点击显示 收起
1 From the Department of Endocrinology and Medicine, Aalborg Hospital, Aarhus University Hospital, Denmark (SA and PL); the Departments of Medicine, (SA and KK) and Surgery (BH), Queen Ingrids Hospital, Nuuk, Greenland; and the Research Centre for Prevention and Health, Copenhagen County, Denmark (TJ).
2 Supported by grants from Greenland Homerule, the Aalborg City Christmas Lottery, the Obel Family Foundation, the Northern Jutland Research Foundation, the Danish Hospital Foundation for Medical Research, and the Region of Copenhagen, the Faroe Islands, and Greenland. 3 Reprints not available. Address correspondence to S Andersen, Department of Endocrinology and Medicine, Aalborg Hospital, Århus University Hospital, Reberbansgade, DK-9000 Aalborg, Denmark. E-mail: stiga{at}dadlnet.dk.
ABSTRACT
Background: Iodine intake in Greenland has been hypothesized to exceed 10 times the recommended amount. The transition from a traditional Arctic society may change the iodine intake, but no field studies have been performed.
Objective: We aimed to ascertain iodine intakes, factors affecting iodine intake in circumpolar populations, and the usefulness of urinary iodine excretion as a biomarker for validation of Inuit food-frequency questionnaires.
Design: Data were collected in a cohort study of 4 Greenland population groups: Inuit living in the capital city, the major town, and settlements in East Greenland and non-Inuit. Supplement use and lifestyle factors were evaluated with questionnaires, and dietary habits were ascertained with a food-frequency questionnaire. Iodine was measured in spot urine samples.
Results: One percent of the population of Greenland was invited, and the participation rate was 95%. Less than 5% of Inuit but 55% of non-Inuit had urinary iodine excretion < 50 µg/24 h. Median urinary iodine excretion declined with the degree of decrease in the traditional lifestyle: it was 198, 195, 147, and 58 µg/24 h among Inuit in settlements, town, and city and in non-Inuit, respectively (P < 0.001). Participants were divided into diet groups calculated from Inuit food frequency. Iodine excretion decreased with increasing intake of imported foods (P < 0.001). In regression models, type of diet and the subject's lifestyle, sex, weight, ethnicity, and intake of iodine-containing supplements affected urinary iodine excretion.
Conclusions: Circumpolar non-Inuit are at risk of iodine deficiency. Departure from the traditional Inuit diet lowers iodine intake, which should be monitored in Arctic societies. Urinary iodine excretion may be a useful biomarker of traditional Inuit food frequency.
Key Words: Diet • food-frequency questionnaire • biomarker • iodine • Inuit • lifestyle changes • Westernization • Arctic • Greenland • cohort study
INTRODUCTION
Iodine intake from the natural diet is low in most populations studied, and iodine deficiency disorders have been reported on all continents. Iodine deficiency may cause a wide spectrum of diseases (1, 2), but it can be prevented by iodine supplementation (3). A high iodine intake, however, also may cause disease (4, 5). Accordingly, the World Health Organization, the United Nations Children's Fund, and the International Council for Control of Iodine Deficiency Disorders have recommended daily iodine intakes of 150 µg for adults and 200 µg for pregnant and lactating women and monitoring of the iodine intake of all populations (6).
The circumpolar Inuit make up one of the few populations considered to have a high iodine intake (7, 8), which is due to a high content of marine food items in the traditional Inuit diet (9-11). The suggestion of excessive iodine intake among Inuit (8) was, however, based on assumptions because no field study was conducted.
We previously measured the iodine content of traditional Inuit food items (12) and found it not to be excessive but clearly higher than that of terrestrial and imported food items. Thus, it may be hypothesized that urinary iodine excretion would be a useful biomarker of Inuit dietary habits. This may prove valuable because of limitations of previously used biomarkers (13).
Danish authorities kept Greenland inaccessible until around 1960. The subsequent transition of Greenlandic society has occurred at different paces in different parts of the country (14), and settlements and towns present today at different degrees of Westernization. In parallel, the Inuit diet has decreased in importance, and the intake of imported food items with a low iodine content (15) has increased (14, 16, 17).
We studied dietary habits and the intakes of both imported foods and Inuit foods as evaluated by a questionnaire and by measuring urinary iodine excretion in subjects living in the city of Nuuk, which is in West Greenland, and in the rural Ammassalik district, which is in East Greenland. We evaluated the effect of a number of factors on iodine intake as estimated by urinary iodine excretion and studied iodine excretion as a possible biomarker of the frequency of consumption of traditional Inuit foods.
SUBJECTS AND METHODS
Area of investigation
Nuuk (64.15N 51.35W) in West Greenland is the capital of Greenland; it has 13 000 inhabitants, of whom 75% are Inuit and 25% are non-Inuit (ie, white). Nuuk was established as a trading post under the Danish crown in 1728 and is now a modern city with access to a wide variety of food items, including fast food, Italian and Thai food, and carry-out food, that are supplementary to the traditional Greenlandic food items. In addition, a wide variety of food items imported from Denmark are available in stores.
The Ammassalik district (65.35N 38.00W) of East Greenland was isolated until 1884 and is still difficult to reach by sea because of pack ice from the northern icecap. It is sparsely populated; there are 2943 inhabitants (93% Inuit) in an area of 243 000 km2. Tasiilaq is the main town of the Ammassalik district, which also includes 7 small settlements. Tasiilaq has one store with a limited selection and 5 small shops. Each of the settlements has one store with a limited selection that depends on access by sea and air. The capital city, town, and settlements are all situated on fjords, which provide access to the sea.
Subjects
Participants were 50–69-y-old men and women, Greenlanders (all Inuit) and non-Greenlanders (all white). Subjects in this age range were selected to obtain a valid representation of the different steps in the transition of Greenlandic society because that group is more homogeneous with respective to changes in lifestyle (13) and tends to have a higher participation rate than would younger subjects. We included persons who were selected and confirmed to be living at the address registered with the National Civil Registration System. The places selected for investigation were the city of Nuuk, the town of Tasiilaq, and the settlements of Tiniteqilaaq, Sermiligaaq, Kulusuk, and Kuummiut in the Ammassalik district (Table 1). For practical reasons, settlements with 15 inhabitants in the selected age group were not included. In Nuuk, names and addresses were obtained from the hospital registration system. A random sample of 480 (25% of the total population aged 50–69 y) was selected. The hospital registration system had not been regularly updated, and, for the investigation in the Ammassalik district, names and addresses were obtained from the National Civil Registration System in which every person living in Denmark, the Faroe Islands, and Greenland is registered. A Greenlander (Inuit) was defined as a person who was born in Greenland and both of whose parents were born in Greenland.
View this table:
TABLE 1. Participant selection and inclusion
Ethical approval by the Commission for Scientific Research in Greenland was obtained before the beginning of this study. All subjects gave informed written consent in Danish or Greenlandic by participant choice.
Investigational procedures
A letter of invitation was delivered to each subject by the local hospital porter or the nursing station attendant. Nonresponders were invited 3 times. The investigation took place at the local hospital or nursing station or, by request, at home visits. A physical examination performed by one of the investigating doctors (SA, PL, or BH) included height without shoes, weight in indoor clothing, and any major disabilities. Participants were interviewed by an interpreter or by one of the investigational physicians, who completed the questionnaire in either Danish or Greenlandic, as appropriate for the subject. Information on age and sex was obtained from the National Civil Registration System. Information on lifestyle patterns and dietary habits was obtained by using questionnaires. Questions were asked as written in the questionnaires. The same interpreter was used in Nuuk, Tasiilaq, and all settlements.
Dietary habits
An interview-based food-frequency questionnaire (FFQ) was used to assess the dietary habits. It included 14 food items, of which 7 were traditional Inuit items (ie, seal, whale, wild fowl, fish, reindeer, musk ox, and hare), and 7 were imported items (ie, cooked meals, potatoes, vegetables, butter, cheese, eggs, and fresh fruit). For each food item, 6 frequency categories were given, ranging from never to daily. Each item was given a frequency score calculated as the average number of days per month it was ingested (13): daily intake = 30.4, 4–6 times/wk = 21.7, 1–3 times/wk = 8.7, 2–3 times/mo = 2.5, 1 time/mo = 1, and never = 0 d/mo. Inuit food items were scored positively, and imported food items were scored negatively. Frequency scores for all food items consumed by each participant were added, and the participants were placed into 1 of 5 diet groups according to a scale on which a score of 100% represented a totally Inuit diet and a score of 0% represented a diet completely made up of imported food. Diet group 1 had Inuit food item scores of >80%, diet group 2 had scores of 60–80%, diet group 3 had scores of 40–60%, diet group 4 had scores of 20–40%, and diet group 5 had scores of <20%.
Dietary habits were validated from 2 cross-check questions: 1) "How many days per week is your main meal from Greenlandic food items?" and 2) "How often do you eat Greenlandic food items at different times of the year?" The same 6 frequency categories were used for each season.
Dairy intake was calculated by the average number of days per month that milk, butter, or cheese was ingested (13), and scores from the 3 dairy products were summed into dairy unit scores. A score of 100% indicated daily intake of all 3 dairy products, a score of 66% indicated an average intake of 20 times/mo, and a score of 33% indicated an average intake of 10 times/mo. Participants were split into 3 groups on the basis of their scores: dairy group 1 had scores of >66%, dairy group 2 had scores of 33–66%, and dairy group 3 had scores of <33%.
The intake of iodine-containing vitamins or mineral supplements was evaluated by asking the frequency of intake. Supplements were presented to one of the investigational doctors for evaluation of iodine content. If no supplement was presented, the iodine content was evaluated on the basis of the interview. From the combined information, participants were divided into 2 groups by whether they took iodine supplements or not. Salt is used for cooking in Greenland, but salt was not iodized at the time of the study.
Urine collection and analysis
At the visit, a spot urine sample was collected from all 535 participants. A subgroup of 36 participants, representing 25% of the Inuit participating in the investigation in Nuuk, was randomly selected for a subsequent 24-h urine collection. All participants were visited at home between 1200 and 1600 by one of the investigational doctors (SA). First, a spot urine sample was collected, and then a 24-h urine collection was initiated after careful instruction. The final urine collection was performed during a second home visit exactly 24 h later. Samples were stored at –20 °C until they were analyzed. One 24-h urine sample was omitted from the calculations because of incomplete collection. The iodine concentrations in the spot urine samples and the 24-h urine samples did not differ significantly (P > 0.1); the median and interquartile range was 75 and 90 µg/L and 75 and 100 µg/L in the spot and 24-h urine samples, respectively, and the difference was 0 and 70 µg/L.
Iodine was determined by using the Sandell-Kolthoff reaction modified according to Wilson and van Zyl (18) as described previously (19, 20). The principle is evaporation and alkaline ashing of the sample, followed by resuspension and measurement of iodine by using spectrophotometric detection of the catalytic role or iodine in the reduction of ceric ammonium sulfate in the presence of arsenious acid. To measure the iodine content, a 1.5-mL sample was used, which gave an analytical sensitivity of 2.0 µg/L. The intraassay CV was 1.5% (>15 µg/L). Recovery of added iodine was >95%. Urinary creatinine was determined by a kinetic Jaffé method (21). The 24-h urinary creatinine excretion measured in Inuit was used to calculate an age-, sex-, and ethnicity-adjusted 24-h urinary iodine excretion in Inuit according to recommendations (22-24). Creatinine excretion in 50–59- and 60–69-y-old Inuit men and women was 1230 and 1112 mg/24 h and 747 and 641 mg/24 h, respectively. In non-Inuit, this same estimation was based on creatinine excretion in an age- and sex-matched group of white Danes (1445 and 1252 mg/24 h in men and 989 and 871 mg/24 h in women) (25). Thus, urinary iodine excretion was expressed in µg/L, as a ratio of iodine (µg) to creatinine (g), and as estimated 24-h urinary iodine excretion by adjustment of iodine:creatinine for age, sex, and ethnicity (22-24).
Statistical analysis
Results are given as medians, with 25th and 75th percentiles. Urinary iodine excretion in groups was compared by using nonparametric statistics: Mann-Whitney U test for comparison of 2 groups, Kruskal-Wallis test for comparing several groups, and Kendall's tau for the relation between groups. Wilcoxon's signed rank test was used to test for difference between iodine in spot urine and 24-h urine samples. For multiple linear regression analysis, urinary iodine excretion data were transformed [ln(x)] because the distribution was positively skewed. Transformation also gave homogeneity of variances [Bartlett's test: P = 0.8 (diet group), P = 0.6 (participant group)]. Linear regression models were used with transformed estimated 24-h iodine excretion entered as the dependent variable. Independent variables entered were: diet group, participant group, sex, weight, ethnicity, and iodine from supplements. The independent variables included in the model explain 37% of the variation in iodine excretion (R2 = 0.37). Random selection of participants in Nuuk was performed by using MEDSTAT software (version 2.12; Astra, Albertslund, Denmark). Data were processed and analyzed using Corel Quattro Pro 8 (version 8.0; Corel Corp, Ottawa, Canada) and SPSS (version 10.0; SPSS Inc, Chicago) software. A P value of < 0.05 was considered significant.
RESULTS
Study cohorts
One percent of the population of Greenland was invited. The participation rate was 95%, and there were minor regional differences (Table 1). The characteristics of the study cohorts are given in Table 2. Seven non-Inuit subjects had one parent born in Greenland, and 94 had neither parent born in Greenland. The non-Inuit subjects were skilled laborers from Denmark, and the group included significantly more men than women (P < 0.001, chi-square test). In addition, there were fewer non-Inuit than Inuit in the 60–69-y-old group, and the mean age of the non-Inuit was significantly (men, P = 0.001; women, P = 0.005) lower than that of the Inuit (Table 2) because some non-Inuit leave Greenland when they retire.
View this table:
TABLE 2. Characteristics of the 4 participant groups representing different steps in the westernization of Greenlandic society
The study cohorts represented different stages of Westernization as illustrated by hunting habits: 54% of Inuit men living in settlements, 24% of Inuit men living in the town, 8% of Inuit men living in the city, and 0% of non-Inuit men made their living by hunting.
Intake of iodine-containing supplements
Non-Inuit were more frequent users of iodine-containing vitamin or mineral preparations than were Inuit living in settlements (Table 2). The intake of iodine-containing supplements in Inuit increased with increasing urban residence (P < 0.001, Kruskal-Wallis test) in parallel with the decrease in traditional lifestyle.
Urinary iodine excretion
Urinary iodine excretion differed significantly (P < 0.001, Kruskal-Wallis test) between participant groups and decreased significantly in parallel with the decrease in the traditional lifestyle (P < 0.001; Kendall's : 0.3) (Table 2).
The use of iodine-containing supplements was an important contributor to iodine excretion among populations in Nuuk, where urinary iodine excretion differed significantly between supplement users and nonusers (non-Inuit: P < 0.001; Inuit: P = 0.001; Mann-Whitney U test). Supplement use did not, however, significantly affect iodine excretion among the population in Tasiilaq (P = 0.43) or the settlements (P = 0.27). Accordingly, the use of iodine-containing supplements may compensate for some of the difference in iodine excretion between participant groups (a shift in Kendall's from 0.33 to 0.30) (Table 2).
Urinary iodine excretion was < 50 µg/24 h in 55% of samples [43% were <50 µg/L, which indicated moderate iodine deficiency (6)] and < 25 µg/24 h in 20% of samples [9% were < 20 µg/L, which indicated severe iodine deficiency (6)] from non-Inuit participants not taking iodine-containing supplements (Figure 1), and only 10% of samples were > 100 µg/24 h [23% were > 100 µg/L, which indicated no iodine deficiency (6)]. No significant difference was seen between non-Inuit from East and West Greenland (P = 0.26).
FIGURE 1.. Distribution of estimated 24-h urinary iodine excretion in Inuit and in non-Inuit for participants not taking iodine-containing supplements (solid line) and for all participants (broken line). Dispersion (variance) differed between Inuit (interquartile range and SD: 203 and 333 µg/24 h; n = 429) and non-Inuit (61 and 85 µg/24 h; n = 93) (P < 0.001, Bartlett's test), as did the median (Inuit: 176 µg/24 h; non-Inuit: 58 µg/24 h; P < 0.001, Mann-Whitney U test). The use of iodine-containing supplements increased iodine excretion significantly among non-Inuit (P < 0.001, Mann-Whitney U test).
Inuit had a higher iodine excretion: < 5% of the samples were < 50 µg/24 h (Figure 1). There were 4% samples with excretion < 20 µg/L, 25% with excretion < 50 µg/L, and 50% with excretion > 100 µg/L, which suggest severe, moderate, and no iodine deficiency, respectively (6). The large dispersion of urinary iodine excretion in Inuit, irrespective of supplement use (Figure 1), was partly due to a difference in urinary iodine excretion between Inuit in Nuuk, Tasiilaq, and the settlements (P = 0.001, Kruskal-Wallis test) (Table 2).
Dietary habits
Inuit food frequency as ascertained from cross-check questions correlated well with the FFQ (Kendall's tau = 0.8, P < 0.001). A distinct difference in dietary habits was seen between the population cohorts (Figure 2). Among Inuit in settlements, 93% lived on mainly traditional Greenlandic food items (diet groups 1 and 2). Among Inuit in Tasiilaq and in Nuuk and among non-Inuit, these proportions were 87%, 58%, and 3%, respectively. These findings corresponded to the general difference in lifestyle (14), to the availability of imported food items, and to the frequency at which the main daily meal was made up of items from a person's own fishing or hunting; the frequency was reported to be 24%, 20%, and 6% among Inuit in settlements, Tasiilaq, and Nuuk and 1% among non-Inuit. No seasonal difference in the frequency of consumption of Inuit foods was reported.
FIGURE 2.. Dietary habits among the 4 participant groups were calculated from an interview-based food-frequency questionnaire. For each participant, frequency scores for 14 food items were calculated and added. Inuit food items scored positively, and imported food items scored negatively. Subjects were placed in a diet group on the basis of scores: diet group 1 had scores of >80%, diet group 2 had scores of 60–80%, diet group 3 had scores of 40–60%, diet group 4 had scores of 20–40%, and diet group 5 had scores of <20% for Inuit food items (n = 241, 110, 80, 60, and 41 in groups 1–5, respectively).
Diet and iodine excretion
Estimated 24-h urinary iodine excretion decreased with a decreasing intake of traditional Inuit foods (P < 0.001, Kendall's : –0.32) to the point of indicating iodine deficiency in diet groups 4 and 5 (Figure 3). The use of iodine-containing supplements increased urinary iodine excretion in the population by 1, 6, 5, 17, and 15 µg/24 h in diet groups 1 through 5 to totals of 204, 148, 104, 86, and 58 µg/24 h, respectively.
FIGURE 3.. Estimated 24-h iodine excretion by diet groups 1–5 in participants not taking iodine-containing supplements (median and 25th and 75th percentiles). Urinary iodine excretion fell with decreasing intake of Inuit food items (P < 0.001, Kendall's : –0.32; n = 482). Diet groups were computed from an interview-based food-frequency questionnaire: diet group 1 had scores of >80%, diet group 2 had scores of 60–80%, diet group 3 had scores of 40–60%, diet group 4 had scores of 20–40%, and diet group 5 had scores of <20% for Inuit food items (n = 241, 110, 80, 60, and 41 in groups 1–5, respectively).
Effect on iodine intake
The factors found to have a major effect on estimated 24-h urinary iodine excretion in the multivariate analysis were diet, intake of iodine-containing supplements, lifestyle, sex, and weight (Table 3). These factors explain 37% of the variation in iodine excretion (R2 = 0.37), a proportion that is high, considering the large spontaneous variation in iodine excretion (24, 26).
View this table:
TABLE 3. Factors affecting the estimated 24-h urinary iodine excretion in regression models1
Milk had no effect on urinary iodine excretion (P = 0.6, univariate analysis), but dairy products as a group contributed to the iodine intake (R2 = 0.016, P = 0.003; univariate analysis), although the effect was limited (1.6%) and not significant in the multivariate analysis (P = 0.36). Nevertheless, the intake was associated with urinary iodine excretion (P = 0.002, Kendall's : 0.103) (Figure 4). Seal, whale, and fish intakes were likely to contribute to the iodine intake individually, but they were not considered individually because they were included as part of the overall diet group. Weight was preferred over BMI because weight was a better predictor of urinary iodine excretion (univariate ßBMI = –0.07, P = 0.15; ßweight = –0.31, P < 0.001). Ethnicity could be an important influence on iodine intake (P < 0.001, univariate analysis; P = 0.047, multivariate analysis), but it was not included because of a high colinearity with lifestyle.
FIGURE 4.. Estimated 24-h iodine excretion by dairy groups in participants not taking iodine-containing supplements. Dairy groups were calculated from the reported frequency of intake of milk, butter, and cheese: dairy group 1 had scores of <33%, dairy group 2 had scores of 33–66%, and dairy group 3 had scores of >66% for dairy (n = 224, 159, and 114 in groups 1–3, respectively). Iodine excretion showed an increasing trend with increasing intake of dairy products (P = 0.002, Kendall's : 0.1), but the effect on iodine excretion was limited [R2 = 0.016, P = 0.003 (univariate analysis); P = 0.36 (multivariate analysis after correction for dietary habits, lifestyle, sex, and weight)].
If estimated 24-h urinary iodine excretion was replaced by iodine concentration in spot urine samples as a dependent variable, the degree of explanation of the variation in iodine excretion was only 12% (R2 = 0.12), and the contribution by weight disappeared (ß = 0.07, P = 0.15).
DISCUSSION
This is the first field study of iodine nutrition among an Inuit population. Rather than the exceedingly high iodine intake that was anticipated (7, 8), we found that urinary iodine excretion among Inuit living in Greenland indicated mere iodine sufficiency (1-3). The intake of traditional Inuit food items had decreased with increasing Westernization, which lowered the iodine intake among Inuit. In addition, the iodine intake differed significantly between population groups, and 77% of urine samples from non-Inuit living in Greenland had iodine concentrations in the range corresponding to iodine deficiency—ie, <100 µg/L (6). The intake of iodine-containing supplements was higher among this low iodine intake group. Finally, urinary iodine excretion may serve as a biomarker for validation of Inuit food-frequency questionnaires in Arctic societies.
Iodine intake is determined principally by the diet (27). Marine food items had a major influence on the iodine intake in some populations (28, 29), and Inuit had a very high intake of fish and marine mammals (9-11). Accordingly, the iodine intake among Inuit was hypothesized to be >10 times that recommended (8). However, the current study showed that the intake of traditional Inuit food items did not lead to the exceedingly high iodine intake expected but rather caused an iodine excretion of 200 µg/24 h, which reflected a high-normal iodine intake (6).
Inuit communities undergo profound changes in transitioning from their traditional hunting lifestyle to a modern, Westernized lifestyle (14, 17). We included 4 population groups at 4 different stages of this transition: rural Inuit in remote settlements in East Greenland, small-town Inuit in the main town on the east coast of Greenland, urban Inuit in the capital city, Nuuk, and non-Inuit, who had the most Westernized lifestyle. The changing lifestyle was associated with decreasing iodine intake, and it may reduce iodine intake in Inuit even to the point of iodine deficiency (6).
Traditional Inuit food items have high contents of contaminants (30-32). Dietary recommendations of a reduced intake of traditional Inuit food items in vulnerable subjects such as pregnant and lactating women (31, 32) are constrained by the sociocultural and nutritional benefits of these food items (30, 32). Nevertheless, the focus on contaminants in Inuit food items is likely to reduce the intake of traditional Inuit foods and thereby further lower iodine intake in groups that are particularly vulnerable to a low iodine intake.
This study included subjects aged 50–59 y to achieve a valid representation of the different steps in Westernization. The decreasing iodine intake seen with Westernization is of concern, particularly with respect to young women, who are likely to be more Westernized than are the older subjects included here. It is pertinent to include younger women in future studies because the iodine requirements increase during pregnancy and lactation (6), and iodine deficiency may have an effect on fetal brain development (1-3).
Dairy products are major sources of iodine intake in other populations (29, 33). It could be speculated that the decrease in iodine intake among Inuit caused by decreased intake of marine food items may be compensated for by an increased intake of iodine-containing dairy products from an imported diet. We found a trend suggesting an effect of dairy products on iodine excretion, but the contribution was limited and clearly not sufficient to obscure the decrease resulting from the reduced intake of Inuit food items.
In other populations, the use of iodine-containing supplements was found to be related more to self-perceived health than to actual iodine intake (34), and it was high even in groups that were not iodine deficient before taking supplements (29, 34). With a relatively high iodine intake among Inuit in Greenland, such a pattern could be of some concern. However, we found that the use of iodine-containing supplements significantly influenced iodine intake only in low-iodine-intake groups.
In multivariate regression models, diet, lifestyle, and iodine intake from supplements were significant determinants of iodine excretion. Together with body weight and sex, these factors may explain 37% of the variation in urinary iodine excretion in the current study, which is a large percentage, considering the high spontaneous variation in urinary iodine excretion (24, 26). Body weight had an effect on 24-h urinary iodine excretion calculated from the concentrations of iodine and creatinine in urine samples. This effect could result from the creatinine correction, because that was not stratified by body weight. When unadjusted urinary iodine concentration was used, the effect of body weight was no longer significant, although the trend was the same. The lack of significance could be due to the greater variance in crude urinary iodine concentration compared with that in the age- and sex-adjusted, creatinine-corrected 24-h iodine excretion (24). The effect of diet, lifestyle, and supplementation suggests that the groups at risk of iodine deficiency include persons who have a low intake of traditional Inuit food items, persons who do not take iodine-containing supplements, and persons who have a Westernized lifestyle. It remains necessary, however, to identify groups at risk of low iodine intake, and a dietary questionnaire is a compelling short-cut.
FFQs are widely used to obtain data on the intake of foods and nutrients because they are easy to distribute and collect, easy to understand, fast to fill in, and of low cost. Such features may be particularly appealing in the sparsely populated Arctic, where clinical investigations are expensive and may entail undesirable challenges. However, validation of a questionnaire is as important as those other characteristics, and it can be done in several ways (13, 35-37). We applied 2 cross-check questions and used urinary iodine excretion as a biomarker because we have previously found a relatively high content of iodine in Inuit food items (12) that, in Greenland, are mostly obtained from the sea. The measurement error for the cross-check questions was not independent of the FFQ, but that for urinary iodine excretion was. Even though questions about portion sizes were not included, the dietary classification in this study seemed valid. Thus, an FFQ may provide a convenient short-cut for identifying groups at risk of iodine deficiency in the Arctics, and urinary iodine excretion proved a useful biomarker of Inuit food frequency.
The validity of the current study was enhanced by 3 factors. First, the same interpreter was used in all areas and among all population groups. This ensured equal understanding and interpretation of the questionnaire between regions and between participant groups. Second, the participation rate was high, and, third, the selection of subjects aged 50–69 y was the basis for a valid representation of the different steps in the transition of Greenlandic society, because this group is more homogeneous with respect to changes in lifestyle than are younger groups (13). However, this cohort study has limitations in evaluating transition because this study represents the first data collection. Proof that differences between population groups in Greenland represent different steps in modernization will have to be obtained from subsequent data collections. Nevertheless, our data are in keeping with other indicators (14).
Urinary iodine excretion reflects dietary iodine intake in that 90% of ingested iodine is excreted in the urine (27). The iodine concentration in spot urine samples is the reference for delineating low iodine intake in population surveys (6). Some have suggested the use of iodine:creatinine (22-24) because that ratio reduces the variation in urinary iodine caused by dilution (22, 23). Applying this ratio introduces the variation of urinary creatinine excretion (38, 39), but that variation can be reduced by comparing homogeneus populations, ie, by stratification (23, 24). Such a step reduces the individual variation in iodine excretion (24), improves the ability of a single urinary iodine value to predict the true iodine excretion (24), and gives a more precise estimate of the actual iodine excretion in a person than does the crude urinary iodine concentration (23, 26). Moreover, it predicts thyroid enlargement and volume more accurately does than urinary iodine concentration (40), and it has been used in several population studies (1, 2, 26, 41). Finally, the higher R2 obtained when using estimated 24-h urinary iodine excretion in this study supports the use of estimated 24-h urinary iodine excretion. We used that value in figures, calculations, and the evaluation of the FFQ, whereas the crude urinary iodine excretion and iodine/g creatinine values were retained in Table 2 for comparison with other studies and with the World Health Organization and International Council for Control of Iodine Deficiency Disorder studies (6).
A single low urinary iodine value does not necessarily indicate iodine deficiency in a person (24). Still, the variance among non-Inuit was low, and an iodine excretion rate as low as that found in this study correlates with an increased risk of thyroid disease (1-3, 6).
The transition of Arctic societies to a more Westernized lifestyle alters iodine intake, and we recommend that iodine intake be monitored and that Arctic populations at risk of iodine deficiency be followed. We also suggest that future studies include whites and Westernized Inuit, and that an FFQ may guide future clinical investigations to groups at risk. Finally, monitoring is important as Greenland is included in the Danish salt iodization program (42).
ACKNOWLEDGMENTS
We gratefully acknowledge Karoline Berglund for her enthusiasm and thorough interviewing of the Inuit participants, and we thank Niels Henrik Bruun for his advice on the statistical analysis. In addition, we are grateful for invaluable support from the Lægeklinikken (Doctors' Clinic) in Nuuk, from Hans Christian Florian Sørensen and the staff at the hospital in Tasiilaq, and from the staffs at the nursing stations in Tiniteqilaaq, Sermiligaaq, Kuummiut, and Kulusuk.
SA was responsible for the conception and design of the study, raising funds, data collection, analysis and interpretation of data, and writing of the manuscript. BH was responsible for data collection and writing of the manuscript. KK was responsible for the study design, translations and explanations, and techniques to improve Inuit participation rate. TJ was responsible for study design and manuscript review. PL was responsible for design of the study, raising funds, data collection, analysis of data, interpretation of data, and writing of the manuscript. None of the authors had any personal or financial conflicts of interest.
REFERENCES