Measuring drought and drought impacts in Red Sea Province 
5. Nutritional status of children in Red Sea Province, November 1985 to November 1987. Mary Cole and Roy Cole 
1. Timing and location of Oxfam Port Sudan Nutritional Surveillance.
Table 5.2. Timing of Oxfam Nutritional Surveillance Cycles 16.
CYCLE BY MONTH 

YEAR 
J 
F 
M 
A 
M 
J 
J 
A 
S 
O 
N 
D 
1985 
1 
1 

1986 
1 
1&2 
2 
2 
2 
2 
3 
3 
3 
3 
3 
4 
1987 
4 
4 
4&5 
5 
5 
5 
5 
6 
6 
6 
6 
The average time taken to survey the whole province was four months. Three surveys were completed in 12 months. Cycles 1 and 4, 2 and 5, and 4 and 6 were carried out at comparable times of year (Table 5.2). Within each cycle, the order in which the districts were visited varied (Table 5.3). The locations of the sites sampled in cycles 16 are shown in Appendix 5.6.
Table 5.3. Districts visited by monfh, NST cycles 16.
MONTH BY CYCLE 

DISTRICT 
1 
2 
3 
4 
5 
6 
DERUDEB 
NOV 
MAR 
SEP 
JAN 
MAY 
SEP 
HALAIB 
JAN 
JUN 
OCT 
JAN 
JUN 
OCT 
HAYA 
DEC 
APR 
AUG 
DEC 
MAR 
SEP 
N.TOKAR 
JAN 
MAY 
OCT 
DEC 
MAY 
OCT 
R.P.S. 
JAN 
JUN 
NOV 
MAR 
JUL 
DEC 
SINKAT 
DEC 
APR 
SEP 
DEC 
APR 
OCT 
S.TOKAR 
JAN 
MAR 
SEP 
JAN 
MAR 
OCT 
Table 5.4. Total number of visits and number of places visited by district, cycles 16.
DISTRICT 
NUMBER OF 
NUMBER OF 
% VISITED 
% VISITED 
VISITS 
PLACES 
> ONCE 
3 TIMES 

DERUDEB 
57 
37 
35 
13.5 
HALAIB 
75 
47 
45 
13 
HAYA 
77 
53 
38 
7.5 
N.TOKAR 
72 
37 
62 
19 
R.P.S. 
67 
49 
29 
8 
SINKAT 
79 
55 
35 
9 
S.TOKAR 
77 
47 
43 
17 
TOTAL 
504 
325 
40 
12 
Oxfam received many complaints from community leaders of repeat sampling and excessive visits by nutritional surveillance teams. Table 5.4 shows the total number of visits made to each district over cycles 16, together with the total number of different places visited. Over all cycles (two years), 40% of places visited were visited more than once, and 12% of places visited were visited three or more times. The maximum number of times any place was visited was four times.
2. Description of the sample.
a. Sample size
Sample sizes by district and cycle for children 75.1115 cm and <75 cm are shown in Tables 5.5 and 5.6 respectively.
Table 5.5. Sample sizes (children >75 cm) by cycle and district.
CYCLE 

DISTRICT 
1 
2 
3 
9 
5 
6 
13 
46 
DERUDEB 
169 
322 
258 
211 
195 
189 
749 
595 
HALAIB 
264 
324 
231 
203 
190 
190 
819 
583 
HAYA 
150 
337 
252 
277 
258 
248 
739 
783 
N.TOKAR 
330 
324 
257 
255 
224 
232 
911 
711 
R.P.S. 
214 
327 
266 
252 
264 
254 
807 
770 
SINKAT 
227 
338 
257 
293 
267 
262 
822 
822 
S.TOKAR 
301 
323 
241 
291 
292 
279 
865 
862 
TOTAL 
1665 
2295 
1762 
1782 
1690 
1654 
5712 
5124 
Table 5.6. Sample sizes (children <75 cm) by cycle and district.
CYCLE 

DISTRICT 
1 
2 
3 
4 
5 
6 
13 
46 
DERUDEB 
22 
34 
42 
35 
46 
51 
98 
132 
HALAIB 
66 
44 
56 
41 
53 
57 
166 
151 
HAYA 
30 
33 
49 
53 
75 
86 
112 
2;4 
N.TOKAR 
89 
41 
44 
53 
77 
70 
174 
200 
R.P.S. 
26 
41 
35 
52 
96 
76 
102 
224 
SINKAT 
40 
40 
44 
65 
66 
68 
124 
199 
S.TOKAR 
88 
46 
59 
70 
70 
81 
193 
221 
TOTAL 
361 
279 
328 
369 
481 
489 
968 
1339 
The percent of the total sample which comprised children less than or equal to 75 cm by district and cycle is shown in Table 5.7. For Red Sea Province as a whole there was a significantly higher percentage of children less than 75 cm tall in cycles 46 than in cycles 13. At the district level there was a significantly higher percentage of children less than 75 cm tall in cycles 46 than in cycles 13 for all districts except Halaib and South Tokar.
Table 5.7. Percent of sample <75 cm by cycle and district.
CYCLE 

DISTRICT 
1 
2 
3 
4 
5 
6 
13 
46 
DERUDEB 
11.5 
9.6 
14.0 
14.2 
19.1 
21.3 
11.6 
18.2** 
HALAIB 
16.7 
8.7 
15.6 
15.3 
23.5 
23.5 
16.9 
20.6 
HAYA 
16.7 
8.9 
11.2 
16.1 
22.5 
25.7 
13.2 
21.2** 
N.TOKAR 
21.2 
11.2 
14.6 
17.2 
25.6 
23.2 
16.0 
22.0** 
R.P.S. 
10.8 
11.1 
11.6 
17.1 
26.7 
23.0 
11.2 
22.5** 
SINKAT 
15.0 
10.6 
14.6 
18.2 
19.8 
20.6 
13.1 
19.5** 
S.TOKAR 
22.6 
12.5 
19.7 
19.4 
19.3 
22.5 
18.2 
19.5 
TOTAL 
17.9 
10.8 
15.7 
17.2 
22.2 
22.8 
14.5 
20.7** 
** Significant difference between cycles 13 and 46 (p<.05)
b. Proportions of males and females in the sample.
The ratios of males to females by district and cycle for children greater than 75 cm and less than or equal to 75 cm are shown in Tables 5.8 and 5.9 respectively.
Table 5.8. Ratios of males:females (children >75 cm) by cycle and district.
CYCLE 

DISTRICT 
1 
2 
3 
4 
5 
6 
13 
46 
DERUDEB 
1.06 
1.37* 
1.10 
1.22 
1.44* 
1.33 
1.20* 
1.32* 
HALAIB 
0.89 
0.98 
0.94 
0.88 
1.18 
0.96 
0.94 
1.00 
HAYA 
1.14 
1.13 
0.97 
1.16 
0.97 
1.25 
1.08 
1.12 
N.TOKAR 
1.17* 
1.13 
1.14 
0.92 
1.20 
1.17 
1.15* 
1.08 
R.P.S. 
1.55* 
1.14 
1.53* 
1.17 
1.28* 
1.17 
1.36* 
1.21* 
SINKAT 
1.39* 
1.28* 
0.85* 
1.02 
1.09 
1.02 
1.15* 
1.04 
S.TOKAR 
1.01 
1.07 
0.94 
1.12 
1.20 
1.05 
1.01 
1.12 
TOTAL 
1.14* 
1.15* 
1.05 
1.07 
1.18* 
1.13* 
1.12* 
1.12* 
* Significantly different from 1.00 at p=.05
In the sample 75.1115 cm, there was a significantly higher proportion of males than females for the whole province in both cycles 13 and 46. By district, Derudeb and Rural Port Sudan had significantly higher proportions of males than females in both cycles 13 and 46, while North Tokar and Sinkat districts had significantly higher proportions of males than females in cycles 13 only.
Table 5.9. Ratios of males:females (children <75 cm) by cycle and district
CYCLE 

DISTRICT 
1 
2 
3 
4 
5 
6 
13 
46 
DERUDEB 
0.83 
0.62 
0.68 
0.75 
0.92 
1.09 
0.69 
0.91 
HALAIB 
0.89 
0.63 
0.60 
1.73 
1.04 
0.68 
1.71 
1.01 
HAYA 
1.50 
1.06 
0.88 
1.04 
1.08 
0.95 
1.07 
1.02 
N.TOKAR 
0.98 
1.28 
1.00 
1.41 
0.97 
1.19 
1.04 
1.15 
R.P.S. 
1.36 
0.78 
1.19 
1.36 
1.13 
1.24 
1.04 
1.22 
SINKAT 
1.00 
1.35 
1.39 
0.81 
0.83 
0.74 
1.24 
0.79 
S.TOKAR 
1.26 
0.92 
0.97 
0.89 
1.50 
1.08 
1.08 
1.13 
TOTAL 
1.07 
0.91 
0.91 
1.07 
1.06 
0.98 
0.97 
1.03 
Despite some apparently large differences between the proportions of males and females less than 75 cm, none were significantly different from a ratio of 1.00 (50% males, 50% females). This is because statistical significance depends not only on the proportions found, but on the size of the sample. The sample sizes for children less than 75 cm were very small. c. Distribution of heights in the sample.
Height distributions for the total sample for cycles 13 and 46 are shown in Figure 5.1. There was a significantly higher percentage of children in the height category 55.165.0 cm in cycles 46 than in cycles 13. There was a significantly lower percentage of children in the height category 95.1105.0 cm in cycles 46 than in cycles 13. No significant differences were found between the percentage of males and females in each height category in either cycles 13 or 46.
3. The nutritional status of children, total province.
Unless otherwise specified, results refer to children 75.1115 cm in height. a. Means and distribution of percent weight for height, total province.
The percentage of children in each percent weight for height class for cycles 13 and 46 are shown in Figure 5.2. For both cycles 13 and 46 the largest percent of the sample fell between 80 and 89.9% weight for height. Males and females are aggregated because no significant differences were found between the percentage of males and females in each percent weight for height class in either cycles 13 or 46. In cycles 46, however, females had a higher mean percent weight for height than males (see Table 5.10).
Table 5.10 Mean percent weight for height and standard deviation by sex, Red Sea Province, cycles 13 and 46.
CYCLE 
MALE 
FEMALE 
TOTAL 

13 
MEAN 
86.68 
87.04 
87.26 
SD 
7.99 
8.12 
8.05 

46 
MEAN 
88.08* 
88.58* 
88.34** 
SD 
7.76 
8.05 
7.90 
* Significant difference between males and females (p<0.05).
** Significant difference between cycles 13 and 46 (p<.051.
For both cycles 13 and 46 the mean percent weight for height for the total sample was above 85% weight for height. The centre of the distribution therefore fell towards the upper limits of the 8089.9% class.
There was an upward shift in the distribution of percent weight for height between cycles 13 and 46. This is reflected in the mean percent weight for height which was significantly higher in cycles 46 than 13 (p<.05). There was a significant decrease in the percentage of the sample in the 8089.9% weight for height class between cycles 13 and 46, and a corresponding increase in the 9099.9% weight for height class over the same periods. Although there was a decrease in the percentage of the sample in the 7079.9% weight for height class between cycles 13 and 46, this was not statistically significant. When all weight for height classes less than 80% weight for height were aggregated, however, the changes between cycles 13 and 46 became significant (see the Results section).
b. Rates of malnutrition, total province.
Children between 70% and 79.9% of their reference median weight for height are classified as "moderately malnourished". Children less than 70% of their reference median weight for height are classified as "severely malnourished". For the purpose of this report the following definitions have been used:
Malnourished  those children less than 80% of their reference median weight for height. This includes both moderately and severely malnourished children.
Severely malnourished  those children less than 70% of their reference median weight for height.
i. Percentage of malnourished children, total province.
The percentage of the total sample who were malnourished together with the 95% confidence interval for the population estimate, is shown in Table 5.11. Provincewide, there was no significant difference between the percentage of males and females who were malnourished. A significantly lower percentage of children were malnourished in cycles 46 than in cycles 13.
Table 5.11. Percent less than 80% reference median weight for height (malaourished) by sex, with 95% confidence intervals, Red Sea Province.
CYCLE 
MALES 
FEMALES 
TOTAL 

% 
95% CI 
% 
95% CI 
% 
95% CI 

13 
16.68 
1.36 
16.65 
1.43 
16.63 
0.49* 
46 
13.72 
1.32* 
12.64 
1.35* 
13.19 
0.47* 
* Significant difference in cycles 13 and 46.
ii. Percentage of severely malnourished children, total province.
The percentage of the total sample who were severely malnourished, together with the 95% confidence interval for the population estimate, is shown in Table 5.12.
Table 5.12 Percent less than 70% reference median weight for height (severely malnourished) by sex, with 95% confidence Intervals, Red Sea Province.
SEX 

CYCLE 
MALES 
FEMALES 
TOTAL 

% 
95% CI 
% 
95% CI 
% 
95% CI 

13 
1.89 
0.43* 
2.74 
0.63* 
2.27 
0.43 
46 
1.36 
0.44 
1.34 
0.47 
1.35 
0.32** 
* Significant difference between males and females (p<.05).
** Significant difference between cycles 13 and 46 (p<.05).
In cycles 13 there was a significantly higher percentage of severely malnourished females than males. There was no significant difference in the percentage of severely malnourished males and females in cycles 46. The percentage of severely malnourished females decreased significantly between cycles 13 and 46, whereas the percentage of severely malnourished males stayed the same.
c. Malnutrition rates by district.
i. Percentage of malnourished children by district.
The percentage of the sample who were malnourished by district and sex, together with the 95% confidence interval for the population estimate, is shown in Table 5.13. There was a significant difference between the percentage of malnourished males and females in South Tokar District in cycles 13. A significantly higher percentage of males than females were malnourished. There were no other significant differences between the percentages of malnourished males and females by district, either in cycles 13 or 46.
Table 5.13. Percent less than 80% reference median weight for height (malnourished) by district and sex, with 95% confidence Intervals for the estimate.
CYCLES 13 
CYCLES 46 

DISTRICT 
MALES 
FEMALES 
TOTAL 
MALES 
FEMALES 
TOTAL 

DERUDEB 
MEAN 
21.57 
21.11 
21.36 
15.04 
18.75 
16.64** 

CI 
4.07 
4.42 
3.00 
3.88 
4.89 
3.05 

HALAIB 
MEAN 
14.90 
15.13 
15.02 
8.25 
6.51 
7.38** 

CI 
3.58 
3.49 
2.50 
3.23 
2.89 
2.19 

HAYA 
MEAN 
20.30 
19.10 
19.76 
18.12 
15.99 
17.11 

CI 
4.12 
4.17 
2.93 
3.79 
3.79 
2.69 

N.TOKAR 
MEAN 
11.50 
14.15 
12.73 
11.65 
11.40 
11.53 

CI 
2.89 
3.39 
2.21 
3.33 
3.44 
2.40 

R.P.S 
MEAN 
10.75 
11.70 
11.15 
10.45 
9.46 
10.00 

CI 
2.87 
3.48 
2.22 
2.98 
3.13 
2.16 

SINKAT 
MEAN 
18.18 
20.94 
19.46 
14.35 
10.45 
12.41** 

CI 
3.86 
4.16 
2.76 
3.43 
3.05 
2.30 

S.TOKAR 
MEAN 
28.28* 
20.00* 
24.16 
14.25 
12.56 
13.46** 

CI 
4.32 
3.86 
2.91 
3.27 
3.29 
2.33 

PROVINCE MEAN 
16.68 
16.65 
16.63 
13.72 
12.64 
13.19** 

CI 
1.36 
1.43 
0.49 
1.32 
1.35 
0.47 
* Significant difference between males and females.
** Significant difference between cycles 13 and 46.
From Table 5.13 the districts in Red Sea Province can be divided into two groups: those that showed a significant decrease in the percentage of malnourished children between cycles 13 and cycles 46, and those that showed no significant difference in percentage of malnourished children between cycles 13 and 46. Derudeb, Halaib, Sinkat and South Tokar Districts all had a significant decrease in the percentage of malnourished children between cycles 13 and 46. Haya, North Tokar and Rural Port Sudan Districts showed no change in the percentage of malnourished children between cycles 13 and 46. This is summarized in Map 5.1.
Map 5.2 shows a comparison of the percentage of the sample who were malnourished in each district for cycles 13 in relation to the provincewide mean percentage of malnourished children for cycles 13. Districts are ranked as above the provincewide mean, equal to the province wide mean, or below the provincewide mean with respect to the percentage of malnourished children. Map 5.3 shows the same information for cycles 46.
It should be remembered that the provincewide percentage of malnourished children was lower in cycles 46 than in cycles 13, and that the relation of each district to the provincewide mean should be interpreted in conjunction with the absolute district figures shown in Table 5.13. A district by district summary combining these data is given below.
DERUDEB: The percentage of malnourished children was higher than the provincewide average in cycles 13, and although it decreased significantly, still remained worse than average in cycles 46.
HALAIB: The percentage of malnourished children was equal to the provincewide average in cycles 13 and decreased significantly, resulting in a lower than average percentage of malnourished children in cycles 46. Halaib District was the only district with average or higher than average percentage of malnourished children to improve significantly between cycles 13 and 46.
MAYA: The percentage of malnourished children was higher than the provincewide average in cycles 13 and did not change significantly in cycles 46. Because the provincewide average percentage of malnourished children decreased between cycles 13 and 46 while the district average for Haya remained the same, the percentage of malnourished children in Haya was worse in relation to the rest of Red Sea Province in cycles 46 than in cycles 13. Haya district was the only district with a higher than average percentage of malnourished children that failed to improve between cycles 13 and 46.
NORTH TOKAR: The percentage of malnourished children was better than average in cycles 13 and did not change significantly in cycles 46. However because the provincewide average percentage of malnourished children decreased in cycles 46, North Tokar moved from being better than the provincewide mean in cycles 13 to being equal to the provincewide mean in cycles 46.
RURAL PORT SUDAN: The percentage of malnourished children was lower than average in cycles 13. The percentage of malnourished children did not change significantly between cycles 13 and 46, but despite the fact the provincewide mean improved, Rural Port Sudan still had a lower than average percentage of malnourished children in cycles 46.
SINKAT: The percentage of malnourished children was higher than average in cycles 13 but decreased significantly and were equal to the average in cycles 46.
SOUTH TOKAR: The percentage of malnourished children was higher than average in cycles 13 but decreased significantly and were equal to the average in cycles 46. The percentage of malnourished males decreased more than the percentage of malnourished females, so the significantly higher percentage of malnourished males than malnourished females found in cycles 13 was not found in cycles 46.
ii. Percentage of severely malnourished children by district.
The percentage of the sample who were severely malnourished broken down by district and sex, together with the 95% confidence interval for the estimate, is shown in Table 5.14.
Table 5.14. Percent less than 70% weight for height by district and sex, with 95% confidence Intervals for the estimate.
CYCLES 13 CYCLES 46
DISTRICT 
MALES 
FEMALES 
TOTAL 
MALES 
FEMALES 
TOTALS 

DERUDEB 
MEAN 
2.21 
4.40 
3.20 
0.88 
2.34 
1.51** 
CI 
1.46 
2.20 
1.29 
1.01 
1.89 
1.00 

HALAIB 
MEAN 
1.77 
2.13 
1.95 
1.03 
0.34 
0.69** 
CI 
1.36 
1.40 
0.97 
1.18 
0.67 
0.69 

HAYA 
MEAN 
2.87 
3.09 
2.98 
1.93 
1.63 
1.79 
CI 
1.71 
1.83 
1.25 
1.35 
1.32 
0.95 

N.TOKAR 
MEAN 
1.44 
1.65 
1.54 
0.54 
0.88 
0.70 
CI 
1.08 
1.24 
0.82 
0.76 
1.01 
0.63 

R.P.S 
MEAN 
0.65 
1.75 
1.12 
1.43 
1.43 
1.43 
CI 
0.75 
1.42 
0.74 
1.16 
1.27 
0.86 

SINKAT 
MEAN 
1.18* 
3.93* 
2.43 
1.44* 
0.50* 
0.97** 
CI 
1.03 
1.99 
1.07 
1.17 
0.70 
0.68 

S.TOKAR 
MEAN 
4.37 
3.95 
4.16 
1.32 
1.23 
1.28** 
CI 
1.96 
1.88 
1.36 
1.07 
1.09 
0.77 

PROVINCE MEAN 
1.89* 
2.74* 
2.27 
1.36 
1.34 
1.35** 

CI 
0.43 
0.63 
0.43 
0.44 
0.47 
0.32 
* Significant difference between males and females.
** Significant difference between cycles 13 and 46.
As noted previously, there was a significantly higher percentage of severely malnourished females than severely malnourished males in cycles 13. At the district level the significantly higher percentage of severely malnourished females than severely malnourished males was found only in Sinkat district, although many other districts showed a similar pattern but did not reach statistical significance, possibly because of small sample size.
No district had a percentage of severely malnourished children which was significantly different from the provincewide mean. Small sample size may have influenced the lack of statistical significance. The tendency was, however, for those districts with the highest rates of moderate malnutrition to have the highest rates of severe malnutrition.
4. Further characterisation of malnutrition in Red Sea Province.
a. Percentage of malnourished children by height category.
The total sample of children of all heights was broken down into height categories. The percent of children in each height category who were malnourished is shown in Figure 5.3. Children less than 55 cm were excluded because of extremely small numbers. No significant differences were found between the percentage of malnourished males or the percentage of malnourished females in any height category in either cycles 13 or 46.
In both cycles 13 and 46 the highest rates of malnutrition were seen in the height category 65.175 cm. High rates of malnutrition were also seen in children 55.165 cm. These results must be interpreted with extreme caution, however, because of the measurement technique used for the majority of children in these height categories. In children aver 75 cm tall highest rates of malnutrition were seen in the category 75.185 cm. Malnutrition rates then declined with height, and the lowest rates were seen in the tallest height category. There was a significant in the percentage of malnourished children in all height categories between cycles 13 and 46. The largest decreases were seen in the height categories 55.1 to 65 cm and 65.1 to 75 cm.
b. Percentage of malnourished children by season.
The results presented above are annual averages. It is important to understand the seasonal variations in malnutrition rates, for example to identify times of high risk, and to construct a framework with which to interpret the results of future surveys carried out at specific times of the year.
As described in the methods section, a season (major wet, minor wet, dry and harvest) was assigned to each case based on ecozone and time of year. Results from this analysis failed to detect seasonal differences; there were no significant differences between the mean percent reference median weight for height or the percentage of malnourished children between any of the seasons. This did not reflect the experience in the field, where May and June were described as the worst months for children's health, and January the best month, irrespective of ecozone. A further analysis was carried out by cycle, which corresponded to months of the year (see Table 5.3).
For the total province, the percentage of malnourished children peaked in September, then decreased in January and remained at this level until peaking again in September (Figure 6.4). Malnutrition rates were therefore worst in September, but improved by January to a level which is sustained for the rest of the year.
When broken down by district and cycle, two patterns emerge: Figure 5.5 shows the percentage of malnourished children by cycle for Halaib and Rural Port Sudan districts, and Figure 5.6 shows the percentage of malnourished children by cycle for the remaining districts of Derudeb, Haya, North Tokar, Sinkat and South Tokar. The latter group of districts followed the pattern described for the province as a whole. In Halaib district there was a steady decrease in the percentage of malnourished children with no seasonal variation apparent. In Rural Port Sudan there was some variation in the percentage of malnourished children, but it was small and not consistent with seasonality.
c. Percentage of malnourished children by settlement type.
The settlements visited by the Oxfam Port Sudan nutrition surveillance teams in cycles 16 were classified as being railway towns, towns, rural settlements or camps (settlements of displaced people that grew up during the drought). Table 5.15 gives the percentage of malnourished children by settlement type for cycles 13 and 46.
Table 5.15. Percentage of children less than 80% reference median weight for height (malnourished) by settlement type, cycles 13 and 46, with 95% confidence Intervals for the population estimate.
CYCLE
SETTLEMENT 
CYCLES 13 
CYCLES 46 

MEAN 
95% CI 
MEAN 
95% CI 

RAILWAY 
17.1 
4.24 
14.0 
3.32 
TOWN 
19.5 
2.66 
10.6 
2.17 
RURAL 
16.2 
1.16 
13.0 
1.13 
CAMP 
24.5 
4.14* 
13.8 
3.86 
* Significantly different from other settlement types (p<.05).
There was a significantly higher percentage of malnourished children in the camps in cycles 13 than in all other types of settlements in cycles 13. There was no significant difference in the percentage of malnourished children in railway towns, towns and rural areas in cycles 13. There was no difference between the percentage of malnourished children in any of the settlement types in cycles 46.
d. Percentage of malnourished children by food security zone.
Red Sea Province was classified into three zones of drought impacts and food security for the years 1987 and 1988. Nutritional surveillance data for 1987 (cycles 46) were analyzed by "food security zone" for the same year. The percentage of malnourished children by food insecurity zone in cycles 46 are shown in Table 5.16.
Table 5.16. Percentage of children less then 80% reference rnedian weight for height by food security zone, cycles 46, with 95% confidence intervals for the population estimate.
FOOD SECURITY ZONE 
N 
MEAN 
95% CI 
HIGH SECURITY 
1891 
11.37 
1.46 
MEDIUM SECURITY 
1535 
12.77 
1.70 
LOW SECURITY 
1029 
15.74 
2.27* 
* Significantly different from other food security zones (p<.05).
There was no significant difference between the percentage of malnourished children in the high and medium food security zones. The percentage of malnourished children in the low food security zone was significantly higher than the percentage of malnourished children in the high and medium food security zones.
e. Percentage of malnourished children and the World Food Programme relief grain ration.
Mean percent reference median weight for height of children 75.1115 cm for each settlement was correlated with the settlement average World Food Programme grain ration per family per day at the time of the nutritional surveillance. Spearman's rank correlation of mean percent weight for height with mean grain relief grain ration gave a correlation coefficient of 0.157 for cycles 13 and 0.008 for cycles 46. These correlation coefficients were not significantly different from zero. There was no correlation between mean percent weight for height and relief grain ration; the ration neither increased or decreased in a consistent pattern as malnutrition rates increased.
f. Variation In percentage of malnourished children within and between settlements.
Oneway analysis of variance (ANOVA) of percent weight for height and settlement for cycles 16 showed that 12.9% of the total variation in percent weight for height could be explained by the relationship of percent weight for height to settlement. Unfortunately, Oxfam Port Sudan did not have the facilities to be able to carry out multiple regression on the nutritional surveillance data, so it was not possible to determine the amount of variance in percent weight for height explained by all the variables examined or their interrelationships.
g. Comparison of Oxfam nutritional surveillance results with Sudan Emergency and Recovery Surveillance System (SERISS) results.
The Republic of Sudan Ministry of Health and USAID conducted four rounds of nutritional surveillance of children under five years old in the six northern regions of Sudan between May 1986 and July 1987. Red Sea Province was included in the surveys. Figure 5.7 shows the comparison of weight for height zscores for children under five years old collected by SERISS, and weight for height zscores for children less than or equal to 115 cm in height collected by Oxfam Port Sudan.
SERISS found consistently higher weight for height zscores than Oxfam Port Sudan; children surveyed by SERISS had a better nutritional status than children surveyed by Oxfam Port Sudan. In the SERISS surveys, however, standard deviations of weight for height zscores were as high as 2.73, whereas standard deviations for weight for height zscores in the Oxfam surveys were close to 1.0. A similar seasonal pattern of changes in weight for height zscores was seen in both surveys. Other SERISS findings which were similar to those of Oxfam Port Sudan included:
1. Nutritional status improved from 1986 to 1987.
2. High malnutrition rates were seen in the 1221 month age group.
3. The smallest unit of analysis (the village council) was the single most important variable determining nutritional status.