F1000Research F1000Research 2046-1402 F1000 Research Limited London, UK 10.12688/f1000research.55269.4 Research Article Articles Factors associated with inadequate urinary iodine concentration among pregnant women in Mbeya region Tanzania.

[version 4; peer review: 1 approved, 1 approved with reservations]

 Lukindo Tedson Conceptualization Formal Analysis Investigation Methodology Writing – Original Draft Preparation Writing – Review & Editing 1 Masumo Ray Conceptualization Formal Analysis Methodology Supervision Writing – Original Draft Preparation Writing – Review & Editing https://orcid.org/0000-0002-3899-5918 a 1 Hancy Adam Data Curation Formal Analysis Methodology Software 1 John Sauli E. Conceptualization Formal Analysis Investigation Supervision Writing – Original Draft Preparation 1 Paulo Heavenlight A. Data Curation Formal Analysis Methodology Writing – Original Draft Preparation https://orcid.org/0000-0002-3162-9982 b 2 Sanga Abraham Conceptualization Methodology Project Administration Supervision Validation c 3 Noor Ramadhan Conceptualization Formal Analysis Methodology Writing – Original Draft Preparation 3 Lankoande Fatoumata Conceptualization Methodology Project Administration Supervision Visualization 3 Towo Elifatio Conceptualization Methodology Project Administration Supervision 1 Leyna Germana H. Conceptualization Formal Analysis Methodology Project Administration Writing – Original Draft Preparation Writing – Review & Editing https://orcid.org/0000-0001-7826-7651 1 Bridge Gemma Writing – Original Draft Preparation Writing – Review & Editing https://orcid.org/0000-0001-7441-9849 d 4 5 Bedi Raman Writing – Original Draft Preparation Writing – Review & Editing https://orcid.org/0000-0002-5804-9355 6 Tanzania Food and Nutrition Centre, 22 Barack Obama Drive, Dar es Salaam, P.O. Box 977, Tanzania Department of Epidemiology and Biostatistics, Muhimbili University of Health and Allied Sciences, Dar es Salaam, MUHAS, P.O. Box 65001, Tanzania The United Nations Children's Fund (UNICEF), The United Nations Children's Fund (UNICEF), Dar es Salaam, Tanzania, P.O. Box 4076, Tanzania Barts and The London School of Medicine and Dentistry, Queen Mary University of London, Garrod Building, Turner Street, Whitechapel,, London, E1 2AD, UK Global Child Dental Fund (GCD fund), King's College London, Norfolk Building, Room G03-G03A, The Global Child Dental Fund, Surrey Street, London, WC2R 2ND, UK King's College London, Norfolk Building, Room G03-G03A, Surrey Street, London, WC2R 2ND, UK rmasumo@yahoo.com heavenlightpaul@gmail.com asanga@unicef.org glbridge1@hotmail.co.uk

No competing interests were disclosed.

 30 4 2024 2021 10 858 18 4 2024 Copyright: © 2024 Lukindo T et al. 2024 This is an open access article distributed under the terms of the Creative Commons Attribution Licence, which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited. Background

Insufficient and above WHO-recommended levels of iodine intake during pregnancy can lead to serious health outcomes. This study aimed to assess median urine iodine concentration and its associated risk factors among pregnant women in the Mbeya region, Tanzania.

 Method

A cross sectional survey involving 420 pregnant women (n=420) aged 15-49, registered in Reproductive and Child Health Clinics was conducted. Socio-demographic and dietary factors were assessed by structured questionnaire and the urine samples were analyzed using the ammonium persulfate digestion method.

 Results

Median urinary iodine concentration (mUIC) was 279.4μg/L and it ranged from 26.1 to 1915μg/L. Insufficient mUIC (below 150μg/L) was observed in 17.14% of participants, sufficient mUIC was 24.29% and 58.57% had mUIC above the recommended level (>250μg/L). Sample women who reported consuming fish in the last 24 hours had an increased risk of insufficient mUIC [Adjusted OR= 2.60 (95%CI 1.31-5.15)] while the risk was lower for those who attended at least primary education [AOR= 0.29 (CI 0.08-0.99)]. Further, sample women resident in Mbarali district, in the oldest age group (35-49) and having a higher socio-economic status were associated with an increased risk of having MUIC above recommended level [AOR=4.09 (CI 1.85-9.010], [AOR=2.51 (CI 0.99-6.330] and, [AOR=2.08 (CI 0.91-4.71) respectively.

 Conclusion

This study demonstrated a significant association between geographical, age and socio-economic factors and median urine iodine concentration above the WHO-recommended level. Further, this study found association between inadequate iodine in diet and insufficient median urine iodine concentration. Therefore, educational programs on iodine intake should be strengthened.

 Iodine deficiency medium urine iodine concentration; pregnant women; socio-demographic and dietary risk factors The United Nations Children's Fund (UNICEF), Tanzania and, the Ministry of Health, Community Development, Gender, Elderly and Children, Tanzania This project was funded by the UNICEF and, the Tanzania Ministry of Health, Community Development, Gender, Elderly and Children. The funders had no role in study design, data collection and analysis, decision to publish, or preparation of the manuscript. Amendments from Version 3

We reorganized the article to improve logical flow and readability, with particular attention to the tables.

 Introduction

Iodine insufficiency is a significant global public health concern. 1 This element is constituent of hormones produced by the thyroid gland namely, triiodothyronine (T3) and thyroxine (T4), 1 , 2 and must be consumed in the diet as it cannot be made naturally in the body. 1 A diet low in iodine results in insufficiency that can occur at any age. If iodine requirements are not met, the thyroid is unable to produce thyroid hormone in sufficient quantities, leading to iodine deficiency disorders (IDD) with associated dysfunctional and developmental abnormalities. 3

The daily iodine intake recommended by the World Health Organization (WHO) is 90 μg for children aged 0-5 years; 120 μg for children aged 6-12 year; and 150 μg for those over 12 years. Pregnant and lactating women the recommended iodine intake is 250 μg daily. 4 By ensuring that individuals have an adequate intake of iodine in their diets, IDDs can be prevented. 5 Although cretinism is the most severe form of iodine insufficiency, minor iodine deficiency can also result in reduced intellectual ability, limited work capacity due to mental and neurological impairment. 4 In 1994, about 1.6 billion people globally (28.9% of global populations) were suffering from iodine deficiency, 11.2 million were affected by overt cretinism, and 43 million people had at least some degree of intellectual impairment. 5

In Tanzania, the most recent figures indicate that more than 40% of the populations live in geographical regions prone to iodine deficiency. 6 Previous studies have reported that the southern highlands of Tanzania were areas with a high prevalence of endemic goiter. 7 , 8 In 1986, Wachter and colleagues investigated the prevalence of goiter before iodine supplementation program in the Southern highlands of Tanzania among 560 schoolchildren aged between 6 to 19 years old and found that 512 children had goitre (a prevalence of 90%). Goitre decreased to 60% by 1995 and further to 17% in 2004. 6 In the 1990s there was a worldwide progress in reducing IDD through universal salt iodization (USI) legislation passed in Tanzania and many other countries. In 2000, more than 28 counties reduced goiter by more than 20% through USI. 9 The Tanzania Demographic and Health Survey (TDHS) indicates that such IDD variations occur largely because of differences in the use of adequately iodised salt (15+ ppm), with households in urban areas more likely to use adequately iodised salt (81%) than those in rural areas (51%) such as the highlands. 9 As recommended by WHO, Iodine Global Network (IGN) and United Nations Children’s Fund (UNICEF), median Urinary Iodine Concentration (mUIC) is considered the most practical biomarker for the assessment and monitoring of iodine nutrition status in the population. 10 According to the National IDD survey conducted in 2004, about 25% of primary school children in Tanzania had UIC below 100 μg/L. 6

For USI, where all the salt meant for human consumption is iodized, is the most practical intervention strategy to increase iodine intake. 11 Evidence from multiple sources indicates widespread of USI, as 68% of households have access to iodized salt. 12 USI is an effective way of delivering iodine to individuals and in term, improving cognition in populations exposed to iodine deficiency. 12 , 13 USI is also affordable, with annual costs of salt iodization estimated at USD 0.02-0.05 per child, while the costs to prevent child death are estimated at USD 1000. There are also large gains in per Disability-Adjusted Life Years (DALYs), at USD 34-36. 14 Before USI, it was estimated that iodine deficiency leads to losses of USD 35.7 billion in the countries affected, which is significant when compared with an estimated cost of USD 0.5 billion for USI, representing a cost benefit ratio of over 70:1. 15

Tanzanian legislation on USI was enacted in 1995. 16 This was later revised in 2010, and now, all salt consumed by animals and humans in the country is fortified with iodine. Enforcement of this legislation is challenging in Tanzania, especially in areas where small-scale salt producers operate. As a result, iodized salt is not widely available. Further, although household coverage with iodized salt is above 80% nationally, the coverage of adequately iodized salt is at 47% across Tanzania.

Although USI is arguably the successful health intervention in global history, legal regulations on salt production are rarely sufficient to guarantee dietary change among rural populations that consume mainly subsistence food products. 17 In 2005 UNICEF documented that in many African countries less than thirty per cent of households consume iodized salt despite universal legislation, and even iodized salt may be insufficient to reduce IDD in populations whose diets contain substantial amounts of iodine-depleting foods. 18 With the respect to the estimated more than 10% of Tanzania households which remain at risk in spite of salt iodization legislation, the magnitudes of micronutrients deficiencies among pregnant women in Tanzania, particularly, in rural areas and low socioeconomic status justify more research. There is a dearth of information regarding the burden of, and factors associated with iodine deficiency among pregnant women. In 2010 and 2015, the TDHS reported that the prevalence of iodine deficiency amongst pregnant women was 54%, however, the results of the TDHS were heterogeneous across the regions. 19 , 20 There are also concerns relating to excessive iodine intake in pregnancy, because although high iodine intakes are well tolerated by most healthy individuals, in some, excess intake can lead to thyroid conditions such as hyperthyroidism, hypothyroidism, and/or thyroid autoimmunity. 21 , 22 As insufficient and excessive iodine consumption in pregnancy can result in negative health impacts, it is imperative to investigate current iodine levels and to assess how current USI interventions affect iodine intake, especially in highland areas of Tanzania. The findings of the present study could be useful especially at regional or local salt facility levels for the future decision making on whether levels of iodine added to fortified salt should be increased or decreased, 22 Given the above, this study aimed to firstly, determine the likelihood of iodine levels being above or at recommended levels in the urine (μg/L) of pregnant women in their second trimester. Secondly, assess the likelihood of IDD or otherwise, differing across socioeconomic groups and locations in Tanzania.

 Methods Study site

Mbeya Region is located in the south-western corner of the southern highlands of Tanzania ( Figure 1). The Region lies between latitude 70° and 90° 31’ south of the Equator and between longitude 32° and 35° east of Greenwich. The economy of Mbeya is based mainly on agriculture. Agriculture contributes most of the Region’s cash income mainly from maize, sorghum, cassava, beans and pigeon peas’ production. Generally, annual rainfall varies from 650 mm in Usangu plains and Chunya to 2600 mm on the Northern shores of Lake Nyasa and in the highlands. The Mbeya region has a population of 2,707,410 and, in 2020 the region had 318 health facilities of which 17 were hospitals, 23 health centers, and 278 dispensaries, with 251 of the health facilities (both government, private and faith-based organizations) providing reproductive and child health services.

Map of Tanzania and Mbeya region. Study design

A cross-sectional survey of 420 pregnant women (gestation age below 28 weeks) registered at the Reproductive and Child Health Clinics (RCH) from the seven districts of the Mbeya Region. This manuscript is part of the project on improving maternal and adolescent nutrition (IMAN) in Mbeya supported by the UNICEF-Tanzania and the Ministry of Health- Tanzania. The study was carried out from September 2020 to October 2020. This study was conducted in 42 RCH across the seven districts of Mbeya region. The allocation of RCH per district for was based to probability proportional to size: Mbeya District Council (n = 11); Chunya District Council (n = 4); Mbeya City (n = 3); Mbarali District Council (n = 8); Kyela District Council (n = 6); Rungwe District Council (n = 7) and, Busekelo District Council (n = 4). The selected RCH clinics were estimated to provide services to approximately 1036 pregnant women.

 Study population

All pregnant women aged 15-49 years who attended the selected RCH clinics within their first and second trimesters (less than 28 weeks of gestation) were invited to participate in the study. A total of 574 pregnant women were eligible, and 420 agreed to take part. Pregnant women who refused to consent and those who were unable to communicate due to illness or taking medication were excluded from the study. Participants in their second trimester, a period during which fetal neurodevelopment is impacted by adequate maternal thyroid function, were included. To eliminate the effects of gestational age on thyroid hormone, participants beyond eight weeks of gestation were excluded.

 Sample size and sampling procedure

The prevalence of iodine deficiency among women of reproductive age reported in Tanzania was estimated at 40%. Based on this figure and the population of pregnant women in this region, a sample size of 574 was calculated using the Lwanga and Lemeshow formula 23 with: margin error of 5%; confidence level of 95%; design effect of 1.5 and; an additional of 10% to account for non-response. Only 420 pregnant women agreed to participate and this sample size was considered satisfactory assuming intracluster correlation coefficient (ICC) of 0.10 and, a power of 80%.

The sampling procedure involved two steps: First, a list of 251 government, private and faith-based health facilities providing RCH services in Mbeya region was obtained and used in a random selection of the health facilities to be involved in the study from each district. Given the sampling frame of health facilities in Mbeya, probability proportional to size was performed to allocate the number of facilities per District for inclusion in the survey. Out of 251 health facilities that offer RCH services (eligibility criteria) in Mbeya , forty two facilities were randomly selected for the study. An additional two reserved clusters were included in the survey. Therefore, a total of 44 health facilities offering RCH services located in the Mbeya region were visited and surveyed.

The second step involved the selection of pregnant women for each selected health facility. An eligibility form was used to list all pregnant women attending RCH services in the selected health facility. The resulting list of pregnant mothers served as the sampling frame for the selection of participants who met the inclusion criteria. Systematic Random Sampling was then carried out by using the list of mothers to randomly select required pregnant women for each facility to participate in the survey based on probability and proportion to size sampling for the specific facility.

 Data collection

Data were collected through interviews guided by a structured questionnaire and, laboratory analysis of urine samples. A standard structured questionnaire was constructed in English and translated into Kiswahili, a language that is spoken by almost 95% of Tanzanians (see Extended data). 24 To ensure the quality of the translation, back-translation was performed by independent translators and reviewed by field staff in Mbeya. Pre-testing was done to evaluate the quality of the translations in terms of comprehensibility, readability, and relevance to assess face validity.

The interviews were administered by a trained Nurse Midwife in face-to-face interviews with participants, before the collection of urine samples. Initial interviews were administrated to determine various social demographic characteristics and dietary factors concerning iodine status, including participant’s age, marital status, education, household assets possessions, socioeconomic status, parity, stage of pregnancy, and dietary habits.

 Urine sample collection and laboratory analysis

A trained Nurse Midwife collected urine samples from consented study participants. The urine samples were collected in a disposable plastic screw caped container of 100 ml. Before urine collection, the approximate volume of urine sample required was pre-marked by a trained Nurse Midwife on urine containers and instructed the participant on collecting her urine in the container. The collected urine sample was stored in the cool-box with a temperature of approximately 2-8 oC for 2-4hours. At the temporary laboratory, the sample was processed, transferred into two 2 ml vials, and then labeled by the trained laboratory technician. The sample into vials was kept stored at -20 oC. All urine samples were shifted to TFNC laboratory (located in Dar es Salaam) for further analysis within one month. The urine samples were analyzed using the ammonium persulfate digestion method, as previously described by Sandell-Kolthoff reaction. 25 TFNC laboratory is registered and successfully participated in the quality assurance program for Ensuring the Quality of Urinary Iodine Procedures (EQUIP) 26 offered by the Centres for Disease Control and Prevention (CDC), Atlanta, Georgia, USA. The assay accuracy was assessed using reference quality-control urine specimens obtained from the CDC. The assay detection limit was <5.0 μg/L with the coefficient of variation <10%, when compared to the reference method. 26

Variables

Outcome/response variable

Median UCI as a response variable was split into three categories as per WHO recommended level of iodine micronutrient. The median urine iodine concentration (mUIC) indicated the level of iodine in urine (μg/L) (see Underlying data). 24

UIC 1 (Iodine <150 μg/L) = Insufficient iodine

UIC 2 (150< Iodine <249 μg/L) = Sufficient iodine

UIC 3 (Iodine >250 μg/L) = Above WHO recommended/excessive iodine intake

Independent variables/predictors

The study includes a set of independent variables to understand the extent and variations between the levels of iodine micronutrients among the participants. Socio-demographic variables assessed included age, residency (district), education level, occupation status, number of pregnancies, visits to the ANC and, upper mid-arm circumference (MUAC), which is the most accurate way to measure fat-free mass outside of a laboratory. Household wealth was also assessed. To do so, durable household assets that indicate wealth such as a radio, television, and telephone were recorded as (1) “available and in working condition” or (0) “not available and/or not in working condition.” Principal component analysis, PCA was then conducted to categorize households into five quartiles of wealth, with 1 being the lowest and 3 the highest. Diet, in specific consumption of certain foods, such as fish, dairy products and processed meat and, refined and baked foods was also assessed among the participants, using 24-hour recalls.

 Data analysis

The data were analyzed using Stata v 15.1(RRID: SCR_012763). Stata is proprietary software but an open-access alternative in which the sequence could have been generated is Microsoft Excel (RRID: SCR_016137). Descriptive statistics were used to summarize the data of study participants. Pearson’s chi-square test and p-values were used to test for the significance of each of the potential risk factors in bivariate analysis. Multinomial logistic regression models were used to adjust for cofounders and predict the true association between the dependent and independent variables. All tests were two-tailed, and the significance level was set at p ≤ 0.05.

 Ethical approval and informed consent

Ethical clearance was obtained from the National Institute for Medical Research (NIMR) with reference number NIMR/HQ/R.8a/Vol. IX/2589 and appropriate authorization was given from the Regional, Council and health facility level. All eligible subjects were given information about the survey and were asked to sign a written informed consent form before participation.

 Results Descriptive of the study participants

The socio-demographic profile of overall sample is shown in Table 1. In this study, 420 agreed to participate (response rate of 73%). More than half of the respondents belonged to age 15-24 years. The mean age of the pregnant women was 25.49 (± 6.37) years. The majority of the respondents (70%) had primary education, two third of the respondents has been pregnant more than once. The household socio-economic composition of the sample shows a better distribution of all categories of respondents with about one third belong to the poorest quintile. More than two third of the respondents were self-employee (84.5%) followed by not employed (11.9%), and formal employee (3.6%). Improved source of water was reported by 71% of the participants. The distribution of respondents according to dietary habit, more than half of the respondents were reported that they consumed fish (68%) and, more than 90% consumed dairy products.

Frequency distribution of the study participant in Mbeya (n = 420).
Variables Category n %
Age group 15-24 215 52.18
25-34 147 35.68
35-49 50 12.14
Education level No formal education 34 8.10
Primary education 301 71.67
Secondary and above 85 20.24
Marital status] Married 238 56.67
Cohabit 133 31.67
Single 39 9.29
Divorced 10 2.38
Occupational status Formal employment 15 3.57
Self-employment 355 84.52
Not employed 50 11.90
Antenatal care center (ANC) visit 1 visit 163 38.81
2-3 visits 226 53.81
More than 3 visits 31 7.38
Residence Chunya District Council 45 10.71
Mbeya District Council 97 23.10
Mbarali District Council 93 22.14
Kyela District Council 50 11.90
Rungwe District Council 68 16.19
Busokelo District Council 33 7.86
Mbeya City 34 8.10
Number of pregnancies Primiglavida 104 24.76
Multiglavida 316 75.24
Type of water source Improved 302 71.90
Unimproved 118 28.10
Mid-upper arm circumference (MUAC) categorization MUAC < 23 cm 16 3.81
MUAC ≥ 23 cm-MUAC < 33 cm 383 90.19
MUAC ≥ 33 cm 21 5.0
Consumption of fish No 287 68.3
Yes 133 31.7
Consumption of Dairy products No 381 90.7
Yes 39 9.8
Consumption of Processed meat No 410 97.6
Yes 10 2.4
Urinary Iodine Concentration (UIC) categorization Insufficient (UIC 0–149 μg/l) 72 17.14
Sufficient (UIC 150–249 μg/l) 102 24.29
Above recommended (>250 μg/l) 246 58.57
Urinary iodine concentration (UIC)

The median UIC in the present study was 279.4μg/L, and it ranged from 26.1-1915μg/L. According to the UIC results, 17.14% of participants had an insufficient iodine intake, 24.29% had sufficient urine iodine concentration, and 58.57% had above the WHO recommended level of iodine in urine ( Table 1).

 Bivariate analysis

Table 2 presents a cross-tabulation of the median UIC status (mUIC) and socio-demographic, economic and dietary factors among pregnant women in Mbeya. Of 215 participants aged between 15-24 years, 17% had UIC (0–149 μg/l) that would be considered insufficient, and 55.8% had UIC (>250 μg/l) above the WHO recommended levels. The residence profile of the sample shows that, Chunya and Mbarali DCs have the highest percentage (above 70%) of the WHO recommended UIC among the pregnant women in Mbeya. On other hand, Rungwe DC had the highest percentage (27.9%) of participants with insufficient urine iodine concentrations. From the 133 participants who had fish in their diet, UIC was insufficient in 23%, sufficient in 19.4%, and 56.9% had above the WHO recommended level.

Predictors of median urine iodine concentrations (mUIC) among pregnant women in Mbeya (n = 420).
Variable Category Insufficient (Urinary Iodine Concentration (UIC) 0–149 μg/l) sufficient (UIC 150–249 μg/l) Above recommended (>250 μg/l) Chi-square P-value
% n % n % N
Age group 15-24 17.2 37 26.9 58 55.8 120 4.0208 0.403
25-34 16.3 24 22.4 33 61.2 90
35-49 14.0 7 16.0 8 70.0 35
Education level No formal education 26.4 9 17.6 6 55.8 19 4.314 0.634
Primary education 15.6 47 24.9 75 59.4 179
Secondary and above 18.8 16 24.7 21 56.5 48
Wealth Index 1 tercile 20.7 29 26.4 37 52.8 74 4.7325 0.316
2 tercile 15.7 22 20.0 28 64.2 90
3 tercile 15.0 21 26.4 37 58.5 82
Marital status Married 18.4 44 22.2 53 59.2 141 2.71 0.838
Cohabit 15.0 20 28.5 38 56.3 75
Single 17.9 7 23.0 9 58.9 23
Divorced 10.0 1 20.0 2 70.0 7
Occupational status Formal employment 20.0 3 33.3 5 46.6 7 4.132 0.388
Self-employment 17.4 62 22.5 80 60.0 213
Not employed 14.0 7 34.0 17 52.0 26
Antenatal care center (ANC) visit 1 visit 17.1 28 23.9 39 58.9 96 3.3699 0.498
2-3 visits 18.5 42 24.7 56 56.6 128
More than 3 visits 6.4 2 22.5 7 70.9 22
Residence Chunya DC 6.6 3 22.2 10 71.1 32 31.987 0.001 *
Mbeya DC 21.6 21 32.9 32 45.3 44
Mbarali DC 11.8 11 13.9 13 74.1 69
Kyela DC 12.0 6 20.0 10 68.0 34
Rungwe DC 27.9 19 25.0 19 47.0 32
Busokelo DC 24.2 8 27.2 9 48.4 16
Mbeya city 11.7 4 32.3 11 55.8 19
Number of pregnancies Primiglavida 16.3 17 22.1 23 61.5 64 0.527 0.768
Multiglavida 17.7 55 25.0 79 57.5 182
Type of water source Improved 18.5 56 23.5 71 57.9 175 0.567 0.457
Unemployed 13.5 16 26.2 31 60.1 71
Mean- upper arm circumference (MUAC) categorization MUAC < 23 cm 12.5 2 18.7 3 68.7 11 0.987 0.912
MUAC ≥ 23 cm-MUAC < 33 cm 17.4 67 24.2 91 58.2 223
MUAC ≥ 33 cm 14.2 3 28.5 6 57.1 12
Consumption of fish No 13.7 38 26.8 74 59.4 164 7.5619 0.023 *
Yes 23.6 34 19.4 28 56.9 82
Consumption of Dairy products No 17.1 60 23.7 83 59.0 206 0.2912 0.865
Yes 16.9 12 26.7 16 56.3 40
Consumption of Processed meat No 17.2 71 23.8 98 58.8 242 2.0475 0.359
Yes 11.1 1 44.4 4 44.4 4
Consumption of refined and baked No 18.9 14 24.3 18 56.7 42 0.2158 0.898
Yes 16.7 58 24.2 84 58.9 204

Represents p value: p < 0.05.

 Multivariate analysis

The multivariate analysis are presented in Tables 3a and 3b. The chi-square model (63.51) was 0.0176, with p < 0.05. Multivariate analysis on factors associated with the low WHO recommendation for mUIC among pregnant women in Mbeya (n = 420).
Independent 95% confidence interval for OR
Variable Category OR Lower bound Upper bound P-value
Consumption of fish No 1
Yes 2.60 1.31 5.15 0.006 *
Consumption of Dairy products No 1
Yes 0.96 0.41 2.28 0.940
Consumption of Processed meat No 1
Yes 0.32 0.03 3.19 0.334
Consumption of refined and baked No 1
Yes 0.79 0.33 1.91 0.609
Residence Mbeya district council (DC) 1
Chunya DC 0.38 0.08 1.65 0.199
Mbarali DC 1.15 0.38 3.44 0.793
Kyela DC 0.85 0.25 2.92 0.809
Rungwe DC 2.43 0.95 6.19 0.061
Busokelo DC 1.79 0.52 6.11 0.351
Mbeya city 0.77 0.18 3.24 0.732
Age group 15-24 1
25-34 1.11 0.50 2.44 0.782
35-49 1.45 0.42 4.98 0.553
Wealth Index 1 quantile 1
2 quantile 1.62 0.62 4.26 0.321
3 quantile 1.28 0.42 3.90 0.663
Education level No formal education 1
Primary education 0.29 0.08 0.99 0.049 *
Secondary and above
Mean upper arm circumference (MUAC) categorization MUAC < 23 cm 1
MUAC ≥ 23 cm-MUAC < 33 cm 1.27 0.18 8.90 0.807
MUAC ≥ 33 cm 1.17 0.10 13.31 0.896
Number of pregnancies Primiglavida 1
Multiglavida 0.83 0.34 2.01 0.683

Represents p value: p < 0.05.

Multivariate analysis on factors associated with the above WHO recommendation for mUIC among pregnant women in Mbeya (n = 420).
Independent 95% confidence interval for OR
Variable Category OR Lower bound Upper bound P-value
Consumption of fish No 1
Yes 1.24 0.71 2.15 0.438
Consumption of Dairy products No 1
Yes 0.90 0.46 1.73 0.754
Consumption of Processed meat No 1
Yes 0.50 0.11 2.31 0.379
Consumption of refined and baked No 1
Yes 0.99 0.50 1.96 0.998
Residence Mbeya DC 1
Chunya DC 2.05 0.84 4.96 0.110
Mbarali DC 4.09 1.85 9.01 0.000 *
Kyela DC 2.15 0.88 5.23 0.089
Rungwe DC 1.49 0.67 3.29 0.321
Busokelo DC 1.55 0.56 4.26 0.390
Mbeya city 1.45 0.54 3.94 0.456
Age group 15-24 1
25-34 1.47 0.81 2.69 0.201
35-49 2.51 0.99 6.33 0.050 *
Wealth Index 1 tercile 1
2 tercile 1.41 0.66 3.03 0.367
3 tercile 2.08 0.91 4.71 0.079
Education level No formal education 1
Primary education 1.04 0.37 2.94 0.929
Secondary and above 1.18 0.35 3.95 0.777
MUAC categorization MUAC < 23 cm 1
MUAC ≥ 23 cm-MUAC < 33 cm 1.01 0.25 4.05 0.985
MUAC ≥ 33 cm 0.74 0.13 4.27 0.745
Number of pregnancies Primiglavida 1
Multiglavida 0.64 0.32 1.26 0.203

Represents p value: p < 0.05.

Table 3a presents a multivariate analysis of the factors related to the low WHO recommendation for mUIC among pregnant women. After taking into account confounding factors, it was found that dietary and socio-demographic factors were significantly associated with the low WHO recommendation for mUIC among pregnant women. The analysis showed that pregnant women who consume fish were more likely to have a low WHO recommendation for mUIC, with an Adjusted OR of 2.60 (95% CI 1.31-5.15). In contrast, pregnant women who had at least primary education were less likely to have a low WHO recommendation for mUIC, with an Adjusted OR of 0.29 (95% CI 0.08-0.99). Table 3a presents a multivariate analysis of the factors related to the low WHO recommendation for mUIC among pregnant women. After taking into account confounding factors, it was found that dietary and socio-demographic factors were significantly associated with the low WHO recommendation for mUIC among pregnant women. The analysis showed that pregnant women who consume fish were more likely to have a low WHO recommendation for mUIC, with an Adjusted OR of 2.60 (95% CI 1.31-5.15). In contrast, pregnant women who had at least primary education were less likely to have a low WHO recommendation for mUIC, with an Adjusted OR of 0.29 (95% CI 0.08-0.99).

Table 3b presents a multivariate analysis of the factors related to the above WHO recommendation for mUIC among pregnant women. After taking into account confounding factors, it was found that only socio-demographic factors were significantly associated with the above WHO recommendation for mUIC among pregnant women. Thus, pregnant women resident in Mbarali DC, aged between 35-49 years and belonging to the highest socio-economic status [Adjusted OR = 4.09 (95% CI 1.85-9.010], [Adjusted OR = 2.51 (95% CI 0.99-6.330] and [Adjusted OR = 2.08 (95% CI 0.91-4.71) respectively were at greatest risk of excess mUIC.

 Discussion

This is the first population-based cross-sectional study to assess the magnitude of iodine status and the association with socio-demographic factors and diet in Tanzanian pregnant women. The findings of the study are important since iodine insufficiency is the most prevalent micronutrient insufficiency, affecting 28.9% of the world population, 27 particularly affecting women living in developing countries. 28 Iodine deficiency disorders in Tanzania are high with the most recent figures indicating that more than 40% of the population in the country lives in geographical regions prone to iodine insufficiency. 6 However, this data is largely outdated, as more recent data as well as the most recent efforts to reduce iodine insufficiency have focused on primary school children in Tanzania, 6 whilst the iodine micronutrient status among pregnant women has been overlooked in recent years. In 2010, a reanalysis of the Tanzania demographic and health survey reported 54% of pregnant women with iodine deficiency. 19

The present research looks for potential socio-demographic and dietary factors associated with levels of mUIC both below and above WHO recommendations. Our study found that residence in Mbarali district, age between 35-49 years, and belonging to high socio-economic status were associated with an increasing odds of pregnant women having excess mUIC levels. This could be explained by the fact that the Mbarali district is home to the Ruaha National Park, which attracts food products to its business distract that are preserved by iodized salt. As previous documented 29 , 30 the frequency and intake of food products preserved or snacks sprinkled with iodated salt were one of the four scenarios of excessive iodine intakes, the other three were: close to salt factories; losses less than expected because of not passing through all the steps in the salt marketing chain, and districts close to large salt processing factories adhering to USI. Thus, it is important to continue monitoring the distribution, packaging and handling iodated salt and, similarly to monitor thyroid function and its associated disorders in this population since excessive iodine is thought to matter most at the time of fetal development.

This study also looked for evidence of the factors likely to be influencing the prevalence of insufficient mUIC among pregnant women in Mbeya. We found that consumption of fish was associated with lower mUIC. This finding could be explained by the fact that iodine levels in freshwater fish may be low and they are not preserved with salt. 6 The poor iodation technologies and supply of potassium iodate in many small and medium salt producers could be the reason behind insufficient iodine observed in this survey. 31 , 32 Moreover, during pregnancy there are variations in the functionality of the thyroid. This can increase the risk of insufficient iodine intake for some mothers. As such, predicting UIC based on usage of iodized salt alone, may not be accurate. 33 – 35 However, other studies have documented that freshwater fish may contain Iodine at levels that can improve daily Iodine intake. 31

In countries with successful USI programs, studies have reported an optimal median UIC in pregnant women. As such, USI remains the most cost-effective strategy for achieving reduced IDD. 36 , 37 However, the full implementation of USI remains a challenge in many sub–Saharan African countries including Tanzania, 38 largely due to the lack of adequate enforcement and, the inadequate monitoring of small-scale salt producers who often do not comply with USI legislation. 6

This analysis also indicated that pregnant women who had a primary school education were at lower risk of iodine insufficiency; however, further studies are needed to investigate this association. 30 Similar evidence has been reported in previous studies among primary school children in Tanzania alerts policymakers to consider adjusting the amount of iodine added to salt along with the obligation of reducing discretionary foods and salt intake. 39 , 40

WHO recommended an increased iodine intake for pregnant women, although evidence is weak. 41 Indeed, detrimental effects from more than adequate and excessive iodine intake have been reported in general populations. 42 – 44 Shi et al. have reported on the associations between UIC and thyroid health among pregnant women and recommend a lower limit for maternal iodine intake during pregnancy than that currently advised by the WHO. 45 This is also an area in need of further investigation. The question remains how much iodized salt pregnant women in Mbarali district should consume, and at what concentration. As our findings illustrate, it is extremely difficult for USI to avoid not only deficiency but also excess, especially in mixed urban and rural settings in areas with complex salt production by small and large producers. 6 , 18

The strength of this study is in the fact that its large population-based sample size managed to demonstrate important factors that could explain factors associated with both excessive and insufficient iodine among pregnant women. However, there were limitations as follows: first, the use of UIC to determine individual iodine status could be limited due to the potential for misclassification of participants because of day-to-day variations. Second, UIC reflects recent iodine intake or exposure rather than chronic individual iodine status. Third, the use of iodized salt was not assessed in this study. Finally, it would have been useful to have a non-pregnant control group to help ascertain whether lower mean UIC concentrations during pregnancy could be attributed to pregnancy itself or the diet.

 Conclusion

This study demonstrated a significant association between geographical factors (residence in the Mbarali district) and median urine iodine concentration above WHO recommended. Further, this study found association between inadequate iodine in diet and insufficient median urine iodine concentration as indicated by the World Health Organization recommendation. Therefore, educational programs on iodine intake should be strengthen as attending at least primary education was found to be a protective factor for insufficient median urine iodine concentration. This study also recommend for further longitudinal studies. Further, attending at least primary education was found to be a protective factor for insufficient median urine iodine concentration. Controlling risk factors through strengthening the USI program to include monitoring excessive iodine exposures will reduce the detrimental effects of iodine during pregnancy. This study also recommend for further longitudinal studies. It illustrates how difficult it is to adjust salt iodation levels so as to avoid both deficiency and excess and the importance of continued monitoring and adjustment of iodation levels at regional and even local levels as needed.

 Data availability Underlying data

Open Science Framework (OSF): Factors associated with inadequate urinary iodine concentration among pregnant women in Mbeya region Tanzania. DOI: https://osf.io/7ysb9/. 24

This project contains the following underlying data:

MBMNS_MUIC10082021: This is the SPSS database file that contained all the laboratory assessment variables for the medium urine iodine concentrations.

This project also contains the following extended data:

Questionnaire English version: This file contains all the questions used to interview pregnant women in Mbeya.

Questionnaire Swahili version: This file is the Swahili version of Questionnaire.

Data are available under the terms of the Creative Commons Zero “No rights reserved” data waiver (CC0 1.0 Public domain dedication).

 Author contributions

Conceptualization, TL, RM., AH. SEJ and GHL; project administration and resources, AS, RN, FK and GHL; formal analysis and writing—original draft, TL, RM, AH, SEJ, HAP, AS, RN, FK, ET and GHL; reviewed and edited the manuscript. GB and, RB. All authors: Reviewed and agreed upon the final manuscript.

 Acknowledgments

We sincerely appreciate health workers assistance in the laboratory data collection and the participation of all pregnant women in this study.

 References Zimmermann MB Jooste PL Pandav CS : Iodine-deficiency disorders. The Lancet. 2008;372:1251–1262. 18676011 10.1016/S0140-6736(08)61005-3 Chung HR : Iodine and thyroid function. Ann Pediatr Endocrinol Metab. 2014;19:8. 24926457 10.6065/apem.2014.19.1.8 PMC4049553 Kapil U : Health consequences of iodine deficiency. Sultan Qaboos Univ Med J. 2007;7:267–272. 21748117 PMC3074887 WHO: Eliminating iodine deficiency worldwide is within reach.GENEVA.2004. 21 December 2004. Reference Source Zimmermann MB Boelaert K : Iodine deficiency and thyroid disorders. Lancet Diabetes Endocrinol. 2015;3:286–295. 10.1016/S2213-8587(14)70225-6 Assey VD : Tanzania national survey on iodine deficiency: Impact after twelve years of salt iodation. BMC Public Health. 2009;9:1–11. 19728863 10.1186/1471-2458-9-319 PMC2749826 Wächter W Mvungi M König A : Prevalence of goitre and hypothyroidism in Southern Tanzania: effect of iodised oil on thyroid hormone deficiency. J Epidemiol Community Health. 1986;40:86–91. 3011945 10.1136/jech.40.1.86 PMC1052495 Joint WHO/UNICEF/ICCIDD consultation : Review of findings from 7-country study in Africa on levels of salt iodization in relation to iodine deficiency disorders, including iodine-induced hyperthyroidism.July 1996, Geneva, Switzerland. https://www.who.int/publications/i/item/WHO-AFRO-NUT-97.5 NBS/Tanzania, N. B. of S.- & Macro, I: Tanzania Demographic and Health Survey 2010. 2011. Assessment of iodine deficiency disorders and monitoring their elimination Third edition A guide for programme managers. Venkatesh Mannar MG Bohac L : Achieving Universal Salt Iodization: Lessons Learned and Emerging Issues. Appiah PK Yanbom CT Ayanore MA : Iodine Content of Salt Use after Years of Universal Iodization Policy and Knowledge on Iodized Salt among Households in the Sissala East Municipality in Upper West Region of Ghana. J Food Quality. 2020;2020. Eastman CJ Zimmermann MB : The Iodine Deficiency Disorders. Endotext. MDText.com, Inc.;2000. Zimmermann MB Andersson M : Assessment of iodine nutrition in populations: past, present, and future. 2012. 10.1111/j.1753-4887.2012.00528.x Gorstein JL Bagriansky J Pearce EN : Estimating the Health and Economic Benefits of Universal Salt Iodization Programs to Correct Iodine Deficiency Disorders. Thyroid. 2020;30:1802–1809. 32458745 10.1089/thy.2019.0719 PMC7757618 Tanzania legislation on USI:1995; accessed 03/08/2021. Reference Source Saha S Abu BAZ Jamshidi-Naeini Y : Is iodine deficiency still a problem in sub-Saharan Africa?: a review. Proc. Nutr. Soc. 2019;78(4):554–566. 30739612 10.1017/S0029665118002859 Assey V Mwanditani R Kimbioka S : IDD NEWSLETTER NOVEMBER 2013 TANZANIA. 2013:9. Mtumwa AH Ntwenya JE Paul E : Socio-economic and spatial correlates of subclinical iodine deficiency among pregnant women age 15-49 years in Tanzania. BMC Nutrition. 2017;3:1–10. 10.1186/s40795-017-0163-1 Ba DM : Factors Associated with Urinary Iodine Concentration among Women of Reproductive Age, 20-49 Years Old, in Tanzania: A Population-Based Cross-Sectional Study. Current Developments in Nutrition. 2020;4. 32462108 10.1093/cdn/nzaa079 PMC7236838 Farebrother J Zimmermann MB Andersson M : Excess iodine intake: sources, assessment, and effects on thyroid function. Ann N Y Acad Sci. 2019;1446:44–65. 30891786 10.1111/nyas.14041 Lee SY Pearce EN : Reproductive endocrinology: Iodine intake in pregnancy-even a little excess is too much. Nat Rev Endocrinol. 2015;11:260–261. 25707780 10.1038/nrendo.2015.28 PMC4590285 Lemeshow S Hosmer DW Klar J : Adequacy of sample size in health studies. 1990. Masumo R : IMAN.2021, August 13. Reference Source accessed 03/08/2021. Reference Source Caldwell KL : EQUIP: a worldwide program to ensure the quality of urinary iodine procedures Springer-Verlag 2005. Accred Qual Assur. 2005;10:356–361. Andersson M Karumbunathan V Zimmermann MB : Global iodine status in 2011 and trends over the past decade. J Nutr. 2012;142:744–750. 22378324 10.3945/jn.111.149393 Glinoer D : The importance of iodine nutrition during pregnancy. Public Health Nutrition. 2007;10:1542–1546(Public Health Nutr). Farebrother J Zimmermann MB Abdallah F : Effect of Excess Iodine Intake from Iodized Salt and/or Groundwater Iodine on Thyroid Function in Nonpregnant and Pregnant Women, Infants, and Children: A Multicenter Study in East Africa. Thyroid (TSI). 2018;28(9):1198–1210. ISSN 1050-7256. 10.1089/thy.2018.0234 Venance MS Martin HD Kimiywe J : Iodine Status and Discretionary Choices Consumption Among Primary School Children, Kinondoni Tanzania. Pediatric Health Med Ther. 2020;11:359–368. 33061741 10.2147/PHMT.S265117 PMC7518783 Eckhoff KM Maage A : Iodine Content in Fish and Other Food Products from East Africa Analyzed by ICP-MS. J Food Composition Analysis. 1997;10:270–282. 10.1006/jfca.1997.0541 Krela-Kaźmierczak I : Is there an ideal diet to protect against iodine deficiency? Nutrients. 2021;13:1–15. 33557336 10.3390/nu13020513 PMC7914421 Glinoer D : Pregnancy and iodine. Thyroid. 2001;11:471–481. 11396705 10.1089/105072501300176426 Marchioni E : Iodine deficiency in pregnant women residing in an area with adequate iodine intake. Nutrition. 2008;24:458–461. 18337060 10.1016/j.nut.2008.01.015 Gowachirapant S : Urinary iodine concentrations indicate iodine deficiency in pregnant Thai women but iodine sufficiency in their school-aged children. J Nutri. 2009;139:1169–1172. 18337060 10.1016/j.nut.2008.01.015 Mao G : Iodine deficiency in pregnant women after the adoption of the new provincial standard for salt iodization in Zhejiang Province, China. BMC Pregnancy and Childbirth. 2018;18:1–7. 30075759 10.1186/s12884-018-1952-5 PMC6091046 Reaching optimal iodine nutrition in Pregnant and Lactating Women and Young Children. Reference Source Ba DM : Factors Associated with Urinary Iodine Concentration among Women of Reproductive Age, 20-49 Years Old, in Tanzania: A Population-Based Cross-Sectional Study. Curr Dev Nutr. 2020;4. 32462108 10.1093/cdn/nzaa079 PMC7236838 Pearce EN Lazarus JH Moreno-Reyes R : Consequences of iodine deficiency and excess in pregnant women: an overview of current knowns and unknowns. Am J Clin Nutr. 2016;104:918S–923S. 27534632 10.3945/ajcn.115.110429 PMC5004501 Leung AM Braverman LE : Consequences of excess iodine. Nat Rev Endocrinol. 2014;10:136–142. 24342882 10.1038/nrendo.2013.251 PMC3976240 Andersson M de Benoist B Delange F : Prevention and control of iodine deficiency in pregnant and lactating women and in children less than 2-years-old: Conclusions and recommendations of the Technical Consultation. Public Health Nutrition. 2007; vol.10:1606–1611. Cambridge University Press. Guan H : Influence of iodine on the reference interval of TSH and the optimal interval of TSH: Results of a follow-up study in areas with different iodine intakes. Clin Endocrinol. 2008;69:136–141. 18042176 10.1111/j.1365-2265.2007.03150.x Sun X Shan Z Teng W : Effects of increased iodine intake on thyroid disorders. Endocrinol Metab (Seoul). 2014;29:240–247. 25309781 10.3803/EnM.2014.29.3.240 PMC4192807 Connolly RJ Vidor GI Stewart JC : Increase in thyrotoxicosis in endemic goitre area after iodation of bread. Lancet. 1970;295:500–502. 4190182 10.1016/s0140-6736(70)91582-5 Shi X : Optimal and safe upper limits of iodine intake for early pregnancy in iodine-sufficient regions: A cross-sectional study of 7190 pregnant women in China. J Clin Endocrinol Metab. 2015;100:1630–1638. 25629356 10.1210/jc.2014-3704 10.5256/f1000research.163953.r270358 Reviewer response for version 4 Greiner Ted 1 Referee Member, Executive Committee; Editor-in-Chief of Association journal World Nutrition (https://worldnutritionjournal.org/index.php/wn), World Public Health Nutrition Association (Ringgold ID: 26716), London, UK

Competing interests: No competing interests were disclosed.

 17 6 2024 Copyright: © 2024 Greiner T 2024 This is an open access peer review report distributed under the terms of the Creative Commons Attribution Licence, which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited. recommendation approve-with-reservations

1. "Table 3a. Multivariate analysis on factors associated with the low WHO recommendation"

Should end with "values below" or "women below" the WHO recommendation, since you are not commenting on the recommendation itself. The same is true for Table 3b.

Similarly, below the tables, " factors related to the low WHO" should be something like "factors related to the group of women who are below the WHO" and similar for the next paragraph regarding those above the recommendation. The paragraph about 3a is repeated a second time.

2. You write that ref. 19 found "54% of pregnant women with iodine deficiency" but the title of that article should be repeated (subclinical). If this is based on goiter rather than UIC, I suggest mentioning this because goiter in Tanzanian adults is likely related more to past than present IDD. Is it correct that this analysis was done in 2010? (It was published in 2017.) Rather than saying recent attention to pregnant women has been overlooked, I suggest saying it has been limited. 

3. "distract" should presumably be district.

4. After saying "In countries with successful USI programs, studies have reported an optimal median UIC in pregnant women." I suggest giving ref. 37 (which is incomplete) if it says that. Then, in comparison, give the median from your study. 

5. In the conclusions, the first mention of "This study also recommend for further longitudinal studies." should be deleted. Instead of "recommend for", I suggest "recommends".

Is the work clearly and accurately presented and does it cite the current literature?

Partly

If applicable, is the statistical analysis and its interpretation appropriate?

Yes

Are all the source data underlying the results available to ensure full reproducibility?

Yes

Is the study design appropriate and is the work technically sound?

Yes

Are the conclusions drawn adequately supported by the results?

Partly

Are sufficient details of methods and analysis provided to allow replication by others?

Yes

Reviewer Expertise:

Public Health Nutrition

I confirm that I have read this submission and believe that I have an appropriate level of expertise to confirm that it is of an acceptable scientific standard, however I have significant reservations, as outlined above.

Masumo Ray Tanzania Food and Nutrition Centre, Tanzania

Competing interests: The authors declare that there are no conflicts of interest.

 31 7 2024

We appreciate the time and effort that you and the reviewers dedicated to providing valuable feedback on our manuscript.

Thank you very much.

Here is a point-by-point response to the reviewers’ comments and concerns.

Results

Comment 1: "Table 3a. Multivariate analysis on factors associated with the low WHO recommendation" Should end with "values below" or "women below" the WHO recommendation, since you are not commenting on the recommendation itself. The same is true for Table 3b.

Similarly, below the tables, “factors related to the low WHO" should be something like "factors related to the group of women who are below the WHO" and similar for the next paragraph regarding those above the recommendation. The paragraph about 3a is repeated a second time.

Response: We agree with this comment and have included your suggestion.

Discussion:

Comment 2: You write that ref. 19 found "54% of pregnant women with iodine deficiency" but the title of that article should be repeated (subclinical). If this is based on goiter rather than UIC, I suggest mentioning this because goiter in Tanzanian adults is likely related more to past than present IDD. Is it correct that this analysis was done in 2010? (It was published in 2017.) Rather than saying recent attention to pregnant women has been overlooked, I suggest saying it has been limited.

Response: We agree with this comment and have rewritten the discussion section to stipulate that 54% of pregnant women had subclinical iodine deficiency.  

Comment 3: "distract" should presumably be district.

Response: Thank you so much for pointing this out and, we have corrected it.

Comment 4: After saying "In countries with successful USI programs, studies have reported an optimal median UIC in pregnant women." I suggest giving ref. 37 (which is incomplete) if it says that. Then, in comparison, give the median from your study.

Response: We agree with this comment and have rewritten the paragraph as suggested.

Conclusion:

Comment 5: In the conclusions, the first mention of "This study also recommend for further longitudinal studies." should be deleted. Instead of "recommend for", I suggest "recommends".

Response: Thank you for pointing this out. We agree with this comment and edited the sentence as suggested.

 10.5256/f1000research.137830.r166505 Reviewer response for version 2 Greiner Ted 1 Referee Member, Executive Committee; Editor-in-Chief of Association journal World Nutrition (https://worldnutritionjournal.org/index.php/wn), World Public Health Nutrition Association (Ringgold ID: 26716), London, UK

Competing interests: No competing interests were disclosed.

 5 4 2023 Copyright: © 2023 Greiner T 2023 This is an open access peer review report distributed under the terms of the Creative Commons Attribution Licence, which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited. recommendation approve-with-reservations

Factors associated with inadequate urinary iodine concentration among pregnant women in Mbeya region Tanzania

This is a useful, reasonably well-done study on an important issue. The OR data referred to in the discussion and any other OR data that was analyzed need to be presented in the Results section. The authors write well in English but it still needs to be proofread and several further corrections made in addition to the comments below:

I think it would be useful to give some of the early provincial school-based survey data from Tanzania showing that Mbeya had one of the highest prevalences of goiter, thus indicating the value of having chosen it for this study.

Is there no more modern reference than your #5? Its data are over 30 years old.

“Although USI is arguably the successful [health] intervention in global history”

“UNICEF documented that [in] many African countries ”

 “whose diets contain sufficient [should presumably read “substantial”] amounts of iodine-depleting foods”

“more than 10% of Tanzania households [which] remain at risk”

“It is also necessary to set upper as well as lower limits for maternal iodine intake to ensure optimal health outcomes are achieved” - This does not seem to be a task related to the present paper. On the other hand, its findings could be useful in future decision making regarding whether levels of iodine added to fortified salt should be increased or decreased, especially if such decisions can be taken at regional or local salt facility levels.

“All pregnant women aged 15-49 years who attended [the selected RCH] clinics within their first and second trimesters ”

“420 agreed to take part, as per the calculated sample size” - The final clause in this statement is premature and in any case incorrect and needs to be reworded or deleted.

“medium urine iodine concentration” Presumably should be “median”?

“Median UCI as a response variable was split into three categories as per WHO recommended level of iodine micronutrient. A new variable called medium urine iodine concentration (UIC) was developed to indicate the level of iodine in urine (μg/L) (see Underlying data).21” - This is unclear and seems unnecessary. Surely each UIC was simply placed into one of the three existing WHO categories.

Is the wealth index in Table 1 really quintiles? If so, which one is completely absent? If it is terciles, which seems likely, remove it from the table, as 33% in each category is literally the definition of a tercile, not a finding.

In Table 2, the n’s are missing from secondary education.

I don’t see the point in Table 3. It cannot be interpreted on its own. It does not even mention fish. Seeing it does not add any information or clarity to what is presented in the text and it only includes a few values, which could be given in the text if important.

“This research looks for evidence of the factors influence excess of UIC above the WHO recommended level.” Suggestion: “evidence of the factors likely to be influencing the prevalence of excess UIC”

“Geographical factors i.e. residence in Mbarali district, and the aged between 35-49 years were play an important role in increasing the risk of pregnant women having excess UIC above the WHO recommended level: [Adjusted OR = 4.09 (95% CI 1.85-9.01)] and, [Adjusted OR = 2.51 (95% CI 0.99-6.33)] respectively.” - This is confusing. It is written as if it’s evidence from the present study, but this was not included in the results section. Then it is referenced as if it is Assey’s finding. The next sentence (which is the one that presumably should be cited for Assey) is also confusing. Why would being closed to earlier access to iodated salt be associated with increased risk of excess?

In this sentence: “The poor iodation technologies and supply of potassium iodate in many small and medium salt producers could be the reason behind consumption of freshwater fish and insufficient iodine observed in this survey” I suggest deleting “consumption of freshwater fish and” since it is illogical.

“This analysis also indicated that pregnant women who had a primary school education were at lower risk of iodine insufficiency [Adjusted OR = 0.29 (95% CI 0.08-0.99)], ” was also not presented in the results section.

Is the work clearly and accurately presented and does it cite the current literature?

Partly

If applicable, is the statistical analysis and its interpretation appropriate?

Yes

Are all the source data underlying the results available to ensure full reproducibility?

Yes

Is the study design appropriate and is the work technically sound?

Yes

Are the conclusions drawn adequately supported by the results?

Partly

Are sufficient details of methods and analysis provided to allow replication by others?

Yes

Reviewer Expertise:

My areas of expertise are infant feeding and combatting micronutrient deficiencies (A, Fe, I) at public health levels. I am co-author on several publications related to IDD in Tanzania and worked on the issue some in the field there, for example a school goiter survey. I was involved with UIC measurement surveys, but am not myself expert on nor have I ever personally utilized laboratory methods in research.

I confirm that I have read this submission and believe that I have an appropriate level of expertise to confirm that it is of an acceptable scientific standard, however I have significant reservations, as outlined above.

Masumo Ray Tanzania Food and Nutrition Centre, Tanzania

Competing interests: We declare no competing interest.

 22 4 2023

We appreciate the time and effort that you have dedicated to providing constructive feedback on our manuscript entitled ‘Factors associated with inadequate urinary iodine concentration among pregnant women in Mbeya region Tanzania’.

Thank you very much.

Here is a point-by-point response to the reviewers’ comments and concerns.

General comment:

Comment 1: This is a useful, reasonably well-done study on an important issue. The OR data referred to in the discussion and any other OR data that was analyzed need to be presented in the Results section.

Response: Thank you very much for pointing this out and, we have incorporated your suggestion throughout the manuscript.

Specific comments:

Comment 1: I think it would be useful to give some of the early provincial school-based survey data from Tanzania showing that Mbeya had one of the highest prevalences of goiter, thus indicating the value of having chosen it for this study.

Response: We agree with this comment and have added more information on the early provincial school-based survey data from Tanzania showing that Mbeya had one of the highest prevalence of goiter.

Comment 2: Is there no more modern reference than your #5? Its data are over 30 years old.

Response: Thanks for pointing this out. We have replaced reference #5 with the recent reference.

Comment 3: “Although USI is arguably the successful [health] intervention in global history”

Response: Thank you very much for pointing this out and we have incorporated your suggestion.

Comment 4: “UNICEF documented that [in] many African countries”

Response: We agree with this comment and incorporated the suggestion.

Comment 5: “whose diets contain sufficient [should presumably read “substantial”] amounts of iodine-depleting foods”

Response: We agree with this comment and replaced the word ‘sufficient’ as suggested.

Comment 6: “More than 10% of Tanzania households [which] remain at risk”

Response: Thank you very much for pointing this out and, we have incorporated your suggestion.

Comment 7: “It is also necessary to set upper as well as lower limits for maternal iodine intake to ensure optimal health outcomes are achieved” - This does not seem to be a task related to the present paper. On the other hand, its findings could be useful in future decision-making regarding whether levels of iodine added to fortified salt should be increased or decreased, especially if such decisions can be taken at regional or local salt facility levels

Response: Thank you for pointing this out and we have rephrased the statement for clarity.

Comment 8 : “All pregnant women aged 15-49 years who attended [the selected RCH] clinics within their first and second trimesters”

Response: We agree with this comment and have incorporated your suggestion.

Comment 9 : “420 agreed to take part, as per the calculated sample size” - The final clause in this statement is premature and in any case incorrect and needs to be reworded or deleted .

Response: Thank you for this suggestion and we have incorporated your suggestion.

Comment 10: “medium urine iodine concentration” Presumably should be “median”?

Response: We agree with this comment and make the correction as suggested 

Comment 11: “Median UCI as a response variable was split into three categories as per WHO recommended level of iodine micronutrient. A new variable called medium urine iodine concentration (UIC) was developed to indicate the level of iodine in urine (μg/L) (see Underlying data).21” - This is unclear and seems unnecessary. Surely each UIC was simply placed into one of the three existing WHO categories.

Response: We highly appreciate the reviewers’ insightful and helpful comments on our manuscript. We have rewritten the statement for clarity.

Comment 12: Is the wealth index in Table 1 really quintiles? If so, which one is completely absent? If it is terciles, which seems likely, remove it from the table, as 33% in each category is literally the definition of a tercile, not a finding.

Response: We agree with this comment and make the correction as suggested.

Comment 13: In Table 2, the n’s are missing from secondary education

Response: Thank you for pointing that out, and we added the missing n’s and percentages

Comment 14: I don’t see the point in Table 3. It cannot be interpreted on its own. It does not even mention fish. Seeing it does not add any information or clarity to what is presented in the text and it only includes a few values, which could be given in the text if important.

Response: We agree with this comment and re-write the description of Table 3.

Comment 15: “This research looks for evidence of the factors influence excess of UIC above the WHO recommended level.” Suggestion: “evidence of the factors likely to be influencing the prevalence of excess UIC”

Response: We agree with this comment and incorporated your suggestion.

Comment 16:  “Geographical factors i.e. residence in Mbarali district, and the aged between 35-49 years were play an important role in increasing the risk of pregnant women having excess UIC above the WHO recommended level: [Adjusted OR = 4.09 (95% CI 1.85-9.01)] and, [Adjusted OR = 2.51 (95% CI 0.99-6.33)] respectively.” - This is confusing. It is written as if its evidence from the present study, but this was not included in the results section. Then it is referenced as if it is Assey’s finding. The next sentence (which is the one that presumably should be cited for Assey) is also confusing. Why would being closed to earlier access to iodated salt be associated with increased risk of excess?

Response: We agreed with the comment and we have re-write the paragraph for clarity.

Comment 17: In this sentence: “The poor iodation technologies and supply of potassium iodate in many small and medium salt producers could be the reason behind consumption of freshwater fish and insufficient iodine observed in this survey” I suggest deleting “consumption of freshwater fish and” since it is illogical.

Response: We agree with this comment and incorporate the suggestion.

Comment 18 : “This analysis also indicated that pregnant women who had a primary school education were at lower risk of iodine insufficiency [Adjusted OR = 0.29 (95% CI 0.08-0.99)],” was also not presented in the results section.

Response: Thank you for pointing that out, and we have re-write the paragraph for clarity.

 10.5256/f1000research.137830.r150166 Reviewer response for version 2 Kinabo Joyce 1 Referee https://orcid.org/0000-0003-1559-9143 Department of Human Nutrition and Consumer Sciences, Sokoine University of Agriculture, Morogoro, Tanzania

Competing interests: No competing interests were disclosed.

 22 9 2022 Copyright: © 2022 Kinabo J 2022 This is an open access peer review report distributed under the terms of the Creative Commons Attribution Licence, which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited. recommendation approve

Urine samples

please indicate time of the day the samples were collected. Was it morning urine or random (i.e. collected at any time of the day)?

Variables

'Median UCI' - misspelt 

Is the work clearly and accurately presented and does it cite the current literature?

Partly

If applicable, is the statistical analysis and its interpretation appropriate?

Partly

Are all the source data underlying the results available to ensure full reproducibility?

Partly

Is the study design appropriate and is the work technically sound?

Partly

Are the conclusions drawn adequately supported by the results?

Partly

Are sufficient details of methods and analysis provided to allow replication by others?

Partly

Reviewer Expertise:

NA

I confirm that I have read this submission and believe that I have an appropriate level of expertise to confirm that it is of an acceptable scientific standard.

 10.5256/f1000research.58830.r141642 Reviewer response for version 1 Kinabo Joyce 1 Referee https://orcid.org/0000-0003-1559-9143 Department of Human Nutrition and Consumer Sciences, Sokoine University of Agriculture, Morogoro, Tanzania

Competing interests: No competing interests were disclosed.

 25 7 2022 Copyright: © 2022 Kinabo J 2022 This is an open access peer review report distributed under the terms of the Creative Commons Attribution Licence, which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited. recommendation approve-with-reservations

Abstract:

Use of the word 'insufficient' should be consistent through out the article.

Conclusion does not state explicitly the factors associated with insufficient/sufficient UIC.

 Introduction:

Provide explanation as to why highland regions have highest prevalence of IDD considering that Tanzania has been implementing salt iodisation programmes for many years?

 Method:

Change 'Ante-natal Care Clinics' to 'Reproductive and Health Clinics' and use it consistently through out the document.

Explain adequately how sampling procedure was carried out.

Was the questionnaire pre-tested before translation?

"interviews were done by health professional" - which cadre?

Provide more information about the conditions for storage of the urinary samples in the field, transportation and storage at the laboratory before analysis and for how long were the samples stored before analysis. What distance from the field to the laboratory? This is important because iodine in urine is know to undergo changes after some time.

Was their any effort to ensure that women at various stages of pregnancy were well represented in the sample? It will also be useful to know of the sample size recruited, what was the stage of pregnancy for each participant? This information is missing.

Why Mbeya region? Provide justification.

 Results:

Area of residence seemed to have an influence on iodine status of pregnant women, it was expected to see a good description of the study areas in this manuscript. The information is missing. Could the authors strive to provide this information?

It will also be useful to know the stage of pregnancy (1st , 2nd, 3rd, trimester) of participants and if there were any association/interaction between area of residence, other factors and UIC.

Presentation of Tables should be improved; instead of brackets, authors could include columns for easy of reading.

What was the consumption of iodised salt? This could be one of the contributing factors to high levels of UIC observed in some areas of the study. But also consumption of foods especially vegetables of the brassica group that are known to have compounds that interact with iodine in the gut.

Clarification about consumption of processed meat.

Clearly stipulate factors that are associated with UIC in this section. Factors are not very clear.

 Discussion:

How were the identified factors influenced iodine status? These have not been adequately articulated in this manuscript.

Was the use of iodised salt investigated in the present study? What is the level of consumption among pregnant women? What is the availability and accessibility to iodised salt in the study areas?

Authors indicated some strengths of the study; large sample size; but also identified limitations? One of the limitations was that they did not investigate use of iodised salt in the study area. This is a serious omission considering that iodised salt contributes to iodine intake and therefore UIC at least for a large part of the population. Additionally, there was no control group, which again raises concern over the design of this study.

Conclusion: It is a repeat of statement of results and not conclusion. Authors should look at the objectives of the study and provide a concrete conclusion as to whether the objectives were attained and what are the implications to individual women and society in general.

Is the work clearly and accurately presented and does it cite the current literature?

Partly

If applicable, is the statistical analysis and its interpretation appropriate?

Partly

Are all the source data underlying the results available to ensure full reproducibility?

Partly

Is the study design appropriate and is the work technically sound?

Partly

Are the conclusions drawn adequately supported by the results?

Partly

Are sufficient details of methods and analysis provided to allow replication by others?

Partly

Reviewer Expertise:

NA

I confirm that I have read this submission and believe that I have an appropriate level of expertise to confirm that it is of an acceptable scientific standard, however I have significant reservations, as outlined above.

Masumo Ray Tanzania Food and Nutrition Centre, Tanzania

Competing interests: No competing interest

 5 8 2022

We appreciate the time and effort that you and the reviewer have dedicated to providing valuable feedback on our manuscript. Thank you very much.

Here is a point-by-point response to the reviewer's comments and concerns.

Abstract

Comment 1: The use  of the word 'insufficient' should be consistent throughout the article.

Response: The word ‘insufficient’ instead of ‘deficiency’ will be used throughout the manuscript.

Comment 2: Conclusion does not state explicitly the factors associated with insufficient/sufficient UIC

Response: The conclusion will carefully be rephrased and show factors associated with insufficient/sufficient UIC.

Introduction:

Comment 1: Provide an explanation as to why highland regions have highest prevalence of IDD considering that Tanzania has been implementing salt iodisation programs for many years?

Response: The explanation has been provided on page 11, discussion section, paragraph 4 ‘ However, the full implementation of USI remains a challenge in many sub–Saharan African countries including Tanzania, largely due to the lack of adequate enforcement and, the inadequate monitoring of small-scale salt producers who often do not comply with USI legislation’.

Method:

Comment 1: Change 'Ante-natal Care Clinics' to 'Reproductive and Health Clinics' and use it consistently throughout the document.

Response: The word 'Ante-natal Care Clinics' will replace 'Reproductive and Health Clinics' throughout the document.

Comment 2: Explain adequately how sampling procedure was carried out.

Response: We agree, and the sampling procedures will be explained in more detail.

Comment 3: Was the questionnaire pre-tested before translation?

Response: Thank you very much for pointing this out. The questionnaire was pre-tested after translation.

Comment 4: Interviews were done by health professional" - which cadre?

Response: Interviews were done by trained Nurse Midwives (we will specify the cadre of health professionals).

Comment 5: Provide more information about the conditions for storage of the urinary samples in the field, transportation, and storage at the laboratory before analysis and for how long were the samples stored before analysis. What distance from the field to the laboratory? This is important because iodine in urine is known to undergo changes after some time.

Response: We agree - more information on the conditions of the urinary samples stored in the field, transportation, and storage at the laboratory before analysis will be added.

Comment 6: Was their any effort to ensure that women at various stages of pregnancy were well represented in the sample? It will also be useful to know of the sample size recruited, what was the stage of pregnancy for each participant? This information is missing.

Response: Thank you very much for pointing this out. The inclusion and exclusion criteria of this study were mentioned on page 4, paragraph 2 ‘ Participants in their second trimester, a period during which fetal neurodevelopment is impacted by adequate maternal thyroid function, were included. To eliminate the effects of gestational age on thyroid hormone, participants beyond eight weeks of gestation were excluded.

Comment 7: Why Mbeya region? Provide justification.

Response: This study is the part of IMAN project piloted in Mbeya and supported by the UNICEF-Tanzania and the Ministry of Health - Tanzania aims to demonstrate how the country can sustainably scale up delivery of a comprehensive package of interventions to improve maternal nutrition (Interventions, Platforms, and Enabling Environment/System strengthening).

Results :

Comment 1: Area of residence seemed to have an influence on iodine status of pregnant women, it was expected to see a good description of the study areas in this manuscript. The information is missing. Could the authors strive to provide this information?

Response: We agree and more information on the area of residence will be added.

Comment 2: It will also be useful to know the stage of pregnancy (1st, 2nd, 3rd, trimester) of participants and if there were any association/interaction between area of residence, other factors and UIC.

Response: The study participants were only pregnant women in the second trimester and the reason behind this was to avoid the effects of gestational age on thyroid hormone.

Comment 3: Presentation of Tables should be improved; instead of brackets, authors could include columns for easy of reading.

Response: We agree with this comment and will remove the bracket.

Comment 4: What was the consumption of iodised salt? This could be one of the contributing factors to high levels of UIC observed in some areas of the study. But also consumption of foods especially vegetables of the brassica group that are known to have compounds that interact with iodine in the gut.

Response: Consumption of iodized salt and vegetables of the Brassica group will be the limitation of the study.

Comment 5: Clarification about consumption of processed meat.

Response: Here we referred to meats that have been preserved by smoking, salting, curing, or addition of chemical preservatives. 

Comment 6: Clearly stipulate factors that are associated with UIC in this section. Factors are not very clear.

Response: We agree with this comment and will rephrase the paragraphs to stipulate clearly the factors associated with UIC.

Discussion:

Comment 1: How were the identified factors influenced iodine status? These have not been adequately articulated in this manuscript.

Response: We agree with this comment and will rephrase the discussion section to articulate clearly on factors that influence iodine status.

Comment 2: Was the use of iodised salt investigated in the present study? What is the level of consumption among pregnant women? What is the availability and accessibility to iodised salt in the study areas?

Response: We have noted these observations and, they will be part of the study's limitations.

Comment 3: Authors indicated some strengths of the study; large sample size; but also identified limitations? One of the limitations was that they did not investigate use of iodised salt in the study area. This is a serious omission considering that iodised salt contributes to iodine intake and therefore UIC at least for a large part of the population. Additionally, there was no control group, which again raises concern over the design of this study.

Response: Thank you so much for pointing this out. We will add more information that clarifies the limitation of our study design.

Comment 4: Conclusion: It is a repeat of statement of results and not a conclusion. Authors should look at the objectives of the study and provide a concrete conclusion as to whether the objectives were attained and what are the implications to individual women and society in general.

Response: We agree, and we will carefully rewrite the conclusion section based on the study objectives.

 Masumo Ray Tanzania Food and Nutrition Centre, Tanzania

Competing interests: None

 23 8 2022

We appreciate the time and effort that you and the reviewer have dedicated to providing valuable feedback on our manuscript. Thank you very much.

Here is a point-by-point response to the reviewers’ comments and concerns.

Abstract

Comment 1: Use of the word 'insufficient' should be consistent throughout the article.

Response: We agree with this comment and have incorporated your suggestion throughout the manuscript.

Comment 2: Conclusion does not state explicitly the factors associated with insufficient/sufficient UIC

Response: Thank you for pointing this out. We agree with this comment and have rewritten the conclusion as suggested.

Introduction:

Comment 1: Provide explanation as to why highland regions have highest prevalence of IDD considering that Tanzania has been implementing salt iodisation programmes for many years?

Response: Agree. We have revised the introduction section to clearly emphasize this point.

Method:

Comment 1: Change 'Ante-natal Care Clinics' to 'Reproductive and Health Clinics' and use it consistently throughout the document.

Response: We agree with this comment and have incorporated your suggestion throughout the manuscript.

Comment 2: Explain adequately how sampling procedure was carried out.

Response: We agree, and have rewritten the sampling procedures.

Comment 3: Was the questionnaire pre-tested before translation?

Response: Thank you very much for pointing this out. The questionnaire was pre-tested after translation.

Comment 4: Interviews were done by health professional" - which cadre?

Response: Interviews were done by a trained Nurse Midwife. We have incorporated this suggestion throughout the manuscript.

Comment 5: Provide more information about the conditions for storage of the urinary samples in the field, transportation and storage at the laboratory before analysis and for how long were the samples stored before analysis. What distance from the field to the laboratory? This is important because iodine in urine is known to undergo changes after some time.

Response: We agree with this comment and have added more information on the conditions of the urinary samples stored in the field, transportation, and storage at the laboratory before analysis.

Comment 6: Was their any effort to ensure that women at various stages of pregnancy were well represented in the sample? It will also be useful to know of the sample size recruited, what was the stage of pregnancy for each participant? This information is missing.

Response: Thank you very much for pointing this out. The inclusion criteria of this study clearly indicated that ‘ To eliminate the effects of gestational age on thyroid hormone, participants beyond eight weeks of gestation were excluded.

Comment 7: Why Mbeya region? Provide justification.

Response: Thank you for pointing this out and we have added the following statement in the manuscript for clarity. This study is part of the project on improving maternal and adolescent nutrition in Mbeya supported by UNICEF Tanzania and the Ministry of Health- Tanzania.

Results :

Comment 1: Area of residence seemed to have an influence on iodine status of pregnant women, it was expected to see a good description of the study areas in this manuscript. The information is missing. Could the authors strive to provide this information?

Response: Thank you for pointing this out and we have added a sub-section of the study site in the methods section.

Comment 2: It will also be useful to know the stage of pregnancy (1st, 2nd, 3rd, trimester) of participants and if there were any association/interaction between area of residence, other factors and UIC.

Response: You have raised an important point here. However, we believe that excluding participants of beyond eight weeks of gestation was more appropriate to avoid the effects of gestational age on thyroid hormone.

Comment 3: Presentation of Tables should be improved; instead of brackets, authors could include columns for easy of reading.

Response: We agree with this comment and have incorporated your suggestion.

Comment 4: What was the consumption of iodised salt? This could be one of the contributing factors to high levels of UIC observed in some areas of the study. But also consumption of foods especially vegetables of the brassica group that are known to have compounds that interact with iodine in the gut.

Response: Thank you for this suggestion. It would have been interesting to explore this aspect. However, in the case of our present study, the assessment of iodised salt and vegetables of the brassica group was out of scope and has been included in the limitation of the study.

Comment 5: Clarification about consumption of processed meat.

Response: Agree. Here we referred to processed meat as that which has been preserved by smoking, salting, curing, or addition of chemical preservatives. 

Comment 6: Clearly stipulate factors that are associated with UIC in this section. Factors are not very clear.

Response: We agree with this comment and have rewritten the result section to stipulate the factors associated with UIC.

Discussion:

Comment 1: How were the identified factors influenced iodine status? These have not been adequately articulated in this manuscript.

Response: We agree with this comment and have rewritten the discussion section to stipulate the identified factors that influence iodine status.

Comment 2: Was the use of iodised salt investigated in the present study? What is the level of consumption among pregnant women? What is the availability and accessibility to iodised salt in the study areas?

Response: You have raised an important point here. However, the use of iodised salt (availability and accessibility) was not investigated in this study, and we have included it as part of the study's limitations.

Comment 3: Authors indicated some strengths of the study; large sample size; but also identified limitations? One of the limitations was that they did not investigate use of iodised salt in the study area. This is a serious omission considering that iodised salt contributes to iodine intake and therefore UIC at least for a large part of the population. Additionally, there was no control group, which again raises concern over the design of this study.

Response: Thank you so much for pointing this out and, we highly appreciate the reviewers’ insightful and helpful comments on our manuscript. It would have been interesting to explore and we have included it in the study's limitations. We will add more information that clarifies the limitation of our study design.

Comment 4: Conclusion: It is a repeat of the statement of results and not conclusion. Authors should look at the objectives of the study and provide a concrete conclusion as to whether the objectives were attained and what are the implications to individual women and society in general.

Response: Thank you for pointing this out. We agree with this comment and have rewritten the conclusion as suggested.