Association between Iron–Folic Acid Supplementation during Pregnancy and Maternal and Infant Anemia in West Java, Indonesia: A Mixed-Method Prospective Cohort Study

ABSTRACT.

The Indonesian government has provided iron–folic acid (IFA) supplementation in response to maternal pregnancy iron-deficiency anemia. However, community-based cohorts on IFA’s effects on maternal and infant anemia are limited. A mixed-method study design with a primary longitudinal cohort was used to observe the association between IFA and anemia in mothers and infants. Iron–folic acid supplementation was observed throughout pregnancy. Anemia status was based on a single hemoglobin assessment using HemoCue Hb 201 + in the second or third trimester of pregnancy for the mother and at birth for the infant. Qualitative data were collected via in-depth interviews (IDIs) and a forum group discussion (FGD). Iron–folic acid supplementation with > 180 tablets throughout pregnancy was associated with lower pregnancy anemia (adjusted relative risk [aRR] = 0.25, 95% CI: 0.092–0.664, P = 0.006) after adjusting for potential confounding variables. Supplementation with IFA was not associated with infant anemia (RR = 1.033, 95% CI: 0.70–1.54, P = 0.873 for 90–180 tablets and RR = 1.07, 95% CI 0.70–1.63, P = 0.774 for > 180 tablets). The IDIs and FGD suggested that IFA and multivitamin content knowledge, IFA consumption monitoring, and paternal involvement were important in IFA supplementation and effectiveness in reducing anemia. Iron–folic acid supplementation was associated with reduced maternal but not infant anemia. Because maternal anemia is associated with infant anemia, an anemia monitoring program for women in early pregnancy is vital in addressing infant health. Paternal involvement was also identified as a major factor in maternal and child health.

INTRODUCTION

The first 1,000 days of life are a critical period that may define future individual well-being.¹ Nutritional anemia remains a concerning nutrition-related public health problem globally, especially in low and lower-middle-income countries.² In 2018, 48.9% of pregnant women in Indonesia had anemia.³ Nutritional problems during pregnancy are associated with offspring health, especially when it links with nutrition in newborns, such as maternal iron and iron storage.^4,5 Maternal anemia before and during pregnancy has been suggested to affect the risk of childhood anemia.^6,7 A previous cohort analysis of the Indonesian population reported that mothers with chronic anemia had an increased risk of childhood anemia.⁶ Moreover, early-life anemia has been reported as the leading cause of mortality in children.⁸

Iron is essential in hemoglobin (Hb) production.⁹ In iron deficiency, the red blood cells (RBCs) have insufficient Hb and a smaller size (microcytic) than normal, resulting in iron deficiency anemia (IDA).⁹ Folic acid, the synthetic form of folate also known as vitamin B₉, is important in RBC maturation. Folate deficiency can lead to anemia due to an immature form of RBC, often called megaloblastic anemia.¹⁰ In addition to RBC maturation, adequate folic acid in early pregnancy also protects against neural tube defects in the fetus.¹⁰

In response to the problem of increasing micronutrient demand in pregnancy and the high prevalence of maternal IDA, the Indonesian government encourages and provides iron folic acid (IFA) supplements to pregnant women. The IFA supplements provided by most community health centers typically contain 60 mg of iron and 400 µg of folic acid, with at least 90 tablets taken during pregnancy.¹¹ However, although the prevalence of morbidity in pregnancy remains high, IFA supplementation is considered low at only 37.7%.³ Moreover, malnourished mothers in Indonesia are a target for supplementary feeding programs, yet compliance is only 25.2%.³ The low reported compliance of IFA supplementation and supplementary feeding amid the high prevalence of anemia in pregnant women indicates a major public health problem.

Maternal supplementations have focused on improving maternal and neonatal outcomes. Several studies on maternal supplementation have examined maternal and child outcomes.^12–14 However, few primary cohort studies have examined early childhood or infancy anemia in the Indonesian population. Prospective studies on pregnancy IFA supplementation and its association with maternal and child anemia are vital to support future policies on maternal IFA supplementation programs. Moreover, this is the first mixed-method cohort study on IFA and its outcome at the community level in Indonesia. Quantitative analysis with a further qualitative approach could provide a better understanding of the reality of IFA program implementation.

MATERIALS AND METHODS

Design.

This study comprised two parts: a quantitative and a qualitative phase. The quantitative phase followed a prospective cohort for a year and was completed before the qualitative phase. The quantitative phase result was triangulated with the in-depth interviews (IDIs) and focus group discussion (FGD) in the qualitative phase. The study phases are presented in Figure 1.

Figure 1.

Cohort timeline. IFA = iron–folic acid.

Quantitative phase.

Study population and setting.

This prospective cohort study was conducted between 2022 and 2023 in three districts with a high incidence of under-5 stunting in Depok, West Java: Pasir Putih, Bedahan, and Pengasinan. The districts selection was based on preselection of districts with high incidence^15,16 and followed by consultation with the Health Office for final area selection. The study population was pregnant women who visited community-based healthcare centers (local term: Posyandu) and their offspring. The participant selection criteria were any pregnant women with a singleton pregnancy who consented to be study participants. Mothers with known pregnancy complications and newborns with birth defects were excluded from this study.

Iron–folic acid consumption was monitored as the independent variable. Maternal pregnancy and infant anemia statuses were monitored as the dependent variable outcomes. Other potential confounding variables included in the analysis were the mother’s sociodemographic characteristics, dietary habits, nutritional status, anemia, IFA-related knowledge, and attitude. Trained enumerators (undergraduate and graduate students in nutrition) collected the data. The questionnaire was validated before data collection with a Cronbach’s alpha of 0.611, which is considered acceptable. The data collection began at the mother’s first pregnancy visit to a Posyandu. Mothers who had visited multiple times before the data collection were asked for their recall on IFA supplementation. Participants were measured for nutritional status and asked to complete a questionnaire about knowledge, attitude, lifestyle behavior, and sociodemographic information. Their subsequent visits monitored IFA consumption and measured the mother’s Hb in the second to third trimester of pregnancy (mean gestational age 26.2 ± 8.2 weeks). The data on the infant anemia outcome was collected by a door-to-door visit after delivery. A minimum sample of 167 women was required based on a sample size calculation using the 87.6% proportion of pregnant women receiving IFA reported in the last national report before data collection.³ The margin of error in this study was 4.3% based on the 95% confidence level and a sample size of 225.

IFA consumption assessments.

Participants were asked about their IFA consumption during their pregnancy. Their reported consumption was confirmed by cross-checking with the Mother and Child Health (KIA) book, a health record completed by health providers for pregnant mothers and their infants. The enumerators also periodically checked the number of IFA left since the previous visit to confirm the actual consumption. Iron–folic acid adherence was grouped based on different recommendations for IFA consumption: < 90, 90–180, and > 180 tablets during pregnancy. Some countries, including Indonesia, recommend a minimum of 90 IFA tablets,¹⁷ whereas the WHO recommends at least 180 IFA tablets throughout pregnancy.^18,19 Consuming at least 90 IFA tablets during pregnancy was used as a standard for IFA adherence in the stratification analysis.

Maternal pregnancy anemia and infant anemia.

Maternal and infant Hb levels were measured using the HemoCue Hb 201 + system (HemoCue, Ängelholm, Sweden). Anemia at pregnancy was diagnosed based on a Hb level of < 110 g/L.²⁰ Due to the variation in participants’ availability, Hb was measured at different gestational ages for each participant during the second to third trimester. For infants, there were different Hb cutoffs for infant anemia: < 110 g/L for term and < 120 g/L for preterm infants from newborn to 2 weeks old, < 100 g/L for infants from 2 weeks to 6 months old, and < 110 g/L for infants older than 6 months.^21–23 There was no adjustment for altitude because this area is < 100 m above sea level.

Possible confounding variables.

Iron–folic acid source; mother’s age, nutritional status, number of delivered children, time between pregnancies, gestational age, antenatal care (ANC) completion (minimum of two ANC visits in the first trimester, one ANC visit in the second trimester, and three ANC visits in the third trimester²⁴), knowledge and attitude about anemia and IFA consumption, and dietary habits; mother’s and father’s education and occupation; household income; and father’s role in supporting childcare and health-related decision-making in the family were included as possible confounding variables. For infant anemia, birth weight (low or normal birth weight) and complementary feeding (already given or still breastfed) were included as possible confounders.

The mother’s nutritional status during pregnancy was determined by measuring mid-upper arm circumference (MUAC) using measuring tape with 1-mm accuracy. A MUAC of < 23.5 cm was considered undernourished.²⁵ The mother’s standing height was measured using a microtoise (GEA Medical SH2A, Tangerang, Indonesia) with 1-mm accuracy. A height of < 145 cm was considered short stature.²⁶ MUAC and height were measured at the same time with Hb measurement. The mother’s pre-pregnancy body mass index (BMI) was calculated based on the preconception anthropometric information in the KIA book (secondary data). A pre-pregnancy BMI of < 18.5 kg/m² was considered underweight.²⁷

The mother’s dietary habits during pregnancy were assessed using a food-frequency questionnaire (FFQ) reporting the last 3 months’ consumption and a single 24-hour dietary recall (24DR) for a weekday. The food portion was given in the form of pictures as references. The 24DR results were checked for the nutritional contents calculated from the Ministry of Health’s Indonesian Food Composition Table²⁸ using the NutriSurvey program. Nutritional content was grouped based on the fulfillment of recommended daily intake (RDI) into inadequate (< 90% of RDI) and adequate (≥ 90% of RDI).^29,30

The anemia and IFA knowledge scoring in the questionnaire comprised 13 questions, with 1 point scored for each correct answer. The questionnaire score could be between 0 and 13. A score ≤ 5 was considered low knowledge. The anemia and IFA attitude scoring were associated with the mother’s usual conduct regarding IFA consumption and food avoidance. The attitude scoring comprised five agree–disagree questions with 2 points for the more favorable answer and 1 point for the less favorable answer. The possible score was from 5 to 10. A score ≤ 8 was considered a negative attitude. Paternal involvement as the health-related decision-maker and in childcare support was compared with others, including relatives, peers, health workers, and the mother alone.

Data analysis.

Data were screened for missing variables. If possible, missing data were reconfirmed with the respondent. After screening, the data were entered into the SPSS statistical software (IBM Corp., Armonk, NY) for analysis. Categorical variable distributions were presented as numbers and percentages. Linear variables are presented as mean ± SD. A P-value < 0.05 was considered statistically significant, and a P-value < 0.1 was considered a tendency for association with a corresponding 95% CI. Relative risk (RR) was calculated to assess the risk each variable contributed to maternal pregnancy and infant anemia. Potential confounding variables included in the multivariable analysis were those with a P-value < 0.25 in bivariable analysis or a conceptual connection with the outcome to get the final multivariable model with adjusted RR (aRR). The final associated variables were checked for the possible effect modifier, with changes of aRR > 10% regarded as a confounding variable to maternal anemia.

Qualitative phase.

The qualitative semistructured IDIs and FGD examined the reason for noncompliance in IFA consumption, monitoring, and evaluation; paternal and external roles in motivating and supporting the pregnancy process; and infant anemia. The IDI participants were purposively selected from the quantitative study’s participants representing the IFA adherence and maternal anemia groups: 1) not adhered to IFA but not anemic, 2) not adhered to IFA and anemic, 3) adhered to IFA and not anemic, and 4) adhered to IFA but anemic. This study did not stratify the groups based on infant anemia because the quantitative result did not show any association between IFA consumption and infant anemia. However, the topic was raised with the healthcare providers. In addition to the mothers, the IDIs involved their spouses (fathers), most frequently visited midwives, and health office representatives based at the district’s community health center. Thirty-seven IDIs were performed among 17 mothers, 12 fathers, one private midwife, two community health center midwives, two community health officers, and three health office staff. The FGD was conducted among 23 cadres representing the three districts. In-depth interviews and FGD were conducted by trained academic interviewers (experienced and trained researchers from Universitas Indonesia) fluent in the local language (Bahasa Indonesia). The IDIs were conducted privately or based on participants’ convenience to obtain honest and unbiased information on their experience, and the FGD was conducted in the local hall. All participants were compensated through a gift for their participation. All IDIs and the FGD were conducted in Bahasa Indonesia and audio-recorded. The audio-recorded interviews were transcribed verbatim and translated into English for publication by the interviewers and cross-checked with two experienced transcribers fluent in Bahasa Indonesia and English. The triangulation involved quantitative data and quotations from IDIs and the FGD.

RESULTS

Quantitative.

The cohort study followed 225 pregnant mothers and their infants at a mean age of 3.3 ± 1.7 months. Anemia was found in 57.3% of the pregnant mothers and 52.0% of their infants. The mothers’ Hb levels were between 75.0 and 142 g/L, with a mean of 108 ± 12 g/L. Their infants’ Hb levels were between 69 and 176 g/L, with a mean of 103 ± 15 g/L. The mothers’ IFA consumption during pregnancy averaged 147.7 ± 61.3 tablets, ranging from 0 to 474. Of the mothers who did not comply with 90 IFA tablets, 34.5% received IFA from the community health center, 41.4% from private midwives, and the rest did not receive any IFA. The distribution of the child-, mother-, and household-related variables are presented in Table 1. Infant anemia was observed in 48.4% of the infants, and pregnancy anemia was observed in 57.3% of the mothers. Among the mothers, 57.8% consumed 90–180 tablets and 27.6% consumed > 180 IFA tablets during pregnancy, with 58.8% obtaining them from private midwife practices and 23.1% from the community health center. Only 11.6% had completed antenatal care from health providers. Among the mothers, 51.6% had low knowledge and 51.1% had a negative attitude toward anemia and IFA consumption. The current population social distribution showed that 86.7% of the mothers and 83.5% of the fathers did not finish high school. More than half of the fathers (52.9%) were unemployed or worked as part-time laborers, whereas 88.4% of the mothers were not working.

Table 1.

Population characteristics (N = 225)

Variable	Frequency
	n	%
Child-related
Sex (n = 225)
Females	109	48.4
Males	116	51.6
Infant anemia (n = 225)
Anemia	109	48.4
Normal	116	51.6
Birth weight (n = 225)
Low birth weight (< 2,500 g)	20	8.9
Normal (≥ 2,500 g)	205	91.1
Mother-related
IFA consumed during pregnancy (n = 225)
< 90 tablets	33	14.7
90–180 tablets	130	57.8
> 180 tablets	62	27.6
Antenatal care (n = 225)
Incomplete	199	88.4
Complete	26	11.6
IFA source (n = 216)
Community health center	50	23.1
Private midwife practice	127	58.8
Mixed and others	39	18.1
Anemia status during pregnancy (n = 225)
Anemia (< 110 g/L)	129	57.3
Normal (≥ 110 g/L)	96	42.7
Mid-upper arm circumference (n = 225)
Undernourished (< 23.5 cm)	12	5.3
Not undernourished (≥ 23.5 cm)	213	94.7
Mother’s height (n = 225)
Short (< 145 cm)	10	4.4
Normal (≥ 145 cm)	215	95.6
Pre-pregnancy BMI (n = 225)
Underweight (< 18.5 kg/m2)	11	4.9
Not underweight (≥ 18.5 kg/m2)	214	95.1
Gestational age (n = 225)
Preterm (< 37 weeks)	22	9.8
Term (≥ 37 weeks)	203	90.2
Maternal parity (n = 225)
Multiparity	183	81.3
Primiparity (first child)	42	18.7
No. of children delivered (n = 224)
> 4 children	5	2.2
≤ 4 children	219	97.8
Age (n = 225)
< 20 and > 35 years	42	18.7
20–35 years	183	81.3
Time between pregnancies (n = 184)
≤ 2 years	15	8.2
> 2 years	169	91.8
Mother’s education (n = 225)
Bachelor’s degree or higher	195	86.7
Graduated high school or under	30	13.3
Mother’s knowledge (n = 225)
Low (< mean)	116	51.6
High (≥ mean)	109	48.4
Mother’s attitude (n = 225)
Negative	115	51.1
Positive	110	48.9
Mother’s occupation (n = 225)
Stay-at-home mother	199	88.4
Working	26	11.6
Household-related
Father’s occupation (n = 225)
Used	106	47.1
Unemployed or part-time labor	119	52.9
Father’s education (n = 224)
Bachelor or higher	187	83.5
Graduated high school or under	37	16.5
The health-related decision-maker in the family (n = 225)
Father	87	38.7
Others	138	61.3
Support for childcare in the family (n = 225)
Father	38	16.9
Others	187	83.1
Family income (n = 223)
Low (IDR < 3,000,000)	109	48.9
High (IDR > 3,000,000)	114	51.1

BMI = body mass index; IDR = Indonesian rupiah; IFA = iron folic acid.

The bivariable analysis between the population characteristics and maternal and infant anemia is presented in Table 2. Iron–folic acid consumption was significantly associated with anemia during pregnancy. Consumption of 90–180 IFA tablets during pregnancy reduced pregnancy anemia by 0.77 times (P = 0.04, 95% CI: 0.62–0.98), and consuming > 180 tablets reduced pregnancy anemia by 0.60 times (P < 0.00, 95% CI: 0.43–0.83) compared with consuming < 90 IFA tablets. In addition to IFA consumption, lower protein consumption was marginally associated with anemia during pregnancy. Maternal pregnancy anemia, fathers’ support for childcare in the family, and nonfathers as health-related decision-makers in the family were significant predictors of infant anemia, whereas IFA source and gestational age were marginally associated with infant anemia.

Table 2.

The association between the population characteristics and infant anemia, neonatal anthropometry, and pregnancy anemia (N = 225)

Variable	Pregnancy anemia			Infant anemia
	%	RR (95% CI)	P-value	%	RR (95% CI)	P-value
Infant
Sex (n = 225)
Boys	61.2	1.15 (0.92–1.45)	0.281	53.4	1.24 (0.94–1.63)	0.122
Girls	53.2	1		42.2	1
Birth weight (n = 225)
Low birth weight < 2,500 g				55.0	1.15 (0.76–1.75)	0.541
Normal (≥ 2,500 g)				47.8	1
Complementary feeding (n = 225)
Already given				45.3	0.90 (0.68–1.20)	0.467
Still exclusively breastfed				50.4	1
Pregnant mothers’
Anemia status during pregnancy (n = 225)
Anemia				55.0	1.39 (1.04–1.86)	0.027*
Normal				39.6	1
IFA adherence during pregnancy (n = 225)
< 90 tablets	75.8	1		48.5	1
90–180 tablets	58.5	0.77 (0.61–0.98)	0.035*	46.9	1.03 (0.70–1.54)	0.873
> 180 tablets	45.2	0.60 (0.43–0.83)	0.003*	51.6	1.07 (0.70–1.63)	0.774
IFA source (n = 216)
Community health center	64	1		60	1
Private midwife practice	50.4	0.79 (0.60–1.03)	0.103	44.9	0.748 (0.556–1.007)	0.056†
Mixed and others	66.7	1.04 (0.77–1.41)	0.796	46.2	0.77 (0.51–1.16)	0.207
MUAC (cm) (n = 225)
Undernourished	50	0.87 (0.49–1.54)	0.820	41.7	0.85 (0.43–1.69)	0.631
Not undernourished	57.7	1		48.8	1
Mother’s height (cm) (n = 225)
Short	80	1.42 (1.02–1.98)	0.122	60	1.25 (0.74–2.12)	0.457
Normal	56.3	1		47.9	1
Pre–pregnancy BMI (kg/m2) (n = 225)
< 18.5	72.7	1.29 (0.88–1.88)	0.231	54.5	1.13 (0.65–1.98)	0.680
≥ 18.5	56.5	1		48.1	1
Number of children delivered (n = 224)
> 4 children	40	0.69 (0.23–2.03)	0.359	20	0.41 (0.07–2.38)	0.203
≤ 4 children	58	1		48.9	1
Maternal parity (n = 225)
Multiparity	59	1.18 (0.85–1.66)	0.372	50.3	1.24 (0.84–1.84)	0.254
Primiparity	50	1		40.5	1
Age (n = 225)
< 20 and > 35 years	64.3	1.15 (0.89–1.50)	0.403	47.6	0.98 (0.69–1.39)	0.906
20–35 years	55.7	1		48.6	1
The gap between pregnancies (n = 184)
≤ 2 years	66.7	1.14 (0.78–1.66)	0.736	53.3	1.06 (0.65–1.74)	0.823
> 2 years	58.6	1		50.3	1
Gestational age (n = 225)
Preterm	54.5	0.95 (0.64–1.41)	0.959	68.2	1.47 (1.07–2.03)	0.051†
Term	57.6	1		46.3	1
Antenatal care (n = 225)
Complete	57.8	1.07 (0.74–1.56)	0.864	47.4	0.91 (0.67–1.24)	0.570
Incomplete	53.8	1		52.0	1
Mother’s education (n = 225)
Bachelor or higher	55.4	0.79 (0.61–1.03)	0.191	48.2	0.96 (0.66–1.42)	0.855
Graduated high school or under	70.0	1		50.0	1
Mother’s occupation (n = 225)
Stay-at-home mother	61.5	1.08 (0.78–1.50)	0.802	50.0	1.04 (0.69–1.56)	0.867
Working	56.8	1		48.2	1
Mother’s anemia and IFA knowledge (n = 225)
Low	53.4	0.87 (0.69–1.09)	0.280	44.8	0.86 (0.65–1.12)	0.264
High	61.5	1		52.3	1
Mother’s anemia and IFA attitude (n = 225)
Negative	55.7	0.94 (0.75–1.18)	0.699	45.2	0.87 (0.67–1.14)	0.324
Positive	59.1	1		51.8	1
Household
Father’s education (n = 224)
Bachelor or higher	56.7	0.91 (0.69–1.21)	0.664	49.7	1.13 (0.85–2.04)	0.253
Graduated high school or under	62.2	1		37.8	1
Father’s occupation (n = 225)
Unemployed or part–time labor	61.3	1.16 (0.92–1.46)	0.249	51.3	1.13 (0.86–1.49)	0.373
Used or own business	52.8	1		45.3	1
Household income (n = 224)
Low (< IDR 3,000,000)	62.4	1.17 (0.93–1.46)	0.228	50.5	1.09 (0.83–1.42)	0.556
High (> IDR 3,000,000)	53.5	1		46.5	1
Support for childcare in the family (n = 225)
Father	52.6	0.90 (0.65–1.25)	0.643	68.4	1.54 (1.18–2.02)	0.007*
Others	58.3	1		44.4	1
The health–related decision maker in the family (n = 225)
Others	58.0	1.03 (0.82–1.30)	0.916	55.1	1.45 (1.07–1.98)	0.012*
Father	56.3	1		37.9	1
Mother’s dietary consumption habit during pregnancy
Red meat
< 4× per week	58.0	1.12 (0.75–1.66)	0.721	48.5	1.01 (0.66–1.56)	0.963
≥ 4× per week	52.0	1		48.0	1
Poultry
< 4× per week	60.3	1.07 (0.84–1.37)	0.679	50.8	1.07 (0.80–1.43)	0.662
≥ 4× per week	56.2	1		47.5	1
Fish
< 4× per week	54.1	0.90 (0.72–1.13)	0.419	42.3	0.82 (0.60–1.13)	0.209
≥ 4× per week	60.3	1		51.3	1
Eggs
< 4× per week	53.2	0.91 (0.68–1.22)	0.631	51.1	1.07 (0.78–1.47)	0.688
≥ 4× per week	58.4	1		47.8	1
Plant-based protein
< 4× per week	62.1	1.09 (0.80–1.49)	0.748	48.3	0.99 (0.66–1.48)	0.965
≥ 4× per week	56.9	1		48.7	1
Fruits
< 4× per week	54.0	0.92 (0.71–1.20)	0.627	50.8	1.07 (0.80–1.43)	0.662
≥ 4× per week	58.6	1		47.5	1
Vegetables
< 4× per week	46.7	0.78 (0.53–1.17)	0.257	53.3	1.11 (0.77–1.60)	0.602
≥ 4× per week	59.6	1		48.2	1
Fortified powdered milk
< 4× per week	60.6	1.10 (0.88–1.39)	0.476	45.0	0.86 (0.66–1.13)	0.271
≥ 4× per week	54.4	1		52.4	1
Snacks
< 4× per week	59.4	1.17 (0.88–1.56)	0.341	47.3	0.97 (0.70–1.33)	0.842
≥ 4× per week	50.9	1		48.8	1
Macro- and micro-nutrient sufficiency
Energy
Inadequate	56.0	0.91 (0.71–1.17)	0.573	45.8	0.86 (0.66–1.13)	0.297
Adequate	61.4	1		53.0	1
Protein
Inadequate	63.8	1.24 (0.99–1.55)	0.089	48.6	1.01 (0.77–1.32)	0.972
Adequate	51.7	1		48.3	1
Fat
Inadequate	53.8	0.92 (0.71–1.19)	0.599	50.8	1.07 (0.80–1.43)	0.930
Adequate	58.8	1		47.5	1
Carbohydrate
Inadequate	58.2	1.07 (0.81–1.40)	0.746	44.1	0.82 (0.63–1.07)	0.143
Adequate	54.5	1		61.8	1
Fiber
Inadequate	57.1	0.86 (0.48–1.53)	0.489	48.9	1.47 (0.47–4.58)	0.455
Adequate	66.7	1		33.3	1
Vitamin B6
Inadequate	56.6	0.97 (0.77–1.22)	0.892	49.7	1.07 (0.81–1.43)	0.634
Adequate	58.5	1		46.3	1
Vitamin B12
Inadequate	60.5	1.13 (0.89–1.42)	0.380	51.9	1.18 (0.89–1.56)	0.255
Adequate	53.7	1		44.2	1
Vitamin D
Inadequate	57.8	1.16 (0.68–1.98)	0.769	48.8	1.14 (0.61–2.12)	0.667
Adequate	50.0	1		42.9	1
Folic acid
Inadequate	58.2	1.07 (0.81–1.40)	0.746	48.2	0.98 (0.72–1.34)	0.913
Adequate	54.5	1		49.1	1
Zinc
Inadequate	56.6	0.94 (0.75–1.18)	0.709	48.2	0.99 (0.75–1.31)	0.933
Adequate	59.5	1		48.8	1
Iron
Inadequate	60.0	1.17 (0.90–1.52)	0.290	45.2	0.83 (0.63–1.09)	0.188
Adequate	51.4	1		54.4	1
Phosphor
Inadequate	59.5	1.05 (0.79–1.39)	0.884	47.6	0.98 (0.69–1.39)	0.906
Adequate	56.8	1		48.6	1

BMI = body mass index; IDR = Indonesian rupiah; IFA = iron folic acid; MUAC = mid-upper arm circumference; RR = relative risk.

^*P < 0.005.

^†P < 0.100.

The final multivariable models are presented in Table 3, while the initial model can be found in Supplemental Material 1. Iron–folic acid consumption was predictive of anemia during pregnancy but not infant anemia in the offspring. After adjusting for potential confounding variables, consuming > 180 tablets during the pregnancy reduced anemia risk by 0.25 times (P = 0.01, 95% CI: 0.09–0.66). In addition to IFA adherence, the mother’s knowledge, the father’s occupation, household income, and protein intake were among the variables contributing to the mother’s anemia during pregnancy. Mother’s pregnancy anemia, preterm birth, and father’s role in childcare support were among the variables contributing to infant anemia. Stratification analysis of IFA adherence (≥ 90 tablets) by mother’s anemia status during pregnancy to infant anemia showed that IFA adherence did not significantly affect infant anemia incidence in anemic mothers (Figure 2). Mother’s anemia and IFA knowledge changes the aRR of IFA consumption > 180 tablets during pregnancy to maternal anemia to more than 10% and therefore is regarded as a confounding variable for IFA consumption association to maternal anemia (Supplemental Material 2).

Table 3.

Final multivariable model

Dependent variable	Independent variables	B	P value	Adjusted RR	95% CI
Mother’s anemia during pregnancy	IFA adherence
	0–90 tablets
	90–180 tablets	−0.797	0.081	0.45	0.18–1.10
	> 180 tablets	−1.396	0.006	0.25	0.09–0.66
	Mother’s anemia and IFA knowledge (lower compared with higher)	−0.548	0.069	0.58	0.32–1.04
	Father’s occupation (unemployed or part-time labor compared with used or own business)	−0.661	0.036	1.94	1.04–3.59
	Household income (lower compared with higher)	0.693	0.028	2.00	1.08–3.70
	Protein intake (inadequate compared with adequate)	0.664	0.023	1.94	1.10–3.44
Infant anemia	Mother’s pregnancy anemia (anemia compared with normal)	0.698	0.014	2.01	1.15–3.50
	Gestational age (preterm compared with term)	0.862	0.080	2.37	0.90–6.21
	Support for childcare (father compared with others)	1.017	0.009	2.77	1.29–5.92

IFA = iron folic acid; RR = relative risk.

Figure 2.

The difference between infant anemia incidence stratified based on iron–folic acid (IFA) adherence and maternal anemia.

Qualitative.

The IDIs and FGD discussed the reason for noncompliance to IFA consumption, monitoring, and evaluation; paternal and external roles in motivating and supporting the pregnancy process; and infant anemia. There was agreement that IFA tablets were widely accessible from the community health center for free, but not every pregnant mother directly received IFA tablets at their first visit. Some reasons the participants gave included that the mother seemed healthy, had no worrying symptoms, had no eating disorder, had already consumed other pregnancy supplements, and was late to their first ANC visit. The gestational age when the mother received their first IFA tablets varied from 3 to 28 weeks. One reason for a late ANC visit was mothers not knowing they were pregnant.

“It was free in the community health center.” (Mother 1, Community Health Center Midwife 1, and Cadres)

“It costs around 80 to 150 [in thousand IDR] including medicines in private midwife.” (Community Health Center Midwife 2)

“Some were lazy to go to the community health center, so they got the IFA late.” (Mother 2)

“Lazy to queue was a barrier … some checked up when they were over 3 months.” (Community Health Center Midwife 1)

“When I knew that I was pregnant, I got IFA from the midwife. So I just started at already 4 months, consumed every day.” (Mother 3)

“I checked at the first month. When I was late, I checked directly and got IFA.” (Mother 17)

“Know when pregnant was at first going the second month, took Folamil Genio, only that … And [my] husband asked the midwife; the midwife said that it already contained iron. So I did not take other supplements.” (Mother 16)

“There is iron. Usually, private doctors focus on folic acid. I saw that there were various types prescribed by doctors, but IFA was still low. Sometimes I offer the mother to switch or consumed to IFA that is given here.” (Community Health Center Midwife 1)

“Half the patients already bought the vitamin themselves.” (Private Midwife)

To monitor IFA supplementation, the private midwives asked for the number of unconsumed IFA tablets before giving additional IFA tablets to be consumed before the next ANC visit. Some interviewed mothers stated that they were given IFA tablets every ANC visits, but sometimes they still had some left from the previous ANC visit, which were then disposed of.

“I asked if the vitamin remained or was empty; if the IFA or calcium was finished, then we give it again, but if she said still have some, we can evaluate.” (Private Midwife)

“It was asked if the previous ones finished or not, and also the month before. … Every check-up the IFA was not finished yet, so it was thrown away.” (Mother 17)

“The midwife asked for the calcium and the other ones for the child’s brain.” (Mother 16)

“That is why I asked why this person still and stayed anemic while we always gave vitamin. I questioned if it was truly taken? Or thrown away? Or kept?” (Community Health Center Midwife 2)

The history of IFA consumption could be monitored from the IFA monitoring sheet in the KIA book. However, during data collection, not every mother had a completed IFA monitoring sheet. The midwives stated that the mothers or their families were asked to complete the IFA monitoring sheet themselves. The cadres stated that midwives should complete the IFA monitoring sheet.

“Actually, in the KIA book, we have the checklist … but we do not know if the mother only marked it or truly consumed it or not. The family should mark them, but if the family did not care just so that the midwife would not scold them, they may just marked all before the visit.” (Community Health Center Midwife 2)

“We have the medical record, so the patient has the pink book [KIA] … there were some checklists filled, but there were some unfilled. If she did not fill, when she visits here, we told her, ‘ma’am, even when you bought them yourself, he checklist in the book still must be filled’.” (Private Midwife)

“Just from the mother’s card. … We only note IFA received and given, did not evaluate how many consumed or not.” (Community Health Center Midwife 1)

“We hope that all pregnant mothers and families read and use the KIA book as the only record.” (Health Office)

Several themes were connected to the mother’s reason for not complying with IFA consumption: IFA side effects; IFA and anemia knowledge; motivation from husband, peers, midwives, doctors, and cadres; and influence from social media. The respondents stated that IFA tablets from the community health centers made them feel nauseous, and some preferred other brands of multivitamins. Responding to the topic of IFA and anemia knowledge, the participants in the low knowledge score group admitted that regardless of their knowledge, they followed the advice to consume IFA tablets from their healthcare practitioners.

“IFA makes me feel nauseous and smells fishy.” (Mother 2, Mother 7, and Mother 15, and Cadres)

“I took Folamil Genio [brand] only because of nausea [when consuming IFA].” (Mother 16)

“I took them all; nothing left, it’s from the doctor.” (Mother 1)

“Just following the midwife’s advice.” (Father 1)

“Because it was advised by the midwife.” (Cadres)

“It was doctor’s advice.” (Cadres)

Support from family, peers, and community was a confounding variable in the quantitative analyses. The IDIs and FGD probed the support from the husband (the child’s father). Some midwives see the fathers’ support as limited to accompanying or waiting for their wives during ANC visits. However, some fathers were already actively involved in gathering information from the midwife. Some fathers were involved but not letting their wives consume IFA tablets from midwives.

“I accompany [for ANC].” (Father 2)

“When checked-up, I usually waited.” (Father 3)

“Every visit to the midwife, my husband accompanies and listened.” (Mother 17)

“[My husband] Gave me a ride and waited outside the community health center” (Mother 1)

“Only 10% [of husbands] involved.” (Community Health Center Midwife 2)

“Father’s role is when the mother checks up, not alone but accompanied. So if there is a problem or something else, it was not only the matter for the mother but the father too. Father has to support. Not only the father, the family too.” (Private Midwife)

“If my husband told me to take the medicine, I directly obey.” (Mother 17)

“Often I remind to take the vitamins. But I took the advice from my parents first. Good to eat soya for hair and young coconut for skin, for health too. … We buy the vitamin ourselves, Folamil [brand].” (Father 7)

“For the vitamin Folamil [brand], my sister-in-law took that when she was pregnant. It is already with blood booster, so do not need to buy again.” (Mother 16)

The mothers received IFA tablets from community health centers, private midwives, or other sources. The representatives from mothers, cadres, and midwives unanimously believed that the difference between the IFA provided by the government and private sources was only in price and prestige but not content. This belief contrasted with the information from the Health Office’s representative, who stated that the quality of IFA tablets provided by the community health center was better. However, the ANC visitation rate was higher among the private than the community health center midwives.

“The same, IFA from the government and the private midwives. But, back to the person, some patients perceive that it is better compared to private midwives. For me, I will ask first what she uses, and if she finds it comfortable, continue.” (Private Midwife)

“For the lower income, they usually take any brands, but with good results. But for the middle and upper, I see that there is the prestige.” (Community Health Center Midwife 1)

“IFA from the Ministry of Health has better quality than the others. But based on the number of ANC visits between the community health center, midwives, hospital, community health center had low ANC rate, more went to private midwives.” (Health Office)

The participants knew about maternal anemia but not child anemia. There was no program for anemia in infants at that time. However, some midwives believed a correlation existed between pregnancy and infant anemia.

“At this time, there is no assessment for babies … no special program for babies. But it will be known when the child was thin, and then the Hb was evaluated and found to be anemic at 1 year old. But there is no rule for babies with thinness to check for Hb.” (Community Health Center Midwife 1)

“Yes, it must be associated if the mother did not get IFA or good nutrition, the baby’s nutrition will be inadequate and anemic if checked.” (Private Midwife)

DISCUSSION

Our study reported the effect of IFA supplementation on pregnant mothers in the community. Iron–folic acid was consumed according to Indonesian government advice of 90 tablets throughout pregnancy by > 80% of pregnant mothers. Pregnancy and infant anemia rates were higher than in previous national reports.^3,31 IFA consumption was a predictor of pregnancy anemia but not infant anemia. Consuming > 180 IFA tablets during pregnancy provided better protection than 90 tablets.

The number of mothers taking IFA tablets was far higher than in the latest national survey (37.7%).³ Despite relatively high IFA compliance, anemia remained a problem in pregnant mothers in the study population, above the reported national prevalence in 2018.³ Other identified confounding variables, including the mother’s knowledge of anemia and IFA, the father’s occupation, income, and protein intake, may alter the IFA and pregnancy anemia relationship. Animal protein is among the dietary sources of iron.³² Protein plays an independent role from iron in erythropoiesis,³³ suggesting the importance of having adequate protein regardless of IFA consumption.

A good socioeconomic condition also significantly supports IFA and the anemia risk independently. A previous systematic review on anemia risk factors among Malaysians also identified noncompliance with iron supplements and low income as contributors to anemia, along with younger age, lower education, unemployment, and high parity.³⁴ Unlike this previous study, our study showed that mothers with lower knowledge were protected from anemia, consistent with another study that reported that anemia knowledge was not associated with anemia incidence.³⁵ One possible explanation from the qualitative exploration was that knowledge did not affect compliance, and mothers with lower knowledge may simply follow the midwife’s advice without second guessing.

Iron–folic acid tablets provided by the government contain more iron. However, the unfavorable side effects of government IFA tablets limit compliance. While some mothers still consumed other brands, the different iron contents between branded and government-provided tablets were one possible contributing factor. There was a gap in IFA content knowledge among the midwives and community members, and this misconception influenced mothers’ perceptions when choosing which IFA tablet to consume. There was also a gap in the obligation of IFA supplementation monitoring. While KIA use was intended to bridge health practitioners, cadres, mothers, and other supporting community members, the obligation to monitor IFA consumption and record it in the KIA book was the health practitioners, not cadres or mothers.³⁶

Although maternal IFA consumption was not associated with infant anemia, maternal pregnancy anemia was a significant factor in infant anemia. This population’s infant anemia prevalence (48.4%) was much higher than that previously reported for infants under 12 months in Indonesia (14.5%).³¹ Mother’s pregnancy anemia and gestational age, the father’s role as the health-related decision-maker, and childcare support were among the factors contributing to infant anemia. The suggested theory was that maternal iron storage contributes to newborn iron storage. However, the demand for iron increases as the pregnancy progress, and IDA in anemic pregnant mothers may prevent adequate iron storage transfer to their fetus.³⁷ Although fetal iron stores develop mainly in the third trimester, a previous study suggested that iron deficiency in the earlier stage of pregnancy had a greater impact on fetal outcomes than in the later stages.³⁷ Another cohort study in the Indonesian population even observed that pre-pregnancy anemia was already an independent risk factor for child anemia.⁶

In populations with a high anemia prevalence, like the current population, mothers may already have had anemia at an early stage or even before pregnancy. A brief period of oral IFA supplementation may not be able to improve the developed anemia, especially when it is caused by the combined effects of chronic undernutrition, poor diet, or infection.³⁷ Although the current IFA supplementation policy also targets adolescent girls, the monitoring and accessibility of IFA supplements for nonpregnant and lactating women are not as rigorous as for pregnant women.³⁸ Wider supplementation access could provide more protection against iron deficiency in high-risk populations.

The Indonesian government’s focus on the first 1,000 days of life already includes IFA supplementation for adolescent girls and pregnant women. However, anemia screening has not been targeted for pregnant women.³⁹ The previous^6,37 and current findings support the view that Hb monitoring is important for child outcomes. Monitoring Hb at the early stage of pregnancy or even earlier for women planning for pregnancy is necessary to provide a well-targeted supplementation intervention and avoid maternal and child anemia.⁶

The WHO recommends the consumption of daily oral IFA (180 tablets) for populations with an anemia prevalence of > 20%.^18,19 Given the current and previous national anemia prevalence reports³ and reflecting on the current finding that adhering to > 180 tablets was associated with better protection, the Indonesian government may need to evaluate its current recommendation of ≥ 90 tablets during pregnancy.¹⁷ The current IFA supplementation must also be accompanied by a robust monitoring and evaluation system. Available tools, such as the KIA book and Posyandu system, could support monitoring and evaluating pregnancy and maternal health status and supplementation but may not be best used. In addition to IFA supplementation, other dietary support for protein adequacy in pregnant women is also important. Indonesia currently has a food supplementary program for undernourished pregnant mothers. However, the program may have a larger impact if also provided to anemic mothers.

Our study also associated the father’s education, occupation, and role in childcare support with maternal and child outcomes. Fathers’ unemployment and lower education contributed to an increased risk of adverse outcomes.^40,41 However, few studies have explored the effect of paternal involvement on mother and child health.⁴² Some studies have shown that the father’s presence was associated with increased birth morbidity.⁴² Our study found that fathers’ involvement in childcare support contributed to increased infant anemia. The current qualitative analyses indicated that the father’s involvement is hoped for and expected. However, their form of involvement was sometimes limited to acting as passive companions. In some cases where the father was actively involved in the mother’s supplementation and intake, their advice and support were not accompanied by proper knowledge of IFA, maternal anemia, and the effect on child anemia. The current quantitative and qualitative findings raise further questions about the father’s role and involvement in supporting maternal and child health. The lack of paternal involvement awareness and father-targeted programs may create gaps in maternal and child education when a male has a predominant role in decision-making and support in the family.⁴³

Our study had some limitations, including the limited area of observations and the possibility of bias from the dietary recall and self-reported FFQ. Because our study only observed three districts in Depok, it may not represent other areas with different socioeconomics and cultural behaviors. Our study also did not assess supplementation history and Hb before and early in pregnancy. Some clinical variables contributing to anemia were also not included, such as inflammation and previous history of other blood-related diseases. Although the dietary recall and FFQ were prone to reporting bias, we minimized the possibility of bias by illustrating the portion size and using a local food composition table.

Our study provides longitudinal observation of IFA consumption during pregnancy and its outcome on maternal pregnancy and infant anemia. Using a mixed-method approach provided objective measures and participants’ subjective insights. Only a few IFA cohort studies have been performed in the Indonesian population, and ours is the first mixed-method study on IFA and maternal and infant anemia. Our analysis also included broad aspects of dietary intake and paternal involvement variables as potential confounders, which have rarely been considered in previous studies. Our study presented some novel knowledge on IFA, maternal anemia, and infant anemia interactions. Although IFA was associated with maternal anemia, maternal anemia was associated with infant anemia. This finding calls for public health policies related to IFA supplementation and response to anemia in a vulnerable population. Moreover, the knowledge of the important role of fathers in supporting maternal and child health could provide insight for future research and interventions in paternal involvement.

CONCLUSION

Iron–folic acid supplementation protects against maternal pregnancy anemia but not infant anemia. However, maternal anemia was an important variable in infant anemia. Iron–folic acid supplementation programs must be available for women at child-bearing age to avoid chronic IDA. Moreover, an actual anemia monitoring program, especially for women in the early trimester, is important in addressing offspring’s nutritional problems. In addition, an education program aimed at fathers is important in supporting their role in maternal and child health.

Note: Supplemental material appears at www.ajtmh.org.

Data Availability

The datasets used and/or analyzed in this study is available from the corresponding author upon reasonable request.