International Journal of Public Health and Epidemiology Research IJPHER

Vol. 7(1), pp. 187-198, January, 2021. © www.premierpublishers.org. ISSN: 1406-089X

Research Article

Household Demographic Predictors of Drug and

Substance Abuse among High School Students in
Kisumu East Sub County, Kisumu County – Kenya
1Orende M Awino*, 1Onguru D Ogungu, 2Odongo D Otieno, 1Muranda, M Agiza , 1Owuocha D
Akoth, 1AtitoR Omolo & 1Ojwang’ H Atieno.
1School of Health Sciences, Jaramogi Oginga Odinga University of Science and Technology (JOOUST) Bondo, Kenya.
P.O. Box 210-40601
2In-depth Research Services Institute of Health and Social Work Nairobi, Kenya.

Purpose: The accessibility, affordability, and consumption of abused drugs by the youths have
attracted great concern among public health personnel. This descriptive cross-sectional study
investigated household demographic predictors of drugs and substance abuse among high
school students in Kisumu East Sub-County, Kisumu County.
Methods: Sample size of 434 was calculated using Yamane formula and the participants selected
through Snow ball, random, cluster and stratified sampling. Descriptive data was summarized
using tables while inferential statistic done using Chi square and logistic regression. Data
collection done through observation and semi- structured questionnaires. P value <0.05 was
considered statistically significant.
Results: Overall, 219 (61.17%) students reported to have engaged in drug and illicit substance
abuse with higher proportion reported among the Christians as compared to Islamic and
Hinduism. Religion, Parental / guardian’s level of education and the person living with the
student were found to significantly predict the abuse of drug and illicit substance among high
school students (p<0.05).
Conclusion: Drug and illicit substance abuse remains a major public health problem among all
age groups worldwide with several negative impacts. Due to its magnitude, it calls for concerted
multi-sectorial effort by all stakeholders for the dream of drug free world to be achieved.

Keywords: Abuse; Demographic; Drug; Illicit substance; High school; Kisumu East sub-county; Predictors; Students.

INTRODUCTION

Globally, it is estimated that drug-related deaths account
for between 0.5 and 1.3 percent of all-cause mortality for
people aged 15-64 years. There are 211,000 drug-related
deaths annually, with younger people being at high risk
(WHO, 2019). According to the world drug report, 37,000
people die annually in Africa due to the consumption of
illegal drugs and substances. The Drug and Illicit
Substance Abuse problems in sub-Saharan African
countries have however been attributed to the spill-over
effect from Europe and the Americas where Africa
occupies the second position worldwide in trafficking and
consumption of illegal drugs (World Drug Report, 2014).
Currently,11.7% of young people aged 15-24 are users of
alcohol, 6.2% use tobacco, 4.7% khat while 1.5% are
users of cannabis. Regrettably, this age bracket
constitutes youths in Kenya some of whom are students in
either secondary schools or higher institutions of learning
thus depriving the country of its energetic population that
could otherwise be producing the best of the economy’s
goods and services (Chesang, 2013). According to
Kisumu county director of education, drugs and substance
abuse is on the increase from 7% in 2009 to 15% in early
2019 with the highly affected area being Kisumu East sub-
county where many students drop out of school due to
drug and substance abuse either through suspension or
health challenges associated with drugs and substance
abuse for instance mental illness and organ damage
(Kisumu County Education Director, 2019). However, it

Household Demographic Predictors of Drug and Substance Abuse among High School Students in Kisumu East Sub County, Kisumu County – Kenya
Orende et al. 188
remains unclear why students still engage in this risky
behavior. It is against this background that the this study
was undertaken to establish the household predictors of
drugs and substance abuse among secondary students
and come up with appropriate solution to the problem in
the area.
Half of the drug abusers in Kenya are aged between 10-
19 years with over 60 percent residing in urban areas and
21 percent in rural areas (NCPD, 2012). However, there
were no rural-urban gender differences in drinking among
adolescents (Gale et al., 2012 ; Chan et al., 2016 ;
Ayebare et al., 2019). Previous studies on drug abuse in
public secondary schools in Kenya revealed that mother’s
level of education determines the likelihood of students
engaging in drug and substance abuse. The study further
revealed that the percentage of students who used alcohol
increased with higher levels of mother’s education
(Chebukaka, 2014). Religious affiliation also shows a
protective effect towards drugs and substance abuse
among adolescents through strict religious teachings that
makes it morally wrong to engage in drug abuse.
Furthermore, adolescents who frequently participate in
religious activities are preoccupied with no idle time thus
deterring them from exposure to substance use (Anyanwu
et al., 2016).
In Kisumu East sub- county some of the demographic
factors influencing drug and substance abuse among high
school students included gender, mother’s educational
level, mother’s marital status at the time of individual’s birth
and number of siblings. The findings further indicated that
more males were abusing drugs and substances than their
female counterparts; students with more than nine siblings
were at a higher risk of engaging in drug and substance
abuse than those with fewer siblings while younger
siblings were introduced in to drugs by older siblings who
were already engaged in to the habit. However, there was
no significant influence of age on students’ drug and
substance abuse (Nyatuoro, 2012). Despite the strategies
put in place by the government of Kenya to prevent the
abuse, the trend still shows continuous increase of drug
abuse among school going youth in the Kenya society
(Nyassy & Kihara, 2008 ; Nyaga & Mwai, 2016). Statistical
information from Kisumu district hospital and provincial
director of medical services indicates increase in mental
and behavioral disorders due to multiple drug use and use
of other psychoactive substances; the high number of out
-patient psychiatric cases in Nyanza could be attributed to
bhang use. In 2010, cases of mental and behavioral
disorders due to multiple drug use and use of other
psychoactive substances among secondary school age
(15-24 years) admitted in the hospital were three which
increased to ten in 2011. Out patients psychiatric cases
alone were 4137 in 2011 of which 90% were from Kisumu
district hospital. The percentage increase is 233% and
continues to increase yearly in Kisumu town (Nyatuoro,
2012). However, there is very little information on the
Household Demographic Predictors of Drug and Substance Abuse among High School Students in Kisumu East Sub County, Kisumu County – Kenya
household predictors for drug and substance abuse
among high school students in the area. This therefore
justified the need for this study that sought to investigate
the household demographic predictors of drug and
substance abuse among secondary school students in
Kisumu East sub-county, Kisumu County so as to help in
awareness creation to the students, community and nation
at large on the dangers of illicit drug and substance use
aimed at achieving positive behavior change , guiding in
formulation and implementation of policies on drugs and
substance abuse among students as well as in the
designing or development of suitable prevention /
intervention strategies aimed at curbing drug and
substance abuse among students and the youth in all
learning institutions in Kenya.
Summary of Methodologies in Literature Reviewed
From the reviewed literature, various methodologies were
employed in collecting data on drug and substance abuse
among high school students based on the study area,
sample size and objectives. In the studies by Abur (2014),
Barasa (2013), King’endo (2010), (Joel 2015), Mukhandia
(2014),Minishi (2017), Nyatuoro (2012) and Njoki (2013),
both probability and non -probability sampling methods
were used with specific techniques being stratified, cluster,
purposive, systematic and random sampling. Furthermore,
descriptive research design was adopted where by data
was collected through questionnaires, key informant
interviews and focused group discussions which were
personally administered by the concerned researchers
after which data was summarized and finally presented
through graphs, pie charts, means, frequency tables and
standard deviations. The inferential statistics used were
chi square, ANOVA and correlation.
However, the methodologies adopted had some few
challenges for instance, during focused group discussion
participants were not comfortable sharing their information
on drugs and substance abuse; they therefore chose to
keep some vital information to themselves for fear of
victimization and associated stigma from fellow students
being that illicit drug ad substance use is punishable by
authority. This therefore limited the amount and quality of
data collected.
Additionally, closed ended questionnaires were used and
administered by the interviewers thus limiting the freedom
of the respondents to freely give in depth information since
some feared eye-contact with the interviewer and not
comfortable answering the questions but would freely do
so if the questionnaires were self-administered and open
ended.
Furthermore data collection process was time consuming
and tiresome being that the researchers had to personally
administer the questionnaires to each respondent rather
than letting them fill questionnaires by themselves upon

giving instructions. Finally, the study sample never
included parents or students family in the study whereby
the focus was majorly on the students and school hence
not getting data on parental and family influence on drug
and substance abuse among students which is very key in
coming up with suitable strategies to address the problem.
In order to avoid the above mentioned shortcomings, this
study therefore adopted a more improved methodology as
outlined below.
METHODOLOGY
This was a descriptive cross-sectional study design on the
household demographic predictors of drugs and
substance abuse among high school students aged
between 15-25 years and learning in secondary schools
within the study area. The study sample was 434
comprising of 394 high school students and 40 household
heads residing within Kisumu East Sub County for not less
than six months. The sample size was determined using
Yamane formula (1967) expressed mathematically thus:
n= N participants in order to prevent any perception of
1+ Ne2
Where: n =sample size required
N = target population
e =maximum acceptable margin of error/ allowable
error %
N= 26,000 e = 0.05
n= 26,000
1+26,000*0.052 = 394.
Adjustment of 10% was made to cover for non -response
bringing the sample size to be 434; 394 students and 40
household heads as key informants. Sampling was done
using snowball, random, cluster and stratified sampling
techniques; the study area was clustered into already
existing heterogeneous 5 wards and village 10 units after
which the household heads were randomly distributed
based on the total number of households per ward while
students distributed per ward based on the total number of
high schools per ward where by the higher the number of
households and high schools per ward, the higher the
proportion of household heads and students allocated. In
order to ensure fair representation of the population, a ratio
of 0.49: 0.51 was applied for male and female participants
respectively giving a based on the total population of
students in the study area as well as total per gender while
the household heads had equal share of 20 each. Snow
ball technique was applied due to the sensitivity of the
study topic to help in getting the required number of
students per ward as well as those students who may be
involved in drug and substance abuse where by the
identified students helped in identifying other high school
students in the area by giving direction on the next
household with high school students being that they know
Household Demographic Predictors of Drug and Substance Abuse among High School Students in Kisumu East Sub County, Kisumu County – Kenya
Int. J. Public Health Epidemiol Res. 189
each other though this was guided by the proportion
allocated per ward.
Reliability and validity of the research instruments was
ensured by pretesting and retesting of the tools in a
different area to determine their correctness, consistency
and ability of meeting the intended purpose. Research
assistants were equally trained on quality data collection
while the respondents taken through the study purpose,
rights and obligation.
Authority to conduct this study was obtained from the
Board of Postgraduate Studies, JOOUST thereafter ethical
approval and clearance from the JOOTRH Ethics Review
Committee and NACOSTI. Permission to conduct the
study in the area was sought from the area local
administration. Prior to enrolment, the purpose of the study
as well as the implications was explained to all prospective
study participants and written informed consent as well as
assent obtained from all willing to take part in the study
before the study commenced. The respondents were
assured of confidentiality of both their information and
identities. Due diligence was accorded the entire data
collection process given the sensitive nature of the study
stigmatization of the participants. Data collection was done
by trained research assistants using semi- structured
questionnaires and observation. The questionnaires were
designed in English with native language translation;
respondents with difficulty filling the questionnaire were
assisted b y t h e researcher and the trained research
assistants in language best understood by them.
Observation was used to collect data on non –verbal
communication displayed through body language such as
facial expression, behavior and assessment of living
condition. The descriptive data was organized and
summarized using tables while inferential statistics using
Chi square and logistic regression. Data from observation
were analyzed through content and narrative analysis. P
value <0.05 was considered statistically significant.
RESULTS
The study achieved 91% response rate of the sample size,
398. Out of the 398 sampled respondents, 358 were
students and 40 were household heads. To provide a bit
more information about the abuse of drugs and illicit
substance among the high school students, the 40
household heads were interviewed as key informants.
Demographic Characteristics of Respondents
A total of 358 students were enrolled in the study, whose
characteristics are presented in Table 1. The majority of
the study participants 217 (60.61%) were aged between
15-17 years. More than half 184 (51.40%) of the
participants were female. The highest number 274
Orende et al. 190

(76.54%) of the respondents were Christians. One were from mixed day secondary schools with only three
hundred and fifty (41.90%) of the students were residents (0.84%) from mixed boarding schools (Table 1). Overall,
of Kolwa Central with a small number 11 (3.07%) came the highest number of 138 (38.55%) of the participants
from Manyatta B. Most of the participants, 229 (63.97%) were students of form two class.

Table 1: Demographic characteristics of students, Kisumu East Sub-County (n=358)

Demographic Characteristics n (%)
Age category
15-17 years 217 (60.61)
18-24 years 135 (37.71)
>=25 years 6 (1.68)
Gender
Male 174 (48.60)
Female 184 (51.40)
Religion
Christianity 274 (76.54)
Islamic 72 (20.11)
Atheist 1 (0.28)
Hinduism 6 (1.68)
Others 5 (1.40)
Ward
Nyalenda A 57 (15.92)
Kajulu 75 (20.95)
Kolwa East 65 (18.16)
Kolwa Central 150 (41.90)
Manyatta B 11 (3.07)
School Category
Boys Day 49 (13.69)
Boys Boarding 15 (4.19)
Mixed Day 229 (63.97)
Girls Day 42 (11.73)
Girls Boarding 20 (5.59)
Mixed Boarding 3 (0.84)
Form
One 69 (19.27)
Two 138 (38.55)
Three 91 (25.42)
Four 60 (16.76)

Demographic characteristics of household heads
Kisumu East Sub-County
To assess strategies used by parents to prevent abuse of
drugs among the students, this study used data from the
key informants. The data for the individual characteristics
of the key informants are presented in the Table 2. Among
the selected key informants, the majority were above 40
years 24 (60%) and most of them 23 (57.50%) were
female. In terms of religion, 29 (72.5%) were Christians
and only 11 (27.5%) were Islamic. The least number of
these respondents were residents of Kolwa East 6 (15%).
Close to three quarters of the household heads were
married 25 (62.5%) and higher proportion of them had a
university education 14 (35%)
(Table 2)

Table 2: Demographic characteristics of household heads Kisumu East Sub-County (n=40)

Demographic Characteristics n (%)
Age category
26-35 years 6 (15.00)
36-40 years 10 (25.00)
>40 years 24 (60.00)
Gender

Household Demographic Predictors of Drug and Substance Abuse among High School Students in Kisumu East Sub County, Kisumu County – Kenya
 Male 17 (42.50)
Female 23 (57.50)
Religion
Christianity 29 (72.50)
Islamic 11 (27.50)
Ward
Nyalenda A 8 (20.00)
Kajulu 11 (27.50)
Kolwa East 6 (15.00)
Kolwa Central 8 (20.00)
Manyatta B 7 (17.50)
Marital Status
Married 25 (62.50)
Single 5 (12.50)
Divorced 3 (7.50)
Widowed 7 (17.50)
Education level
Primary 7 (17.50)
Secondary 7 (17.50)
College 12 (30.00)
University 14 (35.00)

Demographic factors associated with drug and
substance abuse
Drug and substance abuse is associated with various
factors among high students. This study investigated
demographic factors that influence high school students
who abused drugs and substance. Table 3 lists the
distribution of drug and substance abuse against
demographic characteristics of high school students.
Overall, 219 (61.17%) students reported to have engaged
in drug and illicit substance abuse. Higher proportion of
students who abused drugs and illicit substance was
reported among the Christians as compared to Islamic and
Hinduism (78.54% vs 19.63.0%, 78.54% vs 1.83.0%,
p<0.001) respectively. Father’s education, maternal
education, guardian’s level of education and the person
living with the student were other important factors that
significantly determined the abuse of drug and illicit
substance (p<0.001). Factors such age, gender, ward and
the type of family did not report significant association with
abuse of drugs and illicit substance (Table 3).

Table 3: Demographic factors associated with drug and substance abuse among high school students
Demographic Characteristics All N=358 Drug and Substance use p-Value
n (%) Yes (n=219) No (n=139)
n (%) n (%)
Age category
15-17years 217 (60.61) 135 (61.64) 82 (58.99) 0.77
18-24 years 135 (37.71) 81 (36.99) 54 (38.85)
>=25 years 6 (1.68) 3 (1.37) 3 (2.16)
Gender
Male 174 (48.60) 113 (51.60) 61 (43.88) 0.16
Female 184 (51.40) 106 (48.40) 78 (56.12)
Religion
Christianity 274 (76.54) 172 (78.54) 102 (73.38) 0.03
Islamic 72 (20.11) 43 (19.63) 29 (20.86)
Atheist 1 (0.28) 0 1 (0.72)
Hinduism 6 (1.68) 4 (1.83) 2 (1.44)
Others 5 (1.40) 0 5 (3.60)
Ward
Nyalenda A 57 (15.92) 21 ()15.11 36 (16.44) 0.14
Kajulu 75 (20.95) 24 (17.27) 51 (23.29)
Kolwa East 65 (18.16) 33 (23.74) 32 (14.61)
Kolwa Central 150 (41.90) 55 (39.57) 95 (43.38)
Manyatta B 11 (3.07) 6 (4.32) 5 (2.28)
Father's Education

Household Demographic Predictors of Drug and Substance Abuse among High School Students in Kisumu East Sub County, Kisumu County – Kenya
 Primary 36 (10.06) 28 (12.79) 8 (5.76) 0.02
Secondary 35 (9.78) 14 (6.39) 21 (15.11)
College 105 (29.33) 63 (28.77) 42 (30.22)
University 71 (19.83) 47 (21.46) 24 (17.27)
None 111 (31.01) 67 (30.59) 44 (31.65)
Mother's Education
Primary 67 (18.72) 61 (27.85) 6 (4.32) <0.001
Secondary 76 (21.23) 50 (22.83) 26 (18.71)
College 88 (24.58) 38 (17.35) 50 (35.97)
University 27 (7.54) 21 (9.59) 6 (4.32)
None 100 (27.93) 49 (22.37) 51 (36.69)
Guardian's Education
Primary 18 (5.03) 0 18 (12.95) <0.001
Secondary 20 (5.59) 17 (7.76) 3 (2.16)
College 5 (1.40) 5 (2.28) 0
University 18 (5.03) 5 (2.28) 13 (9.35)
None 297 (82.96) 192 (87.67) 105 (75.54)
Family type
Dual/Both parents 204 (56.98) 126 (57.53) 78 (56.12) 0.91
Single parent 111 (31.01) 68 (31.05) 43 (30.94)
Others 43 (12.01) 25 (11.42) 18 (12.95)
Person living with
Both parents 210 (58.66) 133 (60.73) 77 (55.40) 0.02
Father 34 (9.50) 13 (5.94) 21 (15.11)
Mother 71 (19.83) 49 (22.37) 22 (15.83)
Others 43 (12.01) 24 (10.96) 19 (13.67)

Demographic predictors of drug and substance
abuse among high school students
To better comprehend the demographic predictors of drug
and illicit substance abuse, this study analyzed the
independent factors using two models, bivariate and
multivariate logistic models. Table 4 presents the results
from the bivariate and multivariate logistic regressions.
From the bivariate analyses, students aged between 18-
24 years (OR=0.91; 95% CI 0.59, 1.42) and those aged 25
years and above (OR=0.61; 95% CI 0.12, 3.08) had lower
odds of abusing drugs and illicit substance as compared to
those aged between 15-17 years. This study also reveals
that female students (OR=0.73; 95% CI 0.48, 1.12) were
less likely to abuse drugs and illicit substance as compared
to male counterparts. Similarly, Islamic students
(OR=0.88; 95% CI 0.52, 1.50) had reduced odds of
abusing drugs and illicit substance as compared to
Christians; however, students from Hinduism group
(OR=1.19; 95% CI 0.21, 6.59) were more likely to abuse
drug and illicit substance as compared to Christians. With
regard to place of residence, students from Kajulu
(OR=1.24; 95% CI 0.60, 2.56) and Kolwa Central
(OR=1.01; 95% CI 0.54, 1.90) had increased odds of
abusing drugs and illicit drugs as compared to those from
Nyalenda A while those from Kolwa East (OR=0.57; 95%
CI 0.27, 1.17) and Manyatta B (OR=0.49; 95% CI 0.13,
1.79) were less likely to abuse drugs and illicit substance.
Paternal education was associated with reduced odds of
abusing drugs and illicit substance, such that students
whose fathers never attended school (OR=0.44; 95% CI
0.18, 1.04), who had secondary (OR=0.19; 95% CI 0.07,
1.79), college (OR=0.43; 95% CI 0.18, 1.03) or university
degree (OR=0.56; 95% CI 0.22, 1.41) were less likely to
abuse drugs and illicit substance compared to those with
primary education. Likewise, students whose mothers
never attended school (OR=0.09; 95% CI 0.04, 0.24), who
had secondary (OR=0.19; 95% CI 0.07, 0.50), college
(OR=0.07; 95% CI 0.03, 0.19) or university degree
(OR=0.34; 95% CI 0.10, 1.18) were less likely to abuse
drugs and illicit substance. In the similar manner, students
living with guardians who never attended school
(OR=0.32; 95% CI 0.09, 1.13) or had a university degree
(OR=0.07; 95% CI 0.01, 0.34) were less likely to abuse
drugs and illicit substance as compared to those whose
guardians had secondary education. This study confirms
reduced odds of receiving optimal doses for those
students who came from single parents (OR=0.98; 95% CI
0.61, 1.57) and other types of family (OR=0.86; 95% CI
0.44, 1.68). Further, with respect to the person living with
the student, this study reports higher odds of abusing
drugs and illicit substance among students living with their
mothers (OR=1.29; 95% CI 0.72, 2.29) and those living
with other members (OR=1.04; 95% CI 0.56, 1.94) as
compared to students living with both parents. In contrast,
the likelihood of abusing drugs and illicit substance was
lower among students living with their fathers (OR=0.36;
95% CI 0.17, 0.76) (Table 4.3).
The person living with the student was significantly
associated with abuse of drugs and illicit substance after
adjusting for other covariates, such that students living
with their mothers (ₐOR=4.57; 95% CI 1.17, 17.94, p<0.05)
were five times more likely to abuse drugs and illicit

Household Demographic Predictors of Drug and Substance Abuse among High School Students in Kisumu East Sub County, Kisumu County – Kenya
 Int. J. Public Health Epidemiol Res. 193

substance as compared to those living with both parents.
Students living with other members (ₐOR=3.64; 95% CI
0.61, 21.72) were also found to have increased odds of
abusing drugs and illicit substances though the effect was
not significant. Students living with their fathers
(ₐOR=0.50; 95% CI 0.14, 1.77) remained less likely to
abuse drugs and illicit substance after adjustment.
Maternal education remained associated with reduced
odds but with significant effects at the multivariable level;
such that students whose mothers never attended school
(ₐOR=0.08; 95% CI 0.02, 0.29), who had secondary
(ₐOR=0.13; 95% CI 0.04, 0.42), college (ₐOR=0.04; 95%
CI 0.01, 0.14) or university degree (ₐOR=0.17; 95% CI
0.04, 0.76) were less likely to abuse drugs and illicit
substance. In relation to father’s education, students
whose fathers had a college education (ₐOR=2.02; 95% CI
0.56, 7.29) or university degree (ₐOR=3.32; 95% CI 0.82,
13.42), were more likely to abuse drugs than whose
fathers had primary education. Similarly to the bivariate
analyses, students whose fathers did not attend school
(ₐOR=0.50; 95% CI 0.08, 3.11) or had secondary
education (ₐOR=0.70; 95% CI 0.17, 2.92) remained less
likely to abuse drugs and illicit substance after adjusting for
other covariates.

Table 4: Demographic predictors of drug and substance abuse among high school students
Demographic Predictors Drug and Substance use Crude Odds Ratio Adjusted Odds Ratio p-Value
Yes (n=219) No (n=139) (95% CI) (95% CI)
n (%) n (%)
Age category
15-17years 135 (61.64) 82 (58.99) Ref
18-24 years 81 (36.99) 54 (38.85) 0.91 (0.59,1.42
>=25 years 3 (1.37) 3 (2.16) 0.61 (0.12,3.08)
Gender
Male 113 (51.60) 61 (43.88) Ref
Female 106 (48.40) 78 (56.12) 0.73 (0.48,1.12)
Religion
Christianity 172 (78.54) 102 (73.38) Ref
Islamic 43 (19.63) 29 (20.86) 0.88 (0.52,1.50)
Atheist 0 1 (0.72) 0
Hinduism 4 (1.83) 2 (1.44) 1.19 (0.21,6.59)
Others 0 5 (3.60) 0
Ward
Nyalenda A 21 ()15.11 36 (16.44) Ref
Kajulu 24 (17.27) 51 (23.29) 1.24 (0.60,2.56)
Kolwa East 33 (23.74) 32 (14.61) 0.57 (0.27,1.17)
Kolwa Central 55 (39.57) 95 (43.38) 1.01 (0.54,1.90)
Manyatta B 6 (4.32) 5 (2.28) 0.49 (0.13,1.79)
Father's Education
Primary 28 (12.79) 8 (5.76) Ref
Secondary 14 (6.39) 21 (15.11) 0.19 (0.07,1.79) 0.70 (0.17,2.92) 0.63
College 63 (28.77) 42 (30.22) 0.43 (0.18,1.03) 2.02 (0.56,7.29) 0.28
University 47 (21.46) 24 (17.27) 0.56 (0.22,1.41) 3.32 (0.82,13.42) 0.09
None 67 (30.59) 44 (31.65) 0.44 (0.18,1.04) 0.50 (0.08,3.11) 0.45
Mother's Education
Primary 61 (27.85) 6 (4.32) Ref
Secondary 50 (22.83) 26 (18.71) 0.19 (0.07,0.50) 0.13 (0.04,0.42) <0.001
College 38 (17.35) 50 (35.97) 0.07 (0.03,0.19) 0.04 (0.01,0.14) <0.001
University 21 (9.59) 6 (4.32) 0.34 (0.10,1.18) 0.17 (0.04,0.76) 0.02
None 49 (22.37) 51 (36.69) 0.09 (0.04,0.24) 0.08 (0.02,0.29) <0.001
Guardian's Education
Primary 0 18 (12.95) 0
Secondary 17 (7.76) 3 (2.16) Ref
College 5 (2.28) 0 0
University 5 (2.28) 13 (9.35) 0.07 (0.01,0.34)
None 192 (87.67) 105 (75.54) 0.32 (0.09,1.13)
Family type
Dual/Both parents 126 (57.53) 78 (56.12) Ref
Single parent 68 (31.05) 43 (30.94) 0.98 (0.61,1.57)

Household Demographic Predictors of Drug and Substance Abuse among High School Students in Kisumu East Sub County, Kisumu County – Kenya
 Others 25 (11.42) 18 (12.95)
Person living with
Both parents 133 (60.73) 77 (55.40)
Father 13 (5.94) 21 (15.11)
Mother 49 (22.37) 22 (15.83)
Others 24 (10.96) 19 (13.67)
CI- Confidence Interval
Ref- Reference/ baseline category
DISCUSSION
More cases of drug and substance abuse were reported
among Christians compared to other religious faith.
However, it was noted that students from Hinduism were
1.19 times more likely to engage in drug and substance
abuse compared to other religious faith (OR=1.19; 95% CI
0.21, 6.59). This is because most Christian denominations
permit moderate use of socially and legally acceptable
drugs such as wine, alcohol, tobacco and caffeine; this
moderate consumption gradually grows in to addiction.
Moreover, wine is served during Holy Communion
symbolizing blood of Jesus so there is nothing wrong with
consuming wine after all what one consumes is not bad
but only what one does after the consumption that counts
while among the Hindu faith, the Hindu text of ethical rules
Manusmriti allows use of alcohol and drugs to remain
immortal. This is in agreement with finding according to
Anyanwu et al., (2016) religious affiliation shows a
protective effect towards drugs and substance abuse
among adolescents through strict religious teachings that
makes it morally wrong to engage in drug abuse.
Furthermore, adolescents who frequently participate in
religious activities are preoccupied with no idle time thus
deterring them from exposure to substance use.
Students living with their mothers (ₐOR=4.57; 95% CI 1.17,
17.94) were five times more likely to abuse drugs and
substances as compared to those living with their fathers
(ₐOR=0.50; 95% CI 0.14, 1.77). The finding was attributed
to the parenting and nurturing styles; most fathers are strict
and dictatorial in nature and as such would instill severe
form of discipline on the students should they try to engage
in drug and substance abuse; the fear of the punishment
therefore acts as deterrent towards the vice hence the less
likelihood of abuse. In the contrary, most mothers are
lenient and compromising in nature on matters of instilling
discipline on the students especially during their
adolescence/puberty stage due to their rebellious nature
against parental rules to an extent that they are even
willing to physically fight their mothers because they feel
there is nothing serious the mothers can do in terms of
discipline to deter them from the habit. As a result of this,
the mothers tend to withdraw as a way of maintaining
peace and only watch helplessly as the students sink in to
drug addiction. Similarly, those living with the guardians
are equally more likely to engage in to drug and substance
abuse may be because in most cases these students are
total orphans under the guardianship of elder siblings, old
grand- parents or well- wishers who may not have full
control over the students’ behavior for the fear of being
termed “bad person”. They therefore tend to give much
Household Demographic Predictors of Drug and Substance Abuse among High School Students in Kisumu East Sub County, Kisumu County – Kenya
0.86 (0.44,1.68)
Ref
0.36 (0.17,0.76) 0.50 (0.14,1.77) 0.29
1.29 (0.72,2.29) 4.57 (1.17,17.94) 0.03
1.04 (0.56,1.94) 3.64 (0.61,21.72) 0.16
unsupervised freedom all in the name of being “a nice
person.” Unfortunately, these students in turn abuse the
freedom by engaging in drug abuse and other vices being
that they feel their parents had passed on and therefore
nobody should force them to abide by the set rules; to them
any form of strictness and discipline implies nothing but
mistreatment just because their parents are no more. This
is in agreement with study findings by Mukhandia, (2014)
that showed that parenting and nurturing styles by parents
or care givers have fundamental impact on the adolescent
substance abuse especially when they begin to have more
autonomy and opportunities for either prosocial or risky
behaviors since it affects overall child development and
outcomes this is consistent with study findings by
(Mohasoa & Fourie, 2012 ;Kiambi, 2018 &NACADA, 2019)
revealing major role of parenting styles in adolescent’s
likelihood to abuse drugs where by permissive parenting
style contributes to substance abuse because no one
corrects or reprimands such behavior or even guides them
while authoritative parents who do not allow their children
time to relax may frustrate them making them to take
refuge in drugs.
Consistent with previous studies by Chebukaka, (2014
)&Shahraki et al., (2019), parental level of education had
significant influence on drug and substance abuse among
students; higher odds of abuse were realized among
students whose fathers had higher educational level
(ₐOR=3.32; 95% CI 0.82, 13.42) compared to students
whose fathers did not attend school (ₐOR=0.50; 95% CI
0.08, 3.11). However, maternal education showed low
odds for all the categories of educational level. Level of
education has both positive and negative influence on the
student as far as abuse of drugs and other substances is
concerned. The more knowledgeable parents/guardians
are on drug and substance abuse, the more likely they are
able to impart the same knowledge to their children on the
negative effects of drugs and substance abuse deterring
them from engaging in the habit hence lesser odds of
abuse. On the other hand, highly educated
parents/guardians usually have the desire to see their
children follow their footsteps by attaining similar or even
much higher academic qualifications. This therefore drives
them in to exerting a lot of pressure on the students to
achieve academic excellence regardless of their
intellectual capacity. In the long run these children resort
in to drug and substance abuse as a way of managing the
undue influence and pressure from the parents/guardians.
Moreover, such parents/guardians have better access to

well- paying white collar jobs boosting their financial ability
to provide their children with everything they need
including unnecessary amount of pocket money to
entertain themselves and friends. The availability of cash
enable the students to easily access whatever type of
drugs and substance they may wish for without any
struggle. This is in agreement with findings by (Njoki,
2013;Simatwa et al., 2014).
Additionally, white collar jobs are always very demanding
in nature in terms of service hours forcing
parents/guardians to leave home very early and return
back late in the night in order to meet job targets; as a
result they have very minimal contact with their children to
bond, supervise their movement and timely attend to their
various needs whether psychological, social and
academic. When these needs are not timely attended to,
the students may seek help from other people such as
house servants and peers who may misguide and lure
them to take refuge in drugs and other substances as a
way of managing their challenges especially if such people
happen to be drug addicts. This is in agreement with
findings by Gale et al., (2012 );Mukhandia,
(2014);NACADA, (2012) ;Simatwa et al., (2014) revealing
that busy work schedules and long working hours of family
members leads to limited family time for parents to bond
and supervise their children.
Although age had no significant association with drugs and
substance abuse, the majority of the students (61.64%)
aged 15-17 reported higher odds of abuse in the past four
months compared to other age groups. This is contrary to
findings by Nyatuoro, (2012) showing no significant
influence of student’s age on drug and substance abuse.
This increased level of abuse may be linked to the
developmental changes characterized by more
independence from parents, a lot of exploration, struggle
for acceptance and belonging by peers as well as
increased desire for drugs and substance abuse. They
prefer to spend more time with peers than parents. Failure
to attain sense of belonging and acceptance may lead to
feeling of sadness and depression which may push the
student in to abuse of drugs and substances as coping
mechanism. The more controlling the parent is, the more
rebellious the teen is likely to become with more temptation
to engage in drug and substance abuse.
Female students (48.4%) were slightly less likely to abuse
drugs and substance as compared to male counterparts
51.6 %. This concurs with the findings of Nyatuoro, (2012)
where more males were abusing drugs and substances
than their female counterparts. This may be linked to the
rigorous empowerment of the girl child and fight for
equality between males and females that make the
females strive to compete with the males in almost
everything including drug and substance abuse as the
phrase goes “what a man can do, a woman can do even
much better”. For a very long time males have been
Household Demographic Predictors of Drug and Substance Abuse among High School Students in Kisumu East Sub County, Kisumu County – Kenya
Int. J. Public Health Epidemiol Res. 195
associated with a lot of delinquency including drug and
substance abuse while the females viewed as innocent
and law abiding with less chances of engaging in drug
abuse. Unfortunately, this wrong assumption prevents
early detection of the vice among the females with the
reality hitting the parents/guardians hard when it is already
late. Most cases abuse was witnessed among students
from mixed secondary schools in either form one which
was attributed to the transition to secondary school
characterized by feeling of being a young adult free from
too much parental supervision as well as the influence
from the male students who lure them in to drugs and
substance abuse as a way of belonging to the various
groups in school while those who refuse are mistreated.
Even though family type showed no statistical significance,
this study confirmed higher percentage of drugs and
substance abuse among students from dual/ both parent
family 57.53% compared to single parent 43% .This is may
be because in dual family there are two levels of directives
which may be confusing to the children especially if they
are conflicting. Sometimes parents may differ on how the
children should be raised and what rules should govern the
conduct and operations of the family; one parent may be
so strict while the other so compromising giving a loophole
in the child upbringing. The children may maximize on the
weakness of either of the parents to engage in to drugs
and other substances abuse since they know that the
parents are not speaking the same language. Also if either
of the parent is a drug addict, chances of influencing the
children in to the habit is higher compared to single parent
family where the child will be raised according to the rules
and regulations of that existing parent without influence or
compromise from other parent. This is contrary to the
findings by Brown, (2013) that showed youth of single
parent to be at high risk for drug abuse as compared to
those with dual parents.
Consistent with findings by Kasundu et al.,
(2012)&Simatwa et al., (2014) which reported higher odds
of weekly drinking among adolescents in the rural areas
compared to those living in major cities, students from rural
set up (Kajulu) had increased odds of abusing drugs and
substances as compared to those from urban areas
(OR=1.24; 95% CI 0.60, 2.56). This was due to the fact
that the drugs and substances are more readily available
in the area for instance, some homes are brewing dens for
alcoholic drinks as well as growing of bhang as source of
income with students playing the sale agent roles thus
predisposing them to the vice compared to their
counterparts in the urban settings this is in agreement with
findings by Stafström & Agardh, (2013 )&Abbo et al.,
(2016) the culture of brewing alcohol in homes expose the
youth to alcohol consumption at an early age. This is
because most of the cultural practices and ceremonies
such as initiation, marriage ceremonies usually take place
in the rural set ups; during such events, alcoholic drinks
are served to all participants regardless of the age.
Orende et al. 196
Moreover, the ceremonies have unchecked drinking that
goes unchallenged while some parents indulge in drug use
in the presence of their children. Moreover, level of
knowledge/awareness on the dangers of drugs and
substance abuse is usually lower in rural set up compared
to the urban areas since most health promoting activities
are concentrated in the urban settings. Those engaging in
to drugs and substance abuse as well as trade are able to
buy their freedom so cheaply from law enforcers through
bribes compared to urban centers where whistle blowers
and serious institutions fighting the vice are found.
Family history of drug abuse was found to be a contributing
factor in the drug and substance abuse among students,
this is similar to the study findings by Ayebare et al., (2019)
;Bramness, (2014);Chebukaka, (2014); King’endo,
(2010)&Njoki, (2013) where family history of alcohol use
disorders was considered a major vulnerability factor for
both genetic and environmental reasons; biological cause
like family history of illicit drug use may genetically
predispose a person to drug abuse. For instance, a child
born of a drug addict is more likely to have the same habit
being transferred to him or her thus increasing the
tendency of the student to use or abuse drugs frequently.
CONCLUSION
Drug and substance abuse remains a major public health
problem among all age groups worldwide with a
substantial impact on human health and general well-
being ranging from medical, social, economic, educational
and political consequences. If left unaddressed, the vice
will go a long way in frustrating the attainment of the
Sustainable development Goals as well as the vision 2030
which aims at making Kenya an industrial and mid-level
income country. Due to its magnitude, there is a need for
a concerted effort by all stakeholders for the dream of a
drug-free nation to be achieved.
RECOMMENDATIONS
There is need for parents / guardians to balance between
work and family time in order to build strong attachment
with the children and timely address their concerns. This
will ensure sufficient monitoring of their movements and
act promptly should there be anything unusual in their
behavior.
Household Demographic Predictors of Drug and Substance Abuse among High School Students in Kisumu East Sub County, Kisumu County – Kenya
Parents/guardians should love and appreciate their
children just as they are and help them identify and build
their potentials without exerting unnecessary pressure on
them to score highly in school regardless of their
intellectual capability. This will enhance level of self -
awareness and actualization in their children
Ethical Considerations:
Conflict of Interest: The authors declare no significant
competing financial, professional or personal interests that
might have influenced the performance or presentation of
the work described in this manuscript.
Funding: The Authors did not receive any grant for the
study. The expenses for the study were funded by the
corresponding author Mrs. Marceline Awino Orende.
Ethical Approval: Ethical approval was obtained from
Jaramogi Oginga Odinga Teaching and Referral Hospital
ERC and a study permit from National Commission for
Science, Technology and innovation (NACOSTI).
Informed Consent: A written Informed consent and
assent forms for participation and publish were obtained
from respondents and the community members before
starting the study.
ACKNOWLEDGEMENT
Sincere gratitude to God for provision and good health
throughout the manuscript development period, Board of
Postgraduate Studies, JOOUST, the JOOTRH Ethics
Review Committee and NACOSTI for review and approval
of my proposal for this study, the publication committee for
acceptance to publish this manuscript, my Supervisors
and peers at the University extra-mural studies for the
support and guidance towards the completion of this
manuscript, the study area administration and
respondents for the support and permission to conduct the
study in the area and finally my beloved family for their
love, prayers and encouragement throughout this
manuscript development. God bless you all!

REFERENCES
Anyanwu, Ibekwe, & Ojinnaka. (2016). Pattern of
substance abuse among adolescent secondary
school students in Abakaliki. Cogent Medicine,
3(1), 1272160.
Ayebare, Rukundo, Mulogo, Ntaro, & Imelda. (2019).
Prediction of Risky Alcohol Use among Ugandan
Youth: A Case Study of at Marie Stopes and
Youth Reproductive Health Clinics. J Alcohol
Drug Depend 7: 322. doi: 10.4172/2329-
6488.1000322 Page 2 of 4 J Alcohol Drug
Depend, an open access journal ISSN: 2329-
6488 Volume 7• Issue 1• 1000322. youth
reported alcohol use to at least one standard
drink per sitting on average (as defined by
AUDIT) in the, 12, 3.
Bramness. (2014). [Intoxication, drugs, substance abuse
and harmful use]. Tidsskr Nor Laegeforen,
134(20), 1968. doi: 10.4045/tidsskr.14.0905
Brown. (2013). Familial, social, and individual factors
contributing to risk for adolescent substance use.
Journal of addiction, 2013.
Chan, Leung, Quinn, Kelly, Connor, Weier, & Hall.
(2016). Rural and urban differences in adolescent
alcohol use, alcohol supply, and parental
drinking. The Journal of Rural Health, 32(3), 280-
286.
Chebukaka. (2014). Drug abuse among students in
public secondary schools in Kenya, the case of
Vihiga County. International Journal of Social
Sciences and Education, 4(3).
Cheloti. (2013). Effectiveness of the
HeadteachersStrategies Used to Curb drug and
Substance Abuse in Public Schools in Nairobi
County: Kenya.. Unpublished M. Ed. Thesis,
University of Nairobi.
Chesang. (2013). Drug abuse among the youth in Kenya.
International journal of scientific & technology
research, 2(6), 126-131.
Cronbach. (1951). Coefficient alpha and the internal
structure of tests. psychometrika, 16(3), 297-334.
Gale, John, Lenardson, Jennifer, Lambert PhD, & Hartley
PhD. (2012). Adolescent alcohol use: Do risk and
protective factors explain rural-urban differences?
Griffin, Lowe, Botvin, & Acevedo. (2019). Patterns of
adolescent tobacco and alcohol use as predictors
of illicit and prescription drug abuse in minority
Household Demographic Predictors of Drug and Substance Abuse among High School Students in Kisumu East Sub County, Kisumu County – Kenya
Int. J. Public Health Epidemiol Res. 197
young adults. J Prev Interv Community, 47(3),
228-242. doi: 10.1080/10852352.2019.1603672
Kasundu, Mutiso, Sutter, & Mwirigi. (2012). Factors
contributing to drug abuse among the youth in
Kenya: A Case of Bamburi Location.
Kiambi. (2018). FACTORS INFLUENCING DRUGS AND
SUBSTANCE ABUSE AMONG PUBLIC
SECONDARY SCHOOL STUDENTS IN KIAMBU
COUNTY, KENYA. International Journal of
Psychology, 3(1), 1-23.
King’endo. (2010). Incidence and extent of substance
abuse among secondary school students in
Nairobi Province, Kenya: implications for
specialized intervention. Kenyatta University.
Kisumu County Education Director. (2019). Drug and
substance abuse in secondary schools in kisumu
east sub-county
Loke, & Mak. (2013). Family process and peer influences
on substance use by adolescents. International
journal of environmental research and public
health, 10(9), 3868-3885.
McMorris, Catalano, Kim, Toumbourou, & Hemphill.
(2011). Influence of family factors and supervised
alcohol use on adolescent alcohol use and harms:
similarities between youth in different alcohol
policy contexts. Journal of studies on alcohol and
drugs, 72(3), 418-428.
Mohasoa, & Fourie. (2012). Substance abuse amongst
adolescents: a case study of Zeerust, North West
Province, South Africa. New voices in Psychology,
8(1), 30-43.
Mugenda. (2003). Research methods. Quantitative and
qualitativeapproaches.Nairobi. Acts Press.
Mugenda. (2013). O and Mugenda (1999) Research
Method-Quantitative and Qualitative approaches
Nairobi: Act Press.
Mukhandia. (2014). Determinants of drug abuse among
secondary school students in Kenya.A case of
Kabuchai Sub-County.
NACADA. (2012. ). Rapid situation assessment of the
status of drug and substance abuse in Kenya
NACADA. (2019). national survey on alcohol and drug
abuse among secondary school students in
Kenya
NCPD. (2012).
Orende et al. 198

Njoki. (2013). Drug and substance abuse in secondary *Corresponding Author: Orende M Awino,, Jaramogi
schools in Kenya. Acase study of Kiambu Oginga Odinga University of Science and Technology,
County., UNIVERSITY OF NAIROBI. Bondo, Kenya.
Email: awinomerceline49@gmail.com
1
Nyaga, & Mwai. (2016). Contributions of Selected Family Co-Author Email: donguru@jooust.co.ke.
2
Factors to Drug Abuse Among Adolescents in Co-Author Email: dvdotieno@gmail.com
1
Secondary Schools in Manyatta Sub-county, Co-Author Email: mayaagiza@live.com.
1
Embu County, Kenya. Co-Author Email: doriceowuocha@gmail.com.
1
Co-Author Email: atitoraphael@gmail.com.
1
Nyassy, & Kihara. (2008). Kenya risks losing an entire Co-Author Email: ojwanghellen@yahoo.com.
generation to drugs. Sunday Nation.

Nyatuoro. (2012). Environmental and demographic

factors influencing drug and substance abuse
among secondary school students in Kisumu
Town East, Kenya: Doctoral dissertation.
Kenyatta University. Kenya.

Shahraki, Sedaghat, & Fararouei. (2019). Family and

social predictors of substance use disorder in
Iran: a case-control study. Subst Abuse Treat
Prev Policy, 14(1), 17. doi: 10.1186/s13011-019-
0201-x
Accepted 30 November 2020
Simatwa, Odhong, Juma, & Choka. (2014). Substance
abuse among public secondary school students: Citation: Orende M A., Onguru D O., Odongo D Otieno.,
Prevalence, strategies and challenges for public Muranda, M A., Owuocha D A., Atito R O. & Ojwang’ H
secondary school managers in Kenya: A case A. (2020). Household Demographic Predictors of Drug and
study of Kisumu East Sub County. 5(8), 315-330. Substance Abuse among High School Students in Kisumu
East Sub County, Kisumu County – Kenya. International
Stafström, & Agardh. (2013). Socio-economic Journal of Public Health and Epidemiology Research, 7(1):
determinants for alcohol consumption and heavy 187-198.
episodic drinking in a Ugandan student
population. The International Journal of Alcohol
and Drug Research, 1(1), 57-67.

WHO. (2011). Global status report on alcohol use Copyright: © 2021 Orende et al. This is an open-access
article distributed under the terms of the Creative
worldwide.
Commons Attribution License, which permits unrestricted
World Drug Report. (2014). use, distribution, and reproduction in any medium,
provided the original author and source are cited.

Household Demographic Predictors of Drug and Substance Abuse among High School Students in Kisumu East Sub County, Kisumu County – Kenya