Page 37 - JCTR-9-6

Asekenye et al. | Journal of Clinical and Translational Research 2023; 9(6): 398-413 401
The households were, then, sampled using “spinning the bottle” 4. Results
method and a person responsible for meal preparation from
each selected household was interviewed. Cochran’s (1972) 4.1. Demographic characteristics of participants
formula for finite population was used to calculate the sample Results of the demographic characteristics of participants in
size, that is, this study are summarized in Table 1. Out of the 422 eligible
respondents who participated in the study, all were female
2
t pq (persons responsible for meal preparation) and from the sampled
n =
d 2 households. Of these, 253 resided in Ankole sub-region and 169 in
Teso sub-region. The highest number (62.1%) of them were adults
Where t is the value for selected alpha level of 0.025 in each and married (30 – 59 years) and the least (13.5%) were widowed
tail = 1.96 (from the Z table), (p) is the estimated proportion of and elderly (<60 years). Most had acquired a primary level of
the population which has the attributes in question (vegetable education (45.7%) and those with tertiary level were fewer (13%)
consumption and prevalence of hyperglycemia) = 0.5 (since it’s than those without formal education (15.6%). Most respondents
unknown), (q = [1-p]) is the estimate of variance = 0.5 and d is were farmers (71.9%) and the civil servants (8.6%) were less than
the desired level of precision (acceptable margin of error) = 0.05. the non-civil servants (19.3%). Overall, most of the participants
By adding 10% non-response rate, the final sample size was were natives of their respective sub-regions, that is, they were
calculated at 422 individuals. A random sample of 422 households born and had spent more than 30 years in residence.
in the target population was deemed enough to give the confidence
levels of 95%. 4.2. Frequently eaten vegetables in Ankole and Teso sub-regions
of Uganda
3.3. Data collection methods
Figure 2 shows the frequently eaten leafy vegetables in both
3.3.1. Community survey sub-regions.

An interviewer-administered questionnaire for face-to-face Table 1. Demographic characteristics of participants
interviews and focus group discussions (FGDs) was used to obtain Characteristic Sub‑region
information on demographics of the respondents, frequently eaten
vegetables, methods of preparation, and consumption pattern. Ankole Teso Total
Furthermore, the FBG levels were taken from household members Category n Weighted n Weighted n Weighted
%
%
%
(only those 18 years and above) of the sampled households using
a glucometer. Ten women (responsible for meal preparation) were Respondents 253 100.0 169 100.0 422 100.0
engaged in FGDs to assess the community-level information. Age
A pre-test with 25 participants (from each study sub-region) was 18 – 29 65 25.7 38 22.5 103 24.4
conducted to assess language clarity, ability to include information 30 – 59 151 59.7 111 65.7 262 62.1
required, acceptability in terms of length, and the privacy of the ≥60 37 14.6 20 11.8 57 13.5
participants. This also provided the Cronbach alpha of 92% which Total 253 100.0 169 100.0 422 100.0
ascertained the internal validity of the questionnaire whereas the Marital status
test-retest ascertained the reliability of 94%. The recorded data in Single 24 9.5 13 7.7 37 8.8
the questionnaire were thereafter harmonized with the noted and Married 193 76.3 141 83.4 334 79.1
recorded information from the FGDs. Divorced 14 5.5 5 3.0 19 4.5
Field visits were carried out to collect voucher specimens of Widowed 22 8.7 10 5.9 32 7.5
frequently eaten vegetable species in each sub-region. Voucher Education status
specimens were identified and determined by a botanist both at Informal 44 17.4 22 13.0 66 15.6
Mbarara University of Science and Technology and at Makerere Primary 100 39.5 93 55.0 193 45.7
University Herbarium. Secondary 70 27.7 38 22.5 108 25.6
Tertiary 39 15.4 16 9.5 55 13
3.3.2. Statistical analysis Occupation

Data were entered in Microsoft Excel 2016 and exported to the Civil servant 23 9.2 13 7.7 36 8.6
Statistical Package for the Social Sciences version 20 for analysis. Farmers 166 66.1 136 80.5 302 71.9
Descriptive statistics were used to obtain the frequencies and Non civil 61 24.3 20 11.8 81 19.3
percentages of demographic and socioeconomic characters of the servants
participants, the most frequently eaten vegetables, and prevalence Years in residence
of hyperglycemia. Chi-square test, stepwise multi-variate logistic 0 – 10 74 29.6 42 25.1 116 27.8
analysis was used to identify the variables which significantly 10 – 20 47 18.8 42 25.1 89 21.3
impacted and associated with the FBG levels (hyperglycemia 20 – 30 46 18.4 42 25.1 88 21.1
status). >30 83 33.2 41 24.6 124 29.7
DOI: http://dx.doi.org/10.18053/jctres.09.202306.23-00096