A Pilot Study on Iodine in Soils of Greater Kabul and Nangarhar Provinces of Afghanistan

Environ Geochem Health (2009) 31:503–509 DOI 10.1007/s10653-008-9202-9

ORIGINAL PAPER

A pilot study on iodine in soils of Greater Kabul and Nangarhar provinces of Afghanistan

M. J. Watts · C. J. Mitchell

Received: 8 January 2008 / Accepted: 8 August 2008 / Published online: 16 September 2008 © Springer Science+Business Media B.V. 2008

Abstract A robust and rapid methodology for the Introduction determination of iodine by inductively coupled plasma mass spectrometry in environmental samples Approximately 1.9 billion people around the world is presented. Data were initially obtained for the have been estimated to be at risk of iodine deWciency validation of the analytical measurements, using 17 disorders (IDD) [Benoist et al. 2003; World Health commercially available soil reference materials. The Organisation (WHO) 2004]. These estimates were methodology was then tested on soil and water sam- based on WHO surveys in which urinary iodine con- ples collected in Afghanistan where iodine deWciency centrations were measured at less than 100 g/L. and its eVects are reportedly prevalent. Sample collec- Although IDD have been known about for thousands tions were conducted in Greater Kabul; the iodine in of years, the relationship between iodine in the agricultural soils was determined to be in the range of environment and the development of IDD was dem- 1.6–4.2 mg/kg and that in water drawn for drinking onstrated only in the 1930s. Despite concerted world- and irrigation was found to range from 9.9 to wide eVorts to alleviate IDD in the twentieth century, 22.7 g/L. Samples were also collected in a second the incidence of IDD increased by 32% between 1993 region, Nangarhar province, which is located to the and 2003 (WHO 2004). A daily intake of 150 g of east of Kabul, where goitres in the local population iodine per adult is recommended to prevent IDD had been reported. The iodine content in soils and (WHO 2000), and when the physiological require- water at this location was 0.5–1.9 mg/kg and 5.4– ments are not met (<100 g/day), a series of thyroid 9.4 g/L, respectively. The organic content of soils in functional and developmental abnormalities occur Kabul was found to be in the range of 1.9–4.2%; in (Hetzel 1989). Symptoms often occur as goitre, and Nangarhar, organic content ranged from 1.7 to 4.5%. IDD can severely inhibit the mental and physical All of the Afghan soils were slightly alkaline at pH development of children. The recommended daily 7.6–8.2. allowance (RDA) of 150 g per day accounts for the daily amount of hormonal iodine degradation (40– Keywords Environmental strategy · Goitre · 100 g/day) and provides an extra margin of safety to ICP-MS · Iodine deWciency meet the increased demands by natural goitrogens (Delange 1993). In Afghanistan, 500,000 babies are born each year M. J. Watts (&) · C. J. Mitchell with intellectual impairment caused by iodine deW- British Geological Survey, Keyworth, ciency during pregnancy. In countries where the goi- Nottingham NG12 5GG, UK e-mail: [email protected] tre rate is 10% or more of the entire population, as in 123 504 Environ Geochem Health (2009) 31:503–509

Afghanistan (48%), iodine deWciency is estimated to All iodine compounds are readily soluble and the be widespread. This can have the eVect of reducing weathering of rocks and soils, particularly in areas of the national IQ by as much as 10–15% (Stewart et al. high precipitation, results in the release of much of 2003). The impact of vitamin and mineral deWciency their iodine content. The weathering of rocks contrib- in the diet, particularly iron and iodine, on the adult utes little to soil iodine, with the exception of certain population of Afghanistan has been estimated to have high iodine-containing rocks, while the major source reduced the gross domestic product by up to 2% of iodine is from atmospheric deposition. Afghanistan (Micronutrient Initiative 2004). This has occurred is located in the continental interior; consequently, even while it is known that IDD are entirely prevent- any marine iodine circulating in the atmosphere is able through a dietary supplementation of iodine, unlikely to reach the study sites in Greater Kabul and often through the use of iodised salt or the addition of Nangarhar province. The recycling of iodine in a iodine to cooking oil. In Afghanistan, international landlocked country such as Afghanistan is heavily non-governmental organisations have initiated a num- inXuenced by the vegetative recycling processes as ber of programmes to promote the iodisation of salt, well as the soil–plant to atmosphere relationship although no coordinated national scheme has yet (Anke et al. 1995; Aston and Brazier 1979; Johnson been launched. Only 28% of Afghan households 2003a). Soils in Afghanistan tend to be sandy and low have access to iodised salt (ICCIDD 2007). Some in organic material, thereby limiting the ability to trap clinical assessments were conducted to better and retain iodine. Vegetables and animal feed grown understand the iodine status of the Afghan population in these soils tend to be low in iodine content and, (UNICEF_CDC 2002; Ahmad et al. 2005; Oberlin therefore, humans and animals reliant on these crops and Plantin-Carrenard 2007), but there is limited are at risk of developing IDD. overall knowledge, owing to the recent troubled past, This study provides a limited snapshot of the which in turn hinders the planning of any nutritional iodine status of agricultural soils and irrigation and programmes. Ahmad et al. (2005) surveyed 8- to drinking water in two regions of Afghanistan. The 11-year-old school children in Greater Kabul and samples were collected from Greater Kabul and Nan- found that 12% of the children had grade-I goitre, garhar provinces (Fig. 1) where IDD are known to be with a higher prevalence among rural children prevalent from earlier clinical studies (Ahmad et al. (16.5%) than among urban children (7.5%). These 2005; Oberlin and Plantin-Carrenard 2007), but also researchers (2005) also reported that their assessment where there is little information on the link between of the prevalence of sub-clinical iodine deWciency, as environmental iodine and the prevalence of IDD in determined by urinary iodine levels, revealed that the population. The samples were analysed using 68% of the children had mild to severe iodine deW- newly developed methodology involving the use of ciency, with a 75% prevalence among rural children inductively coupled plasma mass spectrometry (ICP- and a 62% prevalence among urban children. MS) that resulted in fast, sensitive and accurate Soils generally contain 0.01–6 mg/kg iodine, measurements, following extraction (soils)/or spiking although in some coastal areas up to 80 mg/kg are (water) of the samples with an alkaline reagent (Watts recorded (Kebatas-Pendias and Pendias 1984). The 2001, 2002, Johnson et al. 2002). highest levels of iodine are found in soils rich in organic matter. Important factors that determine the iodine content of soils are the proximity to the marine Study sites environment and the organic content, which inXu- ences the Wxation of iodine in soil. Johnson (2003a) Sampling sites were selected in cultivated Welds man- compiled literature citations of iodine soil concentra- aged by diVerent farmers. Soil samples were collected tions from more than 2000 papers and classiWed from these plots, and water was collected from streams mean values based on the texture of soil: peat or wells that supplied drinking or irrigation water in (7.0 mg/kg) > clay (4.3 mg/kg) > silt (3.0 mg/kg) > Greater Kabul and in villages close to Jalalabad in Nan- sand (2.2 mg/kg). garhar province, located 120 km east of Kabul. The Iodine minerals include iodides of metals (CuI), Afghan sites were selected based on their direct inter- polyhalides, iodates [Cu(OH)(IO3)] and periodates. action with the source–pathway–receptor, via crop 123 Environ Geochem Health (2009) 31:503–509 505

Fig. 1 Map of Afghanistan, reproduced by courtesy of Henry Holbrook, British Geological Survey, 2008

production and drinking water. Soil samples were Materials and methods collected from a third site in the South West England for comparison. Details of the sites are as follows: Water samples (60 mL) were collected and Wltered on-site using a 0.45-m syringe Wlter. In addition, 1. Greater Kabul. Five water and eight soil samples approximately 0.5 kg of soil was collected with a spade were collected from each of the following sites: from the top 20-cm layer of soil. Samples were dried in Microrayan, Shawr-e-Naw, Qargha, Paghman, a drying oven at 40°C overnight and then sieved using and Istalif. The soils in Shawr-e-Naw and Qargha a nylon mesh to <2 mm. The soil samples were then are underlain by loess, those in Microrayan con- milled to <125 m using an agate ball mill. Loss on sist of alluvial deposition from the Kabul river ignition (LOI) was used for this study as an indicator of and those in Paghman and Istalif are underlain by organic matter content in mainly sandy soils. A posi- conglomerate/sand and gravel or gabbro and tive correlation of 0.915 was achieved for LOI against monzonite (AGS 2005a, b). total organic carbon (TOC) measured for all soils in 2. Nangarhar. Four water and Wve soil samples were this study. The LOI of the soils was determined on 1 g collected from each of three villages to the east of (dry weight) of soil and recorded as a percentage of Jalalabad: Walaity, Geokoski and Kakaram. Soils weight. Soil pH was determined on a 10-g (dry weight) in Walaity and Kakaram are underlain by loess, soil sample to which 10 ml of 0.01 M CaCl solution and those in Geokoski soils are underlain by allu- 2 had been added. The pH of the soil paste was measured vium comprising conglomerate/sand and gravel with a standard Orion pH meter. This method generally (AGS 2005a, b) gives a lower soil pH determination (0.5 U) than water- 3. South West England. Eight soil samples were based methods (Taylor et al. 2005). collected from agricultural soils, largely formed from accumulated plant debris and/or silt deposi- Iodine determination tion from the Tamar river Xoodplain. This study site was located only 20 km from the coastline The methodology for the measurement of iodine in and thus subject to deposition of windblown the water and soil samples was based on the work of iodine. 123 506 Environ Geochem Health (2009) 31:503–509

Table 1 Reference Reference material Iodine Standard n CertiWed material data for iodine measured deviation data (mg/kg) measurements (mg/kg)

GSS-1 (dark-brown soil) 1.8 0.2 7 1.9 § 0.4 GSS-2 (chestnut soil) 1.6 0.2 13 1.8 § 0.2 GSS-3 (yellow-brown soil) 1.3 0.1 7 1.3 § 0.4 GSS-4 (limy soil) 9 0.8 12 9.4 § 1.2 GSS-5 (yellow-red soil) 3.5 0.4 4 3.8 § 0.5 GSS-6 (yellow-red soil) 20.6 1.5 21 19.4 § 1.0 GSS-7 (laterite) 17.3 0.6 5 19.3 § 1.1 GSS-8 (loess) 1.1 0.1 3 1.6 § 0.5 MESS-1 (marine sediment) 18.4 1.8 4 40 SQ MESS-2 (marine sediment) 29.3 2.8 6 nd PACS-1 (marine sediment) 50.2 3.1 10 nd BCSS-1 (marine sediment) 42.7 4.1 11 100 SQ SARM-46 (stream sediment) 4.7 0.2 7 nd SARM-51 (stream sediment) 7.3 0.2 4 nd GBW 07301 (stream sediment) 0.17 0.01 7 nd GBW 08301 (river sediment) 1.3 0.1 6 nd nd, No data available; SQ, semiquantitative data GBW 07303 (stream sediment) 2.2 0.1 3 nd

Watts (2002). Water samples were spiked with 25% Results and discussion tetramethyl ammonium hydroxide (TMAH; Sigma Aldrich, Gillingham, Kent, UK) to result in a Wnal A robust and rapid methodology for the determination solution of 1% TMAH. For soils, 0.25 g (dry of iodine by ICP-MS in environmental samples is weight) of sample was weighed directly into a 15-ml presented, with validation achieved using a range of poly(tetraXuoroethene) bottle to which 5 ml of 5% sediment and soil reference materials (Table 1). The TMAH was added and shaken. Sample bottles, with methodology provided good comparisons with the lids loosened, were placed in a drying oven at 70°C certiWed values for iodine in eight reference materials for 3 h, with bottles shaken at 1.5 h. After 3 h of for which certiWed data were available; the other nine heating, 5 ml of deionised water was added and the reference materials without certiWed values provided bottles centrifuged at 2500 rpm for 20 min. The good precision (§10%) for replicate analyses and supernatant was removed from the top of the sample additional information for a range of other soil and solution and diluted to a Wnal matrix of 1% TMAH. sediment reference materials. All sample solutions were then analysed directly by Table 2 summarises the soil chemistry and iodine a ThermoElectron PQ ExCell ICP-MS instrument content of soil and water samples collected from (Thermo Fisher ScientiWc, Waltham, MA). An inter- each study site. Mean soil-iodine concentrations of nal standard of 10 g/L rhenium in 1% TMAH was 2.4 mg/kg in Greater Kabul and 0.9 mg/kg in Nangar- mixed with the sample solution via a t-piece to mon- har province are signiWcantly lower than the reported itor instrument stability. Drift correction standards worldwide mean of 2.6 mg/kg for inland regions at were employed, as were water-certiWed reference distances greater than 50 km from the coast. The con- materials NASS-4, IAPSO and SLRS-2 for quality trol site in the UK compared closely with the world control checks throughout the analytical run. The mean of 11.6 mg/kg for soils within a coastal zone of accuracy and precision of the method were evalu- less than 50 km from the sea (Johnson 2003a); Fuge ated using 17 soil- and sediment-certiWed reference and Johnson (1986) listed values between 4 and materials at a range of iodine concentrations 8 mg/kg for iodine in all types of soils from around (Table 1). the world. In Derbyshire (UK), goitre was historically 123 Environ Geochem Health (2009) 31:503–509 507

Table 2 Iodine content of Water (g/L) Soil (mg/kg) Organics (%LOI) pH sampled water and soils, plus soil pH and %LOI Range Mean Range Mean Range Mean Range Mean

Kabul 7.9–22.7 15.4 1.0–4.2 2.4 1.9–4.2 3.1 7.6–8.1 7.8 Nangarhar 5.4–9.4 7.6 0.5–1.9 0.9 1.7–4.5 2.7 7.8–8.2 8.0 LOI, Loss on ignition; nd, no data available Devon, UK nd nd 7.3–21.9 14.6 11.7–33.6 14.4 3.5–6.8 4.5 prevalent where soils contained 5.4 mg/kg and sur- the Himalayas with similar topography (<1.0 g/L), face waters 3.9 g/L (Fuge 1989). Soils investigated in where extreme endemic goitre is also commonplace IDD endemic areas in Sri Lanka were found to contain (Day and Powell-Jackson 1972). Iodine values for the iodine at a range of 0.1–10 mg/kg and a mean of Afghan waters were much higher than the UK values 3.1 mg/kg (Fordyce et al. 2000). Based on these values, reported by Fuge (1989) in a goitre-aVected region, the areas sampled in Afghanistan could be considered despite the reported prevalence of goitre in the sam- to be an iodine-poor environment, particularly in Nan- pling areas of this study (Ahmad et al. 2005). In gen- garhar province (0.3–1.9 mg/kg, mean 0.9 mg/kg). eral, surface waters collected in Greater Kabul The pH of soils from Afghanistan were slightly contained higher levels of iodine (7.9–22.7 g/L) than alkaline at pH 7.6–8.2 compared with a pH of 3.5–6.8 those in Nangarhar province (5.4–9.4 g/L), suggest- in the soils from the South West of England. The ing a greater natural input from the water catchment alkaline nature of the Afghan soils is likely to have and possibly urban contamination of surface water in contributed to their low iodine values, as evidence by central Kabul compared to outlying areas and Nan- the more acidic and more iodine-rich soils sampled in garhar province. The latter was particularly noticeable South West England. This factor is likely to eVect in the Kabul river, which is commonly used for the iodine absorption onto Fe and Al oxyhydroxides and dumping of household waste and sewage. When clay minerals, which is greatest under acidic condi- water and soil samples were collected from the same tions and poor in neutral and alkaline soils, which are location, soil and water iodine values exhibited a pos- largely texturally coarse materials in any case itive correlation of 0.865 (P < 0.01). Gbadebo and (Fordyce et al. 2003). However, even with a positive Oyesanya (2005) reported a lower correlation of correlation of 0.737 (Table 3) between soil pH and soil 0.756 (P < 0.05) between the water and soil iodine iodine values at a probability level of 0.05 (P < 0.01), values in Nigeria. Fuge (1989) found no correlation soil pH was not as strong an indicator of iodine mobil- between water and soil values in the UK and sug- ity as the organic matter content of the soils (0.942, gested that water-iodine data could not be reliable in P < 0.01), which inXuences the Wxation of iodine in identifying goitrous areas. soils. The organic matter content of Afghan soils was There is no accepted threshold Wgure in the litera- signiWcantly lower, 1.7–4.5%, than the iodine/organic ture for deWning whether an environment is iodine matter-rich UK soils (11.7–33.6%), which were deWcient. The Iodine status of an area is synonymous formed from accumulated plant debris and/or deposi- with the iodine status of the local population as deW- tion of silt on a Xoodplain. ned by medical parameters. Data for environmental The levels of iodine in Afghan surface waters were studies can be compared against samples from other much higher at 5.4–22.7 g/L than reported values in IDD regions. For example, Afghan soils are lower in iodine content (<2.4 mg/kg) than soils in other Table 3 Correlation between iodine concentration and soil IDD-aZicted regions, such as areas reported in a Sri parameters Lankan study by Fordyce et al. (2000), thereby sup- Parameter Iodine soil Iodine water porting the limited clinical assessments that indicate the prevalence of IDD in Afghanistan. However, Iodine water 0.865 – Johnson et al. (2002) suggested that soil iodine is a Soil organics 0.942 0.217* poor indicator of environmental status owing to the Soil pH 0.737 0.070* high degree of variability of soil properties locally, P < 0.01, *P >0.01 whereas water was found to be a good indication of a 123 508 Environ Geochem Health (2009) 31:503–509 region at risk of IDD in Morocco. However, Tan preparation of the Afghan soil samples and assistance in (1989) found a negative correlation between the preva- obtaining samples from Nangarhar province. This work was undertaken during a UK Department for International Develop- lence of IDD in China and iodine in drinking water. ment funded project for the ‘Institutional Strengthening of the The Afghan study does not follow the same pattern as Afghanistan Geological Survey’. Thanks also are extended to that reported in the Moroccan study because the iodine Mr. Henry Holbrook for the preparation of the map of Afghani- levels in Afghan waters were greater than 5 g/L. The stan. This work is published with the permission of the Director of the British Geological Survey. soil data did suggest that the study areas were of an iodine-poor environment, which does comply with the reported prevalence of grade-I goitre (12%) and sub- References clinical iodine deWciency (68%) in school children living in Greater Kabul (Ahmad et al. 2005). Abrahams, P. W. (2006). Soil geography and human disease: A critical review of the importance of medical cartography. Progress in Physical Geography, 30, 490–512. Conclusion AGS (Afghanistan Geological Survey) (2005a). Geological Map of Quadrangle 3468, Chak Wardak-Syahgerd (509) The conXicting Wndings of this and other studies and Kabul (510) Quadrangles, Afghanistan, AGS Open suggest that the complex factors that control iodine File report (509/510)–1107-A. Available at: http://afghanistan.cr.usgs.gov. Wxation in soils, such as the low organic matter con- AGS (2005b). Geologic Map of Quadrangles 3470 and 3370, tent and alkaline nature of Afghan soils, needs further Jalal-abad (511), Chaghasaray (512) and Jaji-maydan (517) investigation to understand the bioavailability or quadrangles, Afghanistan, AGS Open File report (511/512/ mobility of iodine in soils used for crop production in 517)–1108-A. Available at: http://afghanistan.cr.usgs.gov. Ahmad, A., Parvez, I., & Ud-Din, Z. (2005). Iodine status in Afghanistan. Stewart et al. (2003) also raised ques- children aged 8–11 years in Kabul, Afghanistan. Pakistan tions over the direct correlation of IDD prevalence Journal Medical Research, 44, 105–110. with straightforward total iodine data. There is a need Anke, M., Groppel, B., Muller, M., Scholz, E., & Kramer, K. to survey water and agricultural soils in Afghanistan (1995). The iodine supply of humans depending on site, food oVer and water supply. Fresenius Journal Analytical for total iodine content in more detail, including the Chemistry, 352, 97–101. bioavailable fraction and organic compounds, in order Aston, S. R., & Brazier, P. H. (1979). Endemic goitre, the to better understand the factors that inXuence the factors controlling iodine deWciency in soils. Science of prevalence of IDD. Total Environment, 11, 99–104. Since soils are implicated both directly and indi- Benoist, B., Andersson, M., Takkouche, B., & Egli, I. (2003). Prevalence of iodine deWciency worldwide. Lancet, 362, rectly in the health of humans, geochemical mapping 1859–1860. can stimulate investigations into the causation of Day, T. K., & Powell-Jackson, P. R. (1972). Fluoride, water diseases and, importantly, be instructive in the plan- hardness and endemic goitre. Lancet, 1, 1135–1138. Delange, F. (1993). The disorders induced by iodine deWciency. ning and management of public health responses Thyroid, 4, 107–128. (Abrahams 2006). Current strategies to fortify staple Fordyce, F. M., Johnson, C. C., Navarante, U. R. B., Appleton, foods, such as salt iodisation or direct supplementation J. D., & Dissanayake, C. B. (2000). Selenium and iodine in of iodine, are often patchy and costly. Much of the soil, rice and drinking water in relation to endemic goitre in Sri Lanka. Science of Total Environment, 263, 127–141. fresh produce on sale in Kabul and Nangarhar is Fordyce, F. M., Stewart, A. G., Ge, X., Jiang, J. -Y. & Cave, M. sourced locally, particularly in rural areas. Improved (2003). Environmental controls in IDD: A case study in the availability of environmental data should help in the Xinjiang Province of China, British Geological Survey alleviation of IDD in Afghanistan by stimulating a bet- Technical Report, CR/01/045N. ter employment of environmental strategies, such as Fuge, R. (1989). Iodine in waters: Possible links with endemic goitre. Applied Geochemistry, 4, 203–208. crop biofortiWcation (Xiao-E et al. 2007), iodisation of Fuge, R., & Johnson, C. C. (1986). The geochemistry of iodine— drinking and irrigation water (Maberly et al. 1981; a review. Environmental Geochemistry Health, 8, 31–54. Lim et al. 2006) and simple changes to agricultural Gbadebo, A. M., & Oyesanya, T. M. (2005). Assessment of iodine deWciency and goitre incidence in parts of Yewa practices, such as organic enrichment and ploughing area of Ogun state, southwestern Nigeria. Environmental methods (Aston and Brazier 1979; Johnson 2003b). Geochemistry Health, 27, 491–499. Hetzel, B. S. (1989). The story of iodine deWciency: An interna- Acknowledgements The author wishes to thank the staV of tional challenge in nutrition. New York: Oxford University the laboratories of the Afghanistan Geological Survey for the Press. 123 Environ Geochem Health (2009) 31:503–509 509

ICCIDD (International Council for Control of Iodine DeW- Tan, J. (Ed.). (1989). The atlas of endemic diseases and their ciency Disorders) (2007). 2007 UNICEF report suggests environments in the People's Republic of China. Beijing: global progress against iodine deWciency is slowing. IDD Science Press Newsletter, 23, 10–11. Taylor, H., Vickers, B. P., & Moss-Hayes, V. (2005). Validation Johnson, C. C. (2003a). Database of the iodine content of soils of the procedure for the determination of soil pH, BGS populated with data from published literature. British Internal Report Number IR/05/080. Geological Survey Commissioned Report, CR/03/004N. UNICEF_CDC, 2002, Summary of Nutrition and Health Survey Johnson, C. C. (2003b). The geochemistry of iodine and its Badghis Province Afghanistan. Available at: www.aims. application to environmental strategies for reducing risks org.af/services/seectoral/health/unicef/nutrition_and_health _ from iodine deWciency disorders. British Geological survey_badghis pdf. Accessed 4 Oct, 2007 Survey Commissioned Report, CR/03/057N. Watts, M. J. (2001). Analysis of iodine and selenium in human Johnson, C. C., Strutt, M. H., Hmeurras, M., & Mounir, M. breast milk and infant formula. PhD thesis. University of (2002). Iodine in the environment of the High Atlas Surrey, Department of Chemistry, Guildford, UK. Mountains, Morocco. British Geological Survey Commis- Watts, M. J. (2002). Determination of iodine in geological and sioned Report, CR/02/196N. biological samples by ICP-MS. Paper presented at the 11th Kebatas-Pendias, A., & Pendias, H. (1984). Trace elements in Biannual National Atomic Spectroscopy Symposium, soils and plants. Boca Raton, Florida, USA: CRC Press. Loughborough University, UK. Lim, K. K., Rohana, D. S., Zawiah, A., & Malaimoon, W. M. WHO (World Health Organisation) (2000). Disease eradication/ (2006). An evaluation of the eVectiveness of a water iodi- elimination goals: Iodine deWciency disorders. Available nator system to supply Malaysia. Tropical Biomedicine, at: http://www.who.int/nut/malnutrition_worldwide.htm, 23, 172–178. http://www.who.int/aboutwho/en/disease_er.htm. Access- Maberly, G. F., Eastman, C. J., & Carcoron, J. M. (1981). EVect ed: 4 October 2007 of iodination of a village water supply on goitre size and World Health Organisation (2004). Iodine status worldwide: thyroid function. Lancet, 2, 1270–1272. WHO global database on iodine deWciency. In: B. Benoist, Micronutrient Initiative. 2004. Vitamin and mineral deWciency: M. Andersson, B. Takkouche, & H. Allen (Eds.), Depart- A damage assessment report for Afghanistan. Available at: ment of Nutrition for Health and Development, World http://www.micronutrient.org. Accessed: 1 Oct 2007 Health Organisation, Geneva. Oberlin, O., & Plantin-Carrenard, E. (2007). Goitre and iodine Xiao-E, Y., Wen-Rong, C., & Ying, F. (2007). Improving deWciency in Afghanistan: A case-control study. British human micronutrient nutrition through biofortiWcation in Journal of Nutrition, 95, 196–203. the soil-plant system: China as a case study. Environmental Stewart, A. G., Carter, J., Parker, A., & Alloway, B. J. (2003). Geochemistry Heath, 29, 413–428. The illusion of environmental iodine deWciency. Environ- mental Geochemistry Health, 25, 165–170.

123