IMPROVING ANTENATAL IRON-CONTAINING SUPPLEMENTATION INDICATORS

(1)

IMPROVING ANTENATAL IRON-CONTAINING

SUPPLEMENTATION INDICATORS

REPORT OF KEY INFORMANT INTERVIEWS, ONLINE SURVEY, AND DHS DATA ANALYSES

(2)

(3)

IMPROVING ANTENATAL IRON-CONTAINING

SUPPLEMENTATION INDICATORS

REPORT OF KEY INFORMANT INTERVIEWS, ONLINE SURVEY, AND DHS DATA ANALYSES

(4)

Improving antenatal iron-containing supplementation indicators: a report on key informant interviews, an online survey and DHS data analyses

ISBN 978-92-4-000006-3 (WHO)

© World Health Organization and the United Nations Children’s Fund (UNICEF), 2019

Some rights reserved. This work is available under the Creative Commons Attribution-NonCommercial-ShareAlike 3.0 IGO licence (CC BY-NC-SA 3.0 IGO; https://creativecommons.org/licenses/by-nc-sa/3.0/igo).

Under the terms of this licence, you may copy, redistribute and adapt the work for non-commercial purposes, provided the work is appropriately cited, as indicated below. In any use of this work, there should be no suggestion that WHO endorses any specific organization, products or services. The use of the WHO logo is not permitted. If you adapt the work, then you must license your work under the same or equivalent Creative Commons licence. If you create a translation of this work, you should add the following disclaimer along with the suggested citation: “This translation was not created by the World Health Organization (WHO). WHO is not responsible for the content or accuracy of this translation. The original English edition shall be the binding and authentic edition”.

Any mediation relating to disputes arising under the licence shall be conducted in accordance with the mediation rules of the World Intellectual Property Organization (http://www.wipo.int/amc/en/mediation/rules).

Suggested citation. Improving antenatal iron-containing supplementation indicators: a report on key informant interviews, an online survey and DHS data analyses. Geneva: World Health Organization and the United Nations Children’s Fund (UNICEF), 2019. Licence: CC BY-NC-SA 3.0 IGO.

Cataloguing-in-Publication (CIP) data. CIP data are available at http://apps.who.int/iris.

Sales, rights and licensing. To purchase WHO publications, see http://apps.who.int/bookorders. To submit requests for commercial use and queries on rights and licensing, see http://www.who.int/about/licensing.

Third-party materials. If you wish to reuse material from this work that is attributed to a third party, such as tables, figures or images, it is your responsibility to determine whether permission is needed for that reuse and to obtain permission from the copyright holder. The risk of claims resulting from infringement of any third-party-owned component in the work rests solely with the user.

General disclaimers. The designations employed and the presentation of the material in this publication do not imply the expression of any opinion whatsoever on the part of WHO concerning the legal status of any country, territory, city or area or of its authorities, or concerning the delimitation of its frontiers or boundaries. Dotted and dashed lines on maps represent approximate border lines for which there may not yet be full agreement.

The mention of specific companies or of certain manufacturers’ products does not imply that they are endorsed or recommended by WHO in preference to others of a similar nature that are not mentioned. Errors and omissions excepted, the names of proprietary products are distinguished by initial capital letters.

All reasonable precautions have been taken by WHO to verify the information contained in this publication. However, the published material is being distributed without warranty of any kind, either expressed or implied. The responsibility for the interpretation and use of the material lies with the reader. In no event shall WHO be liable for damages arising from its use.

Printed in Switzerland

Design and cover: Alberto March (Barcelona, Spain)

(5)

LIST OF TABLES

Table 1. Anaemia prevalence in countries in which KII were completed 3 Table 2. Beneficiaries included in national guidelines for iron and/or micronutrient supplement

provision in countries in which KII were completed 5

Table 3. Antenatal iron supplementation indicator used in countries in which KII were completed 6 Table 4. Key results on collection and use of data on coverage/consumption of iron supplements 7

Table 5. Fields of work occupied by respondents 8

Table 6. Types of organization employing respondents 9

Table 7. Status of the national guideline for iron supplement provision in countries

mentioned in the online survey 9

Table 8. Availability of iron supplements at health facilities in countries mentioned in

the online survey 10

Table 9. Satisfaction with how iron supplement coverage/consumption is assessed

in national surveys 11

Table 10. Selected respondents’ suggestions on how questions assessing iron supplement coverage/consumption should be revised in national surveys 11 Table 11. Use of data collected on iron supplement coverage/consumption, based on

respondents’ comments 12

Table 12. Countries included in secondary data analysis of iron consumption data quality 14 Table 13. Iron coverage/consumption questions included in the Pregnancy and

Postnatal Care section of the DHS Model Questionnaire – Phase 7 14 Table 14. Data quality level guidelines used at Statistics Canada 15 Table 15. Iron supplement consumption during pregnancy in countries selected for

DHS data quality analyses 16

Table 16. Descriptive analysis of variables included in DHS data quality analysis, by country 16

Table 17. Evidence of data heaping in DHS datasets 17

Table 18. Reported days of iron tablet/syrup consumption by postpartum women in

three DHS datasets 18

Table 19. Reported days of consumption by recall period among those postpartum women who consumed iron tablets/syrup for at least one day in Afghanistan,

Myanmar and Tanzania 18

(7)

Table 20. Results of linear regression analyses between iron consumption and recall period, number of ANC visits and maternal educational attainment, by country 20 Table 21. Precision of iron consumption data collected in DHS by recall period and country 21 Table C 1. Key informant interview (KII) guide for WHO TEAM exploratory analysis of iron

supplementation data collection and usage 30

Table C 2. List of online survey respondents’ countries 32

Table C 3. List of countries, year and name of most recent survey assessing iron supplement

coverage/consumption 34

Table C 4. Respondents’ suggestions on how questions assessing iron supplement coverage/

consumption should be revised in national surveys 36

Table C 5. Details of how data collected on iron supplement coverage/consumption is used 38 Table D 1. Questions on the age of the most recent child, number of ANC visits and maternal

education included in the Pregnancy and Postnatal Care section of the

DHS Model Questionnaire – Phase 7 40

LIST OF FIGURES

Fig. 1. Flowchart of number of respondents and countries answering each category

of survey questions 8

Fig. 2. Directed acyclic graph (DAG) representing the relationship between antenatal IFA

consumption recall (outcome) and maternal sociodemographic factors and recall period 13 Fig. B 1. Histograms of antenatal iron consumption in DHS datasets 27

(8)

ACKNOWLEDGEMENTS

This technical report is an outcome of research by the WHO-UNICEF Technical Expert Advisory group on nutrition Monitoring (TEAM), which was supported by the Bill & Melinda Gates Foundation.

All TEAM members (in alphabetical order): Mary Arimond, Jennifer Coates, Trevor Croft, Omar Dary, Rafael Flores-Ayala, Edward Frongillo Jr., Rebecca Heidkamp, Purnima Menon, Lynnette Neufeld, Faith Thuita, Sara Wuehler and Wenhua Zhao, for providing feedback on draft findings.

Kuntal Kumar Saha (WHO headquarters), for conceptualization and leading this research;

Sara Wuehler and Rebecca Heidkamp (TEAM), for conceptualization and technical contributions to the development, implementation, interpretation and reporting; Aatekah Owais (Nutrition International, Canada), for conducting this research, data analyses and interpretation, and report writing; and Vrinda Mehra (UNICEF) and Lisa Rogers (WHO headquarters), for their feedback on the draft version of the report.

All key informants and online survey participants, for their time, valuable inputs and thoughts.

(9)

EXECUTING SUMMARY

W

HO-UNICEF’s Technical Expert Advisory group for nutrition Monitoring (TEAM) is leading efforts to propose a global standard indicator on antenatal iron supplementation coverage.

A scoping exercise was carried out in 2017–2018 to assess the feasibility of reporting using the currently defined antenatal iron supplementation coverage indicator. The report concluded that currently available data sources are not sufficient to allow complete reporting with this indicator, and recommended that an interim indicator, defined as “any antenatal iron supplementation”, be used. This recommendation was included in the measurement guidance in the 2017 Global Nutrition Monitoring Framework (GNMF).

The current report builds on these efforts and advances towards a new globally accepted indicator which defines antenatal iron supplementation coverage. It aims to harmonize data collection and reporting by reviewing the use of current antenatal supplementation indicators and to determine whether analyses of available data could provide a better evidence base for the development and validation of a global antenatal iron supplement indicator.

Key informant interviews (N=9) were carried out in late 2018 to early 2019 in eight countries with a varying burden of anaemia prevalence among women of reproductive age. A similar online survey, with respondents (N=145) from 52 countries, was also carried out between 25 March and 22 April 2019 to obtain information on how iron coverage/consumption is assessed in national surveys across the globe and how data are used. Lastly, Demographic and Health Survey (DHS) data from four countries were analysed to identify whether there are measurable determinants governing the quality of reported iron supplement consumption data in the absence of observed consumption data.

Key informants and online survey respondents raised similar issues about how data on iron supplement coverage and consumption among pregnant women are collected and used. The concerns raised related to the potential recall bias due to the long retrospective reporting period, the lack of data on nutrition counselling for pregnant women and other beneficiary groups, and concurrent supply-side issues.

DHS data were assessed based on recall period, number of ANC visits and maternal education for potential measures of data quality including variability, data heaping, frequency of “Don’t know” responses and association with iron consumption modelled using linear regression. Trends were found across recall periods suggestive of a reduction in data quality as the recall period lengthened, but were not statistically or programmatically relevant. Data heaping on multiples of 10 or 30 (30, 60, 90, etc.) was found in 40–75% of country data for those countries that collected data by days rather than months, but given that iron is distributed in batches of 10 or 30 in these countries this kind of data heaping would be expected if all the distributed iron is consumed. The lack of measurable differences in quality must be interpreted in the light of the fact that DHS recall data could not be assessed relative to real-time distribution or consumption data.

Although there were no detectable changes in quality, the findings on data heaping should be reviewed in respect of the following questions. a) Is recall improved by recalling the batches of iron received (e.g. blisters of 10 doses, bottles of 30 doses)? b) Are women correctly reporting that they consumed all iron received, or is heaping due to other factors? c) Are enumerator probes responsible for heaping owing to the method used to assist women in estimating iron consumed, i.e. is heaping driven more by the enumerator than the woman?

Our findings indicate that practitioners across the spectrum of nutrition and public health organizations are aware of the limitations of antenatal iron consumption questions and the recall period. Their expressed concerns indicate an openness to the reduced recall period of the 2017 GNMF indicator (two rather than five years in current DHS modules) and interest in integrating complementary questions and indicators (antenatal nutrition counselling, other beneficiary groups). Future research should seek to quantify more precisely the quality of iron supplement consumption recall data relative to actual iron supplement distribution and consumption, and to use these findings to enhance recall by means of an enhanced survey question methodology.

(10)

INTRODUCTION

W

HO-UNICEF’s Technical Expert Advisory group for nutrition Monitoring (TEAM) is leading efforts to propose a global standard indicator on antenatal iron supplementation coverage.

The primary aim of developing this indicator is to harmonize and improve the reporting of progress in antenatal iron supplementation programmes by WHO Member States.

The 2015 Global Nutrition Monitoring Framework (GNMF) defined the antenatal iron supplementation coverage indicator as the “proportion of women with a birth in the last two years who received or bought iron and folic acid (IFA) supplements for at least six months during their last pregnancy” (1). Several antenatal iron supplementation indicators are currently used for reporting in low- and middle-income countries (LMIC) based on data collected in the Demographic and Health Survey (DHS) women’s questionnaire. These include various indicators in the coverage continuum including delivery of service (e.g. any iron received or purchased), and more specific indicators on the amount consumed by women during pregnancy (e.g. consumed 90+ tablets). Concerns have been raised about the validity of these indicators given that women are reporting on pregnancies completed in the previous five years (2). Other relevant data sources include administrative data systems (e.g. HMIS) for monitoring and reporting on antenatal iron supplementation. However, these systems generally do not follow individual women over time and are limited to distribution, so do not report on specific amounts received or consumed during any given pregnancy.

Due to a lack of consistent definitions for antenatal iron supplementation, the GNMF indicator proposed in 2015 was deferred in order to work up additional guidance on how to operationalize the indicator. WHO-UNICEF’s TEAM commissioned a scoping exercise and feasibility study in eight WHO Member States in 2017–2018, with the aim of assessing the feasibility of reporting on this iron supplementation indicator, along with related country policies and recommendations (3). The study revealed that none of the eight countries – selected on account of their differing economic, regional and anaemia burden profiles – would be able to provide complete information on the iron supplementation indicator as defined by the GNMF.

A simplified indicator was thus proposed in order to avoid burdening Member States with collecting additional/new data. This indicator was defined as “the percentage of women who consumed any iron-containing supplements during a current or past pregnancy within the last two years”, and was included in 2017 GNMF measurement guidance (4).

As part of a 2018–2019 exercise to support the DHS Core Questionnaire review, DataDENT (5) collected examples of iron supplementation questions from research and programme monitoring surveys. In TEAM’s scoping and feasibility study (3), some stakeholders also recommended changes to global household survey core questionnaires included adding multiple micronutrient supplements, identifying the source of iron supplements and whether supplements were free of charge or not.

The current report builds on the above-mentioned efforts to advance towards developing and validating a new indicator by pursuing two aims and their related activities:

1) to identify how iron supplement consumption data are collected (i.e. specific questions asked) and reported/used (i.e. indicator definition) in large-scale nutrition datasets by analysing findings from a series of key informant interviews from eight countries and an online survey completed by public health/nutrition practitioners from 52 countries; and

2) to determine whether it is possible to identify differences in the consistency, quality, and face validity of antenatal iron consumption data collected in four DHS datasets.

(11)

AIM 1: IDENTIFY HOW IRON SUPPLEMENT CONSUMPTION DATA ARE COLLECTED, REPORTED AND USED

METHODS

In early 2019, we carried out a series of key informant interviews (KII) to identify how data on antenatal iron supplementation are collected (i.e. specific questions asked in surveys (Table A1)) and used. An online survey was also carried out to obtain similar information in a wider sample of data users.

KEY INFORMANT INTERVIEWS

Key informant interviews covered similar topics and themes as the WHO-UNICEF TEAM feasibility study (3), but differed in terms of selection criteria. Our criterion for selecting countries to be included in the KII was a random selection of countries with a different prevalence of anaemia, as reported in WHO’s Vitamin and Mineral Nutrition Information System (VMNIS) database (https://

www.who.int/vmnis/en/) and a review by Stevens et al. (6). Seventy-nine countries with data on anaemia¹ prevalence among women of reproductive age² were identified and categorized into high, medium and low anaemia burden, as follows:

• anaemia prevalence > 40%: 30 countries • anaemia prevalence 20–39.9%: 31 countries • anaemia prevalence < 20%: 18 countries.

These cut-offs are in line with those used by WHO for guidance on iron supplement consumption during pregnancy.

We initially randomly selected four countries from each of the three anaemia burden categories, although we were required to replace five countries due to our inability to identify contacts. We finally completed KII for eight countries whose identified contacts responded to our request for participation.

Table 1. Anaemia prevalence among women of reproductive age in countries in which KII were completed

Country Anaemia prevalence among women of reproductive age*

Anaemia prevalence >40%

Tanzania 47.3%

Myanmar 46.5%

Niger 46.0%

Anaemia prevalence 20–39.9%

Colombia 32.8%

Japan 20.8%

Anaemia prevalence <20%

Rwanda 19.2%

Guatemala 13.6%

Afghanistan 12.9%

*Source: VMNIS database, accessed 4 December 2018.

1 Defined as <120 g/L haemoglobin in non-pregnant women and <110 g/L haemoglobin in pregnant women; adjusted by altitude and tobacco use.

(12)

An interview guideline for key informants (KI) (Table C 1) was developed in consultation with TEAM’s iron supplement working group (WG) (see Acknowledgments). Interviews were completed between January and March 2019, and each interview lasted 30 to 40 min.

ONLINE SURVEYS

The questions developed for the KII were also taken up in an online survey which was open for responses from 25 March–22 April 2019. The questions were available in English, French and Spanish.

The online interface was created using SurveyMonkey¹. Separate survey links were created for each of the three languages and disseminated through the WHO Nutrition Listserv.

During data analyses, multiple responses from the same country were manually aggregated.

Findings from online responses were separated from those obtained through direct key informant interviews.

1 www.surveymonkey.com

(13)

AIM 1: RESULTS

KEY INFORMANT INTERVIEWS

Six key informants worked for the government/ministry, while three worked for the WHO country office. A table summarizing the detailed results of the interviews is presented in Appendix A.

NATIONAL GUIDELINE FOR IRON SUPPLEMENT PROVISION TO ADOLESCENT GIRLS, PREGNANT AND/OR LACTATING WOMEN

Key informants from six of the eight countries indicated that a national guideline is currently available for iron supplement provision to adolescent girls, pregnant and/or lactating women (Table 2). Although no national guideline exists for Japan, the country’s obstetric/gynaecological medical association has a guideline for iron/folic acid supplementation for pregnant women. A national guideline is currently under development in Rwanda in collaboration with WHO.

Table 2. Beneficiaries included in national guidelines for iron and/or micronutrient supplement provision in countries in which KII were completed

Country National guideline beneficiaries Iron supplement formulation Anaemia prevalence >40%

Tanzania PW Not mentioned

Myanmar PW and adolescent girls 10–19 years Not mentioned

Niger PW and adolescent girls 10–19 years Not mentioned

Colombia PW, infants <24 months* and adolescents 60 mg

Japan No national guideline Not applicable

Rwanda Children 6–24 months Not applicable

Guatemala PLW and children 0–10 years Not mentioned Afghanistan PW and adolescent girls 10–19 years Not mentioned

*Iron supplements for < 6 months if low gestational weight at birth; MNP for 6–24 months; iron and vitamin A supplements for children 24–59 months.

MNP: micronutrient powder; PW: pregnant women; PLW: pregnant and lactating women.

AVAILABILITY OF IRON SUPPLEMENTS AT HEALTH FACILITIES

In six of the eight countries iron supplements were available free of charge for pregnant women at health facilities where they receive antenatal care (i.e. first point of contact between a pregnant woman and health care provider). In Japan, pregnant women receive a prescription for supply at a pharmacy. The cost is covered by Japan’s universal national health insurance, but pregnant women pay 20–30% of the cost, depending on their age and income level. In Niger, pregnant women received iron supplements free of charge at health clinics with a prescription, or at a charge from pharmacies without a prescription.

COLLECTION OF DATA ON COVERAGE/CONSUMPTION OF IRON SUPPLEMENTS

A nationally representative nutrition survey was carried out in each of the eight countries within the last five years, four of them being a DHS. Surveys in seven countries included assessment of coverage/consumption of iron supplements among pregnant women. The exception was Japan, where only current supplement consumption (including for treatment of anaemia) is assessed for all survey participants aged 20 years and older, regardless of pregnancy. According to the information shared by the KI, only haemoglobin status (distribution and mean (± SD)) of women aged 20 and older is reported, irrespective of birth history (Table 3). In Colombia and Guatemala, iron supplement coverage is also assessed using HMIS, which tracks whether women who present

(14)

for antenatal care are prescribed iron supplements during that visit. Table 3 summarizes the data source and the antenatal iron supplementation indicator used in each of the eight countries.

Colombia was the only country in which the DHS collected data on the number of months, rather than days, during which iron was consumed.

Table 3. Antenatal iron supplementation indicator used in countries in which KII were completed

Country Indicator definition Denominator Data source Anaemia prevalence >40%

Tanzania Took none/any/<60/

60–89/90+ iron tablets/

syrups

Women aged 15–49 with a live birth in the five years preceding the survey

National Nutrition Survey 2018

Myanmar Data not analyzed NA Myanmar Micronutrient and

Food Consumption Survey 2017–2018

Niger Took none/any/<60/

syrups

DHS 2012

Colombia Number of months iron

consumed Women aged 13–49 with a live

birth in the five years preceding the survey

DHS 2015

Japan Distribution and mean (± SD)

of haemoglobin Women aged 20 or older,

irrespective of birth history National Health and Nutrition Survey 2015

Anaemia prevalence < 2%

Rwanda Took none/any/<60/

syrups

DHS 2014–2015

Guatemala Took none/any/<60/

syrups

National Maternal and Child Health Survey 2015

Afghanistan Took none/any/<60/

syrups

Women aged 15–49 with a live birth in the five years preceding

the survey DHS 2015

Key informants from four countries indicated that they were not satisfied with how iron supplement coverage/consumption is assessed in national surveys. Reasons included the long recall period and inability to determine actual consumption of iron supplements by pregnant women. Only the informant from Afghanistan expressed satisfaction with the way iron supplementation coverage/consumption was currently assessed in national surveys. In Afghanistan, the three most recent surveys to assess iron supplementation in pregnant women were the National Nutrition Survey (NNS) in 2013, DHS in 2015 and Afghanistan Health Survey (AfHS) in 2018. We were not able to find the actual questionnaires used in these surveys, but based on the antenatal iron supplementation indicators reported in NNS 2013 (7), AfHS 2015 (8) and AfHS 2018 (9), the questions used to ascertain iron consumption during pregnancy were similar to those used in the DHS 2015 (10).

Additional information. During the interview, respondents were also asked what items of additional information would be useful to know. Suggestions included knowing the dose of the supplement being consumed, inclusion of other groups for whom iron supplementation programmes exist (e.g. adolescent girls) and reasons for poor adherence (e.g. side-effects). Key results for the collection of data on coverage/consumption of iron supplements are summarized in Table 4.

(15)

USE OF DATA ON COVERAGE/CONSUMPTION OF IRON SUPPLEMENTS

Key informants indicated that they used the data collected through national surveys for programme implementation and monitoring, managing the supply chain and improving national policy (Table 4). The key informant from Rwanda indicated that although data from the national survey are not currently used, it would be helpful to have additional data to determine how many women suffer side-effects from iron supplementation.

The key informant from Myanmar indicated that the most recent survey is currently being analysed and data has therefore yet to be used. Once analysis is complete, the informant suggested that it should be used to identify geographical differences in iron supplement coverage and compliance in order to improve programme implementation and delivery.

Table 4. Key results on collection and use of data on coverage/consumption of iron supplements

Country

Nutrition survey conducted in past

5 years Iron supplement questions used

Satisfied with iron supplement

questions How are survey data used

Anaemia prevalence > 40%

Tanzania 1. NNS 2018

2. DHS 2015–2016 Not shared, but those used in NNS 2014 are similar to those used in DHS

No Improve national

policy and programme implementation Myanmar 1. Myanmar

Micronutrient and Food Consumption Survey (MMFCS) 2017–2018 2. DHS 2015–2016

DHS: those used in DHS MMFCS:

1. Have you ever received iron supplements from the health care provider?

2. Do/Did you take the iron supplements provided by the health provider?

Not sure, as data

analysis is pending Not used currently

Niger DHS 2012 Those used in DHS Not answered Not answered

Colombia 1. DHS 2015 2. HMIS 3. Quality of Life (QoL) Survey 2017

DHS: those used in DHS HMIS: indicates if iron supplements were prescribed during ANC visit

QoL:

1. Are you pregnant?

2. If yes, did you receive a prescription for iron, folic acid or calcium supplements?

No Monitor coverage of

iron supplement

Japan Annual National Health and Nutrition Survey

Not applicable Not applicable Not applicable

Rwanda DHS 2014–2015 Those used in DHS No Not used currently

Guatemala 1. National Maternal and Child Health Survey 2015 2. DHS 2014–2015

Those used in DHS No Forecast

procurement and identify areas of concern

(16)

National guidelines:

Q9

[N = 34 from 24 countries]

Demographics:

Q1-Q3 [N = 145 from

52 countries]

Collecting data:

Q4-Q6 [N = 58 from 31 countries]

Country

Nutrition survey conducted in past

5 years Iron supplement questions used

Satisfied with iron supplement

questions How are survey data used

Afghanistan 1. Afghanistan Health Survey 2018

2. DHS 2015 3. Afghanistan Health Survey 2015

AfHS: similar to those used in DHS

DHS: those used in DHS

Yes Programme

implementation, national policy and proposals/reporting

ONLINE SURVEY

A total of 145¹ responses were submitted to the SurveyMonkey link from 52² countries (see Table C 2).

Fig. 1 illustrates the number of respondents, and corresponding countries, answering each question category. Participants from seven of the nine countries included in the WHO feasibility study (3) also responded to the online survey. The exceptions were Brazil and Senegal.

Fig. 1. Flowchart of number of respondents and countries answering each category of survey questions

RESPONDENT DEMOGRAPHICS

Respondents’ fields of work and corresponding organizations, in relation to nutrition and supplements for pregnant women, are summarized in Table 5 and Table 6, respectively. Forty- nine (34.3%) respondents from 21 countries worked for the Government or a ministry. One additional respondent did not provide a country.

Table 5. Fields of work occupied by respondents

Field of work N %

Academic/research 27 19.1

Dietician/nutritionist 5 3.5

Policy development 18 12.8

Procurement/distribution 4 2.8

Programme development 29 20.6

Programme implementation 32 22.7

Programme surveillance 18 12.8

1 Numbers presented in tables vary due to missing responses.

2 Three respondents entered their country as “abc” and “global”, and are therefore not counted.

Iron supplements:

Using data:

(17)

Field of work N %

Other 8 5.7

Total 141 100.0

Table 6. Type of organization employing respondents

Organization N %

Academia 17 11.9

Government/ministry 50 35.0

NGO 40 28.0

Private agency 13 9.1

UN 17 11.9

Other 6 4.2

Total 143 100.0

NATIONAL GUIDELINE FOR IRON SUPPLEMENT PROVISION TO ADOLESCENT GIRLS, PREGNANT AND/OR LACTATING WOMEN

The respondents were also asked if they were aware of an extant national guideline, policy or protocol for iron supplement provision to adolescent girls, pregnant and/or lactating women.

Thirty-four respondents from 24 countries answered this question. As summarized in Table 7, 18 (75%) countries have a currently available guideline for iron supplement provision to adolescent girls, pregnant and/or lactating women.

Table 7. Status of the national guideline for iron supplement provision in countries mentioned in the online survey

Country Anaemia prevalence among

women of reproductive age Guideline status Anaemia prevalence > 40%

India 51* Available

Nigeria 50* Available

Ghana 47.7** Available

Pakistan 43.5** Available

Anaemia prevalence 20 – 39.9%

Malawi 35.3** Available

Nepal 35** Available

Syria 34* Not available

Iran 33.4** Available

Sri Lanka 33* Available

Thailand 32* Available

Fiji 31* Available

Moldova 27.9** Available

Kenya 27* Available

Indonesia 26* Not available

Venezuela 24* Not available

(18)

Country Anaemia prevalence among

women of reproductive age Guideline status

Peru 21** Available

Uruguay 21* Available

Japan 20.8** Not available

China 19.9** Not available

Portugal 18* Available

Philippines 16* Under development

USA 13* Available

Afghanistan 12.9** Available

Mexico 11.8** Available

Sources: *The World Bank, accessed 30 August 2019; **VMNIS database, accessed 4 December 2018.

AVAILABILITY OF IRON SUPPLEMENTS AT HEALTH FACILITIES

Of the 31 respondents from 21 countries who provided information on availability of iron supplements at health facilities where pregnant women receive antenatal care (i.e. first point of contact between a pregnant woman and health care provider), iron supplements are available in 20 (95.2%) countries. Health facilities in 17 (81%) countries provide iron supplements free of charge (Table 8).

Table 8. Availability of iron supplements at health facilities in countries mentioned in the online survey

Iron supplements available at health facility % (n)

Yes, free of charge 81.0 (17)

Yes, at a charge 0.0 (0)

Yes, both 14.3 (3)

No 4.8 (1)

Total 100.0 (21)

COVERAGE/CONSUMPTION OF IRON SUPPLEMENTS

In the online survey, a programming error meant that question #4, whether coverage/consumption of iron supplements was assessed in a national survey, was not asked. However, respondents from 25 countries identified a national survey in the follow-up question which asked when the most recent survey assessing iron supplements coverage/consumption had taken place. In 24 of these counties, a national survey had been carried out between 2014 and 2018 (Table C 3). Moldova is the only country where the most recent survey took place earlier than 2014. For the 27 countries where a survey was not identified by respondents, Google Scholar, PubMed and the VMNIS database were searched to identify the most recent survey (Table C 3).

Of the 58 respondents from 31 countries who answered the question whether they were satisfied with the way in which iron supplement coverage/consumption is currently assessed in national surveys, 45% (n = 26) responded “yes” (Table 9).

(19)

Table 9. Satisfaction with how iron supplement coverage/consumption is assessed in national surveys

Satisfied with iron supplement coverage/consumption assessment % (n)

Yes 44.8 (26)

No 37.9 (22)

Other:

not applicable/no data 12.1 (7)

recall bias 1.7 (1)

lack of antenatal nutrition counseling and information on stock-outs 1.7 (1)

lack of qualitative data 1.7 (1)

Total 100.0 (58)

For the 55% (n = 32) who responded “no” or “other”, 24 (75%) respondents provided suggestions on how questions assessing iron supplement coverage/consumption should be revised. These included shortening the recall period, including other beneficiary groups and inserting questions to assess if, and how, counselling for iron consumption is provided. Selected responses from survey participants is presented in Table 10. The full list of responses is presented in Table C 4.

Table 10. Selected respondents’ suggestions on how questions assessing iron supplement coverage/ consumption should be revised in national surveys

Country Respondents’ suggestions for revision

Ghana “DHS questions are based on recall because it selects mothers who have had a pregnancy/birth in the five years preceding the survey. How would mothers remember the number of IFA tablets consumed for the entire duration of pregnancy five years ago? I seriously think when it comes to IFA intake the question should be posed to currently pregnant mothers to minimize recall bias”.

India “Only pregnant women are covered. Other beneficiary groups are not covered. There is no dedicated survey for anaemia in India”.

Kenya “Include questions on sources of the supplements; total amounts of IFA consumed (e.g. data on mg/day and days consumed); time of start of IFA use; data on reasons for low compliance/late start to IFA use; content disseminated during ANC counselling sessions”.

USA “I would add regarding what is counselled on for anaemia (if counselling takes place), and have responses to find out if the providers actually counsel on benefits, on side- effects, etc.”

ANC: Antenatal care; DHS: Demographic and Health Survey; IFA: Iron and Folic Acid.

Note: verbatim responses edited slightly for clarity.

(20)

USE OF DATA ON COVERAGE/CONSUMPTION OF IRON SUPPLEMENTS

Thirty-five respondents from 20 countries described how they use the data collected. The most common use of data collected on iron supplement coverage/consumption was in monitoring and evaluating programmes and research (see Table 11).

Table 11. Use of data collected on iron supplement coverage/consumption, based on respondents’ comments1

Usage % (n)

Monitoring and evaluation 53.3 (19)

Implementation 16.7 (5)

Policy 16.7 (5)

Research 20.0 (7)

Others 10.0 (5)

1 The full list of respondents’ comments is presented in Table C 5

(21)

AIM 2: DETERMINE WHETHER IT IS POSSIBLE TO IDENTIFY DIFFERENCES IN THE CONSISTENCY, QUALITY, AND FACE VALIDITY OF ANTENATAL IRON CONSUMPTION DATA COLLECTED IN FOUR DHS DATASETS

METHODS

As mentioned in the introduction, the second aim of this survey was to assess whether there are identifiable determinants for the quality of iron supplement consumption data among indicators derived from current questions used in the DHS-7 Core Questionnaire. The purpose of assessing these quality metrics is to identify whether available data affords any insight into how these questions might be modified to produce more accurate reports of antenatal iron supplementation.

We therefore analysed the Demographic and Health Survey (DHS) data from four countries.

VARIABLE SELECTION

Our assumptions for selecting determinants of antenatal iron consumption were based on available research and indicators in DHS data likely to reflect that research. For example, maternal education and nutrition knowledge are associated with a higher consumption of micronutrient (including iron) supplements (11-13). Similarly, women who attend more ANC visits are also likely to consume more iron supplements (12-15). Postnatal maternal recall of events which occurred in the prenatal period is also associated with maternal education (16). Finally, maternal recall of iron supplementation receipt can be poor within 1–2 years of giving birth (2). We therefore hypothesized that the actual duration of recall, number of ANC visits and maternal education may have an impact on the number of tablets consumed and the accuracy of maternal recall. Fig. 2 displays data quality indicators analysed in relation to factors considered to have an impact on consumption or its recall.

Fig. 2. Directed acyclic graph (DAG) representing the relationship between antenatal IFA consumption recall (outcome) and maternal sociodemographic factors and recall period

*Analyzed for association with IFA comsuption recall; ANC: Antenatal care; IFA: iron and folic acid.

*Maternal

education *Total No. of

ANC visits IFA

received

IFA consumption

comsuption recallIFA

*Recall period, i.e. child´s age

(22)

DHS DATA ANALYSIS

Measures of data quality for selected antenatal iron coverage indicators were analysed using Demographic and Health Surveys (DHS) conducted in four countries. Two countries from the high anaemia prevalence (>40%) group, and one country each from the middle (20–39.9%) and low anaemia prevalence (<20%) groups were selected for this purpose (Table 12).

Table 12. Countries included in secondary data analysis of iron consumption data quality

Selected country Anaemia prevalence among

women of reproductive age* DHS year Anaemia prevalence >40%

Tanzania 47.3% 2015–2016

Myanmar 46.5% 2015–2016

Colombia 32.8% 2015

Afghanistan 12.9% 2015

*Source: VMNIS database, accessed 4 December 2018.

Data collected in the DHS women’s questionnaire, administered to all eligible women aged 15–49 years residing within the selected household, are contained in Individual Recode (IR) files.

The IR files for each of the four countries included in the analysis were downloaded from the DHS programme website. The DHS collects data about iron coverage (given or purchased) and consumption (“taken”) from women who have had a live birth in the five years preceding the survey. The following table lists the questions included in the individual woman’s questionnaire (DHS Model Questionnaire – Phase 71):

Table 13. Iron coverage/consumption questions included in the Pregnancy and Postnatal Care section of the DHS Model Questionnaire – Phase 7

Q420 During this pregnancy, were you given or did you buy any iron tablets or iron syrup?

SHOW TABLETS/SYRUP.

Yes………....1 No………2 Don’t know……….8

Q421 During the whole pregnancy, for how many days* did you take the tablets or syrup?

IF ANSWER IS NOT NUMERIC, PROBE FOR APPROXIMATE NUMBER OF DAYS.

Days ……….…I__I__I__I Don’t know………..998

*Colombia collected iron consumption data in months instead of days

These questions were asked in Afghanistan, Myanmar and Tanzania. Colombia collected iron consumption data in months instead of days, using the following question (translated from Spanish): “During the entire pregnancy, for how many months did you take iron?” All women were asked about iron consumption irrespective of whether they had received antenatal care.

1 https://dhsprogram.com/pubs/pdf/DHSQ7/DHS7-Womans-QRE-EN-17Dec2018-DHSQ7.pdf

(23)

We used linear regression to assess the relationship between the number of days reported by the respondents during which iron (tablets/syrup) was consumed (Q421), time since last completed pregnancy (using age of most recent child, Q215), number of ANC visits (Q412) and mother’s educational attainment (Q108–Q109).

Only those women who reported consuming iron tablets/syrup for at least one day (Q421>0) were included in the linear regression analyses. Statistical significance was set at p < 0.05. All analyses were adjusted for survey design, using DHS-assigned weights.

Precision of survey data, as a measure of data quality, was assessed using standard error of mean (SE) and coefficient of variation (CV) calculated around the mean number of days/months during which iron tablets/syrup were consumed, categories of recall period (<1 year, 1 to <2 years, 2 to <3 years, 3 to <4 years , 4 to <5 years), number of ANC visits and mother’s educational attainment (no education, incomplete primary, complete primary, incomplete secondary, complete secondary, higher). SE estimates how much the sample mean deviates from the true population mean. CV describes dispersion of the variable of interest (i.e. mean days/months iron tablets/syrup were consumed) and is calculated using the formula:, where µ is the sample mean.

Higher CV values indicate greater dispersion of the variable. CV values were interpreted as per Statistics Canada guidelines (17), summarized in the table below.

Table 14. Data quality level guidelines used at Statistics Canada

Quality level of estimate Guidelines

1) Acceptable Low CV in the range of 0.0% to 16.5%

2) Marginal High CV in the range of 16.6% to 33.3%

3) Unacceptable Very high CV in excess of 33.3%

CV: Coefficient of variation.

Although antenatal iron tablet/syrup consumption is likely to display heaping at certain values if all the received iron is consumed, we also assessed whether we could identify any differences in data heaping based on the categories of recall period. Stratified mean days/

months of iron consumption for each category of recall period were compared with the overall mean (i.e. for the cumulative five years). For those countries with reported consumption in days rather than months – Afghanistan, Myanmar and Tanzania – evidence of data heaping was also assessed descriptively, by comparing the proportion of women who reported specific number of days, i.e. 30, 60, 90, 120, 150 and 180, against those who reported other numbers.

(24)

AIM 2: RESULTS DHS DATA ANALYSIS

DESCRIPTIVE ANALYSES

Descriptive analyses of mean number of days/months of iron consumed by pregnant women in their last pregnancy, and variables included in DHS data quality analysis are presented in Table 15 and Table 16.

Antenatal iron consumption was highest in Colombia (measured in months) and Myanmar, with Afghanistan reporting the lowest iron consumption. Histograms illustrating iron consumption for each of the four countries are included in Appendix B.

The proportion of women who received at least four ANC visits during their last pregnancy was highest in Colombia (91%) and lowest in Afghanistan (18%). Maternal literacy was also lowest in Afghanistan, with 83% of women reporting no educational attainment. More than 60%

of women in Colombia reported completing secondary school or higher.

Table 15. Iron supplement consumption during pregnancy in countries selected for DHS data quality analyses

Afghanistan Colombia Myanmar Tanzania

Mean days

(± SE) n Mean

months

(± SE) n Mean

days

(± SE) n Mean

days (± SE) n Iron supplement

consumption during pregnancy*

46.2

(± 2.02) 18897 5.4 (±

0.04) 9181 112.1

(± 1.7) 3489 55.8

(± 0.88) 6853 Iron supplement consumption by recall period

<1 year 46.4

(± 2.15) 5479 5.5

(± 0.10) 2136 114.6

(± 3.12) 790 59.3

(± 1.39) 1966 1 to <2 years 50.7

(± 4.23) 5100 5.4

(± 0.09) 2095 117.2

(± 2.68) 811 55.4

(± 1.54) 2018 2 to <3 years 40.5

(± 1.8) 3982 5.4

(± 0.09) 1801 111.3

(± 3.03) 670 53.8

(± 1.84) 1263 3 to <4 years 46.1

(± 4.68) 2152 5.3

(± 0.12) 1682 105.2

(± 3.24) 653 52.8

(± 1.73) 874 4 to <5 years 43.2

(± 3.19) 1328 5.7

(± 0.10) 1466 109.3

(± 3.79) 511 53.1

(± 2.62) 620 Iron supplement consumption by place ANC received

Public health facilities 46.2

(± 2.53) 7354 5.2

(± 0.05) 4652 117.1

(± 1.85) 2168 55.7

(± 0.96) 5762 Private health facilities 50.4

(± 3.24) 2965 5.7

(± 0.07) 4265 127.5

(± 5.46) 262 58.4

(± 6.14) 194 SE: Standard error of mean; *Among those who reported receiving/purchasing iron supplements

Table 16. Descriptive analysis of variables included in DHS data quality analysis, by country

% n % n % n % n

Consumed any iron supplement during

pregnancy 89.3 8108* 95.2 8202* 98.0 3056* 98.0 5548*

Recall period

<1 year 30.5 5770 23.1 2143 22.9 798 29.0 1988

1 to <2 years 28.2 5324 22.8 2112 23.6 823 29.7 2038

(25)

% n % n % n % n

2 to <3 years 21.9 4137 19.6 1819 19.5 680 18.9 1293

3 to <4 years 12.1 2287 18.3 1701 19.1 665 13.1 897

4 to <5 years 7.3 1379 16.2 1507 15.0 524 9.3 637

Mean

(± SE) n Mean

(± SE) n Number of ANC visits 1.9

(± 0.07) 18897 7.2

(± 0.07) 9281 4.8

(± 0.14) 3489 3.7

(± 0.04) 6853 No. ANC visits

<4 81.5 15078 8.7 793 40.7 1409 49.2 3353

≥4 18.5 3432 91.3 8337 59.4 2058 50.8 3467

No. ANC visits % n % n % n % n

<8 97.6 18069 56.9 5196 82.4 2857 98.8 6740

≥8 2.4 441 43.1 3934 17.6 610 1.2 80

Received ANC at

Public health facilities 71.2 7767 52.1 4699 89.1 2204 96.7 5857

Private health facilities 28.8 3140 47.9 4318 10.9 269 3.3 199

Educational attainment

No education 82.7 15622 1.8 169 16.0 559 19.3 1324

Incomplete primary 6.1 1146 7.2 664 26.7 930 12.2 839

Complete primary 2.2 413 8.2 761 18.9 659 52.2 3576

Incomplete secondary 4.4 839 21.7 2015 27.5 959 5.3 362

Complete secondary 2.9 556 28.3 2624 2.5 88 10.1 689

Higher 1.7 321 32.8 3048 8.4 294 0.9 62

SE: Standard error of mean; ANC: Antenatal care; *Number of women who received or purchased any iron tablets/syrup.

DATA HEAPING ANALYSES

We also looked for evidence of data heaping by comparing the stratified mean for each category of recall period with the overall sample mean (Table 17). Our hypothesis was that if women’s recall of iron tablet/syrup consumption did not differ by recall period, the mean days/months of iron consumption stratified by recall period would also not differ from the overall mean days/months of iron consumption. This was not observed. Evidence of data heaping was observed in all four datasets (p <0.05), but the difference was not programmatically significant.

Table 17. Evidence of data heaping in DHS datasets

Mean

days p Mean months p Mean

days p Mean

days p

Overall 46.2

<0.001

5.4

0.0013

112.1

0.0050

55.8

0.0011

<1 year 46.4 5.5 114.6 59.3

1 to <2 years 50.7 5.4 117.2 55.4

2 to <3years 40.5 5.4 111.3 53.8

3 to <4 years 46.1 5.3 105.2 52.8

4 to <5 years 43.2 5.7 109.3 53.1

IMPROVING ANTENATAL IRON-CONTAINING SUPPLEMENTATION INDICATORS

IMPROVING ANTENATAL IRON-CONTAINING

SUPPLEMENTATION INDICATORS

REPORT OF KEY INFORMANT INTERVIEWS, ONLINE SURVEY, AND DHS DATA ANALYSES

IMPROVING ANTENATAL IRON-CONTAINING

SUPPLEMENTATION INDICATORS

REPORT OF KEY INFORMANT INTERVIEWS, ONLINE SURVEY, AND DHS DATA ANALYSES

CONTENTS

LIST OF TABLES

LIST OF FIGURES

ACKNOWLEDGEMENTS

EXECUTING SUMMARY

W

INTRODUCTION

W

AIM 1: IDENTIFY HOW IRON SUPPLEMENT CONSUMPTION DATA ARE COLLECTED, REPORTED AND USED

METHODS

KEY INFORMANT INTERVIEWS

AIM 1: RESULTS

KEY INFORMANT INTERVIEWS

ONLINE SURVEY

AIM 2: DETERMINE WHETHER IT IS POSSIBLE TO IDENTIFY DIFFERENCES IN THE CONSISTENCY, QUALITY, AND FACE VALIDITY OF ANTENATAL IRON CONSUMPTION DATA COLLECTED IN FOUR DHS DATASETS

METHODS

VARIABLE SELECTION

DHS DATA ANALYSIS

AIM 2: RESULTS DHS DATA ANALYSIS