A total of 565 participants took part in the study; 292 from rural and 272 from urban. The median age of participants was 40(26) years and there was no statistical difference between rural and urban participants. The overall median body mass index of participants was 25.6(7.4) with urban dwellers recording a significant higher value. Visceral fat was also significantly higher for urban dwellers compared to rural participants. Calorie, carbohydrate and fibre intakes were significantly higher among rural participants while urban participant consumed higher vitamin A. Other nutrients did not show any significant difference. Table 1 shows socio-demographics, nutritional intake and body composition characteristics of study participants.
Table 1. Descriptive Statistics
DESCRIPTIVE VARIABLES
|
Total
N=565
|
STATISTICS AND BALANCE CHECK
|
|
|
RURAL
n=292
|
URBAN
n=272
|
P-VALUE
|
|
|
n(%)or
Median(IQR)
|
n(%)or
Median(IQR)
|
|
Age
|
40(26)
|
40(26)
|
38(26)
|
0.737
|
Gender
|
|
|
|
|
Male
|
113(19.8)
|
43(14.7)
|
70(25.7)
|
0.001*
|
Female
|
452(79.2)
|
249(85.3)
|
202(74.3)
|
|
Level of Education
|
|
|
|
|
No formal education
|
98(17.2)
|
60(21.4)
|
38(16.7)
|
<0.001*
|
Primary
|
50(8.8)
|
27(9.6)
|
23(10.1)
|
|
Junior secondary
|
176(30.8)
|
120(42.9)
|
56(24.7)
|
|
Senior secondary
|
132(23.1)
|
64(22.9)
|
66(29.1)
|
|
Tertiary
|
53(9.3)
|
9(3.2)
|
44(19.4)
|
|
Weight (kg)
|
65.9(20.2)
|
62.4(18.3)
|
69.9(20.65)
|
<0.001*
|
Height (cm)
|
159.3(9.83)
|
158.0(8.60)
|
160.5(11.2)61.7
|
<0.001*
|
BMI (kg/m2)
|
25.6(7.40)
|
25.0(6.7)
|
26.1(7.65)
|
0.016*
|
Waist circumference (cm)
|
87.4(19.80)
|
87.0(17.55)
|
89.0(22.0)
|
0.115
|
Muscle mass
|
26.9(6.13)
|
26.8(5.05)
|
27.0(6.95)
|
0.526
|
Body fat
|
35.7(15.22)
|
35.2(14.33)
|
36.9(16.35)
|
0.341
|
Visceral fat
|
7.0(4.0)
|
7.0(4.0))
|
7.0(5.0)
|
0.015*
|
Energy intake (Kcal)
|
1434.0(951.1)
|
1601.4(999.04)
|
1330.7(899.84)
|
0.004*
|
Fat (g)
|
42.5(44.08)
|
44.2(50.8)
|
41.0(39.2)
|
0.781
|
Protein (g)
|
42.6(32.5)
|
42.6(32.7)
|
42.9(33.06)
|
0.838
|
Carbohydrate (g)
|
208.0((137.0)
|
223.5(128.8)
|
187.9(137.6)
|
0.001*
|
Fibre (g)
|
18.3(13.3)
|
20.0(13.8)
|
16.9(13.5)
|
0.001*
|
Sugar (g)
|
29.6(41.4)
|
28.0(35.6)
|
33.9(50.04)
|
0.242
|
Folate (µg)
|
225.6(240)
|
234(241.2)
|
216.4(237.0)
|
0.222
|
Iron (mg)
|
9.2(6.6)
|
9.5(6.33)
|
8.8(6.8)
|
0.569
|
Zinc (mg)
|
5.92(4.7)
|
6.1(4.9)
|
5.7(4.7)
|
0.860
|
Vitamin B12 (µg)
|
1.76(3.2)
|
1.92(3.81)
|
1.73(2.64)
|
0.296
|
Vitamin A (µg)
|
120.6(139.8)
|
111.2(121.6)
|
128.8(165.3)
|
0.022*
|
Vitamin E (mg)
|
5.6(6.4)
|
5.6(6.4)
|
5.5(6.3)
|
0.658
|
Saturated fat (g)
|
12.2(15.3)
|
11.1(15.8)
|
13.1(14.6)
|
0.143
|
Monounsaturated fat (g)
|
14.7(18.3)
|
14.8(20.33)
|
14.7(17.0)
|
0.930
|
Polyunsaturated fat (g)
|
7.9(8.4)
|
8.37(9.17)
|
7.3(7.7)
|
0.285
|
Dietary diversity score
|
6.0(2.01)
|
6(3)
|
6(2.01)
|
0.118
|
Some variable responses were missing and therefore does not sum up to 565. *Significant at p-value <0.05
Using BMI, there was no difference in the prevalence of obesity among rural and urban participants. Visceral and body fat cut offs showed a higher prevalence of obesity in urban compared to rural participants. Prevalence of obesity by all parameters was higher among females compared to males with the exception of visceral fat that showed no difference. Table 2 and 3 show the prevalence of obesity by community and gender respectively.
Table 2. Prevalence of obesity by different parameters
Variable
|
Total
|
Rural n=292
|
Urban n=272
|
p-value
|
BMI
|
|
|
|
|
Underweight
|
9(1.6)
|
5(1.7)
|
4(1.5)
|
0.110
|
Normal
|
247(43.3)
|
140(48.3)
|
104(39.4)
|
|
Overweight
|
171(29.9)
|
87(30.0)
|
83(31.4)
|
|
Obese
|
131(22.9)
|
58(20.0)
|
73(27.7)
|
|
Visceral fat
|
|
|
|
|
Normal
|
434(76.0)
|
246(85.1)
|
184(71.0)
|
<0.001*
|
Obese
|
118(20.7)
|
43(14.9)
|
75(29.0)
|
|
Body fat
|
|
|
|
|
Underweight
Normal
Overweight
|
48(8.4)
198(34.7)
130(22.8)
|
29(10.0)
117(40.3)
66(22.8)
|
19(7.3)
80(30.8)
64(24.6)
|
0.023*
|
Obese
|
175(30.6)
|
78(26.9)
|
97(37.3)
|
|
Waist circumference
|
|
|
|
|
Normal
Overweight
|
171(29.9)
107(18.7)
|
95(33.3)
60(21.1)
|
76(33.0)
47(20.4)
|
0.976
|
Obese
|
237(41.5)
|
130(45.6)
|
107(46.5)
|
|
Some variable measurements were missing and may therefore not sum up to 565. *Significant at p-value <0.05.
Table 3. Prevalence of obesity by gender
Variable
|
Male
n=113
|
Female n=452
|
P-value
|
BMI
|
|
|
|
Underweight
|
0(0)
|
9(2.0)
|
|
Normal
|
74(67.3)
|
171(38.4)
|
<0.001*
|
Overweight
|
25(25.7)
|
145(32.6)
|
|
Obese
|
11(10.0)
|
120(27.0)
|
|
Visceral fat
|
|
|
|
Normal
|
84(77.1)
|
347(78.9)
|
0.385
|
Obese
|
25(22.9)
|
93(21.1)
|
|
Body fat
Underweight
|
22(20)
|
26(5.9)
|
|
Normal
Overweight
|
49(44.5)
20(18.2)
|
149(33.8)
110(24.9)
|
<0.001*
|
Obese
|
19(17.3)
|
156(35.4)
|
|
Waist circumference
|
|
|
|
Normal
Overweight
|
81(79.4)
9(8.8)
|
90(21.8)
98(23.7)
|
<0.001*
|
Obese
|
12(11.8)
|
225(54.5)
|
|
*Significant at p-value <0.05.
Participants who were obese by BMI had significantly higher intakes of total calories, fat, carbohydrate protein, fibre and sugar prior to the day of data collection. Using BMI categories, the obese consumed higher quantities of all fats and micronutrients but no difference was observed in dietary diversity among the groups. Table 4–6 shows the difference in nutrient intakes among BMI, waist circumference and body composition categories.
Table 4. Nutrient intakes for BMI, waist circumference and body composition categories.
Variables
|
Energy
|
Fat
|
Protein
|
Carbohydrate
|
Fibre
|
Sugar
|
BMI
|
Median (IQ)
|
Median (IQ)
|
Median (IQ)
|
Median (IQ)
|
Median (IQ)
|
Median (IQ)
|
Underweight
|
1056.5(759.1)
|
34.8(24.6)
|
38.8(30.5)
|
164.4(129.0)
|
11.4(8.4)
|
12.1(24.4)
|
Normal
|
1277.4(849.3)
|
37.9(41.0)
|
38.0(31.6)
|
188.6(123.6)
|
17.6(12.2)
|
25.1(32.5)
|
Overweight
|
1516.5(978.6)
|
45.0(46.0)
|
45.4(35.3)
|
217.6(132.0)
|
19.3(14.0)
|
33.3(54.8)
|
Obese
|
1816.4(1038.1)
|
52.9(52.6)
|
49.5(30.4)
|
255.8(167.3)
|
20.8(15.7)
|
42.1(57.0)
|
p-value
|
<0.001
|
<0.001
|
<0.001
|
<0.001
|
<0.001
|
<0.001
|
Visceral fat
|
|
|
|
|
|
|
Normal
|
1405.9(955.7)
|
39.9(43.0)
|
41.1(33.7)
|
202.9(133.0)
|
18.2(13.6)
|
28.2(41.0)
|
Obese
|
1630.7(1036.7)
|
50.2(53.2)
|
45.4(29.6)
|
236.8(156.6)
|
19.5(13.5)
|
37.5(55.5)
|
p-value
|
0.001
|
0.028
|
0.038
|
0.009
|
0.060
|
0.026
|
Body fat
|
|
|
|
|
|
|
underweight
|
1386.5(944.7)
|
34.3(39.6)
|
43.7(40.6)
|
188.1(132.9)
|
14.5(15.1)
|
21.4(33.3)
|
Normal
|
1313.5(864.8)
|
40.9(45.1)
|
38.0(31.6)
|
195.9(131.5)
|
16.9(12.3)
|
26.0(36.3)
|
Overweight
|
1458.6(912.0)
|
38.5(39.9)
|
48.1(35.6)
|
209.2((131.1)
|
19.9(12.2)
|
35.2(37.6)
|
Obese
|
1676.7(1026.5)
|
47.1(50.4)
|
46.0(27.6)
|
242.2(152.3)
|
19.1(15.3)
|
36.1(57.3)
|
p-value
|
<0.001
|
0.014
|
0.005
|
0.001
|
0.006
|
0.008
|
Waist circumference
|
|
|
|
|
|
|
Normal
|
1384.2(871.7)
|
38.9(42.7)
|
39.3(32.1)
|
196.6(136.5)
|
17.8(13.5)
|
26.1(36.1)
|
overweight
|
1373.0(908.6)
|
39.4(42.9)
|
39.8(41.5)
|
209.1(153.0)
|
18.5(14.4)
|
31.0(41.9)
|
Obese
|
1567.8(1066.6)
|
46.9(49.7)
|
45.9(28.9)
|
222.9(136.8)
|
19.1(14.6)
|
35.0(47.4)
|
p-value
|
0.001
|
0.028
|
0.049
|
0.024
|
0.260
|
0.080
|
Significant p-values are in bold.
Table 5. Differences in micronutrient intakes for BMI, waist circumference and body composition categories.
Variables
|
Folate
|
Iron
|
Zinc
|
Vitamin B12
|
Vitamin A
|
Vitamin E
|
Dietary diversity score
|
BMI
|
Median
(IQR)
|
Median (IQR)
|
Median (IQR)
|
Median (IQR)
|
Median (IQR)
|
Median
(IQR)
|
Median
(IQR)
|
Underweight
|
120(166.7)
|
6.7(5.9)
|
4.3(3.6)
|
1.81(3.1)
|
45.0(101.0)
|
4.5(5.1)
|
6.0(3.0)
|
Normal
|
203.3(205.0)
|
8.4(6.5)
|
5.2(4.7)
|
1.7(3.2)
|
107.6(123.1)
|
4.7(6.4)
|
6.0(3.0)
|
Overweight
|
240.8(270.7)
|
9.8(6.3)
|
6.4(5.0)
|
2.0(3.5)
|
128.9(159.0)
|
5.8(6.7)
|
6.0(2.0)
|
Obese
|
261.6(255.6)
|
11.11(6.8)
|
6.8(4.8)
|
1.7(2.8)
|
133.9(138.8)
|
6.6(5.6)
|
6.0(3.0)
|
p-value
|
0.001
|
<0.001
|
<0.001
|
0.026
|
<0.001
|
0.011
|
0.212
|
Visceral fat
|
|
|
|
|
|
|
|
Normal
|
225.3(243.3)
|
8.9(6.7)
|
5.8(5.0)
|
1.8(3.3)
|
115.0(143.4)
|
5.6(6.5)
|
6.0(3.0)
|
Obese
|
231.6(220.2)
|
10.4(5.6)
|
6.6(4.1)
|
1.8(2.8)
|
133.6(132.8)
|
5.5(5.1)
|
6.0(2.0)
|
p-value
|
0.463
|
0.016
|
0.047
|
0.380
|
0.102
|
0.953
|
0.630
|
Body fat
|
|
|
|
|
|
|
|
underweight
|
193.3(214.7)
|
9.4(6.9)
|
5.6(5.2)
|
1.5(3.2)
|
91.7(148.8)
|
4.9(5.2)
|
6.0(3.0)
|
Normal
|
205.3(215.1)
|
8.5(7.1)
|
5.1(4.6)
|
1.8(3.7)
|
108.3(23.8)
|
4.9(6.9)
|
6.0(2.0)
|
overweight
|
242.4(248.8)
|
9.7(6.3)
|
6.5(5.2)
|
2.1(3.6)
|
129.3(145.7)
|
5.7(6.7)
|
6.0(3.0)
|
Obese
|
246.1(241.1)
|
10.1(6.6)
|
6.6(4.0)
|
1.7(2.6)
|
130.6(147.0)
|
6.4(5.5)
|
6.0(2.0)
|
p-value
|
0.041
|
0.007
|
0.013
|
0.665
|
0.008
|
0.186
|
0.774
|
Waist circumference
|
|
|
|
|
|
|
|
Normal
|
209.5(230.6)
|
9.0(6.9)
|
5.6(4.9)
|
1.8(3.0)
|
113.2(133.7)
|
5.3(6.7)
|
6.0(3.0)
|
overweight
|
214.9(223.0)
|
8.9(6.5)
|
5.9(5.1)
|
1.6(3.6)
|
130.1(147.2)
|
5.3(6.0)
|
6.0(4.0)
|
Obese
|
247.0(251.0)
|
10.0(6.3)
|
6.4(4.8)
|
1.8(3.4)
|
120.6(143.6)
|
5.8(6.1)
|
6.0(2.0)
|
p-value
|
0.098
|
0.058
|
0.160
|
0.489
|
0.356
|
0.310
|
0.511
|
Table 6. Differences in fat intake by body weight category.
Variables
|
Saturated fat (g)
|
Monounsaturated fat (g)
|
Polyunsaturated fat (g)
|
BMI
|
|
|
|
Underweight
|
11.7(11.9)
|
11.9(9.4)
|
6.1(4.2)
|
Normal
|
10.9(13.0)
|
12.8(16.4)
|
6.2(7.6)
|
Overweight
|
12.5(17.5)
|
14.9(19.8)
|
8.4(9.0)
|
Obese
|
15.6(16.2)
|
19.8(23.6)
|
9.6(10.6)
|
p-value
|
0.010
|
0.002
|
0.002
|
Visceral fat
|
|
|
|
Normal
|
12.1(15.0)
|
14.4(17.8)
|
7.5(8.0)
|
Obese
|
13.1(18.4)
|
15.5(19.9)
|
8.9(9.6)
|
p-value
|
0.434
|
0.332
|
0.168
|
Body fat
|
|
|
|
underweight
|
11.7(12.0)
|
13.5(14.9)
|
6.8(7.2)
|
Normal
|
11.0(15.7)
|
13.6(18.7)
|
6.3(7.6)
|
overweight
|
12.2(15.0)
|
13.9(17.1)
|
7.3(8.5)
|
Obese
|
13.6(17.0)
|
16.4(21.5)
|
9.5(9.3)
|
p-value
|
0.346
|
0.080
|
0.009
|
Waist circumference
|
|
|
|
Normal
|
12.0(13.5)
|
13.7(18.5)
|
6.0(3.0)
|
overweight
|
11.6(15.6)
|
13.8(14.7)
|
6.0(4.0)
|
Obese
|
13.0(17.2)
|
16.1(20.9)
|
6.0(2.0)
|
p-value
|
0.370
|
0.074
|
0.064
|
Significant p-values are in bold
Table 7 shows principal component analysis of household food frequency. A total of four components were extracted. The four components explained 30.5% of the total variance. Table 8 shows partial spearman correlation between the four components and body composition measures. Component 2 showed a significant negative correlation with BMI, body fat and visceral fat while component 1 showed a positive correlation with body fat and BMI.
Table 7. Principal component analysis (PCA) of household food frequency
Food items in the PCA
|
Component 1
Diverse diet
|
Component 2
Vegetable convenience pattern
|
Component 3
Snack pattern
|
Component 3 Staple food pattern
|
Eggs
|
0.611
|
|
|
|
Sugar sweetened beverages
|
0.559
|
|
|
|
Milk
|
0.528
|
|
|
|
Processed milk
|
0.509
|
|
|
|
Instant noodles
|
0.498
|
|
|
|
Processed meat
|
0.495
|
|
|
|
meat
|
0.467
|
|
|
|
Commercial bread
|
0.465
|
|
|
0.342
|
Margarine or butter
|
0.445
|
|
|
|
Rice
|
0.414
|
|
|
|
Confectionery
|
0.389
|
|
|
|
Chicken
|
0.382
|
|
|
|
Pasta
|
0.381
|
|
|
|
Organ meat
|
0.364
|
|
|
|
Sweets
|
0.358
|
0.3
|
|
|
Vegetables fried
|
0.596
|
|
|
Diet beverages
|
0.569
|
|
|
Fast food
|
|
0.522
|
|
|
Restaurant meals
|
0.434
|
|
|
Raw vegetables
|
0.3
|
0.337
|
|
|
Salted dried fish
|
0.327
|
|
|
Fried potatoes
|
0.315
|
|
|
Cooked vegetables
|
|
0.644
|
|
Nuts
|
|
|
0.615
|
|
Sugar
|
|
|
0.558
|
|
Breakfast cereal
|
|
0.47
|
|
Tea or coffee
|
0.305
|
|
0.443
|
|
Salty snacks
|
|
0.398
|
|
Fish
|
|
|
|
0.577
|
Legumes
|
|
|
|
0.484
|
Fried fish
|
|
|
|
0.448
|
Fruit
|
|
|
|
0.394
|
Ready to eat meals
|
-0.323
|
|
0.38
|
maize
|
|
|
|
0.363
|
Roots and tubers
|
|
|
0.347
|
Kaiser-Meyer-Olkin Measure of Sampling Adequacy 0.771, Bartlett's Test of Sphericity <0.001
Table 8. Partial correlation between body measures and PCA
Variables
|
Component 1
Diverse diet
r(p-value)
|
Component 2
Vegetable convenience pattern
|
Component 3
Snack pattern
r(p-value)
|
Component 3 Staple food pattern
r(p-value)
|
visceral fat
|
0.090(0.053)
|
-0.142(0.002)
|
0.018(0.695)
|
-0.027(0.567)
|
body fat
|
0.107(0.022)
|
-0.102(0.029)
|
-0.056(0.227)
|
-0.015(0.751)
|
BMI
|
0.099(0.034)
|
-0.136(0.003)
|
-0.008(0.862)
|
-0.009(0.842)
|
waist circumference
|
0.063(0.178)
|
-0.089(0.057)
|
-0.034(0.468)
|
-0.066(0.158)
|
*controlled for age and gender
Table 9 shows a partial spearman correlation adjusted for age and gender between waist circumference, BMI and body composition measures. All variables showed significant correlations but the strongest correlations were between visceral fat and BMI r = 0.905(p<0.001), body fat and BMI r = 0.851 (p<0.001) and BMI and waist circumference r = 0.845(p<0.001).
Table 9 Partial correlation between body composition, waist circumference and BMI.
Control Variables
|
Variables
|
visceral fat
r(p-value)
|
body fat
r(p-value)
|
BMI
r(p-value)
|
waist circumference
r(p-value)
|
Age in years, Gender
|
visceral fat
|
|
0.746(<0.001)
|
0.905(<0.001)
|
0.799(<0.001)
|
|
body fat
|
0.746(<0.001)
|
|
0.851(<0.001)
|
0.766(<0.001)
|
|
BMI
|
0.905(<0.001)
|
0.851(<0.001)
|
|
0.845(<0.001)
|
Table 10 shows multinomial logistic regression for risk factors of central obesity determined by waist circumference. Multicollinearity was checked and carbohydrate and fibre had strong correlations with energy intake and protein respectively and were therefore excluded from the model. The obese group was set as reference for the outcome variable and urban and female were set as reference for the explanatory variables, community and gender respectively. Males had about 22 times odds of being normal compared to females at p-value <0.001. Rural dwellers had odds of about 1.7 times of being normal compared to urban participants and this was significant at p<0.01.
Table 10. Multinomial logistic regression of predictors of obesity
waist circumference category
|
Explanatory variables
|
Odds ratios
|
Normal
|
|
|
|
Energy
|
0.999(0.999-1.000)**
|
|
Protein
|
1.008(0.997-1.019)
|
|
Fat
|
0.996(0.987-1.005)
|
|
Sugar
|
0.998(0.992-1.004)
|
|
Community
|
|
|
Rural
|
1.684(1.039-2.729)**
|
|
Urbana
|
|
|
Gender
|
|
|
Male
|
21.968(10.876-44.373)***
|
|
Femalea
|
|
Overweight
|
Intercept
|
|
|
Energy
|
0.999(0.999-1.000)**
|
|
Protein
|
1.004(0.994-1.015)
|
|
Fat
|
0.999(0.990-1.007)
|
|
Sugar
|
1.002(0.995-1.008)
|
|
Community
|
|
|
Rural
|
1.198(0.741-1.936)
|
|
Urbana
|
|
|
Gender
|
|
|
Male
|
1.991(0.790-5.018)
|
|
Femalea
|
|
*p<0.05, **p<0.01, ***p<0.001. aVariable set as reference. Reference for weight category is obesity. Cox and Snell R-Squared is 0.243 and Naglekerke R-Squared is 0.277.