Basic HTML Version
Journal of Mosquito Research, 2013, Vol.3, No.6, 45
-
53
ISSN 1927-646X
http://jmr.sophiapublisher.com
47
1 Methodology
1.1 Study design
The study was a Community based comparative
Interventional study.
1.2 Study population
Children aged less than five years in two wards in
Kuje Area Council namely Kuje central and
Rubochi were recruited into the study, as intervention
(experimental or study) and control groups
respectively. Kuje central is the headquarters of the
Area Council while Rubochi is over two hours drive
away from Kuje.
1.3 Inclusion/Exclusion criteria
All children less than five years were eligible for the
study. Those recruited were followed up for five
months.
1.4 Sample size determination
The minimum sample size was determined using the
formula for comparative designs below (Kirkwood
and Sterne, 2003).
N = (Zα + Zβ)
2
2p(1-p)
d
2
Where N= minimum sample size for each group
Zα=percentage point of normal distribution
corresponding to the (two sided) significance level. In
this case significance level is 5%, so Zα=1.96
Zβ = power of the test, which is conventionally 80%;
Zβ = 0.84
p= mean proportion of under- fives with malaria in the
control group (55%) (Nsimba et al., 2002; Arnaud et
al., 2005) and intervention group (28%) (From
previous studies, use of ITN’s will reduce the
incidence of malaria by an average of 50%) (Federal
Ministry of Health, 2005a; http://www.rollback-
malaria.org)
p =55 + 28 /2 = 42 or 0.42
1-p= the variance of the proportions = 1 - 0.42 = 0.58
d= the minimum difference to be detected by the study
(20%) = 0.2
Substituting in the formula therefore:
N= (1.96 + 0.84)
2
×2×0.42×0.58=3.819648= 96
0.2
2
0.04
20% was added to take care of Attrition = 96+20=116
Sample size = Experimental group (116) + Control
group (116) = 232.
1.5 Sampling technique
A two staged sampling technique was used.
1.5.1 First stage
From the ten political wards in Kuje Area Council,
Kuje central was randomly selected as the intervention
group by balloting. A second ward, Rubochi with
similar socio demographic characteristics but not too
close to Kuje ward was selected to serve as control.
1.5.2 Second stage
Houses with children less than five years of age from
each of the two selected wards were enumerated. A
total of 285 houses were enumerated in Kuje central
and these formed the sampling frame from which a
sample of 130 households was randomly selected
using computer generated random numbers.
In Rubochi ward, 245 houses were enumerated. These
formed the sampling frame from which 118 houses
were selected using computer generated random
numbers. Mothers of selected children were then
invited to the nearby Primary health care center and
semi structured questionnaires were administered to
them by a team of trained assistants. Where there was
more than one eligible child in a household, one was
selected randomly by balloting. If the child’s mother
was not available, the visit was repeated. If a
participant did not consent, was not available after
three visits, or withdrew from the study she/he was
replaced by another eligible child selected by simple
random technique using the computer generated
random numbers.
1.6 Data collection technique
Data was collected at pre-intervention and at post
intervention using the same instruments.
Data was collected using both the semi structured
interviewer administered questionnaire and the
laboratory investigation results sheet.
1.6.1 Pre-intervention data collection
The questionnaire was used to collect data on the