You are on page 1of 6

Community and International Nutrition

Daily Iron Supplementation Is More Effective than Twice Weekly Iron Supplementation in Pregnant Women in Pakistan in a Randomized Double-Blind Clinical Trial1
Zubia Mumtaz,2 Saqib Shahab, Naila Butt, M. Abdur Rab and Aime DeMuynck
Health Services Academy, West Blue Area, Islamabad, Pakistan.
ABSTRACT In the context of limited effectiveness of iron supplementation programs, intermittent iron supplementation is currently under debate as a possible alternative strategy that may enhance the effectiveness of operational programs. This eld-based trial assessed the outcome of twice weekly iron supplementation compared to daily in Pakistan. A double-blind, randomized, clinical trial was conducted in Northern Pakistan. Anemic pregnant women (n 191) were assigned to receive daily (200 mg ferrous sulfate) or twice weekly (2 200 mg ferrous sulfate) iron supplementation. Hemoglobin was measured at baseline and at 4-wk intervals for up to 12 wk. Serum ferritin was measured at baseline and 8 or 12 wk. Analysis was by intention to treat. The two groups did not differ in age, parity, sociodemographic characteristics, hemoglobin or serum ferritin concentrations at baseline. Women who received iron daily had a greater rise in hemoglobin compared with women who received iron twice weekly (17.8 1.8 vs. 3.8 1.2 g/L, P 0.001). The serum ferritin concentrations increased by 17.7 3.9 g/L (P 0.001) in the daily supplemented group and did not change in the twice weekly group. Daily iron supplementation remained superior to twice weekly supplementation after controlling initial hemoglobin Z-scores and duration of treatment. The body mass index (BMI) modied the effect of daily versus twice weekly iron supplementation. For every unit increase in BMI, the difference between the two treatment groups was reduced by 0.0014 (nal hemoglobin Z-score; P 0.027). We recommend continuation of daily iron supplementation as opposed to intermittent iron supplementation in pregnant women in developing countries. J. Nutr. 130: 26972702, 2000. KEY WORDS:

Downloaded from jn.nutrition.org by guest on November 28, 2010

anemia

iron supplementation

pregnant women

intervention trial

Pakistan

Iron deciency anemia is still a major nutritional and public health problem in developing countries including Pakistan (UNICEF 1994). Pregnant women and preschool children are most commonly affected (Galloway and MCGuire 1994). It is estimated that 50% of pregnant women in developing countries and up to 80% in South Asia have iron deciency anemia (WHO 1998). The consequences of maternal anemia may be serious, with reported associations ranging from preeclampsia to low birth weight and increased risk of maternal deaths (Quillan et al. 1983, Sapre and Joshi 1996, Spinillo et al. 1994). The maternal and child health services of Pakistan distribute oral therapeutic iron supplements through the existing primary health care system. The aim of this program is to reduce the prevalence of anemia in this vulnerable group of women. The effectiveness of this intervention is often poor. Factors include low utilization of services (Pakistan Medical Research Council 1998), lack of compliance (Galloway and MCGuire 1994, Schultink et al. 1993, Simmons and Cook 1993) and varying bioavailability of iron in the iron prepara-

1 The Pakistan Medical Research Council funded this study (Project No. P/61). 2 To whom correspondence should be addressed: Center for Population Studies, London School of Hygiene and Tropical Medicine, 49-51 Bedford Square, London WCIB 3DP.

tions dispensed (Dawson et al. 1998). In the context of ineffective iron supplementation programs, intermittent iron supplementation appeared as a possible alternate strategy. There is concern that compliance is reduced by the undesirable side effects of the iron supplements. These side effects are related to the dosage and formulation of the therapy (Rani et al. 1995). These concerns have led to studies looking at intermittent versus daily iron dosing in an attempt to identify the optimum therapeutic protocol. A study in rats demonstrated that iron supplementation timed to match mucosal renewal is more efcient (Viteri et al. 1995). Iron absorption is suppressed for at least 24 h after consumption of a high iron meal or iron supplement, principally by controlled suppression of intestinal mucosal cell uptake (Fairweather-Tait et al. 1985 and 1986, ONeill-Cutting and Crosby 1987, Solomans et al. 1983). It has been suggested that less frequent supplementation may reduce the transient iron overload and may be as benecial to the subject as routine daily supplementation (Liu et al. 1995). This might also reduce the unpleasant side effects such as nausea and epigastric pain by reducing the iron content in the gut on a day-to-day basis. It has been demonstrated in anemic rats that iron supplements every 3rd d had a similar effect as in those receiving a daily supplement (Wright and Southon 1990). This phenomenon also was true in preschool children in whom the treat-

0022-3166/00 $3.00 2000 American Society for Nutritional Sciences. Manuscript received 14 December 1999. Initial review completed 10 April 2000. Revision accepted 21 July 2000.. 2697

2698

MUMTAZ ET AL.

ment effect of the two dosage schedules was the same after controlling for initial hemoglobin (Schultink and Gross 1995). Studies in pregnant and nonpregnant women suggested that supplementation on a weekly basis was as effective as daily supplementation in improving the iron status of those with moderate anemia (Gross et al. 1994, Ridwan et al. 1996). Currently, the debate on the relative merits of daily versus intermittent iron supplementation revolves around the balance between the less-than-optimum benet of intermittent iron supplementation and the efciency of a large-scale program through the primary health care system. The International Nutritional Anemia Consultative Group, WHO and UNICEF are considering the option of intermittent iron supplementation as a therapeutic protocol that may enhance the effectiveness of large-scale direct supplementation programs (Stoltzfus and Dreyfuss 1998). We argue that this could result in underdosing of unidentied vulnerable iron-decient and anemic subgroups, particularly in developing countries like Pakistan where women are poorly nourished and eat a high phytate diet and where antenatal and obstetric services are mal-distributed, of varying quality, and underutilized (Pakistan Medical Research Council 1998). Our study was designed to test whether twice-weekly iron supplements would be as effective as daily iron supplements in treating anemia in pregnant women in the Pakistani primary health care system. MATERIALS AND METHODS
A randomized, double-blind, placebo-controlled trial was conducted. Between March and November 1998, 191 pregnant women were selected from the maternity wing of the Federal Government Services Hospital, Islamabad, and the Maternal & Child Health Clinic of the Christian Mission Hospital, Taxila. The former is a tertiary care hospital in the capital, whereas the latter is a mission hospital serving a peri-urban and rural population. Sample size estimations indicated that 164 women were required to be able to distinguish a difference between groups in hemoglobin of 5 g/L (5% signicance level, power 90%). Assuming a dropout rate of 20%, we planned to recruit 200 women. Two hundred forty-one women were screened, 191 of whom were eligible and enrolled in the study. The purpose of the study was explained and informed consent was obtained. The selection criteria were as follows: between the age of 1735 y, with an initial hemoglobin of 110 g/L, an uneventful obstetrical history with no history of placenta previa or abruptio placenta, no history of previous cesarean section, parity of not 5 and a xed address at which the subject could be followed up at home. This stringent criterion was to ensure that women with high-risk pregnancies would not be enrolled. The ethics committee of the Health Services Academy, Ministry of Health, approved the study protocol. The women were randomly distributed into two groups, one receiving daily, and the other twice-weekly iron supplementation. Randomization was performed using a random number generator, and each subject was assigned a unique identier. The subjects and the investigator were blinded to the allocation of treatment group (daily versus twice weekly) at initial recruitment and the three follow-up visits. The randomization code was opened only after the follow-up for all patients had been completed. Each woman was given a 4-wk treatment pack and health education regarding the importance of diet in pregnancy. Capsules containing 200 mg of ferrous sulfate (60 mg elemental iron) and 1 mg of folic acid had been especially manufactured and packed in blister packs (Hakimsons Chemical Industries, Karachi, Pakistan). One set contained two capsules of 200-mg ferrous sulfate for days 1 and 4 and placebo for the rest of the days. One mg of folic acid was present in all the capsules. The other set consisted of identical-appearing strips, which contained ferrous sulfate and folic acid for all 7 d. The appearance of the capsules and the blister packs of the two groups was identical. A predesigned, pretested questionnaire was lled out at the initial

visit. Data included age, education levels of the women and their husbands, household income, family size and past obstetric history. Women were weighed on a digital scale accurate to 100 g (Measurement Concepts, North Bend, WA). Height was measured to the nearest centimeter using a height rod (Cardinal, Webb City, MO). Gestational age was determined by the date of the last menstruation and fundal height. The fundal height was taken with a tape measure from the top of the pubic bone to the fundus of the uterus. After a 20 22 wk gestation, the fundal height in centimeters ( 2 cm) was taken as equal to the number of weeks of gestation. In case of a discrepancy, an ultrasound conrmation was done. Venous blood samples were taken for complete blood count at each visit and for serum ferritin at the 1st, 3rd and 4th visits. Blood for serum ferritin was drawn at the 3rd visit in case the women delivered before their nal visit. Women who were late for their follow-up visit were paid a home visit. Complete blood count was done using a Sysmex SF 3000 autoanalyzer (Sysmex Corporation of America, Long Grove, IL). A peripheral lm was made to rule out congenital disorders such as thalassemia minor. Serum ferritin was determined using a commercial kit (Enzymun-Test Ferritin; Roche Diagnostics GmbH, Mannheim, Germany). A validation protocol was followed to conrm the validity and reliability of the laboratory measurements. Blood samples from every 20th woman were distributed in two microvials and analyzed in the same batch. The serum ferritin assay was validated using the ES 300 testing procedure according to the manufacturers instructions. The lower limit of measuring the concentration of ferritin range was set at 10 g/L, with a reference range of 8 140 g/L (women aged 20 25 y). Socioeconomic status was measured as a complex variable consisting of component variables that included education of the subject, education of husband, their occupations, monthly income, type of construction of the house, number of rooms and ownership of transport. A scoring system was developed, with each component variable assigned a weight (Patil 1995). A complex variable reecting socioeconomic status, with a possible range of 0 42, was developed by the addition of all component values. The number of days for which the drugs were dispensed was recorded once a woman returned for her next batch of drugs. An attempt was made to verbally verify compliance, and often the patients were keen to show the used blister packs, thinking that this would ensure dispensation of another packet. This variable thus served as a proxy indicator of compliance with the study protocol and in turn with drug ingestion. The results were analyzed according to the intention to treat principle. To address the complex issue of physiological anemia of pregnancy (which produces a U-shaped distribution of mean hemoglobin with gestational age) and changes in the magnitude of iron demand related to gestational age, hemoglobin Z-scores were calculated. This was done by converting the observed hemoglobin values, subject by subject, to a Z-score relative to an external reference curve adjusted for gestational age (Beaton and MCCabe 1999). The CDC reference hemoglobin values (developed from reported mean concentrations in iron supplemented healthy pregnant women in Europe) were used as the standard (data courtesy of Ray Yip cited in Beaton 1999). The nal hemoglobin Z-score was taken as the outcome measure. A multiple linear regression model was tted to assess the effect of the two treatment types, adjusting for covariates such as initial hemoglobin Z-score, body mass index (BMI), total parity, time since previous pregnancy, socioeconomic status, duration of follow-up in weeks, number of days for which drugs were dispensed and the total dosage of iron given. Two-way interaction terms of treatment and signicant covariates were generated. Kaplan-Meier analysis was performed to determine whether there was a signicant difference in the probability of improvement relative to time in the two groups. The objective of this exercise is to explore difference in treatment effect, controlling for the time from the start of treatment to the time to attain a hemoglobin concentration of 110 g/L. Originally, survival analysis was developed to measure survival time, with death as the outcome, but the methods can be applied to situations in which the outcome is positive such as cure (Collett 1994). Log-rank test was done to assess the equality of success across the two groups (STATA,

Downloaded from jn.nutrition.org by guest on November 28, 2010

DAILY IRON SUPPLEMENT MORE EFFECTIVE IN PREGNANT WOMEN

2699

TABLE 1
Selected characteristics of the women at the beginning of the study1
Population characteristics n Age, y Women with 5 y of school attendance, % Socioeconomic status (SES)2 Parity Time since last pregnancy, mo Weight, kg Height, m Initial hemoglobin, g/L Initial ferritin, g/L Dependants/family, n Gestational age at beginning of study, wk Duration of iron supplementation, wk Daily dose 84 24.2 4.6 24.4 63 19 6.5 1.8 (05) 20.7 (1103) 56.5 9.2 1.55 0.05 93.2 13.9 23.4 29.4 7.6 (230) 21.9 (1032) 10.7 (418) Twiceweekly dose 76 4.0 51 20 5.9 1.6 (05) 16.3 (166) 56.2 10.2 1.55 0.07 95.8 10.0 22.5 33.2 7.6 (222) 21.6 (1030) 10.9 (416)

1 Values are mean SD or range. 2 SES scale was a continuous 0 42 point scale.

version 5.0; STATA, College Station, Texas). Only differences at 0.05 are reported.

RESULTS A total of 191 women were prospectively recruited. Of these, 160 were successfully followed for at least 4 wk and supplemented for an average of 10.9 wk without supervision (Table 1). Fifty-ve percent completed the entire duration of follow-up (i.e., four follow-up visits), although some were as late as 23 wk for their appointments. Fifteen percent of the women recruited did not return for even one visit and were excluded from the analysis. The remaining 30% did not complete the entire 12 wk of planned follow-up due to a variety of reasons including delivery at home, moving to their parental house towards the end of term, expense involved in attending routine antenatal visits, inability of patients and hospital staff to appreciate the importance of regular follow-up, and difculty of identifying correct addresses for home visits if the patient did not keep an appointment. There was no difference between the women who dropped out compared with those who continued in the trial in terms of age, initial hemoglobin, parity and duration of pregnancy (Table 2). Similarly, there was no difference between the women who dropped out in the two treatment groups (41 versus 45). Ninety-ve percent of the women were housewives. Fortythree percent had received no formal education. Their husbands were better educated, although 19% had no formal education. Most of the husbands were skilled or unskilled workers or ofce clerical staff, and 92% belonged to the low income group (earning less than $130 per month, to support an average family size of seven). The two groups did not differ in socioeconomic status. At the start of the study, there was no signicant difference between initial hemoglobin concentrations of the two groups (P 0.076), although mean hemoglobin concentration of the daily group was marginally lower (Table 3). Similarly, serum ferritin concentrations of the two groups did not differ (P 0.90). After an average 10.9 wk of iron supplementation, there was a signicant improvement in the hemoglobin con-

centrations in both groups. In the daily group there was a 17.8 g/L increase in hemoglobin (P 0.001); the comparative increase in the twice-weekly group was 3.8 g/L (P 0.0037). Similarly, the serum ferritin concentrations in the daily group increased by 17.7 g/L (P 0.001), whereas in the twiceweekly group it did not change (P 0.16). Thus, both hemoglobin and serum ferritin increased signicantly more in the daily group than in the twice-weekly group. The nal multiple linear regression model (Table 4) showed a signicant association of nal hemoglobin Z-score with daily versus twice-weekly treatment (P 0.001), initial hemoglobin Z-score (P 0.001), BMI (P 0.02) and the number of weeks that the drug was dispensed (P 0.03). Socioeconomic status, total parity, time since last pregnancy and age were not signicantly associated with nal hemoglobin Z-score. In pregnant anemic women daily iron therapy caused a signicantly greater rise in hemoglobin and serum ferritin than twice-weekly therapy (P 0.001). This difference in the rise of hemoglobin in the two groups persisted after controlling for signicant covariates that include initial hemoglobin Z-score and the duration for which the drugs were dispensed. The BMI modied the effect of daily and twice-weekly iron supplementation. For every unit increase in BMI, the difference between the two treatment groups was reduced by 0.0014 (nal hemoglobin Z-score, P 0.027). Women with a lower BMI had a signicantly greater increase in nal hemoglobin Z-score in the daily group than in the twice-weekly group. As BMI increased, the difference between the two treatment groups narrowed. The model predicted that at a BMI of approximately 30 kg/m2, there would be no signicant difference between the two treatment protocols. The initial hemoglobin Z-score and the duration of supplementation did not interact signicantly with the treatment type (Table 4). If, using WHO criteria, we take hemoglobin 110 g/L as normal, then the Kaplan-Meier analysis showed that women in the daily supplementation group attained normal hemoglobin concentrations in a shorter period of time than the women in the twice-weekly group (P 0.01; Fig. 1). Women in the daily supplemented group achieved normal hematological status in 44 d on average, compared with 73 d in the twiceweekly group. DISCUSSION Intermittent iron dosing to conform to the cycle of intestinal mucosa turnover has been suggested as a method to

Downloaded from jn.nutrition.org by guest on November 28, 2010

TABLE 2
Comparison of subjects lost to follow-up with those who completed the trial
Population characteristics1 n Age, y Socioeconomic status score Parity, n Time since last pregnancy, mo Body mass index, kg/m2 Initial hemoglobin, g/L Dependants/family, n Duration of follow-up, wk Completed 105 24.1 4.0 19.0 6.4 1.7 (05) 19.4 (1103) 23.1 3.7 94.2 12.4 7.3 (230) 12.4 1.45 Lost 86 24.7 4.8 18.0 5.3 1.9 (05) 19.0 (172) 23.7 3.7 94.0 12.9 7.5 (222) 7.8 3.1 P

0.44 0.16 0.38 0.89 0.34 0.90 0.70 0.00

1 All values are mean (range) or SD. 2 P-values are calculated using Students t test.

2700

MUMTAZ ET AL.

TABLE 3
Hematological values at baseline and follow-up contact points in women supplemented with iron daily or twice weekly1
Daily Hemoglobin Baseline First assessment Second assessment Final assessment Serum Ferritin Baseline Last assessment point
1 Values are mean

Twice weekly g/L n 91 76 61 50 n 61 61 23.0 27.6 g/L 95.8 96.2 97.0 100.9 10.6b 10.5b 11.4b 12.3a g/L 33.7 31.5 0.8 14.0* Hg, g/L 3.2 5.3* 11.7* 12.6*

wk 0 4 8 12

n 100 84 68 55 n

92.6 101.6 108.7 113.6

14.1b 15.2a 17.2a 18.3a g/L

0 8 or 12

68 68

23.8 41.6

29.7 34.9

SD. Hg, hemoglobin. * Signicantly different, P 0.05 (Students t test). a,b Values marked a are signicantly greater than values marked b (Students t test, P

0.05).

improve the efciency of uptake of therapeutic doses of iron (Solomons 1995). Previous studies have shown no signicant difference between daily and twice-weekly regimens in children, nonpregnant women and women in early pregnancy (Berger et al. 1997, Gross et al. 1994, Liu et al. 1995 and 1995, Ridwan et al. 1996, Schultink and Gross 1995). These studies have become the basis of promoting twice-weekly iron therapy (Baynes and Cook 1996). However, a critical review of these ve published studies showed that the subjects had low-grade anemia initially and that very small doses of iron were needed for them to reach their optimal hemoglobin concentrations (Hallberg 1998). These may be the reasons that no signicant differences were observed between the two treatment regimens. The only study published to date that compares weekly and daily administration of iron to pregnant women included both anemic and nonanemic women (Ridwan et al. 1996). Although this study found the regimens equally effective, the nal hemoglobin concentrations were lower than any value reported in the many well-controlled studies in developing and developed countries; it appears that the regimens were equally TABLE 4
Multiple regression model comparing the effect of daily vs. twice-weekly iron supplementation regimens in women with nal hemoglobin Z-score1 as the outcome measure
Covariates Treatment type Daily therapy Twice-weekly therapy Initial hemoglobin, Z-score BMI, kg/m2 Duration of ingestion of iron supplements, wk Treatment BMI Treatment initial hemoglobin Treatment duration of ingestion
1 2 3 4

ineffective (Hallberg 1998). The results of our study, in which the daily iron supplemented group had a higher nal hemoglobin Z-score than the twice-weekly group, are consistent with a meta-analysis of 21 published and unpublished studies. Beaton and MCCabe (1999) found, upon pooling the data, that there was a signicant difference between the two treatment groups, with higher nal hemoglobin concentration associated with daily administration of iron. This was especially notable in pregnant women, although it was true across all groups. Signicant differences were not seen in all of the studies, possibly because of limitations of statistical power. Our study subjects consisted of poor illiterate anemic women who were self-selected to attend an antenatal clinic. All the women in this motivated self-selected group were anemic, with an initial hemoglobin of 110 g/L. The hemoglobin response was clearly superior in those who received iron therapy daily. Although the increase in hemoglobin in the twice-weekly group was signicant, it was only 3.8 g/L and was of negligible clinical importance. The gain of 17.8 g/L in the daily group was both statistically signicant and clinically important. Despite large variations in absolute serum ferritin concentrations, they closely followed the hemoglobin concentrations in the two groups. These results offer strong evidence that the daily administration of iron has a greater impact on body iron than twice-weekly supplementation. Equally important, women in the daily supplementation group achieved

Downloaded from jn.nutrition.org by guest on November 28, 2010

-coefcient

P-Value

0.015 ( 0.019, 0.10)2 0.65 (0.5, 0.83) 0.0015 (0.0006, 0.002) 0.0008 (0.0006, 0.002)3 0.0014 ( 0.003, 0.0002)

0.001 0.001 0.002 0.03 0.027 NS4 NS

Adjusted for gestational age. Values in parentheses are the 95% condence intervals. Values are small because the outcome is a Z-score. Not signicant, p 0.05. BMI, body mass index.

FIGURE 1 Kaplan-Meier estimate of time to attain hemoglobin concentration 11 g/L in pregnant women given daily and twiceweekly iron supplements.

DAILY IRON SUPPLEMENT MORE EFFECTIVE IN PREGNANT WOMEN

2701

hemoglobin concentrations 110 mg/L in a shorter time than women in the twice-weekly group. This is important, because time is a limiting factor in our attempts to normalize the hematological status of pregnant women. This study was a double-blind, randomized, eld-based trial that closely resembled real-life conditions in Pakistan. Tablet ingestion was not supervised, an aspect that is not sustainable or cost-effective in large-scale programs. The number of women followed kept decreasing with time, indicating that they had difculty adhering to the treatment protocol. Because randomization was successful, and because there was no difference in the characteristics of dropouts in the two groups, we assume that the groups remained comparable throughout the study. Only one patient mentioned side effects of the tablets as the reason for dropping out of the study; others were mostly social or nancial reasons. The patients often would not come on the exact day and date specied. A home visit was then made, but we were not always successful in contacting the patient. There may be several reasons for the loss to follow-up. Delivery traditionally takes place in the maternal home and some patients moved at the end of term to geographically distant locations, thus ending effective follow-up. One MCH center attended government employees free of cost. Some patients were ctitiously registered under the name and address of a government employee. The poorest group, which included displaced Afghan refugee women, had unreliable addresses. Opportunity costs of taking the time, effort and expense of traveling to the MCH center were perceived to be higher than the potential health benet of the free pills provided. Moreover, at one of the MCH centers, there was a nominal user fee of $0.20 that was not waived for the study. Arrangements had been made for blood to be taken when presented in labor while enrolled in the trial; although some samples were obtained, many women chose to deliver at either their own or their parents home. The analysis of the results was done on the intention-totreat basis, thus including patients who dropped out after the 1st or 2nd visit and did not complete the full 12 wk. The results of this analysis should therefore more closely match the actual impact of an iron supplementation program in poor, predominantly illiterate, anemic pregnant women with a reasonable access to free or subsidized antenatal services. Iron supplementation works in a complex multidimensional context that includes sociocultural, economical and political facets. It is important to look at the effectiveness of otherwise efcacious regimens in eld conditions and not make decisions based on perceived efciencies alone. Pregnant women are routinely anemic due to iron deciency in developing countries such as Pakistan (National Nutrition Survey 1987). The proximate determinants include poor nutrition (Pakistan Integrated Household Survey 1991), frequent and closely spaced pregnancies (Pakistan Fertility and Family Planning Survey 1998), low rate of iron supplementation before and during pregnancy (Tinker and Koblinsky 1993), worm infestations, a high phytate diet (Fox et al. 1998) and concurrent calcium administration (Whiting 1995), a norm in Pakistan. Although the causal pathway of these is fairly well established, there is a complex set of distal determinants, which has an impact on womens health, including a low status of women leading to low self-esteem and a lack of empowerment in decision making regarding their health and reproductive choice (Sattar and Kazi 1997), poverty (Mahbub ul Haq 1997) and misconceptions about the nature and extent of the womens health problems by the policy-makers including professionals and failure to mobilize resources to adequately address the need (De Brouwere et al. 1998). There is an extensive network of

basic health units and community health workers in Pakistan with the purported aim of providing primary health care. Their actual impact on womens health is low, as can be gauged by statistics such as a maternal mortality rate of 340/100,000 live births and a life expectancy of 63 y (UNDP 1996). The interaction between BMI and type of treatment protocol underscores the importance of nutrition status. Women with a low BMI had a signicantly greater increase in nal hemoglobin when taking an iron supplementation pill daily than those supplemented twice weekly. The mean BMI of our study sample was 23.5 kg/m2, and according to FAO, large parts of the Asian populations have low BMI (1994). An inadequate diet is still the most common reason for the high prevalence of anemia in women in Pakistan (Karim et al. 1988). The existence of mucosal block with the administration of oral iron supplements has been challenged (Benito et al. 1998, Hallberg 1998). To suggest an across-the-board change in treatment protocol to reduce possible side effects in poorly nourished, high-risk populations to gain some physiological advantage in iron absorption might be premature. The critical issue of efciency of iron supplementation programs was not addressed in the present study. Some investigators accept that, although daily supplementation may be more efcacious, intermittent supplementation is more efcient and cost-effective in large-scale programs (Schultink and Gross 1999). Further research is recommended to assess the effectiveness and efciency of the two treatment types in different groups at risk under eld conditions. Until then, based on the present study and as long as the current situation prevails, our position is that the current policy of daily iron supplementation in pregnancy should continue and that efforts be made to nd other ways to improve the effectiveness of iron supplementation programs. ACKNOWLEDGMENTS

The authors thank the women who participated in this project for their compliance and patience, the staff of the Christian Mission Hospital, Taxila, and the Maternity Wing of the Federal Government Services Hospital, Islamabad, for their cooperation. We are grateful to Shifa International Hospital Laboratory for their contribution in ensuring quality control of all laboratory measurements and accepting blood samples at all hours of the day or night. A special thanks to S. Siddiqui for his constant guidance at all stages of the project. J. Lall, E. Lall, J.Chaudhry and R. Hameed made it possible for us to collect the data.

Downloaded from jn.nutrition.org by guest on November 28, 2010

LITERATURE CITED
Baynes, R. D. & Cook, J. D. (1996) Current issues in iron deciency. Curr. Opin. Hematol. 3: 145149. Beaton, G. & MCCabe, G. (1999) Efcacy of intermittent iron supplementation in the control of iron deciency anemia in developing countries: an analysis of experience. Final Report to the Micronutrient Initiative. GHB Consulting, Toronto. Benito, P., House, W. & Miller, D. (1998) Comparison of oral and intraperitoneal iron supplementation in anaemic rats: a re-evaluation of the mucosal block theory of iron absorption. Br. J. Nutr. 79: 533540. Berger, J., Aguayo, V. M., Tellez, W., Lujan, C., Traissac, P. & San Miguel, J. L. (1997) Weekly iron supplementation is as effective as 5 day per week iron supplementation in Bolivian school children living at high altitude. Eur. J. Clin. Nutr. 51: 381386. Collett, D. (1994) Modelling survival data in medical research. Chapman & Hall, London. Dawson, E. B., Dawson, R., Behrens, J., DeVora, M. A. & MCGanity, W. (1998) Iron in prenatal multivitamin/multimineral supplements. J. Reprod. Med. 43: 133140. De Brouwere, V., Tonglet, R. & Lerberghe, W. V. (1998) Strategies for reducing maternal mortality in developing countries: what can we learn from the history of the industrialized West? Trop. Med. Intern. Health 13: 771782. FAO (1994) Body mass index. A measure of chronic energy deciency in adults. Food and Nutrition paper 56. Rome: FAO. Fairweather-Tait, S. J., Swindell, T. E. & Wright, A.J.A. (1985) Further studies

2702

MUMTAZ ET AL.
Ridwan, E., Schultink, W., Dillon, D. & Gross, R. (1996) Effects of weekly iron supplementation on pregnant Indonesian women are similar to those of daily supplementation. Am. J. Clin. Nutr. 63: 884 890. Sapre, S. & Joshi, V. (1996) 3 Es for safe motherhood. J. Obstet. Gyn. Fam. Wel. 2: 711. Sattar, Z. & Kazi, S. (1997) Womens autonomy, livelihood and fertility. Pakistan Institute of Development Economics. Islamabad, Pakistan. Schultink, W. & Gross, R. (1995) Effect of daily vs. twice weekly iron supplementation in Indonesian pre-school children with low iron status. Am. J. Clin. Nutr. 61: 111115. Schultink, W. & Gross, R. (1999) Use of daily compared with weekly iron supplementation: apples and pears. Am. J. Clin. Nutr. 69: 739 742. Schultink, W., Van Der Ree, M., Matulessi, P. & Gross, R. (1993) Low compliance with an iron-supplementation program: a study among pregnant women in Jakarta, Indonesia. Am. J. Clin. Nutr. 52: 135139. Simmons, W. & Cook, J. (1993) Evaluation of a gastric delivery system for iron supplementation in pregnancy. Am. J. Clin. Nutr. 58: 622 626. Solomons, N. W. (1995) Weekly versus daily oral iron administration: are we asking the right questions? Nutr. Rev. 53: 326 327. Solomons, N. W., Pineda, O., Viteri, F. & Sandstead, H. H. (1983) Studies on the bio-availability of zinc in human: mechanism of the intestinal interaction on non-heme Fe and Zinc. J. Nutr. 113: 337349. Spinillo, A., Capuzzo, E., Piazzi, G., Nicola, S., Colonna, L. & Iasci, A. (1994) Maternal high-risk factors and severity of growth decit in small for gestational age infants. Early Hum. Dev. 38: 35 43. Stoltzfus, R. & Dreyfuss, M. (1998) Guidelines for the use of iron supplements to prevent and treat iron deciency anemia. International Nutritional Anemia Consultative Group. Washington, D.C. Tinker, A. & Koblinsky, M. (1993) Making motherhood safe. World Bank Discussion Papers 202. The World Bank. Washington, D.C. United Nations Childrens Fund. (1995) Annual Report 1994. p. 52. UNICEF, New York. United Nations Development Program. (1997) Human Development Report 1996. p. 155. Oxford University Press, New York. Viteri, F. E., Liu, X., Tolomei, K. & Martin, A. (1995) True absorption and retention of supplemental iron is more efcient when iron is administered every three days rather than daily to iron-normal and iron-decient rats. J. Nutr. 125: 8291. Whiting, S. J. (1995) The inhibitory effect of dietary calcium on iron bioavailability: a cause for concern? Nutr. Rev. 53: 77 80. World Health Organization. (1998) The World Health Report 1997: conquering suffering, enriching humanity. WHO, Geneva, Switzerland. Wright, A. J. & Southon, S. (1990) The effectiveness of various iron-supplementation regimens in improving the Fe status of anemic rats. Br. J. Nutr. 63: 579 585.

on rats on the inuence of previous intake on the estimation of bio-availability of Fe. Br. J. Nutr. 51: 185191. Fairweather-Tait, S. J. & Minski, M. J. (1986) Studies on iron-availability in man using stable isotope techniques. Br. J. Nutr. 55: 279 285. Federal Bureau of Statistics. (1991) Pakistan integrated household survey. Government of Pakistan. Islamabad, Pakistan. Fox, T. E., Eagles, J. & Fairweather-Tait, S. J. (1998) Bioavailability of iron glycine as a forticant in infant foods. Am. J. Clin. Nutr. 67: 664 668. Galloway, R. & MCGuire, J. (1994) Determinants of compliance with iron supplementation: supplies, side effects, or psychology? Soc. Sci. Med. 39: 381390. Gross, R., Schultink, W. & Juliawati, R. (1994) Treatment of anemia with weekly iron supplementation. Lancet 344: 821. Hallberg, L. (1998) Combating iron deciency: daily administration of iron is far superior to weekly administration. Am. J. Clin. Nutr. 68: 213217. Karim, S. A., Khurshid, M. & Rizvi, J. H. (1988) Anaemia in pregnancy: occurrence in two economically different clinic populations of Karachi. JPMA. 38: 271272. Liu, X., Yang, W. & Song, Y. (1995) Evaluation of the effects of intermittent iron supplements on iron-deciency anemia in children. Chung Hua Yu Fang Hsueh Tsa Chih 29: 34 37. Liu, X.-N., Kang, J., Zhao, L. & Viteri, F. (1995) Intermittent iron supplementation in Chinese pre-school children is efcient and safe. Food Nutr. Bull. 16: 139 146. Mahbub ul Haq. (1997) Human Development Report in South Asia. Oxford University Press, Karachi, Pakistan. National Institute of Health, National Nutrition Survey. (1987) Nutritional Division, Government of Pakistan. Islamabad, Pakistan. National Institute of Population Sciences and the London School of Hygiene and Tropical Medicine. (1998) Pakistan fertility and family planning survey. Ministry of Population Welfare, Government of Pakistan, Islamabad. Pakistan. O Neill-Cutting, M. A. & Crosby, W. H. (1987) Blocking of iron absorption by a preliminary oral dose of iron. Arch. Intern. Med. 147: 489 491. Pakistan Medical Research Council. (1998) National Health Survey of Pakistan: Health Prole of the People of Pakistan. Islamabad, Pakistan. Patil, R. L. (1995) Measurement of household socio-economic status. Demo. Ind. 24: 259 268. Primary Health Care Cell, Federal Ministry of Health. (1996) Second quarterly HMIS Feedback Reports of the Health Management Information System of Pakistan. Government of Pakistan. Islamabad, Pakistan. Quillan, J. P., Kwong, A. & Passmore, P. (1983) An epidemiological investigation of pre-eclampsia and elevated blood pressure among Kampuchean refugee women at Sakaeo Holding Center, Thailand. J. Trop. Med. Hyg. 86: 185191. Rani, N. V., Pandey, J., Das, B., Shruti Talib, V. H., Singh, K. & Bagati, A. (1995) Pregnancy associated anemia and iron: a pilot study. Indian J. Pathol. Microbiol. 38: 293297.

Downloaded from jn.nutrition.org by guest on November 28, 2010

You might also like