring medical

treatment. In 1980, the damage, injury, and lost productivity resulting from accidents cost

the Nation an estimated +83.2 billion. Accident victims are not distributed evenly

throughout the population. Teenagers and young adults have the highest motor vehicle

death rate; fatal falls, which occur primarily in the home, disproportionately affect the

population aged 75 and over; and children 10 years and younger are a high risk population

for bums. Accident prevention programs should be based on epidemiological

documentation o f injury problem areas at the State and local levels. Prevention measures

include public education, skill development, safety engineering, environmental

modification, legislation, regulation, and enforcement13.

An Asian study describes the statistical analysis o f 629 bum patients treated between

January 1989 and August 1990. The analysed data include age, sex, cause o f bum and

mortality in relation to age, cause and extent o f bum injuries. Additional information with

regard to socio-economic status, marital status, place o f bum, family size, type o f bum,

time of accident and time between injury and hospital admission was obtained from

analysis o f 271 o f the 629 patients admitted between January 1990 and August 1990. In

this series, adolescent and young adults (11-40 years) comprised 64.8 per cent o f the

patients. Males formed 54 per cent and females 46 per cent o f the total bums. 82.65 per

cent o f the patients received their bums at home. Most o f the patients belonged to the low

8
or lower middle socio-economic strata and had large families. 95.5 per cent o f the bums

were accidental in nature. Flame bums comprised 67.9% and scalds comprised 16.4% of

the total bums. The overall mortality rate was 48.3%. The mortality rate was

comparatively low in children (20.1%). Flame bums resulted in maximum deaths

(62.1%). In patients with over 40 per cent bums, the mortality was about 80%. There were

no survivors in patients with over 70% TBS A bums in the series14.

Bunk beds are commonly used in American households, but no studies have been done to

determine if they are safe. Selbst S.M. et al, prospectively studied the incidence,

epidemiology, and outcome o f injuries related to bunk beds. They interviewed all patients

with such injuries who presented to the emergency department between February 1987

and February 1988. A control group of children who use bunk beds but who came to the

emergency department for another reason were also interviewed. There were 68 injured

children and 54 controls during the 1-year study period. There were 47 injured children

(70% o f this group) and 26 control children (48% o f this group) younger than 6 years,

which is below the age recommended by the Consumer Product Safety Commission for

bunk bed use. Carpeted floors were significantly more common in the control group, 67%

(36 children) versus 42% (26 children). Injuries occurred most often when the child fell

from the top bed (38 children [58%]), fell off the ladder (7 children [11%]), or fell off the

bottom bed (8 children [12%]). Injuries occurred during sleep (19 children [29%]), getting

in or out o f the bunk bed (13 children [20%]), or playing in or near the beds (28 children

[43%]). O f those injured while asleep, 13 of 19 children were younger than 6 years. Head

injuries accounted for half the trauma (35 children [52%]), and extremities were involved

9
in 16 patients [24%]. The most common injuries were lacerations (27 children [40%]) and

contusions (19 children [28%]), but 8 children (12%) had concussions and 7 children

(10%) had fractures. Six children (9%) required admission to the hospital. Head and face

injuries were significantly more likely if the top bed had no side rails. These data suggest

injuries could be prevented if side rails were mandatory for all top beds, young children

were not permitted to sleep in bunk beds, and all children were encouraged not to use the

beds for play15.

Inadvertent aspiration or ingestion of products in children is a reportable problem. The

National Electronic Injury Surveillance System (NEISS) in America, monitors 119

hospitals for product (nonfood) hazards through a network o f emergency room physicians.

Coins (52%) remain the most frequently ingested object (1988-1989). Pins, including

nails and bolts, are second in rank. Spheres (e.g., balls) are the most common cause of

death. Most deaths (97%) occur in the home. Younger children (13.4 versus 22.7 months)

are more at risk for death. This analysis suggests that shape, as well as availability, is

important in children's injuries and deaths. Accurate reporting to NEISS o f all aspirated

objects remains essential so that product safety and the public's health can be improved.

Food product injuries, however, are not reported to NEISS16.

The primary audience o f poison prevention programs is the parent(s) o f children less than

6 years o f age. Literature review reveals few references assessing other caretakers as risk

factors in childhood poisoning. The frequency and severity o f calls to the poison centre by

nonparental caretakers was studied. A total of 4,205 poisoning cases involving children

10
under 6 years o f age were analyzed. In 11.9% o f the cases the caretaker at the time of

exposure was someone other than the parents and the site o f the exposure was other than

the child's home. O f the 3,702 cases where the exposure occurred while the child was

supervised by the parents in their home, 90.2% were treated in the home and 72.4%

required dilution only. Grandparents represented 39.6% o f caretakers other than the

parents. In these cases 44.6% required treatment beyond dilution, indicating more serious

exposures in this group. Ingestion o f cardiovascular drugs occurred in 12.3% o f calls from

grandparents as opposed to 0.7% o f calls initiated by parents. Poisoning exposures

involving children under six years o f age, where the caretaker is other than the parents,

and the site is other than the child's home, are often more serious. Poison prevention

information programs are needed to reduce the risk factors among this group17.

In order to know the importance and epidemiology o f childhood accidents, a prospective

study was made in a Spanish region. Subsequently, the information was coded and

analyzed at the Murcia University. This study was founded in death certificates and 3,344

questionnaires from children under 7 years old, accidented in 1986 and attended in

emergency departments in that region. Accidents were the main cause o f death in children

aged over one year, and a common cause o f consultation in emergency departments. Its

characteristics are common to most countries, but the author found some differences in

that region. He also recommended more effort in the study and prevention o f childhood

accidents18.

Among calls put through to the Mobile Emergency Unit o f a Paris urban area district for a

11
paediatric problem over a five-year period (n = 630 ), 19% were for a household injury

(with or without physical injury) and 22% (n = 712) were for a physical injury outside the

home. Thus, injuries were the reason for over 40% o f paediatric calls and half o f these

injuries occurred in the child's home. Household injuries were serious, with a pre-hospital

mortality rate o f nearly 5% and intensive care treatment in one-third o f cases. Fifty-five

per cent (60% in boys) o f household injuries occurred in pre-school-age children. Half

these injuries (bums excluded) were physical injuries (66% in boys). Forty per cent of

household injuries (n = 251) were caused by a fall, from over one meter in half the cases.

Most o f these injuries occurred around meal times (75% o f the cases). As compared with

out-of-home childhood injuries, pre-hospital mortality rate was significantly higher and

single injuries significantly more prevalent in household physical injuries. Prevalence of

lesions o f the face, head and neck, chest and abdomen, and pelvis were comparable for

out-of-home and at-home physical injuries; conversely, lesions o f the upper limbs were

more common in household injuries, whereas lesions o f the lower limbs were more

prevalent in out-of-home trauma. Lastly, household physical injuries occurred mainly in

children under five, whereas out-of-home childhood injuries were more common after

five years o f age19.

In Sweden, a total o f 1,590 product-related accidents in the home in children (0-15 years),

presented at two hospitals, were registered in a well defined geographical area to analyse

injury mechanisms, injury panorama and potential risk products. This was done over a

one year period prospectively . The incidence o f accidents in children was 279 per 10,000

inhabitants per year. A tendency o f a smaller incidence with age and a significant higher

12
incidence in boys were found. The most common localisation o f lesion was the upper

extremities (41%) and face and skull (30%). Household furniture and constructional

features of the house caused most accidents and an alarming high number o f bums were

registered20.

Sweden has the lowest childhood injury rate o f any country in the world due to a

combination o f factors including the special characteristics o f Swedish society and an

energetic 35-year campaign. Contributing societal characteristics are a small, relatively

homogenous, health conscious, law-abiding population that values children. Key factors

in the campaign have been support o f trauma surveillance systems and injury prevention

research, ensuring safer environments and products through legislation and regulation,

and a broad-based safety education campaign using coalitions o f existing groups and a

strong factor o f community participation. Emulating the strategies used in the Swedish

campaign would markedly reduce the number o f children killed, injured, and disabled

from trauma21.

Five hundred and eighty-three children (0-18 years old), consisting o f 33.4 per cent o f all

bum inpatients, were admitted to the University o f Alberta Hospitals over an 11-year

period (January 1978 to December 1988). Demographic and outcome variables, in

addition to aetiological factors, were examined. 48.4 per cent o f bums occurred in

children less than 4 years o f age, with males predominating in every age group. Children

had smaller bums, a higher incidence o f scalds, less inhalation injuries and a lower

mortality compared to adult bum patients admitted over the same time period. There was

13
a low incidence of confirmed child abuse by bums (1.4 per cent). High-risk environments

identified were the home (74.6 per cent o f bums) and recreational settings (12.4 per cent

of bums), mainly occurring around campfires. Native children were overrepresented in the

bum population compared to the general population by a factor o f approximately 10:1.

Scald prevention, high-risk environments (home and recreational), high-risk populations

(male and natives) and unsafe practices with flammable liquids (petrol in particular)

should be emphasized in paediatric bum prevention-programmes22.

A prospective study o f 394 burned children (in-patients) up to the age o f 12 years old was

carried out for the period from January 1984 to December 1986. They were categorized

into three age groups, the infants and toddlers 0-2 years, early childhood 3-6 years and late

childhood 7-12 years. In the first two groups scalding was the predominant cause of

injury, while in late childhood there were many more flame bums. Ninety-five per cent o f

the accidents occurred at home and the majority happened in the presence o f parents. The

presence o f parents was not a deterrent to the accident but ensured speedy transport to the

hospital. In this review 3 per cent o f patients sustained more than 50 per cent BSA bums,

there were 12 deaths with a mortality rate o f 3 per cent. An intense campaign to make

parents aware o f the risk factors and their avoidance is required to reduce the number o f

bum accidents23.

In a study done in North Staffordshire, in the U.K, sixty-nine children aged under 15 years

were identified from coroners' records as having died as a result o f an accident between

1980 and 1989. Road traffic accidents (RTAs) accounted for the majority o f cases (n =

38; 55% o f total) and in almost all o f these, the unsafe behaviour o f the child was

14
considered to be at fault. Most fatal accidents occurred between 15.00 and 21.00 hrs and

within 2 km o f the child's home; the majority o f children killed were not supervised by an

adult at the time o f the accident. Considerable variation in mortality within the district

was observed with several areas having a rate significantly higher than the district as a

whole. Head injury was the most commonly recorded cause o f death (n = 37, 53%)

confirming the importance o f head injury as a cause o f childhood mortality. Road safety

educational and engineering measures as well as adequate adult supervision and

awareness could have prevented the vast majority o f these accidental deaths. Coroners

records are a vital and often poorly utilized source o f locally relevant information

regarding childhood accidents which should be o f use to all interested agencies including

child accident prevention groups24.

Within a population-based register of childhood injuries(in a another study in Britain),

641 home accidents to children aged under 2 years were recorded in six hospitals. The

annual rate was higher in children over 1 year (6.6 per 100) than infants (2.7 per 100). A

male predominance was observed only after the first birthday. Most injuries were related

to physical trauma (mainly falls), but bums and poisonings were frequent in the second

year. Most lesions were benign, except in the case o f falls from a height (from a window

or baby furniture). Nineteen per cent o f the children were admitted to hospital. There were

15 skull fractures, two fatal cases and two severe sequelae. Child abuse was seldom

suspected and the reasons for this are discussed. The study o f the circumstances in which

accidents occur leads to the conclusion that the main aspect o f prevention should be

passive protection ensured by the creation of a safe environment (compulsory safety

15
standards for baby furniture, child resistant packaging), but that parents' information and

education should also be developed, with emphasis on knowledge o f children's normal

psychological and motor development and abilities .

In Britain, a study was done in an effort to prevent childhood injuries, whereby the

Hennepin County Bum Centre began a joint venture with a public health agency to reduce

home hazards for children less than 6 years o f age who were treated at the bum centre.

Children were referred to a public health nurse for a home safety assessment. During the

initial visit, child hazards were identified and recommendations were made for reducing

injury risk. Parental compliance with recommendations was evaluated during a second

home visit approximately 1 month later. Home safety assessments were completed in 21

homes, and a total of 131 recommendations were made, Bum prevention

recommendations accounted for 43.5% of the total recommendations, poison control

recommendations accounted for 36.6%, and other injury control recommendations

accounted for 19.9%. Parents complied with 43.5% o f the total 131 recommendations.

Bum prevention recommendations had a compliance rate o f 19%. The program had a

positive effect on reducing home hazards. It was an appropriate response by practitioners

already involved in the care o f children, many o f whom are at risk o f injury26.

16
2.2. HOME ACCIDENTS IN DEVELOPING COUNTRIES

It was long believed, and is still believed by some, that accidents occur only in developed

countries, being the price that has to be paid for industrialisation, technology,

urbanization, and motorization. On the contrary, in the developing countries, accidents

are perhaps just as common, and their consequences are even more serious making the

accident mortality higher in the developing countries. This is due to the differences in

living conditions i.e. rural settings, where there are many dangers associated with an ill-

controlled and frequently hostile environment, in addition to the greater difficulty in

obtaining good medical treatment. These dangers include, unprotected and uncontrolled

dumping o f chemical wastes, solid wastes (such as broken bottles, sharp metals among

other things), uncovered holes, thorns, trees to climb and so on. The health care system is

inadequate, thus morbidity is prolonged and mortality increased.

This phenomenon applies similarly in the rapidly expanding slum areas within which the

so called urban poor live. They are particularly threatened by inadequacy o f social

services, poor environmental sanitation, poor housing and overcrowding.

In a study done in Marigat, Baringo district, Kenya, Oloo found an incidence rate of

7.04% in the under 20 years age-group. This same study showed that bums and scalds

from fire and boiling fluids e.g. open fire, water, porridge, tea, fat and oil; poisoning from

kerosene, medicines, chemicals; falls from pit latrines, ditches and play grounds form

some of the most common injuries among young children. Others are thorn pricks, insect,

snake, or animal bites. In this Baringo study, 58.6% o f the accidents occurred at home

17
and 41.4% outside the home. It was found that the commonest injuries were falls (27%)

followed by bums (26.1%), and that most o f these were in the 0-5 years age-group. O f the

accident victims, only 45% attended health facilities, 88% o f them were treated with or

without follow up and 6% o f them had no treatment. 81% o f the victims recovered

completely without any residual disability and, deformities were caused by accidents due

to falls, snake-bites and bums (extensive scars). Others suffered chronicity e.g. chronic

ulcers from thorn pricks .

In a study done in MCH rural centre o f Kombole in Senegal, in 1978 by Sonakho, 88% of

the accidents occurred at home and, 33% attended health facilities and 20% o f the

accident victims were found to have residual disability in the form o f lasting physical

28
damage .

In Turkey Bertan and Muftir found an incidence o f 21.6% overall in those under 15 years

and a fatality rate o f 6/1000 was found28.

18
 CHAPTER THREE

RESEARCH PROBLEM
RESEARCH PROBLEM

Accidents are a common cause o f illness, loss o f life or disabilities. Government Health

information statistics support this observation and show that children are commonly

involved. It is also not uncommon to read in our local news about children who have

perished in various forms o f home accidents. The current attitude is fatalistic, regarding

such events as inevitable and disregarding the role o f people in accident prevention, and

the fact that most accidents are preventable.

In spite of increasing numbers o f injuries, disabilities, and deaths caused by accidents,

there is scanty information on the involvement o f persons by age and sex-groups, the

place of occurrence and, the outcome particularly among children.

Accidents have been found to account for about half o f all deaths in the 10-19 years age

group and have left many maimed and disabled29. Physical environments, exposure to

hazards and psychosocial stresses including overt child abuse also explain why some

children have recurrent injuries30.

The Daily Nation newspapers revealed that by the end o f one year, approximately eighty

children are reported in the news to have died following house fire accidents, and other

shocking and mostly preventable incidents . Fewer numbers are reported to have perished

in the accidents such as injury due to falls, poisoning, drowning, suffocation, animal or

snake bites and so on31. It is to be remembered that there is some degree o f under­

reporting due to a variety o f reasons and therefore these records cannot be used for

19
scientific purposes but as a general overview o f the extent o f the problem.

The information obtained from New Nyanza Provincial General hospital is non specific

and not very helpful due to the fact that patients presenting to the hospital with any form

o f accidents are lumped together and neither the age nor age-groups as well as types are

indicated. There is also the inability to access to all the records on accidents in children

under five due to the data omitted by some o f the health staff when managing the patients.

This makes analysis very shallow. The closest analysis showed that within a period o f 3

months, about 24 children under five were admitted with various forms o f accidents 3 of

which were victims o f RTA ( Road Traffic Accident) .

However, in the Kisumu district hospital, analysis showed that the number o f children

under five years admitted with bums poisoning and falls with injuries in the months of

April, May, June and July in 1992, were approximately 40. In the same period, 19

children under five were treated in the outpatient department with bums. This does not

give the whole picture of the problem because other hospitals and health centres as well

as clinics do not keep any useful records.

The available records showed that most o f children who suffered the accidental injuries

e.g. bums, were left in the house alone when their parents went to fetch either water,

firewood or other commodities. Others were playing alone in the compound e.g climbing

trees when they fell and got fractured bones.

20
Whatever the type o f injury, the final severity, would to a substantial extent, be

determined by the speed and the efficiency o f the treatment. The children get extensive

injuries and scaring especially from bums. Other forms o f accidents result in

disfigurement and gross disability. The cost of treatment is also a great burden to these

families.

There are many cases that do not reach the health centres because they are treated at the

primary level, some o f which should not have reached health centres end up doing so due

to complications resulting from inappropriate treatment.

Death rate due to accidents is just a tip of the iceberg, as for every accidental death there

are several accidental injuries. Most of these are slight but requiring action by health

services, some are severe and likely to leave permanent after-effects. One child in every 5

to 10 sustains an accidental injury each year33.

3.1. JUSTIFICATION OF THE STUDY

Children are less able than adults to recognize hazardous environments and negotiate

risks. This could be due to their underdeveloped reasoning and psychomotor systems

combined with the fact that they are also curious, exploring and, testing their skills. The

under five years age-group are particularly a priority group as they are the most at risk.

Children in the 0-5 years age-group spend most of their time around their homes and

21
therefore this is where most injuries would take place, as compared to the older children

who go to school, play further away from home and engage in several out-of-home

activities25.

In view of the foregoing, there is evidently a need to prevent home accidents as it has been

well observed that one aspect that singles them out is that many o f them are preventable

by individuals through simple changes in behaviour which, according to Bull’s studies,

may be inculcated in places such as school, and promotion o f use o f safety equipmentj4.

Methods o f prevention o f home accidents or domestic injuries follow logically from

knowledge o f their cause.

Approaches to accident prevention fall into three categories:

a) Education which involves the increase in knowledge o f the problem and its solution, a

change in attitude towards it and an actual change in behaviour,

b) Engineering, the design, manufacture and marketing o f safer products and the design

and planning of the environment in which they are used;

c) Enforcement, the role o f legislation, regulations and standards in accidents prevention.

Accidental injury is one o f the most important epidemics in the whole world today, and is

especially important as a cause of death and disability in childhood. Many environmental

factors are important causes o f accidents, but there is no organisation within which health

personnel, industries and voluntary agencies, can pool their experience and co-ordinate

their efforts to reduce these environmental risks. A joint committee on childhood

22
accident prevention should be formed in this country similar to that which exists in

Sweden26.

This study aims at establishing some facts about home accidents as a public health

problem and to advocate for safe environment and promotion o f education for prevention

and intervention in the occurrence of home accidents in the community. It should also

contribute towards formulation o f specific and appropriate as well as feasible control

programmes for our Kenyan set-up. The best form o f accidents control is the preventative

approach and there is need for most people to discard the popular beliefs and myths that

accidents are inevitable and are an act o f God or witchcraft .

With the changing cultural and socio-economic trends, women who have been at most

home managers, are gradually taking multiple roles in their families, either busy in the

farm or are employed in other jobs to supplement the family income. This then leaves the

children without the optimum parental care hence they become more likely to have

accidents especially within the more disadvantaged groups such as the urban poor

community who have poor housing and sanitation among other socio-economic problems.

3.2 BROAD OBJECTIVE

To find out the extent and determinants of the common home accidents in children under

five years o f age and identify the possible preventive measures against these accidents.

23
3.2.1 SPECIFIC OBJECTIVES

1. To determine the prevalence o f injuries due to home accidents with emphasis on

the three most common ones i.e. bums, poisoning and, falls in the under fives.

2. To determine and compare the socio-demographic, environmental health

circumstances of home accidents in children under five years.

3. To determine the types o f injury, severity o f injury, the immediate intervention,

and the outcome-status o f the victim following the occurrence at home, o f an

accident.

4. To identify and recommend preventive measures that are feasible within the

household and its environment aimed at reducing the 0-5 year, morbidity and

mortality due to accidents.

3.3 STUDY HYPOTHESES

1. The prevalence of home accidents in children 0-5 years is directly related to the

socio-demographic and, physical environmental safety o f the home.

2. The type and severity o f injury, each in itself is correlated with the socio­

demographic and physical environmental safety o f the home.

24
CHAPTER FOUR

STUDY AREA
STUDY AREA

The study was carried out in Winam Division in Kisumu Municipality, Kisumu District.

This area was chosen by the investigator for the simple reasons that there is no record o f a

similar study having been carried out in the area which comprises a variety o f socio­

economic groups. The investigator had good knowledge o f the area to enable easy

accessibility without much difficulty.

4.1. KISUMU DISTRICT-ADMINISTRATIVE UNITS AND BOUNDARIES

Kisumu municipality is located in Winam and Maseno Divisions o f Kisumu District in

Nyanza Province o f Kenya. Kisumu District is divided into eight administrative divisions,

fifty one locations and one hundred and fifty eight sub-locations. The divisions are

Winam, Maseno, Nyando/Kadibo, Muhoroni/Miwani, Lower Nyakach and Upper

Nyakach. Each division is administered by a District Officer (DO), who is also the

chairman o f the Divisional Development Committee (DDC). The divisional headquarters

are fairly accessible in terms of telecommunications and the road network (Source:

Kisumu District Development Plan 1993-1996).

Its strategic location at the focal point o f a communication network makes it the regional

centre for Western Kenya as well as a provincial and district headquarters.

Kisumu District lies within longitudes 33° 20'E and 35° 20'E and latitudes 0° 20'S and 0°

50'S, and covers a total area of 2,660 sq. km., o f which 567 are under water . It is the

second largest o f the four districts that form Nyanza Province. To the South-West is

25
Homa-Bay District; to the South is Kisii District; to the north East is Nandi District; to the

east is Kericho District; to the North-West is Vihiga District; and finally to the West is
«

Siaya District.

4.2. ADMINISTRATIVE DIVISIONS OF KISUMU MUNICIPALITY

Administratively, the municipality straddles Maseno and Winam Divisions. In 1971, the

urban area's boundary was extended adding nearly 400 [Link], to make a total area o f 417

sq. km. of which about 157 is water and approximately 260 sq. km. is land divided into

eight locations {see table 4.1).

Kisumu municipality is the dominant administrative, commercial and industrial centre for

the region and serves a hinterland o f about 31,000 [Link], occupied by over seven million

people.

26
T a b le 4.1. ADMINISTRATIVE SECTIONS OF WINAM DIVISION IN

KISUMU MUNICIPALITY AND NUMBER OF HOUSE HOLDS WITHIN THE

SUB-LOCATIONS.

LOCATION Sub-location No. H/Holds

Kajulu East Kadero 591
Okok 851
Got Nyabondo 562

Kajulu West Konya 1,531

Wathorego 1,052

East Kisumu Swahili (Mkendwa) 123

Dago 710
Kanyakwar 3,787
Kogony 2,693

Central Kisumu Bar 1,169

Nyahera 1,497
Korando 2,765

South West Kisumu Ojola 1,109

Kanyawegi 1,954

Township Southern Kisumu (Milimani) 2,456

Northern Kisumu (Kibuye) 6,632

Kolwa West Nyalenda "B" 4,844

Nyalenda "A" 5,958
Manyatta "A" 9,791
Manyatta "B" 3,986

Kolwa East Mayenya 379

Buoye 665
Nyalunya 1,483
Kasule 1,111
Chiga 1.461
TOTAL 59,160

27
4.3. HEALTH FACILITIES

Health services within the municipality are provided by the following institutions:

• Four hospitals, namely the New and Old Nyanza General Hospitals the Aga Khan

Hospital and Jalaram Hospital all of which provide a full range o f medical services;

• one health centre (Lumumba), which provides services to the population residing in the

old town and the peri-urban areas;

• five sub-health centres (Nyambogo, Chiga, Ojola, Ober Kamoth and Migosi) whose

main three dispensaries (Mosque, the Airport and Nyalenda) offer services

complementary to those provided by health and sub-health centres; and

• the Town Hall clinic and a mobile clinic which provide services similar to those of

dispensaries for Municipal Council staff and the population in the town centre.

While the hospitals are administered by the Government's Medical Officer o f Health and

private doctors, the other health facilities are operated by the Municipal council of

Kisumu staff.

In general, the health facilities and services provided by Kisumu Municipality and those

provided by the district are not adequate to cater for the needs o f the population. The

existing facilities are also being patronised by considerable numbers o f people living

outside the provincial and district boundaries35.

28
4.4. CLIMATE

Kisumu experiences a warm to hot and generally humid climate. Mean annual maximum

temperatures range from 25°C to 30° and mean annual minimum ranges from 9°C to

18°C, resulting in a mean annual temperature o f 23.1 °C for the town. Altitude varies from

1,144 meters on the Kano plains to 1,525 meters in the Maseno area and South Nyakach.

Mean annual rainfall is influenced by elevation and varies from 1,630 mm per year in

Maseno and 1,525 mm per year in Kibos to 1,280 mm per year in Kisumu Town and

1,620 mm per year in Ahero. Rainfall reliability follows a generally similar pattern.

29
 CHAPTER FIVE

METHODOLOGY OF THE STUDY

METHODOLOGY OF THE STUDY

5.1 STUDY DESIGN

This was a cross-sectional survey. The possible predictor variables and outcome variables

were determined at the same time. Data was collected over a period of one month within

the homes sampled as clusters o f households.

5.2. STUDY POPULATION

Children under the age o f five years within the municipality form the study population.

According to 1989 survey the general population comprised o f 229,418 o f the Kisumu

residents. The number o f children o f 0-5 years is estimated to be 20%, i.e. 45,884 children

in total.

The population within the Kisumu Municipality locations comprised o f a cross-section of

people with different demographic and social characteristics. The majority o f them

however, speak the same language o f Dholuo and are culturally similar.

The Kisumu District Development Plan, 1993/96, in its analysis o f 1989 Population

Census data, concluded that the population figures reflected a relatively stable population,

rather than one characterised by high recent migration. This is may be the case in the rural

parts o f the municipality and the old town but the rapid growth o f peri-urban squatter

areas indicate the opposite.

30
A significant number o f people from the surrounding rural areas have joined the

municipal peri-urban community, either as urban informal sector workers or as urban

unemployed.

5.2.1. SAMPLE SIZE

Using the statistical formula for cross-sectional studies36 to determine the sample size (n)

n = puoo- p ) f n - a ) x
52
Where, P= Prevalence o f childhood accidents from a previous local study by Oloo27=10%

5=5, f(l-a ) is a function o f the level o f significance.

f(l-a ) = 3.842 (constant from standard statistical tables ).

10fl00-10) x 3.842

25

= 138 — 140 children.

Taking into account that the number of children under five years on average is probably 2,

this means that the average household number to be visited was 140/2 = 70.

5.2.2 SAMPLING PROCEDURE

The sample was obtained using multistage [Link] A random listing of all the

Locations and Sublocations o f Winam Division was made and used to obtain the sample

required (see table 5.1).

The next level to the administrative locations was the sub-locations. From the

sublocation, the index house for the survey was chosen randomly as the kth house from

the local sub-chiefs house (k being any random number from one upto ten). From here

31
the next nearest house was then visited and the occupants interviewed accordingly until

the required sample size was achieved.

The sampling frame was made o f all children 0-5 years within the municipality and was

obtained from all the households in the following manner.

A total o f 30 clusters, were chosen proportionately and randomly from the population as

follows: Since there were 59,160 households; the first cluster to be chosen was where the

cumulative total o f 59,160/30 = 1972, fell. In this list this was in the sublocation of Got

Nyabondo (see table 5.1). Then the next one was where the total o f (2x1972 = 3934) fell

i.e. in Wathorego and so on. Some sublocations had 2, 3 or 5 clusters chosen from them

depending on its total number of households.

Since the number o f households to be visited was targeted at 700, a cluster then

comprised of 700/30 = 24 households approximately or a total o f 47 children under five

years per cluster.

32
TABLE 5.1: DISTRIBUTION OF CLUSTERS BY SUBLQCATION

LOCATION SUBLOCATION TOTAL SAMPLED CUM.

H/HOLDS CLUSTERS H/H

Kajulu East Kadero 591 591

Okok 851 1442
Got Nyabondo 562 1 2004

Kajulu West Konya 1,531 3535

Wathorego 1,052 1 4587

East Kisumu Swahili (Mkendwa) 123 4710

Dago 710 5420
Kanyakwar 3,787 2 9209
Kogony 2,693 2 11900

Central Kisumu Bar 1,169 13069

Nyahera 1,497 1 14566
Korando 2,765 1 17331

South West Ojola 1,109 1 18440

Kisumu Kanyawegi 1,954 1 20394

Township Southern Kisumu 2,456 1 22850

(Milimani)
Northern Kisumu 6,632 3 29482
(Kibuye)

Kolwa West Nyalenda "B" 4,844 3 34326

Nyalenda "A" 5,958 3 40284
Manyatta "A" 9,791 5 50075
Manyatta "B" 3,986 2 54061

Kolwa East Mayenya 379 54440

Buoye 665 55105
Nyalunya 1,483 1 56588
Kasule 1,111 1 57699
Chiga 1,461 1 59160

T O T A LS 59,160 30

33
A household was defined as a person or persons living together and eating from the same

cooking arrangement.

5.3. INCLUSION /EXCLUSION CRITERIA

5.3.1. Inclusion Criteria

Children were included in the study if they had the following characteristics:

• Age 0-5 years at the time o f the accident.

• Resident within the municipality.

• Willingness of the parents/guardian to participate

5.3.2. Exclusion criteria

• If there was no consent on the part o f the parents to participate in the study and also if

the children were more than five years old.

5.4. ETHICAL CONSIDERATIONS

1. The research was carried out with permission from the Office o f the President and that

of the local authorities i.e. the District heads, Municipal head, the Locational and Sub-

locational chiefs.

2. The Department o f Community Health ethical committee gave the permission to

proceed on with the field work once the research proposal was passed.

3. An informed consent was obtained from all respondents.

4. Children were assessed and those found with any kind o f illnesses were referred for

appropriate medical attention.

34
5. Feedback is to be given to the community under study and recommendations made

through the District headquarters for subsequent action when the final report is ready.

5.5. DEFINITIONS

5.5.1. ACCIDENT

For the purposes of the study, an accident was defined as an unpremeditated event that is

potentially harmful, which may be o f abrupt occurrence, affecting a child, and may

produce recognisable injury. In addition, the accident, may have resulted in some form of

medical intervention by a member o f the family. This depends on the subjective view of

the seriousness o f the accident, on the part of those responsible for the child, and on their

level o f concern, their medical knowledge and social background.

5.6 LOGISTICS

5.6.1 PREPARATIONS FOR DATA COLLECTION.

Permission for research was obtained from the Office o f the President who then

communicated with the D.O Winam. The preparatory visit to Kisumu at the D.O Winam's

office. Once the D.O, was aware of the study, he subsequently informed the chiefs and

assistant chiefs through letters. The investigator then paid visits to the relevant

sublocations to meet the assistant chiefs. The assistant chiefs, who were very co­

operative, then allocated the village elders as guides for the study group.

Meanwhile twelve research assistants, who were enrolled community nurses, were

selected from the Nyanza Provincial General Hospital. These assistants were trained in a

•JNIVER5ITV Uh
35 UBR A P *
three day workshop.

The research group then made a familiarisation tour to identify the boundaries in the areas

under study. We made schedules and informed the guides and assistant chiefs o f the days

the data would be collected in the respective areas.

The questionnaire was then pretested in thirty households under the supervision of the

principal investigator. This took a period o f two days. Following this appropriate

adjustments to the questionnaire were made and its final copies printed out for data

collection.

All other requirements such as transport and stationery were also made ready and the data

collected in a period o f eighteen days.

5.6.2 DATA COLLECTION TECHNIQUES

The main tool o f the study was a semi-structured questionnaire which was administered

by enumerators fluent in Dholuo and Kiswahili as well as in the English language for ease

of communication and to minimise instrumental errors. As already mentioned, the

personnel were trained to administer the questionnaire which had been pretested. The

mothers o f the children, or their guardians (where mothers were absent), were interviewed

once they consented. The interviewees were told what the study was all about and what it

was aiming to achieve.

36
The data was collected under close supervision o f the principal investigator. Every

evening, discussions were held to deliberate and solve any problems as well as to clarify

and correct any ambiguous, wrong and omitted entries by the research assistants.

The total time spent from preparatory visits to the local authorities, the training, pretesting

o f questionnaire and actual data collection was thirty days.

5.7. CONSTRAINTS ENCOUNTERED DURING THE STUDY PERIOD

5.7.1. COMPLIANCE

Few (about 3) respondents from Manyatta "A" Sublocation who became uncooperative,

expressed fears that:

i) The survey endeavoured to establish the number o f children they had so that they can

be told to do family planning.

ii) There was a move to establish their wealth for purposes o f burglary.

iii) Their husbands would not be happy with their decisions to talk to strangers.

Sometimes enumerators had to go back to a household to correct the mistakes or

collect data accordingly. Thanks to the mlangos or miji kumi's who were well known

to the community.

5.7.2. ACCESSIBILITY

Kisumu Municipality besides being very hot and dusty, the survey team had to climb

some hilly areas especially the Got Nyabondo, Ojola and Kanyawegi sublocations. To get

to the next household meant going down a valley and up a ridge. But with determination,

37
the desired objectives were achieved.

5.8. APPROACHES TO ANALYSIS AND PRESENTATION OF DATA

The open-ended questions from the questionnaire were coded. Data collected was entered

into the computer and cleaned for analysis using D-base program. The data was then

translated to the SPSSPC+ (Statistical Package For Social Scientists) programme for

analysis. The Chi-square statistical test o f significance was done to assess significance of

the associations. Pictorial presentations (e.g. pie charts, graphs), were made using the

Lotus 123 computer software.

The circumstances surrounding home accidents that were studied during the survey were

categorised as dependent or independent variables. They could also be viewed in three

levels: The descriptive demographic and social characteristics that existed

(i) before the occurrence o f the accidental injury;

t

(ii) during the occurrence o f the accidental injury as well as describing the accident; and

(iii) those following the injury and hence leading to the outcome.

DEPENDENT VARIABLES:

These were those representing the extent o f accidents namely;

• The occurrence o f home accidents

• Prevalence o f home accidents

• Type, severity and outcome o f the accidents

38
INDEPENDENT VARIABLES:

These were representing the possible determinants o f home accidents and other

circumstances surrounding these accidents namely;

• Age o f child

• Sex o f child

• Supervision during play

• Type o f caretaker - age and education level

• Parents age, education and occupation

• Place of accident

• Action taken following injury (type and time taken to act)

• Presence o f risk factors in the environment such as ; Scattered litter in compound, Safe

custody of medicines, chemicals and harmful objects, Use o f home as work place or

workshop as in carpentry and commercial fish flying, Protection o f fire place (whether

open or not)

• Type o f cooking fuel

• Lighting system

• Type o f housing (includes number o f rooms, house size, persons per household and

building materials)

CONFOUNDING VARIABLES:

The confounding variables which were controlled for were the Socio-demographic

characteristics o f parents and other caretakers namely;

39
Age, Sex, Education, Occupation, Residential location.

Elements o f the data collected consisted of;

1. Basic demographic data from the households e.g. name, age, sex, number o f siblings,

birth place, (or location o f origin), area o f residence.

2. who is the caretaker o f the child.

3. conditions o f the physical environment.

4. History of any accidental injuries over the last twelve months: place of the accident;

circumstances and causative agent.

5. Type and severity o f accident.

6. What action was taken and type of treatment given.

7. What was the residual state or outcome of the child following the injury: Immediate

sequelae and site o f wounds.

8. Composition o f the family: Parent(s) marital status; Parents occupation; Parents

education and residential status.

40
CHAPTER SIX

RESULTS
RESULTS

The results presented here are as follows:

• Background characteristics o f child, parents caretakers and the environment

considered to be vital as determinants o f home accidents either directly or

indirectly as risk factors in the particular home set up;

• The extent o f Home Accidents is reflected in the analysis o f variables related

to accidents such as the prevalence, the different types o f accidents, their

severity, type of action taken following accidents and, outcome status o f the

victims;

• Analysis o f associations between variables and accident occurrence which

were subsequently tested by the Chi-square and t tests o f significance.

6.1. SOCIO-DEMOGRAPHIC CHARACTERISTICS OF THE STUDY

SAMPLE

Out o f 774 households that were visited, 1,465 children were found in the age-group

0-5 years. This therefore formed the study sample for Home Accidents in children 0-5

years. The socio-demographic characteristics o f the child, parents and caretakers are

presented in table 6.1.

6.1.1. CHILD’S DEMOGRAPHIC CHARACTERISTICS

The ages o f the children were recorded in months and then re-grouped into five

classes o f one year (12 months) intervals. The mean age was found to be 28.1 months,

(SD = 18.5). Table 6.1 shows the sex distribution in all the age groups.

41
TABLE 6.1 AGE/SEX DISTRIBUTION OF CHILDREN.

AGE IN SEX
MONTHS

MALE FEMALE TOTAL

n % n % n %

0-11 186 12.7 180 12.3 366 25.0

12-23 131 8.9 168 11.5 299 20.4

24-35 141 9.6 121 8.3 262 17.9

36-47 120 8.2 121 8.3 241 16.5

48-60 139 9.5 158 10.8 297 20.3

TOTAL 717 48.9 748 51.1 1465 100

(X2 = 6.8 DF = 4 p = 0.15 )

The highest p ro p o rtio n o f children was found in 0-11 months age group comprising

25% of the 0-5 y ea-r olds, followed by the 12-23 months (20.4%), the 24-37 months

(17.9%) and the 36 -47 m onths (16.5%). There was a gradual decrease in proportion

of the children re p re s e n te d in the study sample as the age increased upto 16.5% in the

4th year (35-47montrhs) then the 48-60 months were 20% showing a spike.

SEX d i s t r i b u t i o n

The sex distributioi n o f the children was as follows: 748 (51.1%) females and 717

(48.9%) males w itl h a m ale:fem ale ratio o f 1:1.04 (see table 6.1). There was no

significant differencxe observed in the distribution o f males and females within all the

a8e-groups (p = 0.1 “ 5).

42
6.1.2. PARENTS CHARACTERISTICS

The variables analysed below were: Age, educational, occupational, residential and

marital status o f parents.

Age

The fathers' as well as mothers' ages were recorded and summarised in age groups

with 10 year intervals as in table 6.2 below.

TABLE 6.2 DISTRIBUTION OF PARENTS BY AGE.

AGE IN COMPLETE MOTHER FATHER

YEARS

n % n %

15-24 381 27.2 32 2.7

25-34 816 58.2 589 50.0

35-44 172 12.3 424 35.9

45-54 26 1.9 100 8.5

>55 6 0.4 34 2.9

TOTAL 1401 100 1180 100

TABLE 6.3 SUMMARY STATISTICS ON AGE FOR THE PARENTS

M EAN n* SD* RANGE

Mothers' age (years) 28.5 1401 6.4 15 - 59

Fathers' age (years) 35.2 1180 7.9 18-74

n * = Number; SD * = Standard deviation

^°st of the parents were between 25 and 34 years old. The mean age for the mothers

was 28.5 years, while for the fathers was 35.2 years (see table 6.2 & 6.3).

43
PARENTS EDUCATION LEVEL AND OCCUPATION

This information for both parents was obtained from mothers who were the main

respondents. In some cases the mothers did not know the fathers occupation nor

education hence the lower response observed for the fathers as compared to the

mothers (see table 6.4 & 6.5).

Education of parents

The education o f the parents was classified in 3 groups as none, Primary , secondary

and higher education stati as in the table 6.4.

TABLE 6.4 EDUCATION OF PARENTS

LEVEL OF EDUCATION MOTHER FATHER

n % n %

None 105 7.2 36 2.6

Primary 878 60.3 548 40.1

472 32.4 783 57.3

Secondary & Higher

TOTAL 1455 100 1367 100

(X2 = 284.8 df= 4 p = 0.000)

Most o f the parents had some education. The fathers were generally more educated

with 57.3% o f them having secondary and higher education levels as compared to the

mothers (32.4%). The larger proportion (60.3%) o f the mothers had primary

education, and 7.12% had no education. O f the fathers, 40% had primary education

while 2.6% had no education. This difference between the education o f the father and

lhat of the mother was found to be statistically significant (See table 6.4).

44
Occupation

The parents was classified in three occupation groups o f Professional (e.g. Teacher,

Doctor, Nurse, Lawyer, Secretary, Clerks and so on); Business, Craftsman and artisan;

Housewife, Farmer or Unemployed as in table 6.5.

TABLE 6.5 OCCUPATION OF PARENTS

OCCUPATION MOTHER FATHER

n % n %

Professional 140 9.7 539 39.0

Business, Craftsman, 604 41.8 706 51.0

Artisan

Housewife, Farmer, 702 48.5 138 10.0

Unemployed

TOTAL 1446 100 1383 100

(X2 = 110.7, d f= 4, p = 0.0000)

Most o f the mothers were either housewives, farmers or unemployed (48.5%) while

41.8% were in the business group. The professional mothers were only 9.7%.

The fathers were employed mainly in business, crafts or as artisans (51%). The

professionals were 39.0% and the farmers or unemployed were 10%. The father's

occupation was higher in category than that o f the mother (See table 6.5) .

Marital Status of Child's Mother

Although the marital status was analysed for both parents, only the marital status of

lhe mother is used to represent that o f the parents since the marriage is a unit for both

m°ther and father and mothers have known to be closer with their children as will

also be seen below.

TABLE 6.6 MARITAL STATUS OF CHILD'S MOTHER

MARITAL STATUS MOTHER

n %

Single 43 2.9

Married 1378 94.4

Separated and/or Divorced 17 1.2

Widowed 22 1.5

TOTAL 1460 100

The majority o f parents were married with 94.4% o f mothers married and 99.1%

fathers married. Only 2.9% o f the mothers were single while single fathers were only

0.4%. Very few were either separated, widowed or divorced (See table 6.6).

Residential Status of Parents

The residential status o f both the parents was analysed and grouped into three as,

resided permanently (always lived together) with their children; Non permanent (lived

separated from their children); Partly permanent (lived together with their children

some o f the time) as in table 6.7.

TABLE 6.7 RESIDENTIAL STATUS OF PARENTS

RESIDENTIAL STATUS MOTHER FATHER

n % n %

.Permanent 1211 83.6 1082 78.5

.Non permanent 27 1.9 22 1.6

Partly permanent 211 14.6 274 19.9

JfOTAL 1449 100 1378 100

More o f the mothers (83.6%) resided permanently (always lived together) with their

children as compared to the fathers (78.5%). Few (14.6% for the fathers and 19.9% for

the mothers) were partly permanent (lived together some o f the time) and even fewer

(1.9% for the mothers and 1.6% for the fathers) were non permanent (residing

separately). The two parents residential stati are statistically related (See table 6. 7).

6.1.3 CARETAKER

The caretaker was defined as the person who routinely supervised the child as he or

she played. The characteristics o f caretaker analysed were: Category (or relationship to

the child ), the age and educational status.

Category of Caretaker

Mothers were the majority o f caretakers (79.1%) o f children, followed by the maid

(10.8%), grand parents (4.0%), older children (3.0%), other relatives (1.6%), and the

last in rank was the father (1.5%) (see table 6.8).

Age of Caretaker

For all the caretakers, the ages were recorded in three categories as, less than 13 years;

13-18 years; and above 18 years as in table 6.8.

47
TABLE 6.8. DISTRIBUTION OF CARETAKER BY AGE

CARETAKER OF CHILD AGE RANGE IN YEARS TOTAL

< 13 13 - 18 > 18

Father 0 (0%) 0 (0%) 21 (100%) 21 (1.5%)

Mother 0 (0%) 16(1.5%) 1081(98.5%) 1097(79.1%)

Maid 18(12.0%) 73 (48.7%) 59 (39.3%) 150(10.8%)

Grand parents 0 (0%) 0 (0%) 55 (100%) 55 (4.0%)

Older children 23 (54.8%) 4 (9.5%) 15(35.7%) 42 (3.0%)

Other relatives 5 (22.7%) 3(13.6%) 14 (63.6%) 22(1.6%)

All caretakers 46 (3.3%) 96 (6.9%) 1245(89.8%) 1387(100%)

The majority o f caretakers (89.8%) were over 18 years, while those in the 13 - 18 year

age group were 6.9%. Persons under 13 years taking care o f children formed only

3.3%.

The majority (98.5%) o f the mothers as caretakers were over 18 years old. The maids

were 13-18 years in most cases (48.7%), 39.3% were above 18 years while only 12%

were under 13 years. For the older children (siblings), the majority were under 13

years (54.8%), 35.7% were above 18 years while 9.5% were 13-18 years. Other

relatives were mainly (63.6%), followed by those under 13 years (22.7%) and 13-18

years were only 13.6%. The fathers and grandparents were all above 18 years as

expected especially for grand parents (see table 6.8).

Education Level of Caretakers

^formation on the educational status of the caretakers were categorised in four levels

summarised in three levels as in the table 6.9.

48
TABLE 6.9 DISTRIBUTION OF CARETAKERS BY EDUCATION LEVEL.

CARETAKER EDUCAT ION LEVEL TOTAL

NONE PRIMARY SECONDARY

OR HIGHER

Father 2 5 14 21
(9.5%) (23.8%) (66.7%)

Mother 68 742 285 1095

(6.2%) (67.8%) (26.0%)

Maid 14 112 18 144

(9.7%) (77.8%) (12.5%)

Grand parents 29 14 4 47
(61.7%) (29.8%) (8.5%)

Older children 8 21 0 29
(27.6%) (72.4%) (0%)
Other relatives 3 10 11 24
(12.5%) (41.7%) (45.8%)

All caretakers 124 904 332 1360

(9.1%) (66.5%) (24.4%) (100%)

(X2= 226, d f= 10, p = 0.0000)

The table shows that most o f the caretakers (66.5%) were o f primary education level.

This distribution pattern can be seen throughout all the caretaker categories except in

the one for the fathers where most (66.7%) were o f secondary or higher education

levels.

The mothers who were the main caretakers o f the children other than the maid or

house girl were also mainly educated upto primary level (67.8%).

The majority o f maids (77.8%) were educated upto primary level, 12.5% upto

secondary or higher levels and 9.7% had no education at all.

49
The grandparents had 61.7% with no education at all, 29.8% with primary education

and 8.5% with secondary or higher education.

The older children or siblings had 27.6% with no education, 72.4 % with primary

education while none with secondary or higher education.

Other relatives had 12.5% o f them not educated, 41.7% upto primary level and 45.8%

secondary or higher level. A small proportion (<10%) in all the groups had no

education at all {see table 6.9).

50
ENVIRONMENTAL CHARACTERISTICS

The characteristics considered were the main factors in the environment that play a

major role in childhood accidents. These were: area o f residence; type o f housing;

source o f artificial light during darkness; source o f cooking fuel; whether the fire place

is protected or not; safe keeping o f medicines and chemicals away from children;

presence or absence o f harmful objects (scattered litter) in the compound; whether

home is used as a working place or workshop as in carpentry among others.

Area of Residence

The areas o f residence for the children were categorized into four groups as shown in

the table 6.10 below.

TABLE 6.10 AREA OF RESIDENCE

AREA OF RESIDENCE n %

Rural 224 15.4

Urban 913 62.9

Periurban 298 20.5

Slums 17 1.2

TOTAL 1452 100

Most of the homes (62.9%) were in the urban region, 20.5% were in the peri-urban

and 15.4% in the rural. The least were the slum dwellers who formed 1.2%.
T y p e s o f H o u s in g

The houses were classified in descending order o f quality and durability o f building

material as permanent (stone or cemented walls and floor); semi-permanent (mud plus

cemented walls and/or floor) and traditional (grass-thatched mud hut). The

proportions represented by the houses are as seen in figure 1 below.

FIG: 1 TYPE OF HOUSES

TRADITIONAL
9%

64%

The semi-permanent houses were the majority (64%) followed by the permanent

(27%) and the traditional mud hut was last (9%).

Types of Artificial Lighting

The types o f artificial lighting routinely used by the respondents are as shown in figure

2.

52
FIG: 2 TYPE OF ARTIFICIAL LIGHTING

The hurricane lamp is the most commonly used source o f light (51%) followed by the

very dangerous traditional open flame lamp locally known as Nyangile (36%).

Electricity is used by a mere 13% o f the households.

TABLE 6.11 SOURCE OF COOKING FUEL

COOKING FUEL n %

Electric cooker 43 2.9

Gas cooker 62 4.2

Charcoal stove 819 56.0

.Kerosene stove 258 17.6

Firewood 280 19.2

total 1462 100

r — ——
most frequently used source o f fuel was found to be the jiko or charcoal stove

53
cooker together formed 7%.

RISK FACTORS IN CHILD'S HOME

The enumerators assessed the environment for presence or absence o f risk factors by

observation and the findings are as indicated in table 6.12

TABLE 6.12 PRESENCE OR ABSENCE OF RISK FACTORS IN THE HOME.

ENVIRONMENTAL CHARACTERISTICS (Risks) Present Absent

n % n %

Open fire place (n = 1461) 865 59.2 596 40.8

Unsupervised playing (n = 1457) 847 58.1 610 41.9

Custody o f medicines and chemicals within easy 430 29.4 1031 70.6
reach for the children
(n = 1461)

Sharp instruments (e.g knives) custody within easy 528 36.1 934 63.9
reach for the children
(n = 1462)

Home as working place or work shop 222 15.3 1231 84.7

(n = 1453)

Litter scattered around compound 460 31.7 992 68.3

(n = 1452)

Crowding status (n = 1459) 534 36.6 925 63.4

Each type o f risk was considered independently and most o f the risk factors analysed

were found to be absent in the majority of the homes except the ‘Open fire place’

found in 59.2% o f homes and ‘unsupervised playing’ found in 58.1% o f them. These

two risk factors were found to be present in more than half o f the homes.
6.3. HOME ACCIDENTS

6.3.1. Prevalence of injuries due to home accidents

Out o f the 1465 children studied, the number that was reported to have suffered

injuries due to accidents in the year prior to the study period was 456. The prevalence

was therefore 32.1 %.

TABLE 6.13 INJURIES DUE TO HOME ACCIDENTS

TYPE OF INJURY. n %

Bum /Scald 136 30.3

Poisoning 44 9.8

Falls 189 42.1

Others 80 17.8

All accidents 456 32.1

Out of the injuries, 42% occurred due to falls. The next commonly occurring injury

was bums and scalds (30%), followed by poisoning (10%). Other types o f injuries i.e.

foreign bodies in the airways, drowning, suffocation and others formed 18% (see

Table 6.13).

Severity of Home Accidents

^formation on the type of accident was obtained from the mother and her opinion

rec°rded as to whether the accidents was mild, moderate or severe in nature. The

S itin g information has been shown in figure 3.

 FIG: 3 SEVERITY OF ACCIDENTS

Out of all the accidents, 219 (48.9%) o f the injuries that occurred, were o f moderate

nature; the mild ones were 141 (31.5%); the severe ones were 80 (17.9%) and 8

(1.8%) could not be graded by the mothers.

Out of the 80 severe injuries that occurred, 31 (38.8%) o f them were due to bums; 27

(33.8%) due to falls ; 7 (8.8%) due to poisoning and 15 (18.8%) due to others such as

foreign bodies in openings. The moderate injuries were predominated by falls, then

Urns>other accidents grouped together and lastly poisoning. The mild injuries were

Predominated by falls then bums, others and lastly poisoning in that order. For the
injuries that could not be graded on severity, poisoning was leading, then bums

followed, then others. No falls were in this group. The trend for each type o f injury

was as follows:

On enquiry from the respondents, 68 (50%) o f the bums and scalds that occurred were

rated as moderate, 34 (25%) were mild in nature while 31 (23%) were severe and 2%

were not known.

For the poisonings, 26 (59%) were moderate, those classified as mild or severe were

each 7 (16%) and, 4 (9%) o f them had severity not known.

With falls, 85 (45%) were mild, and those moderate were 76 (40%), leaving 27 (15%)

as severe.

Out o f the other types o f accidents, 49 (16%) were moderate, the mild and the severe

ones were each 15 (19%) and only 1 (1%) had severity not known by the respondents.

MOTHER’S OPINION ON PREVENTION OF THE ACCIDENTS.

Mothers were asked whether they thought the accidents could have been prevented or

not. Out o f all accidents, 246 (55.4%) were said to be preventable while 198 (44.6%)

not preventable according to the mothers.

each accident thought preventable, the corresponding rates were obtained by cross

filiatio n and compared as follows: Bums and scalds (68%) were leading, followed

57
by poisoning (58%), others (52%), and lastly falls (47%). For the other types of

accidents, 52% were thought not preventable (p=0.007).

6.3.3 HOME ACCIDENTS AND CHILDS SOCIO-DEMOGRAPHIC

CHARACTERISTICS

TABLE 6.14 ACCIDENT RATES IN THE 0-5 YEARS

AGE GROUP n ACCIDENTS PER ACCIDENT RATES

(MONTHS) AGE GROUP (%)

0-11 366 56 15.9

12-23 299 98 33.7

24-35 262 95 37.5

36-47 241 101 43.7

48-60 297 106 36.4

TOTAL 1465 456 31.1

(X2 = 63.25, d f = 4, p=0.000)

Age and Accidents

For the whole sample, the number o f accidents increased with age and most accidents

(106) were found to have occurred in the 48 - 60 months age-group while the least

(56) in the 0 -11 months. Further analysis was done and the rates o f accidents in each

Particular group obtained. The findings are shown in table 6.14 above.

Of children under one year, 15.9% are likely to suffer an accident in a period o f one

year. So are 33.7% o f children 1-2 years, 37.5% children 2-3 years, 43.7% children 3-

^ years, and 36.4% o f children 4-5 years. As age increases, children are more prone to

58
accidents. This relationship is statistically significant ( p - 0.000), therefore age is an

important determinant o f whether a child gets an accident or not (see also figure 4).

FIG. 4 ACCIDENTS BY AGE OF CHILDREN

120 45

40
100
M 35

W
9 80 30
o ui «
C5 £
< 25 £ 2
60 S |
o
20 2
ui O
|
a. ra
z
40 15 ~

oI- 10
20
5

0 0
0-11 12-23 24-35 36-47 48-60
AGE GROUPS (MONTHS)

ITOTAL NO. ACCIDENTS ■PERCENTAGE

The specific accidents that occurred were distributed across all the age groups as

shown in table 6.15 below.

TABLE 6.15 TYPES OF ACCIDENTS IN THE VARIOUS AGE GROUPS

AGE BURNS POISONING FALLS OTHERS ALL

GROUP IN ACCIDENTS
1 months
Hm ] 19 (14%) 6(13.6% ) 15(7.9%) 14 (17.5%) 54 (12%)
te rT 27(19.9% ) 11 (25%) 47 (27.9%) 12(15% ) 97 (21.6%)
24-35 36 (26.5%) 5(11.4% ) 40 (21.2%) 14 (17.5%) 95 (21.2%)
36-47 26(19.1% ) 10(22.7% ) 43 (22.8%) 21 (26.3%) 100 (22.3%)
48-60 28 (20.6%) 12 (27.3%) 44 (23.3%) 19(23.8%) 103 (22.9%)
1 £ 2 a i~ 136(30.3% ) 44 (9.8%) 189 (42.1%) 80(17.8% ) 449(100% )
(>-2 = 14, d f= 12, p = 0.000)

59
The children under one year were more likely to suffer from bums and scalds,

followed by falls, followed by other accidents combined then poisoning.

The subsequent age groups (2-3yrs, 3-4yrs, 4-5yrs) had identical trends where they

were more likely to suffer from falls, followed by bums and scalds, followed by other

accidents and lastly poisoning.

Analysis o f the specific accident types revealed that Bums and scalds were

predominated by those 2-3 years (26.5%); Poisoning by those 4-5 years (27.3%); Falls

by those 1-2 years; ‘Others’ by those 3-4 years (26.3 %)(see table 6.15).

TABLE 6.16 TYPE OF ACCIDENT BY SEX

1 SEX BURNS & POISONING FALL OTHERS ALL

SCALDS ACCIDENTS

MALE 74 21 95 40 230
(54.4%) (47.7%) (50.3%) (50%) (33.6%)

FEMALE 62 23 94 40 219
(45.6%) (52.3%) (49.7%) (50%) (30.8 %)

TOTAL 136 44 189 80 449

(30.3%) (9.8%) (42.1%) (17.7%) (32%)

(X2 = 0.9, df=3, p = 0.2)

Of the girls, 30.8% had accidents while 33.6% o f the boys had accidents. This finding

was not statistically significant (p= 0.2 ). Further analysis was done for each type of

accident and sex o f child was controlled for, and it was found that for all the accidents

except poisoning, the accident rate for males was always a fraction higher, though the

observed small difference was not statistically significant. This means that a child’s

Sex is not a significant determinant o f the type o f accident that occurred. In both sexes,

60
the falls were the leading cause of accidents followed by bums, poisoning and then

other accidents (see table 6.16).

PARENTS CHARACTERISTICS AND ACCIDENT OCCURRENCE

Cross tabulations were done on the accident occurrence and mothers education level

and the following was the result:

FIG. 5: ACCIDENT OCCURRENCE BY EDUCATION OF MOTHER

NO PRIMARY SECONDARY
EDUCATION EDUCATION & HIGHER
EDUCATION LEVEL

The relative prevalence of accidents was found to increase as the level o f education

increased. Thus, those not educated had 29% o f their children with accidents; 32%

children from the primary and 32.6% in those whose mothers had secondary and

higher school education (%2= 0.5, d f = 2, p = 0.8) (see fig.5).

 FIG: 6 ACCIDENT OCCURRENCE AND OCCUPATION OF
MOTHER

PROFESSIONAL
OCCUPATION

On looking at occupation, it was worth noting that the professional group had the

highest (34.8%) prevalence o f accidents in their children, while the farmers and

unemployed mothers had the lowest prevalence (29.3%) and the rate in the business

group was 34.4% (x2 = 4.4, df = 2, p = 0.11) (see fig. 6).

CHARACTERISTICS OF CARETAKER AND ACCIDENT OCCURRENCE

The table below shows the different rates o f accident occurrence in the different age

groups o f caretakers.

TABLE 6.17 ACCIDENT BY AGE OF CARETAKER.

ACCIDENT < 13 YRS 13-18 YRS > 18 YRS TOTA

.OCCURRENCE n % n % n % n %
Yes 15 33.3 38 39.6 373 27.8 426 31.7
No 30 66.7 58 60.4 828 68.9 916 68.3

total 45 100 96 100 1201 100 1342 100

(X - 3.03661, d f= 2, p value = 0.2).

62
The 13- 18 years age group having the highest rate (39.6%) followed by those under

13 years (33.3%) and lastly those above 18 years (27.(%). This was not a statistically

significant finding (p = 0.2). Further analysis was done to compare the pattern of

accidents rates in the different caretaker categories before and after controlling for

their ages and the rates obtained were as follows:

TABLE 6.18 ACCIDENT OCCURRENCE RATES IN THE DIFFERENT

CARETAKER CATEGORIES CONTROLLING FOR AGE

AGE CARETAKER
GROUP
(YEARS) ALL
CARE­
FATHER MOTHER MAID GRAND OLDER OTHERS TAKERS
PARENT CHILDREN

< 13 0 0 5/18 9/22 1/5 15/45

p = 0.5 (27.8%) 0 (40.9%) (20%) (33.3%)

13-18 0 5/18 31/73 2/4 0/2 38/96

p = 0.4 (31.3%) (42.5%) 0 (50%) (39.6%)

> 18 9/18 304/1043 23/56 28/55 1/15 8/14 373/1201

p = 0.000 (50%) (29.1%) (41.1%) (50.9%) (6.7%) (57.1%) (27.8%)

TOTAL 9/18 309/1059 59/147 28/55 12/41 9/22 426/1342

p = 0.0023 (50%) (29.2%) (40.1%) (50.9%) (29.3%) (40.9%) (31.7%)

In the first two age groups, there was no significant finding except in the above 18 years

age group where the highest rate of accidents occurred in children being taken care o f by

other relatives (57.1%) followed by the grandparents (50.9%), father (50%), maid

(41.1%), mother (29.1%) and lastly, those being cared for by older siblings (6.7%) (p =

0. 000 ) .

63
6.3.4 ENVIRONMENTAL CIRCUMSTANCES AND ACCIDENT

OCCURRENCE

Area of residence and accident occurrence

Accidents were analysed in relation to the area o f residence and the corresponding

rates shown in fig. 7.

FIG: 7 ACCIDENT BY AREA OF RESIDENCE

AREA OF RESIDENCE

Out of all the accidents, 64% occurred in the urban area, 25% in the peri-urban and

slums, and 11% in the rural areas (p = 0.000).

When the area specific prevalence rates were calculated, the peri-urban and slums

were leading with 42.1%, followed by the urban with 31.6%, and the rural areas with

22% [X2 = 22.67, d f = 3, p - 0.0000 (see f i g 7)].

64
H o u s in g a n d a c c id e n ts

The rates o f accidents in the different types o f housing were calculated and compared

as in fig. 8.

FIG. 8: ACCIDENTS IN RELATION TO TYPE OF HOUSE

60

PERMANENT SEMI- TRADITIONAL

PERMANENT
TYPE OF HOUSE

Of all the accidents that occurred, 67.6% were from those living in semi-permanent

houses, 20.6 % from those in permanent houses and 11.8% from those in the

traditional mud huts.

Figure 8 shows the actual prevalence o f accidents in these house types rather than the

distribution above. The traditional grass thatched mud huts were leading with 34.9%,

Allowed by the permanent houses (33.2%), and the semi-permanent houses (31.3%)

^ ese observations were found not to be statistically significant ( x 2 = 0.95, df=2,

r
P=0.6).

65
ENVIRONMENTAL RISKS AND ACCIDENT OCCURRENCE

Enumerators observed whether risk factors were present or absent in the homes.

Accident rates were subsequently analysed by doing individual cross-tabulations for

each risk factor found. Table 6.19 shows a combination o f the resultant 2x2 tables,

and shows the numbers o f accidents observed, and the accident rates for these risks.

The levels of significance are as indicated for each risk.

TABLE 6.19 ENVIRONMENTAL SAFETY AND ACCIDENT OCCURRENCE

RATES

ENVIRONMENTAL RISKS
CHARACTERISTICS

ABSENT: n (%) PRESENT: n (%) Significance

Open fire place 179 (31.1) 276 (32.9) p = 0.61

n = 1416

Unsupervised playing 183 (30.7) 271 (33.2) p = 0.32

n = 1412

Custody of medicines and 319(32.1) 135(31.9) p = 0.94

chemicals n = 1416

Custody of sharp 293 (32.6) 162 (31.3) p = 0.61

instruments (e.g knives) n = 1417

Home as working place or 392 (33.0) 62 (28.2) p = 0.16

work shop n= 1408

Litter scattered around 319(33.4) 134 (29.6) p = 0.16

compound n= 1407

Crowding status 305 (34.0) 147 (28.5) p = 0.03

n = 1414

The accident rates (ranging from 28.2% to 34%) were not markedly different in the

different situations and there was no statistical significance in most o f the findings

except for the crowding status where the rate of accidents was higher (34% vs. 28%)

in the houses classified as not crowded (%2 = 4.5, p = 0. 03, d f = 1). The crowding

status was assessed by enumerators as more than four people sleeping in a room, and

66
 about two thirds of all the sampled houses, in the study, were assessed as crowded.

Place of accidents

Analysis was done to determine the place with the highest risk for childhood accidents

and the results obtained are as shown in table 6.20.

TABLE 6.20 ACCIDENT TYPE BY PLACE OF ACCIDENT

TYPE OF KITCHEN LIVING AND COMPOUND TOTAL

ACCIDENT BED ROOM

n % n % n % n %

Bum/Scald 67 50.4 49 36.8 17 12.8 133 30.8

Poisoning 17 38.6 26 59.1 1 2.3 44 10.2

Falls 8 4.4 30 16.4 145 79.2 183 42.4

Other 13 18.1 11 15.3 48 66.7 72 16.7

All accidents
105 24.3 116 26.9 211 48.8 432 100
(X2 = 199, d f = 4, p = 0.0000).

Out of all the injuries from all the accidents combined, 51% occurred within the house

(26.9% in the living and bed rooms and 24.3% in the kitchen). Those occurring out o f the

house structure (outdoors) were 49%.

Most bums occurred in the kitchen (50.4%) followed by the living(sitting) and

bedroom (36.8%), and least in the compound (12.8%).

Most poisonings occurred in the living and bedroom (59.1%), followed by the

kitchen (38.6%), and lastly out in the compound(2.3%).

Falls were most likely to occur in the com pound (79.2%), followed by the living and

bedroom (16.4%), then lastly the kitchen(4.4%).

Other accidents were most likely to occur in the com pound (66.7%), followed by the

kitchen (18.1%), and lastly the living and bedroom.

Compound

In the compound, the most likely accident was falls (145), followed by other accidents

(48), then bums and scalds (17), then poisoning (1).

Living and Bedrooms

In the living room and bedroom, the majority ( 49) o f accidents were due to burns

and scalds, 30 due to falls, 26 due to poisoning and 11 due to others.

Kitchen

In the kitchen, 67 o f the injuries that occurred were due to bums or scalds, 17 were

due to poisonings, 8 due to falls, and 13 due to other accidents.

There is a statistically significant relationship between the type o f accident and the

place where the accident occurred (p<0.05) (see table 6.20).

68
 a c c id e n t o c c u r r e n c e a n d l i g h t i n g s y s t e m s

TABLE 6.21 ACCIDENT OCCURRENCE BY THE LIGHTING SYSTEM

a c c id e n t ELECTRICITY HURRICANE OPEhI FLAME OTHER TOTAL

LAMP PAREAFIN
LAM1 >

N % N % N % N % N %

YES 67 14.8 250 55.2 134 29.6 2 0.4 453 32.2

NO 113 11.8 474 49.7 364 38.2 3 0.3 964 67.8

TOTAL 180 12.8 724 51.5 498 35.4 5 0.4 1407 100

(:12= 10.4, df= 3, p= 0.01539).

Table 6.21 shows that o f the 453 accidents, 250 (55.2%) were in those using hurricane

lamps, 134 (29.6%) in those using the open flame paraffin lamp, 67 (14.8%) in those

using electricity, and other types were only 2 (0.4%).

ELECTRICITY HURRICANE LAMP OPEN FLAME LAMP &

OTHERS
DOMESTIC LIGHTING

□ burns ■ p o is o n in g dfall bothers

On the other hand, the accident rates shown in table 6.21 can be summarised for all

the accidents combined as follows; 67/180 (37.2%) o f those using electricity are likely

k 69
to have an accident as compared to 250/724 (34.5%) o f those using hurricane lamp,

135/498 (27.0%) o f those using the open flame paraffin lamp and other methods o f

lighting. This relationship is statistically significant.

The specific accident rates shown in figure 9 are follows:

It was found that 52.2% o f bums occurred in those using hurricane lamps compared to

35.3% and 12.5% o f those using the open flame lamp and electricity respectively.

It was found that 72.8% o f poisoning occurred in those using hurricane lamps

compared to 13.6% and 13.6% o f those using the open flame lamp and electricity

respectively.

With falls, it was also found that 52.4% occurred in those using hurricane lamps

compared to 33.2% and 14.4% o f those using the open flame lamp and electricity

respectively.

With other accidents, it was also found that 60% o f falls occurred in those using

hurricane lamps compared to 21.2% and 18.8% o f those using the open flame lamp

and electricity respectively.

Further analysis in an attempt to find out the type o f injury mechanism in these

different situations, time o f accident and method of injuries were controlled for. In the

day time it was found that 50.9% o f the injuries in those with hurricane lamps, were

due to striking by person, object or animal; 23.9% due to bums or scalds whereas

70
7.5% and 4.9% were due to paraffin and foreign body ingestion respectively. The rest

were other accidents (p< 0.05).

In the night time when the lighting is dependent on these lamps, it was found that 15%

o f the injuries in those with hurricane lamps, were due to striking by person, object or

animal; 70% due to bums or scalds whereas 5% and 5% were due to paraffin and

foreign body ingestion respectively. The rest were other accidents (p >0.05). The same

pattern was mirrored in those using the open flame lamp. This however, was found to

be not statistically significant.

SOURCE OF COOKING FUEL AND ACCIDENTS

TABLE 6.22 ACCIDENT BY SOURCE OF COOKING FUEL

A C C I­ E L E C T R I­ GAS CHARCOAL K ER O SENE F IR E W O O D TOTAL

C IT Y
DENT

n % n % n % n % n % n %

YES 12 2 .6 20 4 .4 258 5 6 .7 73 4 .0 92 2 0 .2 455 3 2 .1

NO 31 3 .2 40 4 .2 537 5 5 .8 184 1 9 .1 170 1 7 .7 962 6 7 .9

TOTAL 43 3 60 4 .2 795 5 6 .1 257 1 8 .1 262 1 8 .5 1417 100

(X2 = 3.14, df= 4, p = 0.54).

The distribution o f accidents were as follows:

O f the accidents that occurred, 56.7% used charcoal for cooking, 20.2% use firewood,

4.4% use gas, 4% use kerosene and 2.6% use electricity. The prevalence o f accidents

in those who use firewood burners is 35.1%, those using gas is 33.3% those using

charcoal burners is 32.5%, those using kerosene 28.4%, and those using electricity

27.9%.

71
The different types of accidents were controlled for and it was found that:

With charcoal stove users, the most likely accident was falls, followed by bums,

poisoning, then others;

With firewood stove users, the most likely accident was bums and scalds, then falls, then

others, then poisoning;

With kerosene stove users, the most likely accident was falls, then bums, then poisoning

and others;

With electricity and gas combined, the most likely accident was falls, then bums and

scalds and others and lastly poisoning. All the above, however were not found to be

statistically significant (x2 = 15, d f = 9, p = 0.09).

6.35 THE COURSE OF ACTION TAKEN FOLLOWING ACCIDENTS

TABLE 6.23. DISTRIBUTION OF ACCIDENTS BY TYPE OF ACTION TAKEN.

I ACCIDENT NONE OUTPATIENT INPATIENT HOME TRADITIONAL ROW

REMEDY HEALER TOTAL

n % n % n % n % n % n %

Bums/ i 0.7 50 36.8 51 37.5 34 25.0 0 0 136 30.4

Scald

Poisoning 0 0.0 20 45.5 18 40.9 4 9.1 2 4.5 44 9.8

Fall 8 4.3 63 33.9 24 12.8 82 43.9 10 5.3 187 41.8

Other 0 0 27 33.8 21 26.3 32 40.0 0 0.0 80 17.9

All Accidents 9 2.0 160 35.8 114 25.5 152 34.0 12 2.7 447 100

On considering all the accidents together, most o f them (35.6%) were managed in an

72
outpatient facility. The second course o f action was home remedy or self treatment of

34.2% o f the accidents, then in-patient management o f 25.6% and lastly 2.7% o f the

accidents by traditional healer, 2.5% were given other treatments, 9 (2%) were not

acted upon, while none (0%) were managed by prayers (see table 6.23).

The different types o f accidents were considered individually for course o f action

following injuries and the results were as follows (see table 6.23 & 6.24)\

TABLE 6.24. SUMMARY OF COURSE OF ACTION TAKEN ON THE SPECIFIC

TYPES OF ACCIDENTS.

ACCIDENT TREAMENT HOME NO ACTION

TYPE FROM HEALTH TREATMENT
INSTITUTION
Bum/Scald 101 (74.3%) 34 (25.0%) 1 (0.7%)

Poisoning 40 (90.0%) 4(9.1% ) 0 (0.0%)

Fall 97 (51.9%) 82 (43.%) 8 (4.3%)

Other 48 (60.0%) 8 (4.3%) 0 (0.0%)

All Accidents 286 (64.0%) 152 (34.0%) 9 (2.0%)

From the above finding, it is evident that the type o f accident determines the action

taken following injury.

Of the 136 bums and scalds, 37.5% were treated as in patients, 36.8% were treated as

out patients in an institution, 25% given home remedy, and 0.7% were not treated.

At least something was done about all the 44 poisonings where 20 (45.5%) were given

73
out patient treatment, 18 (40.9%) were treated as in patients, 4 (9.1%) were given

home remedy and 2 (4.5%) treated by a traditional healer.

O f the falls, 43.9% were given home remedy, 33.9% were treated as out patients, 12%

as In - patients in institutions, 10 (5.3%) were treated by traditional healers, and 8

(4.3%) had no treatment.

Other accidents such as drowning, suffocation and insect or animal bites were

considered collectively. It was found that 33.8% were outpatients, 26.3% hospitalised

as inpatients and 40% given home treatments.

TREATMENT GIVEN

The type o f treatment given can be seen to be interrelated with the severity of

accidents and this is considered separately as follows:

Home Treatment

Most o f the home treatments were given to falls (43%), followed by bums (25%). Few

home treatments were given to poisonings (9.1%) and other accidents (see table 6.23

& 6.24).

Admissions (In-Patients!

The majority o f admissions (in patients) were due to bums and scalds, followed by

falls, poisonings, and others.

74
 O u tp a tie n ts

The majority o f out patients were falls, followed by bums, poisonings, and others.

No T reatm ent

Out o f the 9 that were not treated, 8 (89%) were falls and 1 (11%) bum or scald.

These differences are statistically significant.

TABLE 6.25 TIME TAKEN TO ACT FOLLOWING SPECIFIC ACCIDENT

OCCURRENCE

TYPE OF TIME TAKEN TO ACT FOLLOWING ACCIDENTS

ACCIDENT
IMMEDIATI DELAYED ROW TOTAL
(WITHIN 2HRS.) (AFTER 2HRS.)

n % n % n %
Bum/Scald 125 91.9 11 8.1 136 30.6
Poisoning 42 95.5 2 4.5 44 9.9
IT )
OO

Fall 85.4 27 14.6 185 41.6

Other 68 85 12 15 80 18.0
All Accidents 393 88.3 52 11.7 445 100

There was good response to all the accidents because 88% were acted upon

immediately, i.e. within two hours. For poisoning, two hours was considered too

prolonged for action so it was separately analysed as follows:

TABLE 6.26 POISONING : TIME OF ACTION

0-30 min 30 min - 2hr. MORE THAN TOTAL

TWO HOURS.

n % n % n % n %

34 77.3 8 18.2 2 4.5 44 100

O f the poisonings, 77% were acted on in 30 minutes, 18% between 30 minutes and two

75
 hours, and only 2 (4.5%) more than 2 hours.

O f the accidents leading to immediate actions, falls were the majority (40%) followed

by bums (32%), poisonings (11%) and others (17%).

O f those with delayed actions, again falls predominated (52%), followed by bums

(21%), poisoning (5%) and others (23%). This relationship was not statistically

significant.

6.3.6 OUTCOME OF ACCIDENTS

TABLE 6.27 OUTCOME OF ACCIDENTS

TYPE OF FULL RESIDUAL DEATH TOTAL

ACCIDENT RECOVERY SCARRING
AND
CHRONICITY

n % n % n % n %

BURN/SCALD 113 29.4 23 41.8 0 0 136 30.3

POISONING 33 8.6 5 9.1 6 66.7 44 9.8

FALLS 165 42.9 23 41.8 1 11.1 189 42.1

OTHER 74 19.2 4 7.3 2 22.2 80 17.8

ALL 385 85.7 55 12.2 9 2.0 449 100

ACCIDENTS

(X2 = 41.63496, d f= 6, p = 0.0000)

O f all the 449 accidental injuries, most resulted in full recovery (85.7%). However

12.2% had residual damages and chronic problems, while 9 (2%) resulted in death1.

Out o f the burn or scald injuries, 83.1% recovered fully while 16.9% had residual

The information on the dead children was obtained only from the house holds that were visited (those with children under five years).

76
damages or scarring and no deaths were reported.

In the poisonings group 75% recovered fully, 11% had residual damage or chronicity,

and 6 (13%) died.

The majority of those with falls recovered fully (87.3%), 12.2% had residual damage

or chronicity, and 1 (0.5%) resulted in death.

There is a statistically significant relationship between type o f accident and outcome .

Most o f the children (41.8%) with residual scarring and chronic problems were

victims o f bums and poisoning (41.8%). O f those who recovered fully, the majority

had falls (42.9%). O f the deaths, 66% were due to poisoning, 11% due to falls and the

other 2 deaths (22%) were observed in the group o f other accidents (see table 6.27).

The deaths were analysed and two thirds found to have resulted mainly from

poisoning. Out o f the 6 deaths that occurred due to poisoning, 5 (83.3%) were due to

ingestion o f medicines and 1 (16.7%) due to poisoning from a sting or snake bite.

There was one death resulting from a fall and the method o f injury was the impact o f

being thrown and getting struck by an object when a couple (the parents) had a fight.

The method o f injury for the two other accidental deaths were from drowning (1) and

one from foreign body ingestion (x2 = 12.6, df=6, p=0.005).

77
 CHAPTER SEVEN

DISCUSSION, CONCLUSION AND

RECOMMENDATION

/
DISCUSSION. CONCLUSION AND RECOMMENDATIONS

7.1. DISCUSSION

This descriptive study on the extent and determinants o f home accidents in children

under five years was carried out with the intention o f getting some vital information. The

information sought was mainly related to socio-demographic and environmental health

circumstances o f home accident in children under five. Parental characteristics, child's

characteristics as well as environmental factors were considered as vital in the search for

determinants o f accidents in the particular home set up.

7.1.1. BACKGROUND CHARACTERISTICS OF THE STUDY SAMPLE

A. CHILDREN S DEMOGRAPHIC CHARACTERISTICS

The demographic characteristics o f the 1465 children studied were found to conform

with the general trend o f Kenyan children as seen in the population pyramid o f the same

age-group. This was seen when the children were regrouped into five groups o f one year

intervals and the frequency was found to decrease with age. The most were the infants (0-

1 years) who formed 25% o f the 0-5 year olds. However the last group containing the 4-

5 year olds showed an exceptional spike from the trend possibly due to an artefact or

pooling o f the children whose ages were unknown by the respondents (p = 0.15). This

was not a statistically significant finding (see table 6.1).

78
The children’s sex distribution in all the five age- groups, was found to have slightly

more girls (51%) than the boys (49%) giving a male: female ratio o f 1:1.04. This was

however not a statistically significant finding (p=0.15).

B. PARENTS DEMOGRAPHIC CHARACTERISTICS

The parents were mostly between 25 and 34 years with the mothers, in most cases,

proving younger than the fathers. The mothers had a mean age o f 28.5 years and the

fathers 35.2 years. This is the most economically active period in both parents lives. Both

parents had some education and very few had no education at all. For the mothers, two-

thirds had primary school education and one-third had higher education. The fathers had

significantly better education with one third having primary and two-thirds having higher

than primary education (see table 6.4) (p = 0.000).

Analysis o f the mothers occupation revealed that about half (48.5%) were either

housewives, farmers or unemployed; while 41.8% engaged in business, crafts or

artisanship. Only 9.7% had professional jobs such as teaching, medicine, law and so

forth. For the fathers, the picture was different and 39% o f them had professional jobs

and 51% were businessmen, craftsmen or artisans. The unemployed fathers were only

10%. The proportion of the fathers reduced as one descended down the list o f professions

while that o f the mothers increased (see table 6.5). This was found to be a statistically

significant relationship.

79
Almost all the parents (> 90%) were found to be married and, 83.6% o f the mothers as

compared to 78.5% o f the fathers reside together with their children showing a complete

family unit for most o f these children (see table 6.6 and 6.7).

The socio-economic profile o f the study population as described above by the education

and occupation o f parents is o f low to middle class. This low socio-economic status

(though beyond the scope o f this study) has been shown by researchers such as Gupta M.,

to contribute to increased incidence o f bum accidents in children14. On comparison o f

characteristics between the two parents, it was found that they are statistically related.

Subsequently, the mothers’ characteristics have been used to represent that o f the parents.

C. CARETAKERS OF CHILDREN

The type o f caretakers were analysed for their characteristics such as age and educational

status each o f which were categorised in three groups as in table 6.8 & 6.9. In the

majority o f homes, the leading caretaker was the mother (79.1%), followed by the maid

(10.8%), the grand parents (4.0%), older children or siblings (3.0%), other relatives

(1.6%) and lastly the father (1.5%).

Since the mothers have to do household chores, feed the family and do farm work,

children may not get the maximum supervision possible from their mothers. This is a

major risk to children because the mothers may not concentrate on guiding their children

as they play (see table 6.8).

80
For all the caretakers together, about 90% were over 18 years (the mothers, whose ages

had already been discussed, are also included).

The majority (49%) o f maids were found to be adolescents (i.e. aged mainly between 13

and 18 years), 39% were above 18 years, while 12% were under 13 years. Because of

their age and hormonal changes, these adolescent maids are apt to labile moods,

irrational behaviour and may easily get distracted and therefore not perform the duty of

child care as necessary.

More than half (54.8%) o f older children taking care o f their siblings were under 13

years meaning they were children themselves taking care o f children. Other relatives

were mainly (63.6%) over 18 years old and the rest were adolescents (13.6%) or children

(22.7%).

For the fathers and grandparents, they were all over 18 years as would be expected.

The grandparents are a special risk as caretakers because o f the advanced ageing process

going on, leading to deterioration of their senses and reflexes (see table 6.8).

The education level for most of the caretakers combined was primary education (66.5%).

The mothers education status was found to be lower than the fathers.

The maids were mainly educated upto primary school level (78%), only few (12%) upto

secondary or higher and about 10% with no education at all. This low education status

may be an important determinant of their effectiveness in child-care.

81
The grandparents had the lowest level of education with 61.7% having no education at

all. Only about 29.8% had primary education and 8.5% had secondary education. The

other relatives and older siblings were mainly of primary education status (see table 6.9).

Contrary to this study, the low education status o f these caretakers would be expected to

contribute an unknown amount o f danger to the children as far as risks to accidents are

concerned. This was attributable to an interplay o f other factors beyond this study.

D. ENVIRONMENTAL CHARACTERISTICS

Out o f the environmental characteristics studied it was found that on the geographical

area o f residence, the majority (62.3%) in the sample were in the urban region; 20% in

the peri-urban; 15.4% in the rural and 1.2% from the slums (see table 6.10). This

portrays a predominantly urban community which is according to the sampling.

The type o f housing as priorly defined, was found to be mainly the semi-permanent

(64%) one followed by the permanent (27%) and lastly the traditional grass thatched one

(9%) (see fig. 1). These semi-permanent houses are mostly found in the peri-urban, slums

and rural areas o f Kisumu and few in the urban area. The traditional house being simpler

and cheaper to construct is predominant in the rural areas while the permanent house, in

the urban and to a lesser extent in the peri-urban and rural areas. The type o f housing

generally depends on the economic capability of the individuals.

82
It was found that the most commonly used source o f cooking fuel was the charcoal stove

(56%) followed by firewood (19%), kerosene stove (18%) then gas and electric cookers

combined formed 7% (see table 6.11).

The most commonly used source of light was the hurricane lamp (51%) followed by the

open flame lamp (36%) and electricity (13%) (see table 6.11). These findings show that

most children are exposed to the risks posed by these stoves firewood and charcoal) and

lighters (hurricane and open flame lamps) and are likely to get bums and scalds from

them because they are vectors o f direct heat and are most often placed within reach to the

children.

Findings from the analysis o f risk factors supports this where 59.2% o f the homes had

open or unprotected fireplace. Other risks analysed showed that 58.1% homes had

children 0-5 years playing without routinely being supervised; 29.4% homes had

custody o f medicines and potentially harmful chemicals within easy reach o f the

children; Sharp instruments were within easy reach o f the children in 36.1% o f the

homes; litter was scattered around the compound in 31% o f homes; 36.6% o f the

households were assessed as crowded and only 15% had homes being used as work place

such as workshops (see table 6.12).

Some o f the aforementioned risk factors may contribute to all the accidents while some

such as the sharp instruments, medicines and chemicals within reach o f children and

open fireplace are specific to accidents as cuts, poisoning and bums respectively.

83
7.1.2. THE EXTENT AND DETERMINANTS OF HOME ACCIDENTS

A. PREVALENCE OF HOME ACCIDENTS

In this study o f 1465 children, it was found that 456 (32%) o f the children under five had

accidents over the last one year. This figure o f 32% is higher than the already stipulated

10 % from the study done five years earlier in Baringo District, Kenya by Oloo M.

among the 0 - 1 8 year old children . The study done in Turkey also came up with an

incidence rate o f 21.6% in those under fifteen years o f age . This could be due to the

differences in age group studied, environmental factors and the difference in perceptions

o f accidents. In the Baringo District study it was found that most o f the accidents in

persons under twenty years o f age, are contributed to by those five years and below.

B. TYPE OF ACCIDENTS

The most commonly occurring type o f accident was found to be falls with injuries

(42%); followed by burns and scalds (30%); poisoning (10%) then others (18%). This

finding concurs with that o f Oloo in the Baringo District study (see table 6.13).

C. SEVERITY OF HOME ACCIDENTS

Most (48.9%) o f the injuries were rated as being o f moderate severity. The mild injuries

formed 31.5%, while the severe ones were 17.9%. O f the 80 severe injuries that

occurred; 38.8% were due to burns, 33.8% due to falls and 8.8% due to poisoning. This

trend o f finding is similar to that by Oloo M. in the Baringo Study27.

When each type o f accident was considered alone it was found that injuries due to burns

and scalds were more likely to be severe (22.8%) than those due to poisoning (15.9%)

84
and falls (14.4%). Injuries due to falls (45%) were more likely to be mild as compared to

bums/scalds (25%) and poisoning (16%). Poisonings were more likely to be moderate

(60%) as compared to bums/scalds and falls (40%) (see figure J).

This shows that even though falls predominate, they are more trivial in nature than the

other accidents such as bums and poisoning.

D. MOTHERS’ OPINION ON ACCIDENT PREVENTION

For those whose children got accidents, the mothers were asked if they thought the

accident could have been prevented and 55% o f the responses were in the affirmative,

34% were thought not preventable and 11% responses were ‘unknown’. Further analysis

showed that the mothers thought Bums and Scalds were most preventable (68%),

followed by Poisoning (58%) and Falls (51%). This was a statistically significant finding

(p=0.007). This attitude is encouraging for any intended accidents prevention

programme because it allows for direct interaction with mothers at the grassroots level.

The few (34%), who are negative and more so those (11%) who are undecided as to

whether accidents are preventable can, through education and advocacy, be convinced to

have positive attitudes towards accident prevention and actually do something about it.

85
E. CHILD’S CHARACTERISTICS AND ACCIDENTS OCCURRENCE

i). AGE

In the 0-5 years age-group alone, it was found that as age increases, children were more

at risk o f suffering from accidents because, presumably, they became more adventurous

and exploring as they developed. Children 3-4 years were the most at risk o f getting an

accident with a rate o f 43.7% as compared to children 2-3 years (37.5%); 4-5 years

(36.4%); 1-2 years (33.7%); and least at risk were the 0-1 years. This can be explained by

the fact that the infants are not yet developed enough to go exploring as older children

are, and also that they almost more often have supervision by a caretaker (see figure 4).

This relationship was statistically significant (p = 0.00), therefore age is an important

determinant o f accident occurrence. This was also shown by the Baringo District study

by Oloo27 and the Swedish study which found an increasing incidence with age and a

predominance in boys20 (N.B. sex is not a significant accident determinant in this study).

The different age groups were analysed for susceptibility to the various accidents and a

special pattern was observed. The children 2-5 years were most likely to suffer from falls,

bums and scalds, other accidents (such as suffocation and drowning) then poisonings.

The 0-1 years had a different pattern because they were found to be more likely to suffer

from a bum accident followed by falls, then others accidents and least poisoning.

When the type o f accident was controlled for, it was found that with bums and scalds, the

2-3 years age group were leading; for poisonings, the 4-5 years were leading; as for falls,

the 1-2 years were leading; and as for other accidents, the 3-4 years were leading. These

86
age groups mentioned contribute the largest to the respective accidents and this is a

statistically significant finding.

ii). SEX

It seemed as though boys get more accidents (33.6%) than girls (30.8%) and that a

smaller proportion of the accidents in total was contributed to by girls (219), but this

difference was not found to be statistically significant in this age-group (p = 0.8). Girls

are therefore as equally likely to get an accident as boys are (see table 6.16). When sex

was controlled for, the analysis revealed that either boys or girls were most likely to

suffer falls, bums and scalds and, lastly poisoning, in that order.

F. PARENTS’ CHARACTERISTICS AND ACCIDENT OCCURRENCE

The relative prevalence o f accidents in children 0-5 years was found to increase with

mothers’ level o f education with the lowest being in the children whose mothers were

not educated (29%) and highest in those whose mothers had secondary and higher

education (32.6%). The less educated mothers were more likely to be at home than

employed in formal jobs and they were more likely to take care o f their children

themselves (see figure 5). This trend was also observed with the occupation status.

The unemployed mothers or housewives had the lowest prevalence while the highest was

in the business, crafts or artisan group (see figure 6). The unemployed mothers had

comparatively better opportunity to take care o f their children directly as compared to

mothers in the professional and business groups. The latter two groups were likely to

leave their children in other people’s hands. The relationship between the mothers’

87
education or occupation and accident occurrence were each not found to be statistically

significant.

G. CARETAKERS AND ACCIDENT OCCURRENCE

When ages o f all the caretakers, were analysed for accident occurrence, it was found that

the 13-18 years had the highest (39.6%) accident rates in their children followed by those

< 13 years (33.3%) and lastly those >18 years (27.8%) (see table 6.17). This is possibly

due to their inexperience and lack o f commitment resulting from their stage in

development which tends to be o f playful nature and lacking in concentration and

commitment.

The rate o f all the accidents combined in the group with other relatives as caretakers was

the highest (57.1%), then the grandparents (50.9%), closely followed by the father

(50%), followed by the maid (41.1%) as compared to the mothers (29.1%) and older

siblings (6.7%). O f the maids, those less than 18 years had more than 60% o f their

children suffer accidents. This was however not a statistically significant finding (see

table 6.18 ). The high rate o f childhood accidents for those with grand parents as

caretakers, corresponds with the finding that more serious poisoning than other

caretakers in a study done in the under 6 years of age by Werozek C. et al l7. This means

they are not the best caretakers as they have mostly reduced reflexes due to age.

88
H. ENVIRONMENT AND ACCIDENT OCCURRENCE

i). AREA OF RESIDENCE

The area o f residence with most accidents in total was the urban one which contributed to

64% o f all the accidents, followed by the peri-urban (25%), and lastly the rural areas

(1!%). This finding was statistically significant (p< 0.05). This can be explained by the

fact that the sample had a similar distribution.

However further analysis showed that the highest prevalence rates o f accidents was found

in the peri-urban and slum areas (42.1%), followed by urban (31.6%) and lastly the rural

area (22%) (see figure 7). The higher rates observed are possibly due to the

environmental disadvantages inherent in the peri-urban and urban areas as compared to

the rural areas with less apparent risks. It is also worth noting that, in order to get the

most impact, one would start an accident control program in the urban area.

ii). HOUSING AND ACCIDENTS

Most accidents occurred in those with semi-permanent houses (67.6%), fewer in the

permanent (20.6%) and 11.8% in traditional huts which is common in the rural areas

Again this relationship may not be a causative one but a result o f certain characteristics

o f the lifestyles and environmental factors interacting and affecting persons living in the

particular areas and the different types of houses.

The relative prevalence o f accidents was found to be highest (34.9%) in the traditional

(grass thatched mud hut), followed by 33.2% in the permanent houses, and 31.3% in the

semi-permanent (see fig. 8). The rates, though not statistically significant, may be

89
attributable to an interplay of many socio-economic factors.

iii). ENVIRONMENTAL RISKS AND ACCIDENTS

In this study, analysis was made of some particular hazards found in the homes. It was

found that the accident rates were slightly higher in those homes with observed risks as

open fire places and where children played without supervision although this was not

statistically significant. But surprisingly, the rates were found to be lower in the houses

with risk factors such as: unsafe custody o f medicines and harmful chemicals, unsafe

custody o f sharp instruments, home being used as a working place such as workshop,

litter scattered around compound and crowded houses (see table 6.19). This was also not

statistically significant.

The accident rates were not markedly different in the different situations (ranging from

28.2% to 34%) and that there was no statistical significance in most o f the findings

except for the crowding status where the rate o f accidents was higher in the houses

classified as crowded.

The crowding status was assessed by enumerators as more than four people sleeping in a

room, and about two thirds o f all the sampled houses were assessed as crowded. O f all

the accidents, 28% were found to occur in the houses classified as crowded and 34% in

those not crowded. This was a statistically significant finding (p< 0.05, y2 = 4.5, d f = 1).

Just as in the homes without apparent risks, the homes classified as not crowded may be

posing unknown danger to young children and thus contributing to the higher incidence

o f accidents as compared to those classified as crowded. This relationship is not obvious

because the available rooms in crowded residences or few roomed houses are normally of
90
multipurpose use as kitchens, sitting room, and bedroom. This situation is normally

associated with congestion and thus exposure to hazardous environment. This type of

situation is most commonly found in the peri-urban and slum areas in towns. In some

cases, the cooking is done indoors or out in the compound alternately. Children are also

more likely to play out o f the house.

iv). PLACE OF ACCIDENT

Out o f all the accidents that occurred, 51% were indoors (27% in the living and

bedrooms and 24% in the kitchen). Those outdoors comprised 49%, making the house a

more risky environment for the child than the outdoors one. This higher prevalence of

accidents indoors was found to be statistically significant.

The house has been found to be a high risk environment as concerns childhood accidents

occurring at home, in other studies as well . In the study done in Paris, by Gaillard the

injuries that occurred in the household were found to have a higher pre-hospital mortality

rate as compared to those out o f the compound and were found to occur more commonly

in the under five years 19

Bums and scalds occurred mostly in the kitchen (50%), poisonings in the living and

bedrooms (59%), and falls mainly in the compound (79%). Other accidents such as

drowning and suffocation occurred mainly in the compound (67%). This was a

significant finding and therefore, in order to curb bums and scalds, poisoning and even

falls, children should stay away from the house and play outside in carefully and

deliberately made safer environments or play grounds. There should be supervision o f the

children where they play which should also be clean and appropriate.

91
v). TYPE OF LIGHTING

The highest number o f all the accidents was contributed to by those using hurricane

lamps (55.7%), followed by the open flame lamp and other types o f lighting such as

sticks or papers (29.8%), and lastly, in those using electricity (14.5%) and lastly those

using other lamps (0.4%) (see table 6.21).

The highest rate of accidents however, was found to be in those using ‘other’ methods of

lighting (40%), followed by those using electricity (37%), then the hurricane lamp (35%),

and lastly the open flame paraffin lamp (see figure 9). These percentages comparable to

those o f the relationship between accidents and the areas o f residence, possibly due to

interactions o f the socio-economic factors within the study sample where, urban dwellers

have access to electricity and may also use hurricane and the open flame lamps, the rural

- the open flame lamps and hurricane lamps and to a lesser extent electricity. The

findings are statistically significant.

When the specific accidents were analysed, it was found that:

The highest rate of bums occurred in those using the open flame lamp, followed by the

hurricane lamp and then electricity; The highest rate o f falls was also found in the group

using open flame lamp and others, followed by the electricity and then hurricane lamps;

The highest rate o f poisoning was found in those using the hurricane lamps, followed by

electricity and then open flame lamps and others (see figure 9).

Further analysis done, to find out the type o f injury mechanism in these different

92
situations, time o f accident and method of injuries were controlled for. In the day time it

was found that 50.9% o f the injuries in those with hurricane lamps, were due to striking

by person, object or animal; 23.9% due to bums or scalds whereas 7.5% and 4.9% were

due to paraffin and foreign body ingestion respectively. In the night time when the

lighting is dependent on these lamps, it was found that 15% o f the injuries in those with

hurricane lamps, were due to striking by person, object or animal; 70% due to bums or

scalds whereas 5% and 5% were due to paraffin and foreign body ingestion respectively.

The same pattern was mirrored in those using the open flame lamp. This pattern suggests

that these lamps contribute directly to the bums, (p < 0.05).

vi). SOURCE OF COOKING FUEL AND ACCIDENTS

The most number o f accidents was observed in those using charcoal stoves for cooking

(56.7%), followed by those using firewood stoves (20.2%), then gas (4.4%), then

kerosene and lastly electricity (2.6%). These proportions are more or less similar to that

o f the sampled population and the distribution can be attributed to that (see table 6.22).

The highest accident occurrence rates were however observed in those using firewood

(35.1%), then those using gas (33.3%), charcoal (32.5%), kerosene (28.4%) and lastly

electricity (27.(%). The children from families using firewood and gas are the most at

risk o f getting accidents than in the other groups. However this observed relationship was

not statistically significant.

Analysis o f the relationship between accident occurrence and the various sources of

cooking fuel revealed a trend similar to the one already observed for the whole sample

93
where for each type o f fuel, the most likely accident to occur was falls, followed by bums

and scalds, followed by poisoning and other accidents. The only difference observed was

in the case o f firewood where bums and scalds were leading followed by falls, followed

by ‘other accidents’ and lastly poisoning. This finding could be attributable to the fact

that it is the most open method o f cooking and can easily be accessed even by young

children. This was however, not statistically significant.

7.1.3. COURSE AND TIME OF ACTION FOLLOWING ACCIDENTS

The course o f action taken following injury mirrors the severity o f the accidents and has

socio-economic implications on the family and health system in general.

About 88% o f accidents were acted upon immediately (within two hours) which showed

that parents did not take these accidents lightly. The type o f accident was not found to

determine the speed o f action taken following an injury (see table 6.2 5 & 6.26).

For all the accidents considered together and individually, it was found that the majority

(35.6%) were treated as outpatients in a public health institution, whereas an almost

equal proportion (34.8%) were given home remedy. The ones which required admission

into an institution were about a quarter (25.5%) and these possibly represented the more

serious ones. Those not acted upon, and probably the most trivial, were very few (2%).

There was a different picture seen when each type o f accident is considered separately. In

the case of bums, the leading course of action was inpatient management, followed by

94
outpatient, home remedy and no treatment. This shows that for the bums the severity was

significant and give a lot o f strain on the health system.

For poisoning the leading action was outpatient treatment (45.5%), followed closely by

inpatient treatment (40.9%), then home remedy (9.1%) and traditional healer (4.5% ).

For the falls, the highest tendency was to give home remedy, followed by outpatient

treatment then inpatient treatment was last. This further proves the trivial nature o f the

falls as compared to the other types o f accidents (see table 6.23 & 6.24).

It was noted that o f the accidents that led to admissions and outpatient treatments,

bums/scalds and falls were the leading; those that were treated as out patients, bums

were still leading together with poisoning. This shows the seriousness o f these accidents

as compared to falls which are leading in the treatments given at home and those that

were not treated at all. This means that poisoning and bums and scalds have the most

implications on the health system in general. These were all statistically significant

findings.

95
7.1.4. OUTCOME OF HOME ACCIDENTS

In this study o f the 0-5 years it has been observed that on for all the accidents together,

86% had full recovery and for specific accidents on average, 80% recovered fully, 12%

had residual damages and chronic problems while the fatality rate observed was higher

depending on the type o f accident.

Just as in the study in Baringo, over 80% o f the injuries led to full recovery while 20%

had residual disability and lasting physical damage. The accident fatality rate was

observed at 6/100028.

Bums and scalds and falls contributed the most to residual damages and scarring. These

meant a certain amount o f chronic problems and therefore additional burden after the

initial recovery or partial recovery.

For the poisonings, the fatality rate was 13% or 130/1000 which is quite high. These

deaths resulted from ingestion o f medicines and snake bite. Poisonings made up two

thirds o f all the deaths. As for the falls it was 0.5% or 5/1000. This was one child who

died as a result o f the impact o f being thrown during his parents fight.

Two other accidental deaths resulted from drowning in a drum o f water and foreign body

ingestion. The information on these deaths however, was limited because it was only

obtained from the houses where there were other 0-5 years old children (see table 6.27).

The bums and scalds resulted in no deaths during the period o f the study. These were

statistically significant findings.

96
 . . CONCLUSION
1 2

The study established the prevalence o f home accidents as 32% which is higher than the

already stipulated in the previous study done in Baringo Kenya27 five years before which

was 10%. The three most commonly occurring types o f injuries were due to falls (42%),

followed by bums and scalds (31%), and poisoning (10%) for all the age groups, and

therefore they have the most impact on the health system in general.

Age o f children was found to be associated with accident occurrence and to the type of

accidents that occurred with the 2-3 years age group being o f special risk o f getting bums

and scalds and the 4-5 years age group at risk o f poisoning. The older a child became, the

more prone to accidents he got. The age group found to be o f special risk was the 3-4

years (with the highest rate o f 43.7%). Gender or sex was not found to predict one's risk

o f getting an accident.

The type o f caretakers posing the highest risk to child accidents were relatives and

grandparents, and 13-18 year old maids.

Accidents occurring within the house structure predominated (51%) over those occurring

in the compound (49%). The kitchen was found to be a very dangerous place as for all

the accidents.

Bums formed approximately 87% o f the household injuries ( bums that occurred in the

kitchen were 50.4% as compared to falls (21%) and poisoning (98%).

97
Injuries in the compound were predominated by those due to falls.

About one third o f homes were found to have risk factors (each considered separately)

such as: unsafe custody o f medicines and harmful chemicals, unsafe custody o f sharp

instruments, home being used as a working place such as workshop, litter scattered

around compound and crowded houses. Accident rates were slightly higher in those

homes with observed risks as open fire places and where children played without

supervision.

The lighting system was found to be a statistically significant contributory factor. The

children from families using hurricane lamps, firewood and gas were the most at risk of

getting accidents than in the families using other methods o f lighting and cooking

respectively.

The residential areas with leading accident rates were the peri-urban and slums.

Most o f the accidents (80%) were trivial (mild to moderate in nature). The severe injuries

were contributed to mainly by bums/scalds and falls in equal proportion. Though if a

child got a bum or scald, the chances that the accident would be severe are 1.5 times

higher than if he or she got a fall (i.e. 23% versus 15% respectively).

For all the accidents, 64% required institutional management (25.5% resulted in

admissions to health care institutions and 38.5% treated as out-patients), where most of

the admissions were due to bums and scalds (46%).

98
The time o f action in most cases was immediate and appropriate as 88% o f them were

acted upon within two hours ( or 30 minutes in the case o f poisoning).

The outcome o f injuries due to home accidents was satisfactory as most (85.7%) resulted

in recovery; while 12.2% had residual damages and chronic problems and 2% resulted in

death. The deaths occurred due to poisoning (67%) and falls (11%) and others (22%).

Bums and scalds and falls contributed the most to residual damages and scarring and

subsequently chronic problems and additional burden after the initial recovery or partial

recovery. The highest fatality rate was found to be from poisoning (13%); followed by

other accidents (5%); and falls (0.5%). There were no deaths reported with bums.

On accidents prevention, most mothers (55%) thought that accidents were preventable

while 34% were thought not preventable and 11% did not know. This is a positive

attitude and is encouraging because it shows that a good percentage would be receptive

to intervention programmes to curb accidents at the grassroots level and, something can

actually be done by tackling the problems as found from analysis in this study, o f the

correlated factors and the background of the population (see recommendations).

99
7.3. RECOMMENDATIONS

1. This study indicates that accidents are more prevalent than already known and are

viewed as preventable by many mothers and that the next course o f action should be

an intensified campaign on health education and preventive measures on childhood

accidents as a whole.

2. Children should be given some supervision which is specific and appropriate to their

age and developmental stages.

3. Children, especially those above 1 year, should be encouraged to play out o f the house

in deliberately made safer environment and should specifically not be allowed in the

kitchen.

4. Fireplaces should be protected within the households in case unguided children find

their way near the fires.

5. Since the modem mother has to play various roles one o f which is a family co-bread

winner, caretakers such as maids should be given simple appropriate education on

safety and child development.

6 . Certain risk factors such as medicines and harmful chemicals should be particularly

observed and emphasis made to keep them out o f reach as poisonings resulting from

these caused the highest fatality rate.

100
7. There should be more nation-wide and specific research to enable formation of

national policies and guidelines on the feasible aspects o f childhood accident

prevention, especially more so relating to the low socio-economic status mothers who

have too much in their hands and cannot adequately watch on children as they play

and also do not have the means to employ caretakers. Active prevention o f home

accidents would also go along way in lessening the burden they cause on our health

system

8 . There should be more specific qualitative research locally on the significant

contributing risk factors in the environment such as residential areas and housing,

lighting systems, kitchens, fireplaces and cooking fuel sources to find out which

methods are best used in preventive activities and programs. Meanwhile, using the

findings from earlier studies to formulate age-specific safety education that is

repetitive and individualised with active parental participation to increase knowledge

and certain safety practices around homes.

9. Much can be achieved by actively involving communities in the aspect o f accident

prevention in the peri- urban and slum areas terms o f Community Participation from

the planning stages through the implementation right to the monitoring and evaluation

stages. Preventive measures such as defining appropriate caretakers can be supervised

at the grassroots level by community health workers who have been trained and who

would therefore give repeated health education and monitor things like storage of

medicines and at the same time advocate for adequate and safe play facilities outdoors

rather than indoors for children to reduce household accidents.

101
REFERENCES
 REF ERENCES

1. WHO., Accidents in childhood and adolescence. 1991: xi.

2. Manciaux M., RomerC.J., Accidents in childhood and adolescence. 1991:
xi.
3. Romer C. J., Injury Prevention Programme, WHO. 1991.
4. Romer C.J., Manciaux M., Prevention of accidental injuries in childhood
and adolescence. Bulletin of the International Pediatric Association, 6:
243-249(1985).
5. Marcusson H., Oehmisch W. Accidents mortality in childhood in selected
countries of different continents, 1950-1971. World health statistics report,
30(1): 57-92 (1977).
6. WHO: New trends and approaches in delivery of maternal and child care.
WHO. Tech. Repor. Series, No. 600, 1976.
7. Epidemiology of childhood injuries. American Journal of Diseases of
Children. [JC:3gs] 136(5):399-405, May 1982.
8. Minchom P.E., Sibert J.R., Newcombe R.G., Bowley M.A. Does health
education prevent childhood accidents?: Postgraduate Medical Journal.
[JC:pfx] 60(702):260-2, Apr. 1984.
9. Dershewitz R.A., Williamson J.W., Prevention of childhood household
injuries: a controlled clinical trial. American Journal of Public Health.
[JC:3xw] 67(12):1148-53, Dec. 1977.
10. Kelly B., Sein C., McCarthy PL., Safety education in a pediatric primary
care setting: Pediatrics. [JC:oxv] 79(5):818-24, May 1987.
11. Cason C.G. A study of scalds in Birmingham : Journal of the Royal
Society of Medicine. [JC:jx1] 83(11 ):690-2, Nov. 1990.
12. Rfidah E.I., Casey P.B. Tracey, J.A.; Gill, D.; Childhood poisoning in
Dublin. Irish Medical Journal. [JC:gxd] 84(3):87-9, Oct. 1991.
13. Anonymous. Health protection; Accident prevention and injury
control:Pub!ic Health Reports - Hyattsville. [JC:qja] Suppl:80-93, Sep-Oct
1983.
14. Gupta M., Gupta O.K., Yaduvanshi R.K., Upadhyaya J. Burn
epidemiology: the Pink City scene. Burns. [JC:afc] 19(1 ):47-51, Feb
1993.
15. Selbst S.M., Baker M.D., Shames M. Bunk bed injuries. American
Journal of Diseases of Children. [JC:3gs] 144(6):721-3, Jun, 1990.

102
16. Reilly J.S., Walter M.A. Consumer product aspiration and ingestion in
children: analysis of emergency room reports to the National Electronic
Injury Surveillance System. Annals of Otology, Rhinology & Laryngology.
[JC:5q2] 101 (9):739-41, Sep 1992.
17. Wezorek C., Dean B., Krenzelok E. Accidental childhood poisoning:
influence of the type of caretaker on etiology and risk : Veterinary &
Human Toxicology. [JC:xbv] 30(6):574-6, Dec 1988.
18. Cervantes P.A., Borrajo G.E. et al. Importance of childhood accidents.
Results of a regional study. An-Esp-Pediatr; 32(6):493-8. Jun 1990.
19. Gaillard M., Herve C. Emergency medical care and severe home
accidents in children. Study of 630 cases over 5 years. Their significance
in traumatic accidents. Ann-Pediatr-Paris. 38(5): 311-7. May 1991.
20. Lindbald B.E., Terkelsen C.J. et [Link] related childhood accidents. A
survey of 1590 cases. Scand-J-Soc-Med. 18(4): 269-71. Dec 1990.
21. Bergman A.B., Rivara F.P. Sweden's experience in reducing childhood
injuries. Pediatrics. 88(1): 69-74. Jun 1991.
22. Ryan C.A., Shankowsky H.A., Tredget E.E. Profile of the paediatric burn
patient in a Canadian burn centre. Journal: Burns. [JC:afc] 18(4):267-72,
Aug 1992.
23. LariA.R., Bang R.L., Ebrahim M.K., Dashti H. An analysis of childhood
burns in Kuwait. Journal: Burns. [JC:afc] 18(3):224-7, Jun 1992.
24. Bannon M.J., Carter Y.H., Mason K.T. Causes of fatal childhood
accidents in North Staffordshire, 1980-1989. Archives of Emergency
Medicine. [JC:acg 9(4):357-66, Dec 1992.
25. Tursz A., Lelong N. et al. Home accidents to children under 2 years of
age. Paediatr-Perinat-Epidemiol. 4 (4): 408-21. Oct, 1990.
26. Sullivan M., Cole B. et al. Reducing child hazards in the home. A joint
venture in injury control. J-Burn-Care-Rehabil. 11(2): 175-9. Mar-Apr;
1990.
27. Oloo M. Home accidents in childhood and adolescence. M. Med Thesis.
Oct, 1988.
28. Hazoume F., Bertan M., MuftirY., Sonako A., Mello Jorge (de), Schmidt, B.
J., Bucher, J. F. WHO: Communication to the study group on assessment
of country surrveys on accidents in childhood, Ankara. 24-25, November
1982, IRP/ADR 216, 1742M, 1983.

103
29. Fiedman H. L., The health of adolescents and youth: a global overview.
Wld. Hlth. stat. Quart., 38 (3): 256, 1985.
30. Martin H. L. Antecedents of burns and scalds in children. Brit. J. Med.
Psychol; 43: 39, 1970.
31. Personal Review. Accidents and child abuse folders. Daily Nation
newspapers archives, Nation, Centre Nairobi. 1989- 1992.
32. Hospital records on accidental injury. The paediatric ward, NPGH, Kisumu,
April, May, June 1992.
33. Manciaux M. Accidental injuries in the young: from epidemiology to
prevention. Effect. Hlth. Care, 2 (1):21, 1984.
34. Bull J., Prevention of childhood injuries.(1975).
35. The Kisumu District socio-economic profile. UNICEF. 1990.
36. Lwanga S.K., Lemeshow S., Sample size determination in health studies..
A practical manual. WHO, Geneva. 1991.

104
ANNEXES

I, II, & III

Q U E S T IO N N A IR E HOME A C C ID E N T S ANNEX I

S e r ia l n u m b e r -----------------------------------------------------------------------------------

E n u m e r a t o r ' s n a m e ----------------------------------------------------------------------------

1. Name o f R espondent (m other, g u a rd ia n o r fa th er)

2. D ate o f in te r v ie w ----------------------------------------------------------------

3. Id e n tific a tio n of h /h o ld (H o m e ste a d a n d h o u s e No.

4. V illa g e /E sta te ----------------------------------------------------------------------

5. S u b lo c a tio n of h / h o l d -------------------------------------------------------

PA REN TS' P A R T IC U L A R S (to be en tered in th e ta b le /r e g iste r o f p a r e n ts .)

6. P a r e n ts/G u a r d ia n s o c c u p a tio n
1. P r o fe s s io n a l (e .g tea ch er) 2. C r a ftsm a n /a rtisa n
3. H o u se w ife 4. B u sin e ss 5 . Farm er
6 . U n e m p lo y e d 7 . O t h e r s s p e c i f y . 1*

7. M a r ita l s t a t u s o f p aren t:
1. S in g le 2 . M a rried 3. S ep arated
4 . D iv o rc ed 5 . W id o w e d .

8. E d u c a tio n a l s t a t u s :
1 . None 2. 0-8 y ea r s(p r im a ry ) 3. Secondary
3. P ost Secondary i . e (U n iv e r sity , C o lle g e , e.t.c)

9 . A re t h e p a r e n t s perm anent r e s id e n t s in th e house?

1. Y es. 2 . No. 3. P a r tly .

10. Who l o o k s a fter th e c h ild r e n as th ey p la y (i.e caretak er)?

1. F a th er, 2 . M oth er 3 . M a id
4. E ld e r ly gran d p a ren ts. 5. O ld er c h ild r e n
6. O t h e r s p e c i f y ---------------------------------------------------------------------------

11. W hat i s h is/h e r ed u c a tio n a l sta tu s?

1. None 2 . 0 - 8 y e a r s (p r im a r y ) 3. Secondary
4. P ost Secondary i . e (U n iv e r sity , C o lle g e , e . t . c )

12. W hat i s th e ca r eta k er's age? I |

|____i
1. <13 y e a r s 2. 13-18 years 3. A b o v e 18 y e a r s .

105
ANNEX I 106

R E G IS T E R OF ALL PA REN TS, G U A R D IA N S AND ADULTS IN THE H O U S E -H O L D

RELATIONSHIP
DATE OF MARITAL EDUCATIONAL RESIDENTIAL TO HEAD OF ETHNIC
No NAME BIRTH SEX OCCUPATION STATUS STATUS STATUS H/HOLD ORIGIN
 107

ANNEX I

R E G IS T E R OF ALL C H IL D R E N IN THE H O U S E -H O L D

NAME OF.
MID ARM 1 . FATHER MARITAL
DATE RELATIONSHIP PLACE CIRCUM­ 2 . MOTHER STATUS PARENTS' EDUCAT­
OF TO HEAD OF OF ETHNIC FERENCE 3. GUARDIANOF OCCUPA­ IONAL
No NAME SEX BIRTH H/HOLD BIRTH ORIGIN ; 5YRS 4. CARETAKER
PARENTS TION STATUS
 ANNEX I
13. Have t h e c h ild r e n <5 y e a r s fa lle n sic k over th e la st 1 YEAR? i .

1. Y es. 2. No.

14. I f y e s , w hat was t h e i l l n e s s i n y o u r o p in io n ?

1. A c c id e n t. 2. M a la ria 3. D ia rr h o e a
4. O t h e r s p e c i f y --------------------------------------------------------------------------------------------
L i s t t h e i l l n e s s e s i n o r d e r o f o c c u r r e n c e i f m ore t h a n o n c e .

Q15. Q16.

£□□□□□□□□
=00000000
a. __________________________n □
b. __________________________□ n
c. __________________________n □
d. __________________________□ n
e. ________________________________________________ □ □

f. __________________________□ n
g- ____________________________________________ □ □

h. __________________________□ n
15 W hat c o u r s e of a c tio n was t a k e n fo llo w in g th e illn e ss?

1. Taken t o t h e h e a l t h c e n t r e / d i s p e n s a r y ( o u t - p a t i e n t )
2. H o sp ita liz a tio n (in p a tie n t).
3. T r a d itio n a l h e a le r .
4. S e l f m e d i c a t i o n (hom e rem ed y).
5. P rayers.
6. N one.
7. O t h e r s s p e c i f y ----------------------------------------------------------------------------

16. How l o n g a f t e r r e a l i s i n g t h e i l l n e s s was t h e a c tio n tak en ?

1. <30 m in . 2. 30 m i n . - 2 h rs. 3. 2-12 hr
4. 1 2 - 2 4 h rs. 5. 1-3 days 6. >3 d a y s

17. If h e a l t h f a c i l i t y was u s e d , how f a r w as i t ?

1. < lk m 2 . l-2 k m 3 . 3 -4 k m 4. >4km
5. Not a p p lic a b le .

18. W er e y o u s a t i s f i e d w ith th e trea tm en t g iv e n in th e h e a lth

fa c ility ?
1. Y es. 2. No.

19. What m e a n s d o y o u u s e t o g e t t o t h e h e a l t h f a c i l i t y ?
1. W a lk 2. B ic y c le 3 . M otor v e h i c l e
4. O t h e r s p e c i f y . -------------------------------------------------------------------------------

20. W hat i s t h e le n g t h o f tim e ta k e n t o th e reach

th e n e a r e st se r v ic e sta tio n ?
1. <1 h r 2. 1-2 h rs 3. >2 h r s

21. Has t h e r e been a d eath o f any c h ild under th e

age o f f iv e in t h is house ?
1. Y es 2. No
3. D o n ' t know

108
 ANNEX I

22. W hat w a s t h e c a u s e o f d e a t h ?
A c c id e n t 2. M a la r ia
H co in vo
Fever 4. D ia rr h o e a
D o n ' t know
O t h e r s , s p e c i f y _____________________________

109
 110

ANNEX I
ACCIDENTS

23. Has t h e r e b een an a c c id e n t in th e h o u s e h o l d am ong t h e c h ild r e n under f i v e years ?

1. Y es 2. No
If y e s, f i l l th e ta b le b elo w a s you an sw er f o l l o w i n g th e q u estio n s:

CHILD NAME OF CHILD DATE OF Q24 Q25 Q26 Q27 Q28 Q2 9 Q3 0 Q31 Q32 Q33 Q34 Q35 Q36
REGISTER ACCIDENT
NUMBER

24. W hat t y p e o f a c c i d e n t w a s i t ?
1. B urns 2. P o iso n in g 3. F a ll
4. O t h e r s , S p e c i f y ___________________________________

25. D esc rib e t h e m ethod o f in ju r y .

1. S tr u c k b y o b j e c t , p e r s o n o r a n im a l 2. S c a ld from h o t w a ter, h o t t e a , u j i or soup.

3. F o r e ig n b ody. 4. S tr a n g u la tio n . 5. B ite /stin g
6. D ro w n in g /S u b m er sio n 7. P a r a ffin p o iso n in g . 8. M e d ic in e P o iso n in g
9. I n s e c tic id e s /O t h e r c h e m ic a ls. 110
0 O th er, S p e c if y ____________________________
 ANNEX I

26. W hat w a s t h e r e su ltin g in ju r y ?

1. Burns a n d /o r s c a ld , 2. S u p e r fic ia l a b ra sio n s

3. F ra ctu re, 4. D eep f l e s h w ounds, 5. C uts.
6. In te r n a l/S y ste m ic in ju r ie s
7. O th ers s p e c i f y , e . g co m b in a tio n o f th e a b o v e.

D esc rib e oth er e v e n tu a litie s.

What w a s t h e a n a to m ica l site of th e in ju ry ?

1 1
1. H e a d /N e ck 2. U pper lim b s . 3. Lower l i m b s .
4. Trunk 5. E yes, 6. E ars, 7. N ose,
8. M ou th .

What w a s t h e se v e r ity of th e in ju r y ?
1 1
1. M ild 2 . M od erate. 3. Severe 4. D on' t k n ow

29. W hat c o u r s e of a c tio n was ta k e n fo llo w in g th e a c c id e n t ?

1. Taken t o th e H e a lth C e n t r e /D is p e n s a r y (O u t-p a tie n t)

2. H o s p ita lis a tio n (In -p a tie n t)
3. T r a d itio n a l h e a le r
4. S e l f M e d i c a t i o n (Home R em ed y)
5. P rayers. 6. None
7. O th ers, S p e c ify

30 . How l o n g a fter r e a lisin g th e a c c id e n t was th e a c tio n tak en ?

1: < 30 m i n . 2: 30 m i n . - 2 h r s . 3 : 2 - 1 2 h rs.
4 : 12 - 24 h r s . 5: 1 - 3 days. 6 : > 3 d ays.

31. Who w a s t a k i n g care of th e c h ild at th e tim e o f th e a c c id e n t ? . i

1. Parent 2. R e la tiv e 3. O th er c h i l d
4 . M aid 5. No o n e
6 . O th er, S p e c ify

32 . W here d i d th e a c c id e n t ta k e p la c e ?
1 1
1. K itc h e n 2. L i v i n g Room 3 . Bedroom
4. Out i n t h e co m p o u n d /G a rd en 5. D iffe r e n t hom e.
6. O ther, S p e c if y

33. At what tim e d id th e a c c id e n t ta k e p la c e ?

1. M o rn in g 2. L u n c h tim e 3. A fte r n o o n /E v e n in g
4. S u p p e rtim e 5. A fter su p p e r / bed t i m e lI

Ill
 ANNEX I

34. W hat w a s t h e cause o f th e a c c id e n t in your o p in io n ?

1. C a r e le ssn e ss 2. F a u l t y eq u ip m en t
3. S tu b b orn C h ild 4. D o n ' t know
5. O t h e r , S p e c i f y ____________________________________________

35. C o u ld t h e a c c id e n t have been p reven ted in your o p in io n ?

1. Y es 2. No. 3. D o n ' t know.

If yes, h o w ? ____________________________________________________________________

36. What w a s t h e fin a l ou tcom e o f th e a c c id e n t ?

1. F u ll reco v ery
2. R esid u a l d i s a b i l i t y / e x t e n s i v e sc a r r in g
3. D eath 4 . C h r o n ic ity .

SOCIO-ECONOMIC STATUS

37. What i s th e sa la r ie d fa m ily in co m e p e r m onth ? (Kshs.)

1. 0-700 2. 800-1,500 3. 1,500-3000

4. >3000 5. D on' t know

38 . What p r o p e r t y i s o w n e d b y t h e f a m i l y ?
1. C a t t l e - N u m b e r ______________________________________
2. F arm l a n d - S t a t e a c r e - a g e ______________________
3. C o m m e r c i a l b u s i n e s s - t y p e ( s ) __________________
4. M o t o r V e h i c l e s - n u m b e r a n d t y p e ______________
5. B i c y c l e _________________________________________________
6. R a d i o - What p r o g r a m m e s d o t h e y l i s t e n t o ?

39. Do t h e y r e a d N e w s p a p e r s ?

1. Y es 2. No. 3. O th er, S p e c ify

40. W hat i s th e system of lig h tin g when d a r k ?

1. E le c tr ic ity 2. H u r r i c a n e p a r a f f i n lam p 3. T r a d itio n a l

p a r a f f i n lam p ( N y a n g i l e ) 4. O t h e r s p e c i f y _____________________

41. W hat i s th e re g u la r source of fu el for c o o k in g ? □ □ n □

1. E le c tr ic cooker. 2 . G as. 3. C harcoal 4. K erosene 5. F irew o o d

6. O t h e r , S p e c i f y ____________________________________________________________

42. W hat i s t h e w a t e r s o u r c e f o r t h e h o u s e / h o l d ?
1 . P ip e d w a te r . 2 . B o r eh o le 3 . W ell 4 . R iv er
5. O ther, s p e c i f y 1

43. How i s t h e h o u s e h o l d w a s t e d i s p o s e d ?
1 . B u rn in g 2 . C om post p i t 3 . O pen d u m p in g

112
 ANNEX I

4. M u n icip a l d u s t b in s .

44. Is i t w ith in ea sy reach o f th e ch ild r e n ?

1 . Y es 2 . No

45. D o e s t h e h om e h a v e t h e f o l l o w i n g c h a r a c t e r i s t i c s ?
(T o b e o b s e r v e d , t h e n e n t e r 1 . Y e s 2 . No)

1. U n p ro tec te d m e d ic in e s or c h e m ic a ls
w ith in th e c h ild 's reach .

2. K n iv es and h a rm fu l to o ls left w ith in reach

3. U n p ro tected fir e p la c e .

4. U n su p erv ised p la y .

5. Home b e i n g u s e d a s a w o r k p l a c e
e . g C a r p e n tr y , C o m m ercia l f i s h fr y in g etc.

6. S ca ttered h arm fu l litte r a ll over th e com pound.

46. D oes any one in th e h o u se ta k e a l c o h o l i c d r in k s ?

1 . Y es 2 . No 3 . D o n ' t k n o w

47. I f y e s , how m uch ?

1. S o c ia lly a c c e p ta b le le v e ls
2. E x c e ssiv e
3 . D o n ' t know

HOUSING AND ROOM ALLOCATION

48. What i s th e siz e of th e house (in m etre sq . ) ?

49. How m a n y r o o m s d o e s th e house have ?

(A lso in d ic a te b elo w )

K itc h e n 1 S ittin g 1 Bedroom 1 L a trin e 1 G arden1

50. W hat a r e t h e h o u s i n g m a t e r i a l s ? 1— 1 i__l

(In d ic a te by tic k in g ) 1. 2. 3. 4 . 5. 6. 7.
1. T h atch ed r o o f 2 . Muddy w a l l 3 . S em i-p er m a n e n t w a l l s i .e .
E arth + C em en t 4 . E a rth f l o o r 5 . C em en ted , wood o r s t o n e w a l l s
6. C orrugated ir o n r o o f 7 . C em en ted o r w ood en f l o o r .

51. How d o t h e c h ild r e n sle e p in order o f th e room s ?

113
 ANNEX I

D escrib e and g i v e a sim p le s k e t c h b e lo w :

52. A re t h e room s 1. C row ded ( i - e m ore t h a n

fo u r sle e p in g in a room ), 2. Not crow ded ?

53. W hat i s t h e Zone o f r e s i d e n c e o f t h e h o u s e h o l d ?

1. R ural 2 . U rban 3. P e r i-u r b a n 4 . S lu m s.

114
BACKGROUND INFORMATION ON KISUMU Annex III

GEOGRAPHY

K i s u m u D i s t r i c t l i e s w i t h i n l o n g i t u d e s 33° 2 0 'E a n d 35° 2 0 ' E

and latitudes 0° 2 0 ' S a n d 0° 5 0 ' S, and covers a total area

of 2660 sq. km. ,of which 567 square kilometres are under

water . It is the s e c o n d l a r g e s t of t h e f o u r d i s t r i c t s t h a t

form Nyanza Province. To the southwest is South Nyanza

District; to the S outh is K i s i i D i s t r i c t ; to the nor t h e a s t

is N a n d i District; to the east is K e r i c h o D i s t r i c t ; to the

northwest is K a k a m e g a D i s t r i c t ; a n d f i n a l l y to t h e w e s t is

Siaya District.

The District lies in a d e p r e s s i o n that is p a r t of a large

lowland. It s u r r o u n d s t h e N y a n z a Gulf, a p r o t r u d i n g p a r t of

Lake Victoria at t h e h e a d of w h i c h is K i s u m u Town. East of

Kis u m u T o w n are the Kano p l a i n s o c c a s i o n a l l y b r o k e n by low

ridges and rivers. The processes associated with the

formation of the Rift Valley are believed to have

influenced some of thhe notable physical features, for

example the scarps in the north, east and South and the

associated hill slopes and the p i e d m o n t p l a i n s that spread

across the vast Kano plains.

TOPOGRAPHY

The D i s t r i c t can be d i v i d e d into three t o p o g r a p h i c a l zones;

The Kano Pla i n s , the upland areas of Nyabondo, and the

midland areas of Maseno. The upland area is comprised of

ridges which rise gently to altitude 1835 m above sea

116
 ANNEX I I I

level, while the Kano plains lie on the floor of the Rift

Valley, which is a flat stretch bordered to the north and

east by the escarpment.

The N y a bondo Escarpment forms the s o u t h e r n b o u n d a r y of the

Kano Plains and is far from the municipality. The Nyando

Escarpment forms t h e n o r t h e r n b o u n d a r y of the plains, with

the N y a n d o a n d N a n d i hills l y i n g to t h e n o r t h - e a s t a n d the

w e s t e r n c o n t i n u a t i o n of t h e s e h i l l s form i n g the K a j u l u and

Kisian ranges up to M a s e n o in t h e n o r t h - w e s t .

The flattest parts of the municipality occur within the

Kano Plains. The Kano Plains are p r o n e to annual flooding,

especially the lower Kano plains and particularly in the

Nyando Valley. Notably, t h e 1 988 l o n g r a i n s c a u s e d a lot of

flooding which destroyed a lot of crops and livestock as

well as evacuating the residents. This virtually brought

all activities to a standstill. Otherwise the area is w e l l

suited for sugar, maize and sorghum, and has a high

potential for irrigation whereby rice does well. Zero­

grazing is a n o t h e r p o t e n t i a l in t h e area.

KISUMU TOWN

Kisumu is a n o l d t o w n w h i c h w a s established in 1901. It is

surrounded on the southern side by Lake Victoria. Since

1971 when the town was given municipal status and its

boundaries extended, there has been a rapidly expanding

117
 ANNEX I I I

informal squatter zone.

Due to its origins, Kisumu Town has over the years relied

heavily on administrative and commercial development to

sustain its g r o w t h . There are six major roads that connect

kisumu to its regions and the nation; the main one being

the Kisumu-Kericho-Nakuru-Nairobi-Mombasa road, among the

other bitumen roads.

Apart f r o m t h e s e b i t u m e n roads, t h e r e a r e a l a r g e n u m b e r of

secondary roads connecting Kisumu to the various parts of

its a r e a of influence.

Other forms of transport include, rail, lake and, air

transport. Kisumu also has good posts and

telecommunications services, with the Kenya Power and

Lighting company supplying the old town and, to a lesser

extent, the peri-urban and rural areas of Kisumu

municipality with electricity.

SO C IO -EC O N O M IC PR O FIL E

Table 9.0: P o te n tia l L and u s e

Aaro-Ecological zones* by Divisions (sa. km.)

D iv is io n UM, LM, LM, LM, LM. O th e r

M ase n o 0 26 149 2 36 7 13

N yando 47 0 52 431 1 29 8

W inam 0 0 0 38 56 0

M u h o ro n i 24 6 33 7 0 0 0

118
 ANNEX I I I

UM3 = Marginal Coffee Zone

LM3 = Lower Midland Sugar Cane Zone

LM2 = Marginal Sugar Cane Zone

LM3 = Lower Midland Cotton Zone

LM4 = Marginal Cotton Zone

Other = UM, (13) A N D U M 2 (8), Coffee/Tea Zones

S o u r c e : District Agricultural Office.

Land use within the municipality is dominated by

agriculture and residential, commercial and industrial

housing.

There are varied types of soils within the municipality.

Less than one t h i r d of the d i s t r i c t is c l a s s i f i e d as b e i n g

of h i g h p o t e n t i a l . T h i s l a n d is l a r g e l y l i m i t e d t o t h e d e e p

well d r ained soils of K a j u l u a n d t h e K a n o P l a i n s area east

of Kibos and lying between the Kisumu-Kibos and Kisumu-

Kericho roads (su b j e c t to adequate drainage during the

rainy sea s o n an d to i r r i g a t i o n d u r i n g the d r i e r p e r i o d s of

the y e a r ) . T h e b a l a n c e of t h e m u n i c i p a l i t y f a l l s w i t h i n the

low potential areas of the district.

Livestock and Crop Production

There are four major activities in the a g r i c u l t u r a l sector

of Kisumu Municipality: crop production, livestock

production, fisheries and forestry. Crop production, ranks

119
 ANNEX I I I

first in o r d e r of importance and involves the growing of

mai z e , beans, sorghum, rice, finger millet, cotton and

s u g a r cane. The m u n i c i p a l i t y has c o n s i derable potential for

horticultural, crops, bananas, citrus, pawpaw, mangoes,

pineapples and vegetables.

Maize and cotton are the crops occupying the largest area

of cultivated land in W i n a m a n d M a s e n o divisions.

There are stati s t i c s that show that in 1 9 8 8 o n l y 10% of the

l a n d in t h e n e w m u n i c i p a l i t y w a s c a s h c r o p p e d , a n d t h a t the

area under cash crops in the old municipality was

negligible. The p r oblems inhibiting increased agricultural

production in Kisumu Municipality include inadequate

advisory services, inadequate farm inputs, poor marketing

systems, adverse weather conditions, pests and diseases,

p o o r soils and flooding. T hese will c o n t i n u e to impede crop

p r o d u c t i o n in t h e f o r e s e e a b l e fut u r e . T h e p e o p l e of K i s u m u

M u n i c i p a l i t y w i l l c o n t i n u e to d e p e n d o n o f f - f a r m a c t i v i t i e s

for their livelihood. The Government recognizes this f act

and has initiated programmes in the area to increase

employment and help fight poverty.

Livestock Production

The latest records of livestock are data from 1983, there

were 65 grade cattle and 19,111 local cattle in the new

municipality, and no grade cows and only 4 8 of the local

120
 ANNEX I I I

variety in the old municipality. The effective stocking

rate was 85 livestock per sq. km in the new municipality

a n d 4% in t h e o l d m u n i c i p a l i t y .T h e g r a d e c a t t l e in t h e a r e a

are mainly Zebu cattle which are kept for both meat and

m i lk. Grade c a t t l e a r e m a i n l y f o u n d in t h e c o o l e r a r e a s of

Maseno and W i n a m divisions.

Employment

A l t h o u g h a g r i c u l t u r e is t h e m a i n s t a y of t h e a r e a ' s e c o n o m y ,

it is not developed well enough to employ the whole of

western Kenya's working population. This results in the

in f l u x of people into the urban centre in search of work,

but there are not enough jobs there for those seeking

employment. The unemployment situation in Kisumu

Municipality is p a r t i c u l a r l y a l a r m i n g .

A study done on the Kisumu water supply and sanitation in

1985, r e v e a l e d that t h e t o t a l n u m b e r of p e o p l e employed in

the formal sector was 11%.

Compared to Nairobi and Mombasa, Kisumu has been faring

poorly in industrial development. According to Employment

and Earnings in Modern Sector Reports ( C B S , 1979), Kisumu

has the lowest e m p l o y m e n t p r o s p e c t s c o m p a r e d to N a i r o b i and

Mombasa.

A l t h o u g h a b r e a k d o w n of m a l e / f e m a l e ratios is n o t p o s s i b l e

121
 ANNEX I I I

for employment in Kisumu Municipality, it can be assumed

that because of the bias of equipping more men than women

with marketable ski l l s , there are more men than women

employed in t h e m u n i c i p a l i t y ' s formal sector.

The informal sector in Kisumu Municipality, like nearly

everywhere in Kenya, has gained increasing importance,

offering employment opportunities for a large number of

people.

Water and Sanitation

K i s u m u M u n i c i p a l i t y s t a n d s at t h e s h o r e s of L a k e V i c t o r i a -

one of the largest fresh water lakes in the world-yet her

residents have always experienced water problems. The

residents' demand for w a t e r has outstripped the available

supplies. But it is the peri-urban areas where water

problems are [Link] residents are then forced to buy

water e x p e n s i v e l y or fetching it from a far distance.

According to the K i s u m u W a t e r S u p p l y a n d S a n i t a t i o n Study,

V o l . I I B , 1 9 8 5 87% of K i s u m u ' s o l d t o w n u r b a n p o p u l a t i o n is

served by a water-born sewerage system. Of the remainder,

those living in l o w p o p u l a t i o n d e n s i t y a r e a s are served by

septic tanks and house-holds within the high population

density use low-cost sanitation methods, usually pit

latrines.

UNIVERSITY Oh ft4 IR£»'r

• •3RAP*'
122
 ANNEX I I I

Incomes and Expenditure

In Kisumu Municipality, incomes vary from earnings under

Kshs.215 per month to incomes in e x c e s s of Kshs.6,000 per

month f o r a p e r i o d of f i v e yea r s .

As per the Government's stipulated minimum wage in 1985

(KShs.540 p e r month), 50% of t h o s e e m p l o y e d l i v e d b e l o w or

close to the p o v e r t y line.

Currently, Kisumu Municipality has an area of 270 sq. km.

M o s t of t h e l a n d in t h e e x p a n d e d a r e a s is in p r i v a t e hands.

123