Even today nearly 1.5 billion people all over the world are affected by iron deficiency anemia (IDA). In India alone, depending on age and sex, IDA has been reported to range from 38-72%(1), majority of them being women and children. The IDA prevalence rate beyond the age of 6 years(2), increases in girls. This could be due to certain factors like menstruation, gender discrimination in intra-household food allocation and early marriage leading to early pregnancy. Of the total population, adolescent girls form 22% and estimates suggests that about 25-50% girls become anemic by the time they reach menarche(2,3).

A study on adolescents in China revealed that 61.8% of girls were anemic(4). Another study(5) showed that iron deficiency was more prevalent in females than males, the highest rate 

being in teenage girls. Thus, in these adolescent girls, the mothers-to-be, who will usher the next generation, pregnancy only serves to aggravate their pre-existing anemia. Under the circumstances, the need to combat anemia in adolescent girls had been suggested by the Nutrition Foundation of India(6). Preventing fetal brain damage due to maternal anaemia in early pregnancy(7) support a strategy that ensures that women have a satisfactory Hb status (11 g/dl as recommended by WHO) even at the beginning of conception.

Various studies had indicated that with continued daily administration, iron absorption could decrease due to "tiredness" of the intestinal mucosa(8,9). According to a study, absorption from a single dose of iron reduces from 30-40% on the first day to as low as 3-6% after a few days of continuous daily administra-tion(9). Studies carried out on pre-schoolers support that iron supplementation once or twice a week, increased their hemoglobin status significantly(10,11).

The bioavailability of iron is low in predominantly cereal-based diets because of their high phytate content. It has been suggested that the bioavailability of iron may be signi-ficantly improved by ascorbic acid supple-mentation(12). As little as 50 mg of ascorbic acid can double iron absorption when taken with a meal(13). The comparative efficacy of iron/folate supplementation with and without the addition of vitamin C to the supplement would seem to need further investigation.

The present study was therefore conducted to obtain baseline data on hemoglobin (Hb) levels of adolescent girls belonging to low socio-economic groups; investigate the com-parative efficacy of once `weekly' and `daily' administration of iron-folate tablets with respect to impact on the Hb levels; and find out the effect of added ascorbic acid supplementation 

investigated. The subjects were divided into the following three treatment groups: Group 1: One tablet of iron-folate (100 mg elemental iron + 500 mg folate) once weekly; Group 2: One tablet of iron-folate Vitamin C (100 mg elemental iron + 500 mg folate + 25 mg vitamin C) once weekly; Group 3: One tablet of iron-folate (100 mg elemental iron + 500 mg folate) daily.

A prospective completely randomized experimental design was utilized for the conduct of the study(14). The subjects in both the urban as well as rural areas were randomly categorized into three experimental/control groups. In the school the tablets were distributed weekly to the subjects by a research worker immediately after lunch time so that it was not taken on an empty stomach. The intake of the tablets by the subjects was assured by a helper. In the rural area, two field workers were responsible for the distribution and ensuring intake of tablets by the subjects only after brunch.

Statistical Analysis

The data were analyzed for both rural and urban group of children(14). The mean±SD of height (%) and weight (%) by grades of Hb values were compared and tested. The gain in Hb levels over periods of study at 3rd and 6th months were studied in all the subjects who participated in the intervention study. The Mean±SD of improvements in hemoglobin for the scales of Hb were found and tested for differences between all the study groups. Analysis of variance and multiple comparison `t' test procedures were utilized(14). The prevalence of anemia with Hb levels below 12 g/dl was found and compared between study groups. Chi square test for significance of differences between proportions was utilized(14). Stepwise multiple regression analysis with multiple corelation co-efficient was utilized to trace the best set of variables associated with hemoglobin levels(14).

Anthropometry

The mean heights and weights of urban and rural girls at different ages were compared with National Center for Health Statistics [NCHS](15) and National Nutrition Monitoring Bureau [NNMB](16) values. The urban sample had a better anthropometric status than the rural one. The results indicated that the rural subjects were shorter and lighter than their urban counterparts. Urban subjects in the study had larger heights and heavier weights whereas the rural girls had lower heights and lower weights than those in the country-wide study of NNMB(16). Also, both rural and urban subjects were significantly shorter and lower in weight than the international standards [NCHS](15). This was true for almost all the age groups.

Hemoglobin Profile

At the baseline, the hemoglobin status of the urban girls was better than their rural counter-parts, who were largely out of school. As can be seen from Table I, the percentage of girls with Hb <12 g/dl in the urban sample was 61.9% as against 85.4% in the rural sample (p<0.001).

Thus, the rural girls were worse off, in comparison to the urban sample, with regard to both the anthropometry and the hemoglobin status. It was observed that the girls with better grades of height (%) or weight (%) had higher hemoglobin value (p<0.05) than those with poor grades. In other words, prevalence of anemia was lower in those who were taller or heavier than those who were shorter or lower in weight for given age. BMI was also positively correlated with the hemoglobin values. The best set of variables associated with hemoglobin in order were height, BMI and age. The multiple corelation coefficient with these variables was 0.260 (p=0.002). The correspondence between anthropometric status and the hemoglobin profile of the two communities would suggest that poor Hb status was a part of general malnutrition attributable to multiple nutrient deficiencies.

Intervention

Of 705 subjects with base-line data, 31.9% had Hb levels of >12 g/dl and were excluded from the intervention study. Of the remaining 479 who were anemic, only 270 (56%) expressed willingness to participate in the study (their features, with respect to height, weight and Hb, were similar to those who were unwilling), 169 in urban Delhi (55, 57 and 55 in groups 1, 2 and 3) and 103 in Rajasthan (28, 38 and 37 in groups 1, 2 and 3). The data of the subjects who were available for all periods of follow-up were taken for the statistical analysis, that is, 164 subjects from the urban sample (group 1, 2 and 3_54, 55 and 55 respectively) and 82 girls from the rural counterpart (groups 1, 2, and 3_22, 32 and 28 respectively) 246 in all.

Effect Of Intervention

The change in the hemoglobin levels brought about by iron/folate supplementation with/without Vitamin C was investigated both in the urban and rural subjects_after three months and after 6 months from the beginning of supplementation (Table II). The Fe/folate supplementation given weekly brought about appreciable change in the mean hemoglobin level of urban subjects, the improvement being 0.62 g/dl at 3 months and 0.79 g/dl at 6 months, as compared to -0.06 g/dl and 0.63 g/dl in the control group; the impact at 3 months was statistically significant (p<0.05). The addition of Vitamin C resulted in significantly greater improvement, namely 1.05 g/dl at 3 months and 1.17 g/dl at 6 months. In rural subjects also Vitamin C supplementation had a significant effect but only by 6 months. Daily Fe/folate was more effective than once-weekly Fe/folate, the mean increase being 0.99 g/dl and 0.71 g/dl at 3 months (p >0.05) and 1.57 g/dl and 0.98 g/dl at 6 months (p <0.05).

Though Table II shows an increase in the Hb levels of the subjects, this however does not indicate the extent to which the iron deficiency anemia had been eradicated. A brief picture of this data is presented in Table III where subjects of both the rural and urban areas have been combined to see the overall effect of supplementation, on the prevalence of anemia. It will be seen from Table III that: (i) the response obtained with Fe/folate given daily was superior to that obtained with Fe/folate supplemented weekly; and (ii) with the inclusion of vitamin C once-weekly supplementation the results obtained were as good as those with daily administration of Fe/folate.

The prevalence of anemia was more or less same in all the age groups of children studied; however, it was more prominent in the subjects who were short and low in weight. Similar findings were noted in rural Hyderabad pre-school children(2). It may be reasonable to 

argue that children who are severely mal-nourished showed a better absorption of micronutrients, including iron, on supple-mentation.

In the present study, there were some non-respondents in both urban and rural areas. The characteristics of the respondents and non-respondents were compared with the available baseline information. The results indicated no 

differences between respondents and non-respondents (p >0.05). Thus, the two groups were similar with respect to height, weight and hemoglobin indicating that the large non-response did not introduce any bias into the comparisons.

Though, the increase in Hb levels achieved through daily iron/folate supplementation was greater than that achieved through once weekly 

supplementation, it is important to note that the response obtained even with once-weekly administration was considerable. From the point of view of feasibility, cost and compliance, once-weekly administration appears suitable as a public-health measure for the control of anemia, especially for the developing countries.

Trials with intermittent doses have been carried out in China, Guatemala, Indonesia, Malaysia, Mali and the USA among various groups including adolescent girls(17). A Chinese study has shown that weekly supplementation with 120 mg iron was more effective than a daily 60 mg dose which in turn was as effective as daily 120 mg dose(17). Another study revealed that weekly 180 mg iron dose was less effective than a daily 60 mg dose(17). A proposed meta-analysis of the above referred approach is required to finally come to a conclusion about the degree of effectiveness of the intermittent dosage.

This study was conducted in two distant parts of North India. The initial phase (first 3 months) of the study was carried out in autumn (September, October, November), during which period, green leafy vegetables were in poor supply and very costly; they did not form a part of the diet of the poor. The second phase of the study (after 6 months) was done in winter (December, January, February) when besides being available in abunance, GLVs were cheap 

and could be afforded almost daily by the poor. The availability of fruits and vegetables rich in vitamin C (e.g., guava, oranges, cabbage, tomatoes) was also high in the winter.

The dietary intake of Vitamin C along with iron rich sources enhances the absorption of iron significantly(13). The results of the study also appear to suggest that anemia in poor-cereal eating population is as much (if not more) due to deficiency of Vitamin C as to deficiency of iron. The difference in the response of the subjects getting Vitamin C supplementation with Fe/folate as compared to those receiving only Fe/folate seems to highlight the above said point. Thus, inclusion of Vitamin C in Fe/folate as part of a Public Health Operation may be recommended. For daily administration, it is a costly affair and thus may not be feasible. On the other hand, in case where administration is once-weekly, this may still be possible. However, we wish to emphasize that the need to improve the diet through increased intake of GLVs and fruits is even more important in the long run to combat anemia.

Thus, the results obtained related to the reduction in prevalence of anaemia with once-weekly Fe/folate with or without Vitamin C supplementation are of interest. Such supplementation program my be carried out through schools and child-welfare centers such as Integrated Child Development Services 

(ICDS) in India. This important program may turn out to be of less fatigue, better productivity and work capacity, improved immune function, reduced risk of reproductive failure and maternal mortality.

The authors are grateful to Dr. C. Gopalan for his guidance and advise and to Dr. Sharda Jain for her help in the field work. The co-operation of the teachers and students of Government Comp (m) Girls Sr. Sec. School No. I, R.K. Puram, Sector II, New Delhi in the urban area and the field staff of the rural unit supported the study.

Contributors: AS co-ordinated the study, collected the data and drafted the manuscript. KP collected the samples and estimated Hb. KVR performed the statistical analysis.

Funding: Nutrition Foundation of India
Competing interests: None stated.