Hostname: page-component-586b7cd67f-t7czq Total loading time: 0 Render date: 2024-11-26T03:01:32.667Z Has data issue: false hasContentIssue false

Status of zinc nutrition in Bangladesh: the underlying associations

Published online by Cambridge University Press:  06 June 2016

Sabuktagin Rahman*
Affiliation:
Nutrition and Clinical Services Division, International Centre for Diarrhoeal Disease Research (ICDDR,B), 68 Shaheed Tajuddin Ahmed Sharani, Mohakhali Dhaka 1212, Bangladesh
Tahmeed Ahmed
Affiliation:
Nutrition and Clinical Services Division, International Centre for Diarrhoeal Disease Research (ICDDR,B), 68 Shaheed Tajuddin Ahmed Sharani, Mohakhali Dhaka 1212, Bangladesh
Ahmed Shafiqur Rahman
Affiliation:
Nutrition and Clinical Services Division, International Centre for Diarrhoeal Disease Research (ICDDR,B), 68 Shaheed Tajuddin Ahmed Sharani, Mohakhali Dhaka 1212, Bangladesh
Nurul Alam
Affiliation:
Health Systems and Population Studies Division, ICDDR,B, 68 Shaheed Tajuddin Ahmed Sharani, Mohakhali Dhaka 1212, Bangladesh
A. M. Shamsir Ahmed
Affiliation:
Nutrition and Clinical Services Division, International Centre for Diarrhoeal Disease Research (ICDDR,B), 68 Shaheed Tajuddin Ahmed Sharani, Mohakhali Dhaka 1212, Bangladesh School of Public Health, Faculty of Medicine and Biomedical Sciences, The University of Queensland, Herston, QLD, Australia
Santhia Ireen
Affiliation:
Nutrition and Clinical Services Division, International Centre for Diarrhoeal Disease Research (ICDDR,B), 68 Shaheed Tajuddin Ahmed Sharani, Mohakhali Dhaka 1212, Bangladesh
Ireen Akhter Chowdhury
Affiliation:
UNICEF, Bangladesh
Fatima Parveen Chowdhury
Affiliation:
Ayesha Memorial Medical College, Dhaka, Bangladesh Institute of Public Health Nutrition, Dhaka, Bangladesh
S. M. Mustafizur Rahman
Affiliation:
Institute of Public Health Nutrition, Dhaka, Bangladesh Micronutrient Initiative, Dhaka, Bangladesh
*
*Corresponding author: Dr S. Rahman, fax +880 2 9827101, email [email protected]

Abstract

Bangladesh is a country with a high burden of micronutrient malnutrition. Stunting affects 41 % of children aged under 5 years. Zn is one of the key micronutrients that is associated with stunting. The present study, as part of the national micronutrient survey 2011–2012, revealed for the first time the nationally representative prevalence of Zn deficiency and determined the associations of the condition. A cross-sectional ‘nationwide’ survey was conducted in pre-school-age children (6–59 months; PSAC) and non-pregnant non-lactating women (15–49 years; NPNLW). Multistage random sampling was done in 150 clusters; fifty in each of the rural, urban and slum strata. Data were analysed on 662 PSAC and 1073 NPNLW. Serum Zn was assayed by atomic absorption spectrophotometry. Zn deficiency was defined as serum Zn of <9·9 and <10·1 µmol/l in PSAC and NPNLW, respectively. The national prevalence of Zn deficiency was 44·6 and 57·3 % in PSAC and NPNLW, respectively. In PSAC, it was 29·5, 48·6 and 51·7 %, respectively, in urban, rural and slum strata. Household expenses (β = 0·13; P = 0·007), Hb (β = 0·10; P = 0·005), intake of animal-source Zn (β = 0·096; P = 0·02) and asset score (β = 0·11; P = 0·03) were positively associated with serum Zn in NPNLW. Residence in an urban area (β = 0·33; P = 0·03) and intake of plant-origin Zn (β = −0·13; P = 0·038) determined higher and lower status of Zn in PSAC, respectively. Zn deficiency was highly prevalent in Bangladesh, and it was principally related to inadequate quality of diet. To improve Zn nutrition, Bangladesh needs to strengthen research and programmes related to Zn biofortification, fortification and phytate-reducing technologies in the food system in the short and medium term. In addition, promotion of animal-source Zn for all is important in the long run.

Type
Research Article
Creative Commons
Creative Common License - CCCreative Common License - BY
This is an Open Access article, distributed under the terms of the Creative Commons Attribution licence (http://creativecommons.org/licenses/by/4.0/), which permits unrestricted re-use, distribution, and reproduction in any medium, provided the original work is properly cited.
Copyright
Copyright © The Author(s) 2016

Zn is one of the most important trace elements involved in human metabolism. It is implicated in all major biochemical metabolism and plays multiple roles in the maintenance of genetic material, including transcription of DNA, translation of RNA and cellular division( Reference Brown, Rivera and Bhutta 1 ). Zn supplementation produces highly significant positive responses in linear growth and weight gain( Reference Bhutta, Black and Brown 2 ). On the other hand, Zn deficiency is associated with stunted linear growth and diminished immune function. Supplementation of the mineral is also associated with lessening of diarrheal duration and respiratory tract infection in children( Reference Bhutta, Black and Brown 2 ). Childhood stunting in children aged under 5 years is widely recognised as a proxy indicator of Zn deficiency. However, not until the present study was conducted in 2011–2012 were nationwide data of Zn deficiency available in Bangladesh, a country with a high prevalence of stunting in children aged under 5 years (41 %)( 3 ). The present study has revealed for the first time in Bangladesh the national estimates of subclinical Zn deficiency and the underlying associations of the condition.

Methods

Sampling and study population

The present study is a part of the national micronutrient survey 2011–2012 of Bangladesh. A multi-stage random sampling procedure was applied to select the study participants from the rural, urban and the slum domains distributed all over the country. In the first stage, 150 clusters (fifty clusters in each domain) from the 15 000 clusters of the Bangladesh Multiple Indicator Cluster Sampling (MICS)( 4 ) frame were randomly selected. In the second stage a segment of fifty households was randomly chosen from each of the selected clusters and in the final stage, twenty households were selected randomly from the fifty-household segment; these twenty households formed the definitive sampling frame from which the required number of study participants was selected.

The sample size (Supplementary Table S3) for assessing Zn deficiency was representative of the stratum; however, combining all the three strata together and with application of population weight, nationally representative estimates were obtained. Data were collected in two population groups – pre-school-age children (PSAC; 6–59 months) and non-pregnant non-lactating women (NPNLW; 15–49 years).

Data and blood sample collection

Data on socio-economic status (SES), household possession of assets, household construction material, household food insecurity, morbidity of children and food intake of the study participants were analysed. SES was assessed by constructing the wealth index( Reference Gwatkin, Rustein and Johnson 5 ). Household food insecurity was assessed by the Household Food Insecurity Access Scale (HFIAS) for measurement of food access( Reference Coates, Swindale and Bilinsky 6 ). In order to assess food consumption, a semi-quantitative FFQ was used taking into consideration commonly consumed Bangladeshi foods( Reference Cheng, Yan and Dibley 7 Reference Buch Andersen, Perez-Cueto and Toft 10 ). The semi-quantitative FFQ assessed consumption over the preceding 7 d of the survey. The tool was administered by the field data collectors. For NPNLW, the study respondent provided the response; however, for PSAC, the mother or primary caretaker responded. The tools did not require the respondents to have any reading or writing skills necessary. The respondent was asked to tell about the number of portions (servings) of particular foods that she and/or her child had consumed over the past 7 d. She was required to indicate the portion size (g, ml) of the consumption from the standardised food photographs or commonly used household utensils, which were displayed and explained at the time of the interview. The number of portions consumed over the past 7 d multiplied by the average size of the portions yielded the total weekly consumption (g, ml). Raw food weight was calculated by using appropriate conversion factors( Reference Ali 11 ). The nutrient value for Zn was calculated per 100 g of raw food consumed using an updated Food Composition Table on Bangladeshi food( Reference Islam, Khan and Akhtaruzzaman 12 ). The prevalence of Zn intake inadequacy was measured by the estimated average requirement (EAR) cut-point method( Reference Brown, Rivera and Bhutta 1 ). For children, we defined dietary Zn inadequacy using the EAR for 2- to 3-year-old children (<2 mg/d)( Reference Brown, Rivera and Bhutta 1 ). For women, dietary Zn inadequacy was calculated by using lower bioavailability with an unrefined, cereal-based diet (<7 mg/d)( Reference Brown, Rivera and Bhutta 1 ). Phytate is a chemical substance present in plant foods which acts as a chelator of minerals, including Zn. The inhibitory effect of phytate on Zn absorption appears to follow a dose-dependent response, and the phytate:Zn molar ratio of the diet has been used to estimate the proportion of absorbable Zn( Reference Brown, Rivera and Bhutta 1 ). To classify populations according to mixed or cereal-based diet types, the phytate:Zn molar ratio of foods or diets was calculated according to the methods described in the technical document of the International Zinc Nutrition Consultative Group( Reference Brown, Rivera and Bhutta 1 ). Blood samples were collected in a centralised location, such as school, health centre, non-governmental organisation (NGO) office, developed as temporary sample collection centres. The selected survey respondents were given a token number indicating their names and identification and guided to appear at the collection centres. Venous blood was collected in trace element-free venoject tubes. The blood tubes were placed in a cool box and allowed to clot. The whole blood was centrifuged (Portable Centrifuge, 3000–3500 rpm) at the field collection centre and the serum was aliquoted in cryovials (trace element free) using a disposable pipette. The cryovials were stored in a −20°C freezer. Aliquoted serum samples were transferred in dry ice to the Nutritional Biochemistry Laboratory at the International Centre for Diarrhoeal Disease Research, Bangladesh (ICDDR,B), and stored in a −70°C freezer. The study received ethical approval from the Institutional Review Board of ICDDR,B. Written informed consent was taken from all study participants.

Biochemical analysis

We assayed serum Zn by atomic absorption spectrophotometry (Shimadzu AA-7000). C-reactive protein (CRP) and α-1-acid glycoprotein (AGP) were analysed by sandwich ELISA (Dynex Technologies Inc.). Hb was assessed using HemoCue Hb 301 (Hemocue AB) on venous blood.

Adjusting for infection

Serum Zn level was adjusted for infection by estimating biomarkers of infection – CRP and AGP. We performed the adjustment for elevated levels of CRP (>10·0 mg/l) and AGP (>1·0 g/l) by deriving the correction factors following the methods described by Thurnham et al. ( Reference Thurnham, McCabe and Haldar 13 ) and Engle-Stone et al. ( Reference Engle-Stone, Haskell and Ndjebayi 14 ).

Quality control

All biochemical analyses were carried out in the nutritional biochemistry laboratory of ICDDR,B. To control the quality of the laboratory analyses pooled serum and commercial quality-control (QC) material (Bi-level serum toxicology control; UTAK Laboratories, Inc.) were used for Zn. Pooled serum had the assigned value against the standard reference material. The pooled serum was stored in a freezer (−20°C) and analysed with every batch of samples along with the commercial control material. If the values of these preliminary analyses had fallen within the assigned ranges for the QC, the regular QC assays (pool and commercial) for all samples were performed. The mean values for the pool and the commercial QC along with standard deviations and 95 % CI were calculated. For Zn, the CV against the pooled serum, QC-normal and QC-high were 4·5, 4·1 and 4·0 %, respectively. Precinorm protein and Precipath protein (Roche Diagnostics GmbH), as QC, were used to check the accuracy and precision of CRP and AGP. The CV for CRP and AGP measurement were 3·9 and 5·9 %, respectively. We participated in the VITAL-EQA (Centers for Diseases Control) for CRP.

Sample size

The sample size to estimate Zn undernutrition was calculated based on the prevalence of low serum Zn observed in a contemporary study in Bangladesh( Reference Arsenault, Yakes and Hossain 15 ). The required sample size for PSAC and NPNLW was 969 and 1500, respectively, taking into consideration an attrition factor of 1·2 and a design effect 2·0. However, the final analysed sample number was 662 and 1073, respectively, taking into consideration the availability of both the biomarkers of infection (CRP and AGP) for analysis, in order to adjust for infection. In addition, losses in sample number occurred due to inadequate serum in aliquots, haemolysis, and error in labelling of sample identification by the field staff. This resulted in some overall loss of precision in the prevalence estimates: 10·3 % precision instead of the stipulated 7 % in PSAC and 6·2 % precision in NPNLW in place of 6 % (Table 1 and Supplementary Table S3).

Table 1. Prevalence of zinc deficiency*

(Number of subjects, percentages and 95 % confidence intervals)

PSAC, pre-school-age children; NPNLW, non-pregnant non-lactating women; CRP, C-reactive protein; AGP, α-1-acid glycoprotein.

* Zn deficiency is defined as serum Zn level of <9·9 µmol/l in PSAC and <10·1 µmol/l in NPNLW( Reference Brown, Rivera and Bhutta 1 ).

Adjusted for elevated CRP (>10 mg/l) or elevated AGP (>1 g/l) by mathematical correction( Reference Islam, Khan and Akhtaruzzaman 12 , Reference Thurnham, McCabe and Haldar 13 ).

Estimates weighted to represent at the population level.

§ Age 6–59 months.

Age 15–49 years.

Statistical analysis

We performed the statistical analyses using STATA 10.0 SE (StataCorp LP). We calculated the key proportion estimates with a 95 % CI; the mean estimates were calculated with standard deviations through one-way ANOVA. Pearson's χ2 test was used to compare proportions. Projected estimates of the population in the three strata (rural, urban and slum) in 2011 were 122·6, 23·3 and 5·5 million, respectively( 16 , 17 ). A total of fifty primary sampling units (PSU) were selected from each stratum, hence resulting in differential selection probabilities and representation across the strata. Sampling weights were applied to the households in each stratum to compensate for the differential representation and to derive weighted national estimates combining the estimates of the three strata. Prior to performing statistical tests, we treated the variables with skewed distribution with logarithmic transformation to convert into normal distribution. Statistical analyses, e.g. proportion estimates, one-way ANOVA and multivariate regression analyses were performed on weighted data. Through using the cluster option in STATA, the analyses, including the regression models, were adjusted for clustering of data at the unit of the randomisation level (PSU). The relevant socio-economic, demographic and dietary intake predictor variables which are logically presumed to have association with serum Zn were entered in univariate linear regression or bivariate analysis. If there was a significant association with serum Zn at P < 0·05 and it had the expected sign mark (i.e. plus or minus sign of the coefficient according to the logical expectation), the variables were selected for the multivariate regression( Reference Gelman and Hill 18 ). Interaction analyses were performed taking into consideration the variables that logically appear to be interacting in determining serum Zn status. The predictor variables, which are assumed to have interaction, were entered in the initial regression analyses as an interaction term along with the interacting predictors. If the regression coefficient for the interaction term was non-significant (P ≥ 0·05), it was omitted; however, in cases where the interaction term was statistically significant (P < 0·05), it was retained and entered into the final multivariate model( Reference Mason, Gillenwater and Pugh 19 ).

Cronbach's α reliability coefficients for internal consistency of data were calculated considering the variables of SES, dietary consumption, household food insecurity and serum level of Zn. Cronbach's α in data was 0·8179 and 0·8159, respectively, in PSAC and NPNLW, which is suggestive of good internal consistency( Reference George and Mallery 20 ).

Results

The national prevalence of Zn deficiency in pre-school children was 44·6 % (Table 1). It was 29·5, 48·6 and 51·7 %, respectively, in the urban, rural and slum strata. Of the NPNLW, 57·3 % suffered from Zn deficiency at the national level; the prevalence being highest in women living in slums (66·4 %). The national mean of serum Zn was 10·2 and 10·04 µmol/l in PSAC and the NPNLW, respectively. The mean level of serum Zn appeared to be higher in the urban stratum than in the slum stratum in both the populations (P < 0·001, P = 0·017, respectively) (Table 2).

Table 2. Mean zinc concentration in serum (μmol/l) §

(Mean values and standard deviations)

PSAC, pre-school-age children; NPNLW, non-pregnant non-lactating women; CRP, C-reactive protein; AGP, α-1-acid glycoprotein.

Mean value was significantly different from that for slum: *P = 0·017, ** P = 0·012, *** P < 0·001.

††† Mean value was significantly different from that for rural (P < 0·001).

Adjusted for elevated CRP (>10 mg/l) or elevated AGP (>1 g/l) by mathematical correction( Reference Ali 11 , Reference Islam, Khan and Akhtaruzzaman 12 ).

§ Estimates weighted to represent at the population level.

Age 6–59 months.

Age 15–49 years.

‡‡ One-way ANOVA.

The prevalence of inadequate intake of Zn in PSAC was 32·6 % at the national level. The respective estimates for rural, urban and slum domains were 33·0, 28·7 and 44·1 %. In NPNLW, 91·1 % had inadequate intake of Zn at the national level; by and large the estimates remained at similar levels across the strata (Table 3). The national average consumption of dietary Zn in PSAC was 3·1 mg/d, the urban intake being higher than in slums (3·2 v. 2·6 mg/d; P = 0·004). At the national level mean intake of Zn in NPNLW was 4·2 mg/d. The urban women consumed higher amounts than their peers from slums (4·6 mg v. 4·1 mg/d; P = 0·001; Table 3). Fig. 1 further depicts the relative intake profile.

Fig. 1. Comparative intake of dietary zinc and phytate in urban and slum strata. Intakes of total zinc and animal-source zinc were significantly higher in the urban stratum than in the slum stratum in both the pre-school-age children (PSAC) and non-pregnant non-lactating women (NPNLW) populations: 22·4 mg/7 d (urban) v. 18·2 mg/7 d (slum) (P = 0·004) and 9·3 mg/7 d (urban) v. 7·3 mg/7 d (slum) (P = 0·005), respectively, for total and animal-origin zinc in PSAC. However, intake of phytate was at similar levels: 1567 mg/7 d (urban) v. 1533 mg/7 d (slum) (NS) in PSAC. Similar profiles of the intake were observed in NPNLW. Significantly different from slum: * P = 0·01, ** P = 0·004, *** P = 0·005, **** P = 0·001.

Table 3. Dietary intake of zinc and prevalence of inadequacy of zinc intake

(Mean values and standard deviations; percentages with their standard errors)

PSAC, pre-school-age children; NPNLW, non-pregnant non-lactating women.

Mean value was significantly different from that for slum: * P = 0·004, ** P = 0·001.

One-way ANOVA.

Inadequate Zn intake: <2 mg/d in PSAC; <7 mg/d in NPNLW( Reference Brown, Rivera and Bhutta 1 ).

Less than 1·0 and 6·4 % of the NPNLW aged 15–18 and 19–49 years, respectively, met the RDA for Zn( 21 ). In PSAC, 44·7 % of the 2- to 3-year-olds and 11·4 % of the 4- to 5-year-olds met the RDA( 21 ) (Table 4). Mean intake of Zn was 2·8 and 3·2 mg/d in 2- to 3- and 4- to 5-year-old groups, respectively; it was 4·0 and 4·3 mg/d in NPNLW aged 15–18 and 19–49 years, respectively. Intake of animal-source Zn was 1·13–1·15 mg/d in PSAC and 1·28–1·33 mg/d in NPNLW according to age subgroups (Table 4). Intake of animal-origin Zn progressively increased as household SES and household expenses were higher; an increasing level of food insecurity was associated with progressively decreasing intake in children and women. However, intake of phytate remained at similar levels irrespective of SES and household food security status (Table 5).

Table 4. Intake of zinc v. RDA

(Mean values with their standard errors; percentages with their standard errors)

PSAC, pre-school-age children; NPNLW, non-pregnant non-lactating women.

* Institute of Medicine( 21 ).

D = C/A × 100.

F = E/C × 100.

Table 5. Intake of animal-source zinc and phytate by socio-economic status (SES) and household food insecurity

(Mean values and standard deviations)

USD, US dollars; NPNLW, non-pregnant non-lactating women; PSAC, pre-school-age children; FS, food secure; MI, moderate insecure; SI, severe insecure.

a,b,c Mean values within a column within a category with unlike letters were significantly different (P < 0·001).

* Mean value was significantly different from that for FS (P < 0·001).

Mean value was significantly different from that for SI (P = 0·001).

One-way ANOVA.

According to dietary patterns of ‘mixed’ and ‘cereal-based’ diets, inadequacy of Zn intake in women was pervasively prevalent at 90 and 100 %, respectively. As per dietary type mean intakes of Zn were 4·5 and 2·4 mg/d, respectively (P < 0·001; Table 6). In accordance with the absorption pattern for mixed and cereal-based diets( Reference Brown, Rivera and Bhutta 1 ), the amount of absorbed Zn was 1·41 and 0·56 mg/d which fulfilled 75·8 and 30·1 % of physiological requirements, respectively. Intake of animal-source food, according to mixed and cereal-based diets was: meat, 27·6 v. 6·7 g/d (P < 0·001); fish, 58·6 v. 22·5 g/d (P < 0·001); and eggs, 11·7 v. 3·4 g/d (P < 0·001), respectively. Intake of cereals was 337·1 and 457·4 g/d, respectively (P < 0·001).

Table 6. Zinc status in non-pregnant non-lactating women sorted by dietary type and food intake

(Percentages with their standard errors, and mean values and standard deviations)

Significantly different from cereal-based diet: * P = 0·32, ** P < 0·001.

Dietary Zn intake <7 mg/d (estimated average requirement) in women( Reference Brown, Rivera and Bhutta 1 ).

Absorption in women of 31 and 23 % for mixed and unrefined cereal-based diets, respectively( Reference Brown, Rivera and Bhutta 1 ).

§ D = C × 0·31 and D = C × 0·23 for mixed and cereal based diets, respectively( Reference Brown, Rivera and Bhutta 1 ).

Physiological requirement of absorbed Zn in adult women( Reference Brown, Rivera and Bhutta 1 ).

F = D/E × 100.

†† Phytate-Zn molar ratio (4–18)( Reference Brown, Rivera and Bhutta 1 ).

‡‡ Phytate-Zn molar ratio (>18)( Reference Brown, Rivera and Bhutta 1 ).

By and large, there was an upward gradient in serum levels of Zn concurring with progressively increased intake of dietary Zn in PSAC; however, in NPNLW that trend was not observed (Table 7).

Table 7. Dietary intake of zinc and serum levels of zinc

(Mean values with their standard errors)

PSAC, pre-school-age children; NPNLW, non-pregnant non-lactating women.

* One-way ANOVA.

Table 8 shows the results of the multivariate linear regression analyses stating associations of serum Zn status in the studied populations after controlling for the relevant covariates. Multivariate linear regression analyses suggested that household expenses (β = 0·13; P = 0·007), Hb (β = 0·10; P = 0·005), intake of animal-source Zn (β = 0·096; P = 0·02) and asset score (β = 0·11; P = 0·03) determined higher levels of serum Zn in NPNLW. Non-agricultural occupation of the household head was nearly positively associated with serum Zn in NPNLW (β = 0·07; P = 0·054). Intake of plant-source Zn determined the lower status of Zn in NPNLW (β = −0·13; P = 0·001). Residence in an urban area was associated with higher serum Zn in PSAC compared with their peers from slum areas (β = 0·33; P = 0·03). Intake of plant-origin Zn was associated with lower status of Zn in PSAC (β = −0·13; P = 0·038), while intake of animal-source Zn was nearly associated with higher levels of serum Zn (β = 0·13; P = 0·07).

Table 8. Multivariate regression determining zinc level in serum in pre-school-age children (PSAC) and non-pregnant non-lactating women (NPNLW)

HFIAS, Household Food Insecurity Access Scale.

* Interaction terms – (occupation of household head × intake of animal-source Zn), (occupation of household head × intake of plant-source Zn), (asset score × intake of animal-source Zn), (household expenses × intake of animal-source Zn), (stratum × intake of animal-source Zn), and (household expenses × educational status of women) were separately assessed through initial regression equations, along with component predictors. However, none of the interaction terms reached statistical significance level at P < 0·05 except for (household expenses × occupation of household head) (P = 0·049), which was considered in the final multivariate model; however, it had a non-significant interaction (P = 0·44) and as such was removed from the model.

Interaction terms – all the interaction terms assessed for the PSAC were tested in NPNLW. In the initial regression analyses none had significant interaction except for the household expenses × intake of animal-source Zn (P = 0·03). It was considered in the final multivariate model where it is reported to have significant interaction (P = 0·01).

Discussion

The finding of 44 % prevalence of subclinical Zn deficiency in preschool children was close to the contemporary stunting prevalence of 41 % in Bangladesh( 3 ), reaffirming the notion that the stunting estimate can be a proxy indicator of Zn undernutrition in children. The dietary intake of Zn in the present study (Table 3) was consistent with a contemporary study which reported intake of 2·5 mg in young children and 5·5 mg in women in two subdistricts( Reference Arsenault, Yakes and Hossain 15 ). The same study reported the prevalence of inadequate intake of Zn to be 22 and 94 % in children and women, respectively, which approximately corresponds to our findings (Table 3).

In PSAC, increasing intake of Zn to roughly twice the amount (from 3·0 to 6·1 mg) was associated with a 7 % increment in serum Zn level (from 9·74 to 10·42 µmol/l; Table 7). A recent meta-analysis pooling all eligible randomised controlled trials of Zn supplementation modelled that doubling of Zn intake was associated with a 9 % increment in serum Zn levels( Reference Moran, Stammers and Medina 22 ). Hence our observation of a 7 % increase from this cross-sectional national survey is by and large consistent with the experimentally designed global model predicting Zn supplementation and its reflection in serum status. This is an important finding in relation to the designing of Zn promoting interventions for children in the country. In NPNLW, the approximated doubling of intake (from 2·9 to 6·0 mg, from 4·0 to 7·9 mg; Table 7) was associated with a less marked increase in serum Zn (1·2 and 4·6 %, respectively). Somewhat inconsistent Zn status in NPNLW in relation to dietary intake perhaps has resulted from the fact that Zn status varies with exercise, stress, starvation and timing of food intake( Reference Ashwell, Lambert and Alles 23 ). Bangladeshi women are likely to experience these in day-to-day affairs. However, these factors were not explored in the survey.

The multivariate regression showed intake of animal-source Zn was associated with a higher level of serum Zn in NPNLW (Table 8). The underlying mechanism is unclear; however, other studies reported animal protein being associated with a greater percentage absorption of dietary Zn( Reference Sandström 24 , Reference Lönnerdal 25 ).

Household expenses which signify household spending capacity were associated with a higher level of serum Zn in NPNLW. The underlying mechanism into this is not clear; however, it might be attributed to the higher consumption of higher bioavailable animal-origin Zn by women from households with higher financial status (Supplementary Table S1). Multivariate regression showed that Hb level was positively associated with serum Zn in NPNLW (Table 8). Complementing this, the prevalence of subclinical Zn deficiency was found to be higher in anaemic women than in non-anaemics (Supplementary Table S1). This might be explained by common dietary sources of Zn and Fe or the possible role of Zn in erythropoiesis( Reference Gibson, Abebe and Stabler 26 Reference Folin, Contiero and Vaselli 28 ).

Multiple regressions showed that the PSAC living in urban areas had higher levels of serum Zn compared with slum areas. It is in agreement with a lower prevalence of subclinical Zn deficiency in urban PSAC than their peers in the slums (Table 1). The underlying reason is difficult to explain; however, we assume that a 23·1 and 27·4 % higher intake of total and animal-source Zn, respectively, might have accounted for the higher status of serum Zn in urban PSAC (Fig. 1). On the other hand, in NPNLW, urban residence was not associated with higher status of Zn (Table 8). Underlying this is perhaps the intakes of total and animal-source Zn which were only marginally higher (13·7 and 17·2 %, respectively; Fig. 1) in urban women, while the intake of phytate was at similar levels.

Household asset score determined higher status of Zn in NPNLW. This could be plausibly explained by differential intakes of higher bioavailable animal-origin Zn which was about 2·5 times higher in the ‘richest’ households than their peers from the ‘poorest’ (Table 5). However, in PSAC the asset score was not a predictor of Zn status. The reason for the differential relationship can be explained by the bioavailability of Zn, an issue that perhaps affects women more in Bangladesh. This is evident from the data in Table 5; for example, in the ‘poorest’ SES, while women's consumption of animal-source Zn was 33 % higher than in the ‘poorest’ PSAC (0·8 v. 0·6 mg/d), the intake of phytate (Zn chelator) was 207 % higher in the poor women, compared with the poor young children (3388 v. 1636 mg/7 d). Thus, the PSAC are likely to be affected less than the women in relation to bioavailability of Zn as a result of a lower consumption of phytate. This might lead to a relatively favourable bioavailability of Zn in PSAC across a wide range of SES.

Dietary intake of phytate was at similar levels across the SES or household food security status in both the populations (Table 5). This is perhaps one of the fundamental issues behind the population-wide high prevalence of subclinical Zn undernutrition in Bangladesh.

The strength of the study lies at the consideration of multiple domains – rural, urban and slum, where status of Zn nutrition and attribution of underlying correlates, such as, SES and food consumption status were different, and thus providing a representative status of Zn nutrition of the country. Although the semi-quantitative FFQ has been increasingly used and validated in other countries( Reference Cheng, Yan and Dibley 7 Reference Buch Andersen, Perez-Cueto and Toft 10 ), it is new in Bangladesh, and hence, it was not validated in Bangladeshi populations. While this is a limitation of the study, when comparing the intake data in the survey with another contemporary study in the country( Reference Arsenault, Yakes and Hossain 15 ) in nearly comparable population groups, the results appear to be similar (Table 3, Supplementary Table S2). Although the semi-quantitative FFQ has been validated for Zn( Reference Cheng, Yan and Dibley 7 , Reference Zhang, Qiu and Zhong 8 ); it is not directly validated for phytate, which is a limitation of the study. However, the tool is validated for cereals, beans and dietary fibre( Reference Cheng, Yan and Dibley 7 , Reference Zhang, Qiu and Zhong 8 ), and these food components are the major sources of phytate.

The timing of blood sample collection in relation to the last meal is important for assessing serum Zn status( Reference Arsenault, Wuehler and López de Romaña 29 ). However, we could not arrange a time protocol for sample collection due to the limited amount of time to complete data collection in the clusters. Also, because the respondents visited the collection centres according to their convenience in the day, this was a limitation as well.

In conclusion, subclinical Zn deficiency is pervasively prevalent in Bangladesh in vulnerable populations. The worst hit by the condition are women, people living in lower SES and in the slums. A high burden of Zn undernutrition is related to the intake lower than the requirements, especially that of the higher bioavailable animal-source Zn and the high intake of plant-origin phytate-bound Zn. Bangladesh needs to strengthen research and programmes related to Zn supplementation, biofortification, industrial fortification and phytate-reducing technologies in the food system in the short and medium term, along with promotion of animal-source Zn for all in the long run.

Supplementary material

The supplementary material for this article can be found at http://dx.doi.org/10.1017/jns.2016.17

Acknowledgements

The study acknowledges support of UNICEF Bangladesh. We thank the national technical committee for guidance into the study design, data collection and data analysis. We thank the local government officials, field-level staff members of the Institute of Public Health Nutrition, Government of People's Republic of Bangladesh, and the survey participants from all over Bangladesh.

The study was funded by UNICEF Bangladesh (grant number: SC/2008/0475-, GC/2005/6012-01). UNICEF Bangladesh was involved in the conceptualisation and designing of the study and review of the manuscript but had no role in data collection, data analysis, data interpretation and manuscript writing.

S. R., T. A., A. S. R., N. A., I. A. C. and A. M. S. A. conceived and designed the study. S. R., T. A., F. P. C. and S. M. M. R. acquired the data. S. R. and N. A. analysed the data. S. R., A. S. R. and N. A. interpreted the data. S. R. drafted the manuscript. T. A., S. I., A. S. R., N. A., S. M. M. R. and I. A. C. contributed to redrafting and finalising the manuscript. S. R. has full access to all the data in the study and had final responsibility for the decision to submit for publication. All authors read and approved the final manuscript.

There are no conflicts of interest.

References

1. Brown, KH, Rivera, JA, Bhutta, Z, et al. (2004) International Zinc Nutrition Consultative Group (IZiNCG) technical document #1. Assessment of the risk of zinc deficiency in populations and options for its control. Food Nutr Bull 25, 1 Suppl. 2, S99S203.Google Scholar
2. Bhutta, ZA, Black, RE, Brown, KH, et al. (1999) Prevention of diarrhea and pneumonia by zinc supplementation in children in developing countries: pooled analysis of randomized controlled trials. J Pediatr 135, 689697.CrossRefGoogle ScholarPubMed
3. National Institute of Population Research and Training (NIPORT), Mitra and Associates, and ICF International (2013) Bangladesh Demographic and Health Survey 2011. Dhaka, Bangladesh and Calverton, MD: NIPORT, Mitra and Associates, and ICF International.Google Scholar
4. BBS/UNICEF (2010) Multiple Indicator Cluster Survey (MICS) 2009, Progotir Pathey, Volume I: Technical Report. http://www.unicef.org/bangladesh/MICS-PP-09-v10.pdf (accessed May 2016).Google Scholar
5. Gwatkin, DR, Rustein, S, Johnson, K, et al. (2000) Socioeconomic Differences in Health, Nutrition, and Population in Bangladesh, HNP/Poverty Thematic Group Working Paper. Washington, DC: The World Bank.Google Scholar
6. Coates, J, Swindale, A & Bilinsky, P (2007) Household Food Insecurity Access Scale (HFIAS) for Measurement of Food Access: Indicator Guide, version 3. Washington, DC: Food and Nutrition Technical Assistance III Project.Google Scholar
7. Cheng, Y, Yan, H, Dibley, MJ, et al. (2008) Validity and reproducibility of a semi quantitative food frequency questionnaire for use among pregnant women in rural china. Asia Pac J Clin Nutr 17, 166177.Google ScholarPubMed
8. Zhang, H, Qiu, X, Zhong, C, et al. (2015) Reproducibility and relative validity of a semi quantitative food frequency questionnaire for Chinese pregnant women. Nutr J 14, 56.CrossRefGoogle ScholarPubMed
9. Tang, Y, Liu, Y, Xu, L, et al. (2015) Validity and reproducibility of a revised semi-quantitative food frequency questionnaire (SQFFQ) for women of age-group 12–44 years in Chengdu. J Health Popul Nutr 33, 5059.Google ScholarPubMed
10. Buch Andersen, T, Perez-Cueto, FJA, Toft, U, et al. (2016) Relative validity and reproducibility of a parent-administered semi quantitative FFQ for assessing food intake in Danish children aged 3–9 years. Public Health Nutr 19, 11841194.CrossRefGoogle ScholarPubMed
11. Ali, SMK (1991) Conversion Factors and Dietary Calculation. Dhaka: University of Dhaka, Institute of Nutrition and Food Sciences.Google Scholar
12. Islam, SN, Khan, MNI & Akhtaruzzaman, M (2010) A Food Composition Database for Bangladesh with Special Reference to Selected Ethnic Foods. Dhaka: Institute of Nutrition and Food Sciences, University of Dhaka.Google Scholar
13. Thurnham, DI, McCabe, LD, Haldar, S, et al. (2010) Adjusting plasma ferritin concentrations to remove the effects of subclinical inflammation in the assessment of iron deficiency: a meta-analysis. Am J Clin Nutr 92, 546555.CrossRefGoogle ScholarPubMed
14. Engle-Stone, R, Haskell, MJ, Ndjebayi, AO, et al. (2011) Plasma retinol-binding protein predicts plasma retinol concentration in both infected and uninfected Cameroonian women and children. J Nutr 141, 22332241.CrossRefGoogle ScholarPubMed
15. Arsenault, JE, Yakes, EA, Hossain, MB, et al. (2010) The current high prevalence of dietary zinc inadequacy among children and women in rural Bangladesh could be substantially ameliorated by zinc biofortification of rice. J Nutr 140, 16831690.CrossRefGoogle ScholarPubMed
16. Bangladesh Bureau of Statistics (2007) Population Census–2001, National Series. Vol. 1: Analytical Report. Dhaka: Bangladesh Bureau of Statistics, Planning Division, Ministry of Planning, Government of Bangladesh.Google Scholar
17. Center for Urban Studies, National Institute of Population Research and Training & MEASURE Evaluation (2006) Slums of Urban Bangladesh: Mapping and Census 2005. Dhaka and Chapel Hill, NC: CUS, NIPORT and MEASURE Evaluation.Google Scholar
18. Gelman, A & Hill, J (2007) Data Analysis Using Regression and Multilevel/Hierarchical Models, p. 69. New York: Cambridge University Press.Google Scholar
19. Mason, J, Gillenwater, K, Pugh, R, et al. (2012) Practical Analysis of Nutritional Data (PANDA), Section 3. http://www.tulane.edu/~panda3/FS/Computing/regression/interact.htm (accessed March 2015).Google Scholar
20. George, D & Mallery, P (2003) SPSS for Windows Step by Step: A Simple Guide and Reference, 11.0 Update, 4th ed. Boston, MA: Allyn & Bacon.Google Scholar
21. Institute of Medicine (2002) Dietary Reference Intakes for Vitamin A, Vitamin K, Arsenic, Boron, Chromium, Copper, Iodine, Iron, Manganese, Molybdenum, Nickel, Silicon, Vanadium, and Zinc. Washington, DC: National Academies Press.Google Scholar
22. Moran, VH, Stammers, AL, Medina, MW, et al. (2012) The relationship between zinc intake and serum/plasma zinc concentration in children: a systematic review and dose–response meta-analysis. Nutrients 4, 841858.CrossRefGoogle ScholarPubMed
23. Ashwell, M, Lambert, JP, Alles, MS, et al. (2008) How we will produce the evidence-based EURRECA toolkit to support nutrition and food policy. Eur J Nutr 47, 216.CrossRefGoogle ScholarPubMed
24. Sandström, B (1992) Dose dependence of zinc and manganese absorption in man. Proc Nutr Soc 51, 211218.CrossRefGoogle ScholarPubMed
25. Lönnerdal, B (2000) Dietary factors influencing zinc absorption. J Nutr 130, 1378S1383S.CrossRefGoogle ScholarPubMed
26. Gibson, RS, Abebe, Y, Stabler, S, et al. (2008) Zinc, gravid, infection and iron but not vitamin B-12 or folate status predict haemoglobin during pregnancy in southern Ethiopia. J Nutr 138, 581586.CrossRefGoogle ScholarPubMed
27. de Jong, N, Ampong Romano, AB & Gibson, RS (2002) Zinc and iron status during pregnancy of Filipino women. Asia Pac J Clin Nutr 11, 186193.CrossRefGoogle ScholarPubMed
28. Folin, M, Contiero, E & Vaselli, GM (1994) Zinc content of normal human serum and its correlation with some hematic parameters. Biometals 7, 7579.CrossRefGoogle ScholarPubMed
29. Arsenault, JE, Wuehler, SE, López de Romaña, D, et al. (2011) The time of day and the interval since previous meal are both associated with young children's plasma zinc concentrations in Peru and Ecuador, and therefore affect estimates of the populations’ risk of zinc deficiency. Eur J Clin Nutr 65, 184190.CrossRefGoogle ScholarPubMed
Figure 0

Table 1. Prevalence of zinc deficiency*†‡(Number of subjects, percentages and 95 % confidence intervals)

Figure 1

Table 2. Mean zinc concentration in serum (μmol/l)‡§(Mean values and standard deviations)

Figure 2

Fig. 1. Comparative intake of dietary zinc and phytate in urban and slum strata. Intakes of total zinc and animal-source zinc were significantly higher in the urban stratum than in the slum stratum in both the pre-school-age children (PSAC) and non-pregnant non-lactating women (NPNLW) populations: 22·4 mg/7 d (urban) v. 18·2 mg/7 d (slum) (P = 0·004) and 9·3 mg/7 d (urban) v. 7·3 mg/7 d (slum) (P = 0·005), respectively, for total and animal-origin zinc in PSAC. However, intake of phytate was at similar levels: 1567 mg/7 d (urban) v. 1533 mg/7 d (slum) (NS) in PSAC. Similar profiles of the intake were observed in NPNLW. Significantly different from slum: * P = 0·01, ** P = 0·004, *** P = 0·005, **** P = 0·001.

Figure 3

Table 3. Dietary intake of zinc and prevalence of inadequacy of zinc intake(Mean values and standard deviations; percentages with their standard errors)

Figure 4

Table 4. Intake of zinc v. RDA(Mean values with their standard errors; percentages with their standard errors)

Figure 5

Table 5. Intake of animal-source zinc and phytate by socio-economic status (SES) and household food insecurity(Mean values and standard deviations)

Figure 6

Table 6. Zinc status in non-pregnant non-lactating women sorted by dietary type and food intake(Percentages with their standard errors, and mean values and standard deviations)

Figure 7

Table 7. Dietary intake of zinc and serum levels of zinc(Mean values with their standard errors)

Figure 8

Table 8. Multivariate regression determining zinc level in serum in pre-school-age children (PSAC) and non-pregnant non-lactating women (NPNLW)

Supplementary material: File

Rahman supplementary material

Tables S1-S3 and References

Download Rahman supplementary material(File)
File 16.2 KB