Changing Iodine Status and the Incidence of Thyroid Disease in Mainland China: A Prospective 20-Year Follow-Up Study

Abstract

Background:

We aimed to assess the long-term effects of the transition in iodine status on the incidence of thyroid disorders over 20 years of follow-up.

Methods:

The original prospective cohort study, started in 1999 (n = 3761), classified three regions in north China based on iodine status (insufficient iodine, more than adequate iodine, and excessive iodine, respectively) for 5 years. Subsequently, participants were followed for up to another 15 years to assess the long-term effects of shifts to adequate iodine on the incidence of thyroid disorders. Panshan transitioned from insufficient to adequate iodine, and Huanghua transitioned from excessive to more than adequate iodine. Both regions were compared with Zhangwu, in which iodine status changed from more than adequate to adequate iodine (from 214 to 167.2 μg/L). A cluster sampling method was used to select participants in the three regions. Participants completed questionnaires and underwent thyroid ultrasonography. Urinary iodine concentrations (UICs), serum thyroid hormone concentration, and thyroid antibodies were measured.

Results:

When the iodine status changed from insufficient to adequate (with the median UIC increasing from 88 to 141.9 μg/L), the incidence density of subclinical hyperthyroidism, positive thyroperoxidase antibody, positive thyroglobulin antibody (TgAb), and goiter decreased significantly (p < 0.05 for all). Additionally, the cumulative incidence of subclinical hypothyroidism was significantly lower compared with the region where the iodine status changed from being more than adequate to adequate (1.9% vs. 6.0%, p < 0.001). When the iodine status changed from excessive to more than adequate (median UIC from 634 to 266.7 μg/L), a significant decrease in the incidence density of subclinical hyperthyroidism, positive thyroid antibodies, positive TgAb, and goiter (p < 0.05 for all) were also found. However, an increase in thyroid nodule incidence density (17.26 vs. 28.25 per 1000 person-years, p < 0.001) was seen.

Conclusions:

The incidence of thyroid disorders (except for thyroid nodules) stabilized or decreased among adults in the three communities from year 5 to year 15 of follow-up. Appropriate iodine fortification is safe and effective over the long term. Restoring urinary iodine to appropriate levels reduces population risk for thyroid disorders.

Introduction

Thyroid disorders are common, with abnormal thyroid function or morphology occurring in up to 20% of individuals.¹ Thyroid hormone acts on all the nucleated cells of the human body and regulates growth and metabolism. Iodine is vital for the synthesis of thyroid hormone.² Therefore, iodine intake affects the spectrum and incidence of thyroid disorders. Starting in the 1990s, the International Council for the Control of Iodine Deficiency Disorders (now the Iodine Global Network), United Nations Children's Fund, and World Health Organization (WHO) have led global efforts to eliminate iodine deficiency disorders, using universal salt iodization (USI) as the primary strategy.³ It is estimated that this has led to a 75% reduction in the prevalence of iodine deficiency disorders worldwide.⁴ These changes will inevitably bring about changes in the types and incidence of thyroid disease.⁵

The association between iodine nutrition and thyroid disease has been studied previously, but most of these studies were cross-sectional or were cohort studies assessing the effects of iodine fortification or supplementation on a single disease entity.^6

–11 Compared with repeated cross-sectional studies, a prospective cohort design provides more clarity related to temporal sequence and causality. In addition, the impact of iodine status changes on thyroid disease is particularly important for the implementation of iodine supplementation policy, but evidence from long-term follow-up studies on the spectrum of thyroid disease is sparse.

In addition, the incidence of thyroid disorders following a reduction in iodine nutrition is poorly understood. Most previous studies were based on the observation of iodine supplementation in individuals with iodine deficiency; reports examining a reduction in excessive iodine nutrient intake have been lacking.^12
–14 Considering that urinary iodine concentration (UIC) in some countries has exceeded the optimal intake level, it is necessary to explore the effects of reducing iodine intake on population with thyroid disease risk.¹⁵

Our group completed five-year follow-up studies on iodine nutrition and thyroid disease (Iodine-induced thyroid disorders study) in three communities with different iodine intakes in 1999 and 2004.⁶ To assess the long-term effects of the transition in iodine status on the incidence of thyroid disorders, we have now completed the prospective 20-year follow-up of this cohort.

Materials and Methods

Study design and study population

As previously described, three communities with different iodine intake levels were selected for a long-term follow-up study in 1999.⁶ These three regions, Panshan, Zhangwu, and Huanghua, were classified as having insufficient iodine, more than adequate iodine, and excessive iodine, respectively, during 1999 to 2004. The inhabitants in Panshan traditionally consumed locally produced salt (iodine content <3.4 mg/kg) and thus had long-term mild iodine deficiency.⁶ Residents in Huanghua had excessive iodine intakes due to high drinking water iodine content (96 to 228 μg/L).⁶ After 2004, the government implemented various policies intended to optimize iodine nutrition, including the improvement of drinking water in the Huanghua area and the introduction of salt iodization in the Panshan area, which resulted in the reduction of UIC in Huanghua and increased UIC in Panshan. The WHO classification was utilized to assess the iodine status (Table 1).³

Table 1.

Diagnostic Criteria for Thyroid Disorders and Iodine Status

No.	Thyroid disorder	Diagnostic criteria
1	Overt hyperthyroidism	TSH <0.3 mIU/L; fT4 > 24.5 pmol/L; and/or fT3 > 6.3 pmol/L
2	Subclinical hyperthyroidism	TSH <0.3 mIU/L, fT3 and fT4 within the reference ranges
3	Overt hypothyroidism	TSH >4.8 mIU/L and fT4 < 10.3 pmol/L
4	Subclinical hypothyroidism	TSH >4.8 mIU/L and fT4 within the reference range
5	Positive TPOAb	TPOAb ≥50 IU/mL
6	Positive TgAb	TgAb ≥40 IU/mL
7	Positive thyroid antibodies	TPOAb ≥50 IU/mL or TgAb ≥40 IU/mL
8	Goiter	Thyroid volume >25.6 mL for men or >19.4 mL for women
9	Thyroid nodule	Nodules >5 mm in diameter
10	Insufficient iodine	UIC <100 μg/L
11	Adequate iodine	UIC = 100–199 μg/L
12	More than adequate iodine	UIC = 200–299 μg/L
13	Excessive iodine	UIC >299 μg/L

The reference range for TSH is 0.3–4.8 mIU/L; for fT4, 10.3–24.5 pmol/L; for fT3, 2.3–6.3 pmol/L; for TPOAb, 7–50 IU/mL; and for TgAb, 10–40 IU/mL.

fT3, free triiodothyronine; fT4, free thyroxine; TgAb, thyroglobulin antibody; TPOAb, thyroid peroxidase antibody; TSH, thyrotropin; UIC, urinary iodine concentration.

In the baseline study in 1999, a cluster sampling method was used to select participants in the three regions. We used a two-stage sampling method. In the first stage, three different iodine status regions were selected from north China, and in the second stage, a sample of individuals was randomly recruited from each of the selected regions.¹⁶ The inclusion criteria were as follows: aged 14 years or older and living in the community >10 years. Pregnant women and those who were using oral contraceptives were excluded. In Panshan, Zhangwu, and Huanghua, the age range of respondents was 14 to 73, 14 to 80, and 14 to 73 years in 1999, respectively. Each participant completed an oral questionnaire. Both palpation and B-mode ultrasonography of the thyroid were used to assess for goiter, and samples of urine and blood were collected from each participant after an overnight fast.

Original participants were followed up locally in 2004 and 2019. The same protocol was utilized in the 2019 follow-up study, which included 625 subjects in Panshan, 839 subjects in Zhangwu, and 634 subjects in Huanghua (Supplementary Fig. S1). Due to residential relocation and migrant work, attrition rates in Panshan, Zhangwu, and Huanghua were 35.2%, 35.2%, and 33.5%, respectively.

Research protocols were approved by the institutional review board (IRB) of the First Hospital of China Medical University (IRB [2019]122). All participants provided written informed consent after receiving a comprehensive explanation of the research protocols.

Laboratory methods

The same assay methods and assay kits for testing thyroid function were used for the baseline and follow-up studies. The instrumentation and personnel were the same at baseline and follow-up.

Serum levels of thyrotropin (TSH), thyroid peroxidase antibody (TPOAb), and thyroglobulin antibody (TgAb) were measured in all subjects. Serum levels of free thyroxine (fT4) and free triiodothyronine (fT3) were measured only if TSH was outside the population-based reference range. The reference ranges for TSH, fT4, fT3, TPOAb, TgAb, and thyrotropin receptor antibody were 0.3–4.8 mIU/L, 10.3–24.5 pmol/L, 2.3–6.3 pmol/L, 7–50 IU/mL, and 10–40 IU/mL, respectively, as previously described.^6,17 TSH, TPOAb, TgAb, fT4, and fT3 were measured by chemiluminescence immunoassay (Diagnostic Products Corporation).

The colorimetric ceric ion–arsenious acid ash method based on the Sandell–Kolthoff reaction was used to assess the urinary iodine excretion in all subjects at baseline and at the 2004 follow-up. UIC was determined using inductively coupled plasma mass spectrometry (ICP-MS) at follow-up in 2019. Previous studies have demonstrated that the ICP-MS methods and the Sandell–Kolthoff reaction are congruent, and the test results of these two methods are directly comparable.^18,19 Trained observers used the same equipment (Medison's model SA600 with 7.5-MHz linear transducers) to perform thyroid ultrasonography at both the baseline and follow-up studies. Further details of the laboratory tests can be found in the Supplementary Data.

Reference ranges

We established the reference range for serum TSH levels (0.3–4.8 mIU/L) based on the National Academy of Clinical Biochemistry guidelines, using the 2.5th to 97.5th percentile for log-transformed TSH levels of 2503 survey participants who had no thyroid antibodies, goiter, nodules, or personal or family history of thyroid disease.²⁰ Although the reference ranges of TPOAb and TgAb provided by the test kit manufacturer were 35 and 40 IU/mL, respectively, we determined that the cutoff values for TPOAb and TgAb in our population were 50 and 40 IU/mL, respectively, as previously reported.^6,17

The reference range for thyroid volume was established using the mean (+2 standard deviations [SDs]) thyroid volume of 392 subjects who did not have thyroid disorders, a family history of thyroid disorders, antibodies, or evidence of goiter or nodules on B-mode ultrasonography from a region with adequate iodine intake.^6,16 The diagnostic criteria for thyroid diseases and iodine status are listed in Table 1.^3,6,17 Additional endpoints were defined based on participants' use of levothyroxine, antithyroid drugs, or other thyroid treatments during the interval between evaluations in the study.

Statistical analyses

Characteristics of the study population were presented as number (proportion) for qualitative variables, and median (interquartile range) and mean (SDs) for quantitative variables, as appropriate. Histogram, normal probability plot, and Kolmogorov–Smirnov test were used to comprehensively evaluate normally distributed data. Differences in continuous variables were compared by using one-way analysis of variance or Kruskal–Wallis one-way analysis of variance. Bonferroni correction was applied to all multiple comparisons. For categorical variables, the chi-square test or Fisher's exact test was used for group comparisons. Logistic regression analysis was used to investigate the associations between transitions in iodine status and the development of thyroid disorders, with the Panshan area as the reference group. A multiple linear regression correction approach was used to calculate the incidence density of thyroid disorders adjusted by age.²¹ All analyses were conducted using SAS 9.4 (SAS Inst., Inc., Cary, NC, USA), and all reported p-values are two-sided, with p < 0.05 considered statistically significant.

Results

Of the original 3761 study participants in 1999, 1103 were recruited from Panshan, 1584 were recruited from Zhangwu, and 1074 were recruited from Huanghua. In 2019, 1306 participants had been lost to follow-up, 357 participants had died, and the remaining 2098 were assessed for outcomes (Supplementary Fig. S1). Characteristics of participants at baseline, the 5-year follow-up, and the 20-year follow-up are shown in Table 2. The median UIC increased from 88 to 141.9 μg/L among the schoolchildren in Panshan, decreased from 214 to 167.2 μg/L in Zhangwu, and decreased from 634 to 266.7 μg/L in Huanghua between 2004 and 2019. Therefore, Panshan was considered to have transitioned from insufficient to adequate iodine intake, Zhangwu was considered to have transitioned from more than adequate to adequate iodine intake, and Huanghua transitioned from excessive to more than adequate iodine over the study period. The median TSH levels were 2.29 mIU/L in Huanghua, 2.11 mIU/L in Zhangwu, and 1.54 mIU/L in Panshan in 2019.

Table 2.

Demographic Characteristics and Iodine Intake at Baseline (1999) and at Follow-Up (2004 and 2019)

Characteristic	Panshan			Zhangwu			Huanghua
Characteristic	1999 (n = 1103)	2004 (n = 884)	2019 (n = 625)	1999 (n = 1584)	2004 (n = 1270)	2019 (n = 839)	1999 (n = 1074)	2004 (n = 864)	2019 (n = 634)
Sex, n (%)
Males	284 (25.7%)	212 (24.0%)	137 (21.9%)	375 (23.7%)	290 (22.8%)	188 (22.4%)	275 (25.6%)	207 (24.0%)	125 (19.7%)
Females	819 (74.3%)	672 (76.0%)	488 (78.1%)	1209 (76.3%)	980 (77.2%)	651 (77.6%)	799 (74.4%)	657 (76.0%)	509 (80.3%)
Age, years
Mean (SD)	36.2 (12.6)	41.8 (11.9)	56.2 (10.3)	39.8 (13.2)	45.4 (12.3)	59.5 (10.7)	37.0 (12.8)	43.2 (12.0)	56.8 (10.1)
Median (IQR)	35 (28–44)	41 (34–50)	55 (49–64)	38 (30–48)	44 (36–53)	59 (52–67)	36 (29–46)	42 (36–51)	56 (50–65)
UIC, μg/L
Schoolchildren, n (%)	60 (33.3%)	60 (33.3%)	60 (33.3%)	60 (33.3%)	60 (33.3%)	60 (33.3%)	60 (33.3%)	60 (33.3%)	60 (33.3%)
Median	84	88	141.9	243	214	167.2	651	634	266.7
IQR	33–99	37–102	90.3–247.5	167–479	189–458	123.5–281.9	521–844	511–897	162.5–393.9
Adults, n (%)	1103 (29.3%)	884 (29.3%)	625 (29.8%)	1584 (42.1%)	1270 (42.1%)	839 (40.0%)	1074 (28.6%)	864 (28.6%)	634 (30.2%)
Median	103.1	97	114.3	375.8	350	175.3	608.3	635	260.1
IQR	60.5–188.5	46–137	66.0–164.9	236.8–562.4	213–505	118.5–253.5	470.2–768.6	511–897	166.2–370.5
Iodine in drinking water, μg/L
Median	8.2	10	4	4.6	7.8	4.8	201.8	201.8	190.3
IQR	1.4–17.0	8.1–12.4	1.7–15.0	2.4–13.9	6.6–10.0	3.9–8.6	174.7–260.5	167.6–205.7	179.9–224.4
Iodine in salt, mg/kg	<3.4	<3.4	18.8	54.5 ± 3.6	45.6 ± 5.9	18.5	23.3 ± 5.6	25.0 ± 10.0	0.4
TSH
Median	1.13	1.08	1.54	1.36	1.43	2.11	2.07	1.58	2.29
IQR	0.73–1.69	0.76–1.59	1.01–2.23	0.79–2.12	0.83–2.33	1.33–3.31	1.35–3.03	1.08–2.39	1.53–3.47

IQR, interquartile range; SD, standard deviation.

Over the period when the iodine status shifted from insufficient to adequate in Panshan, from 2004 to 2019, the cumulative incidence of subclinical hypothyroidism was significantly lower (1.9% vs. 6.0%, p < 0.001) compared with Zhangwu (Table 3). In addition, the incidence density of subclinical hyperthyroidism (3.07 vs. 1.19 per 1000 person-years, p = 0.04), positive thyroid antibodies (9.87 vs. 5.13 per 1000 person-years, p = 0.008), positive TPOAb (6.30 vs. 3.44 per 1000 person-years, p = 0.04), positive TgAb (7.25 vs. 3.26 per 1000 person-years, p = 0.005), and goiter (5.47 vs. 0.28 per 1000 person-years, p < 0.001) significantly decreased between 1990–2004 and 2004–2019 in Panshan (Table 4). Compared with Zhangwu, Panshan had a significant decrease in the cumulative incidence of subclinical hypothyroidism (adjusted odds ratio [OR] = 0.25; confidence interval [CI] = 0.12–0.49; p < 0.001) between 2004 and 2019 (Supplementary Table S1).

Table 3.

Cumulative Incidence of Thyroid Disorders in the Three Regions, 2004–2019

Thyroid disorder	Panshan	Zhangwu	Huanghua	p-for difference
Overt hyperthyroidism, % (n/N)	1.7 (9/524)	1.2 (9/740)	1.0 (5/524)	0.535
Subclinical hyperthyroidism, % (n/N)	1.7 (9/516)	1.2 (9/725)	1.2 (6/520)	0.668
Overt hypothyroidism, % (n/N)	1.3 (7/527)	1.9 (14/745)	3.3 (17/512)	0.07
Subclinical hypothyroidism, % (n/N)	1.9 (10/528)^a	6.0 (43/715)	8.8 (43/490)	<0.0001
Positive thyroid antibodies, % (n/N)	7.5 (34/453)	7.2 (45/628)	9.6 (42/436)	0.312
Positive TPOAb, % (n/N)	5.1 (24/473)	6.6 (44/662)	7.9 (37/466)	0.206
Positive TgAb, % (n/N)	4.6 (22/479)	4.3 (29/678)	5.7 (26/460)	0.553
Thyroid nodule, % (n/N)	37.9 (128/338)	35.5 (209/588)	42.4 (186/439)	0.08
Goiter, % (n/N)	1.3 (5/387)	1.9 (13/671)	2.1 (10/482)	0.66

Significant difference compared with Zhangwu at p-value <0.025.

Table 4.

The Incidence Density of Thyroid Disorders in the Three Regions Between 1999–2004 and 2004–2019

Thyroid disorder	Panshan	Zhangwu	Huanghua
Overt hyperthyroidism, cases per 1000 person-years [CI]
Incidence density, 1999–2004	2.08 [0.95–3.95]	1.93 [1.00–3.38]	1.40 [0.51–3.06]
Incidence density, 2004–2019	1.11 [0.55–2.29]	0.77 [0.40–1.64]	0.61 [0.23–1.57]
p-Value	0.3	0.08	0.31
Subclinical hyperthyroidism, cases per 1000 person-years [CI]
Incidence density, 1999–2004	3.07 [1.63–5.25]	4.74 [3.18–6.81]	2.81 [1.45–4.92]
Incidence density, 2004–2019	1.19 [0.51–2.09]	0.87 [0.34–1.41]	0.80 [0.26–1.55]
p-Value	0.04	<0.001	0.01
Overt hypothyroidism, cases per 1000 person-years [CI]
Incidence density, 1999–2004	0.46 [0.05–1.65]	0.48 [0.10–1.40]	0.71 [0.14–2.07]
Incidence density, 2004–2019	0.88 [0.35–1.92]	1.24 [0.68–2.17]	2.20 [1.29–3.65]
p-Value	0.64	0.18	0.08
Subclinical hypothyroidism, cases per 1000 person-years [CI]
Incidence density, 1999–2004	0.69 [0.14–2.01]	6.19 [4.38–8.50]	8.12 [5.59–11.40]
Incidence density, 2004–2019	1.23 [0.61–2.49]	3.78 [3.19–6.11]	5.55 [4.60–8.83]
p-Value	0.53	0.04	0.14
Positive thyroid antibodies, cases per 1000 person-years [CI]
Incidence density, 1999–2004	9.87 [6.98–13.55]	11.91 [9.17–15.20]	12.86 [9.45–17.10]
Incidence density, 2004–2019	5.13 [3.29–6.64]	4.93 [3.25–5.00]	6.59 [4.39–8.24]
p-Value	0.008	<0.001	0.002
Positive TPOAb, cases per 1000 person-years [CI]
Incidence density, 1999–2004	6.30 [4.07–9.30]	9.15 [6.83–11.99]	8.31 [5.68–11.73]
Incidence density, 2004–2019	3.44 [2.09–4.85]	4.52 [3.09–5.71]	5.37 [3.62–7.07]
p-Value	0.04	<0.001	0.1
Positive TgAb, cases per 1000 person-years [CI]
Incidence density, 1999–2004	7.25 [4.86–10.41]	8.56 [6.33–11.32]	11.31 [8.22–15.18]
Incidence density, 2004–2019	3.26 [1.64–4.15]	3.09 [1.55–3.57]	4.08 [1.99–4.80]
p-Value	0.005	<0.001	<0.001
Thyroid nodule, cases per 1000 person-years [CI]
Incidence density, 1999–2004	18.80 [13.95–24.78]	17.72 [14.19–21.86]	17.26 [13.20–22.17]
Incidence density, 2004–2019	25.25 [21.06–30.02]	23.70 [20.59–27.14]	28.25 [24.33–32.61]
p-Value	0.09	0.02	<0.001
Goiter, cases per 1000 person-years [CI]
Incidence density, 1999–2004	5.47 [2.99–9.18]	4.10 [2.50–6.33]	6.74 [4.40–9.87]
Incidence density, 2004–2019	0.28 [1.07–5.04]	0.86 [1.35–4.30]	0.67 [1.82–5.79]
p-Value	<0.001	<0.001	<0.001

The incidence density of thyroid disorders in the three regions between 2004 and 2019 was adjusted by age by using a multiple linear regression correction approach.

CI, confidence interval.

In Zhangwu, where the iodine status shifted from more than adequate to adequate, the incidence density of subclinical hyperthyroidism (4.74 vs. 0.87 per 1000 person-years, p < 0.001), subclinical hypothyroidism (6.19 vs. 3.78 per 1000 person-years, p = 0.04), positive thyroid antibodies (11.91 vs. 4.93 per 1000 person-years, p < 0.001), positive TPOAb (9.15 vs. 4.52 per 1000 person-years, p < 0.001), positive TgAb (8.56 vs. 3.09 per 1000 person-years, p < 0.001), and goiter (4.10 vs. 0.86 per 1000 person-years, p < 0.001) decreased significantly between 1990–2004 and 2004–2019 (Table 4). However, the incidence density for thyroid nodules (17.72 vs. 23.70 per 1000 person-years, p = 0.02) increased in 2004–2019 compared with 1999–2004. Compared with Panshan, Zhangwu had a significant increase in the cumulative incidence of subclinical hypothyroidism (adjusted OR = 3.12; CI = 1.54–6.30; p = 0.002) (Table 5).

Table 5.

Logistic Regression Analysis of the Associations Between Different Iodine Status Transition and the Development of Thyroid Disorders Between 2004 and 2019

Thyroid disorder	Panshan		Zhangwu		Huanghua
Thyroid disorder	OR [CI]	p-Value	OR [CI]	p-Value	OR [CI]	p-Value
Overt hyperthyroidism
Unadjusted model	1 (Reference)		0.71 [0.28–1.79]	0.46	0.55 [0.18–1.66]	0.29
Age- and sex-adjusted model	1 (Reference)		0.77 [0.30–1.97]	0.58	0.56 [0.19–1.69]	0.31
Subclinical hyperthyroidism
Unadjusted model	1 (Reference)		0.71 [0.28–1.80]	0.47	0.66 [0.23–1.86]	0.43
Age- and sex-adjusted model	1 (Reference)		0.75 [0.29–1.92]	0.55	0.67 [0.24–1.88]	0.44
Overt hypothyroidism
Unadjusted model	1 (Reference)		1.42 [0.57–3.55]	0.45	2.55 [1.05–6.21]	0.04
Age- and sex-adjusted model	1 (Reference)		1.45 [0.58–3.65]	0.43	2.56 [1.05–6.23]	0.04
Subclinical hypothyroidism
Unadjusted model	1 (Reference)		3.32 [1.65–6.66]	<0.001	4.98 [2.48–10.03]	<0.001
Age- and sex-adjusted model	1 (Reference)		3.12 [1.54–6.30]	0.002	4.98 [2.47–10.05]	<0.001
Positive thyroid antibodies
Unadjusted model	1 (Reference)		0.95 [0.60–1.51]	0.83	1.31 [0.82–2.11]	0.26
Age- and sex-adjusted model	1 (Reference)		0.96 [0.60–1.54]	0.86	1.31 [0.81–2.10]	0.27
Positive TPOAb
Unadjusted model	1 (Reference)		1.33 [0.80–2.22]	0.27	1.61 [0.95–2.74]	0.08
Age- and sex-adjusted model	1 (Reference)		1.36 [0.81–2.29]	0.24	1.60 [0.94–2.73]	0.08
Positive TgAb
Unadjusted model	1 (Reference)		0.93 [0.53–1.64]	0.8	1.24 [0.70–2.23]	0.46
Age- and sex-adjusted model	1 (Reference)		0.91 [0.51–1.61]	0.74	1.24 [0.69–2.23]	0.47
Thyroid nodule
Unadjusted model	1 (Reference)		0.91 [0.69–1.19]	0.48	1.21 [0.90–1.61]	0.21
Age- and sex-adjusted model	1 (Reference)		0.80 [0.60–1.07]	0.13	1.15 [0.85–1.55]	0.36
Goiter
Unadjusted model	1 (Reference)		1.51 [0.53–4.27]	0.44	1.62 [0.55–4.78]	0.38
Age- and sex-adjusted model	1 (Reference)		1.56 [0.55–4.46]	0.41	1.65 [0.56–4.86]	0.37

OR, odds ratio.

During the time period when the iodine status changed from excess to more than adequate in Huanghua, the incidence density of subclinical hyperthyroidism (2.81 vs. 0.80 per 1000 person-years, p = 0.01), positive thyroid antibodies (12.86 vs. 6.59 per 1000 person-years, p = 0.002), positive TgAb (11.31 vs. 4.08 per 1000 person-years, p < 0.001), and goiter (6.74 vs. 0.67 per 1000 person-years, p < 0.001) significantly decreased between 1990–2004 and 2004–2019 (Table 4). However, the incidence density for thyroid nodules (17.26 vs. 28.25 per 1000 person-years, p < 0.001) increased in 2004–2019 compared with 1999–2004. Compared with Panshan, Huanghua had a significant increase in cumulative incidence of overt hypothyroidism (adjusted OR = 2.56; CI = 1.05–6.23; p = 0.04) and subclinical hypothyroidism (adjusted OR = 4.98; CI = 2.47–10.05; p < 0.001) between 2004 and 2019 (Table 5). In addition, compared with Panshan and Zhangwu, Huanghua had a significant increase in the cumulative incidence of thyroid nodules (adjusted OR = 1.32; CI = 1.04–1.68; p = 0.02) (Supplementary Table S2).

Discussion

In the present study, we investigated the long-term effects of changes in iodine nutrition on the incidence of thyroid disorders over 5 and 20 years of follow-up in three cohorts from north China with different levels of iodine intake. When iodine nutrition was within the appropriate range (either following a period of iodine excess or mild iodine deficiency, or with more than adequate iodine), we demonstrated a decrease in the incidence density of thyroid disorders, such as subclinical hyperthyroidism, thyroid antibody positivity (especially TgAb), and goiter. The incidence density of overt hypothyroidism remained stable. However, the incidence density of thyroid nodules increased when iodine intakes changed from being excessive or more than adequate to being adequate.

It is known that iodine intake has a large effect on rates of thyroid disease in the population.^7
–9 Even small changes in iodine intake can affect the risk of developing thyroid-related diseases.²² The median UIC among school-age children increased in Panshan and decreased in both Zhangwu and Huanghua from 2004 to 2019. These shifts were due to government implementation of iodine adjustment measures based on previous local iodine nutrition status. Because our own research in these regions and the results of other domestic and foreign studies have indicated harmful effects of both iodine deficiency and iodine excess, the national health administration department in China has repeatedly adjusted the iodine content of table salt.⁶

In the iodine-deficient Panshan area, education on iodine deficiency disorders has been strengthened, and local residents have replaced non-iodized salt with iodized table salt. In the Huanghua area, where iodine concentrations are high in water sources, the supply of iodized salt was stopped, and drinking water iodine levels were mitigated. Although the iodine content of the local water supply remains high, residents now use bottled purified water rather than local water for drinking. Thus, the iodine status improved in these regions. In addition to the influence of iodine nutritional factors, the improvement in public awareness of health and disease prevention and medical conditions in recent years may partly explain the decline in the incidence density of most thyroid diseases.²³

The current study observed a decline in the incidence density of overt hyperthyroidism over time in all three regions, although there was no statistically significant difference among the regions. The incidence density of subclinical hyperthyroidism was significantly decreased. Previous studies have suggested that hyperthyroidism can transiently increase after iodine fortification, with recovery after 3 to 5 years.^24,25 Our follow-up study confirmed that despite the previous iodine nutritional status being mild deficiency, more than adequate, or excess, the incidence density of subclinical hyperthyroidism will significantly decrease as long as the final iodine nutrition level reaches an appropriate level.

We also found that the cumulative incidence of subclinical hypothyroidism was significantly higher in Zhangwu and Huanghua compared with Panshan. Iodine fortification may lead to a higher incidence of hypothyroidism.¹⁰ Similarly, our previous study indicated that more than adequate iodine or excessive iodine intake was associated with a higher incidence of subclinical hypothyroidism compared with insufficient iodine over five years of follow-up.⁶ Our hypothesis was that despite eventually achieving appropriate iodine nutritional status, individuals who had previously been in a state of excessive or more than adequate iodine for an extended period of time may experience the cumulative effects of iodine, leading to higher TSH levels compared with those who had previously been in a lower iodine nutritional state. However, a national study in mainland China previously reported that there was no significant difference in subclinical hypothyroidism prevalence between participants with UIC <100 μg/L and those with UIC = 100–199 μg/L.⁷

A study from Italy similarly documented that the prevalence of subclinical hypothyroidism in a iodine-deficient area was no different from that observed in iodine-sufficient regions.²⁶ The differences in results between our study and others might be due to differing research methodology and populations, with differences in iodine status before and during the implementation of USI. Hence, even when optimal iodine intakes are eventually achieved, the pre-iodine transition status significantly affects the risk of the development of subclinical hypothyroidism. In addition, our study confirmed that the long-term incidence density of subclinical hypothyroidism decreased over time when the population iodine status changed from more than adequate to adequate.

When iodine intake increases, the risk of inducing iodine excess may lead to the development of autoimmune thyroid disease. In fact, the sudden and excessive increases in iodine intake may result in the occurrence of thyroid antibodies in populations.²⁷ Although this did not reach statistical significance, we found that the cumulative incidence of positive thyroid antibodies, a risk factor for subclinical hypothyroidism, was higher in the Huanghua area than in the other regions. Long-term excessive iodine intakes may continue to have an impact on thyroid autoimmunity in the future, regardless of the eventual iodine nutritional status. We observed a decreased incidence density of thyroid antibody positivity as iodine status was optimized between the first and second follow-up periods in the three regions. Previous studies have focused mainly on the impact of sudden or periodic increases in iodine intake populations with baseline moderate iodine deficiency on rates of thyroid autoimmunity.

In contrast, we examined the long-term effects of transitioning from insufficient to adequate iodine, from more than adequate to adequate iodine, and from excessive to more than adequate iodine. A prior follow-up study in Denmark showed that the incidence of thyroid autoimmunity increased after 4–5 years of mandatory iodine fortification, which improved population status from mild/moderate iodine deficiency to optimal iodine intake.¹⁰ Similarly, another follow-up study in Slovenia showed that after 10 years of iodine fortification, the iodine status changed from mild deficiency to sufficiency, and the incidence of TPOAb positivity significantly increased.²⁸ We hypothesize that the decreased incidence of thyroid antibody positivity that we observed may be related to the length of follow-up and the optimal iodine intake achieved.

In addition, we observed a decreased incidence of goiter over time. This was consistent with a prior 10-year cohort study as well as two interventional studies that showed a significant decrease in thyroid goiter in Slovenia after an increase in salt iodization.²⁸ The current study found a high cumulative incidence of thyroid nodules in the Huanghua area, although this did not differ significantly from the other regions, suggesting that previous iodine nutritional status may have a potential impact on the development of thyroid nodules. We demonstrated an increase in the incidence density of thyroid nodules in Huanghua, which could not be explained by improved thyroid ultrasound technology because the same resolution (7.5 MHz) was used to evaluate thyroid nodules with the same diagnostic criteria (>5 mm in diameter) over time.

The prevalence of thyroid nodules in China was found to be as high as 20.43% in our previous nationally representative cross-sectional study.⁷ The prevalence of thyroid nodules decreased significantly with increased UIC.⁷ A previous study also indicated a significant increase in the prevalence of thyroid nodules following reduced iodine intake in the national population.⁹ The significant increase in the incidence density of thyroid nodules in the Huanghua area in this study may be partially related to the intentional decrease in iodine intakes in this area. However, the change in iodine nutrition may not be the only reason for the increased incidence of thyroid nodules, and further research is needed to explore the reasons for this increase.

The major strength of our study is that, to the best of our knowledge, it is the first to describe the changing incidence of thyroid disorders among populations with changing iodine status over a 20 year follow-up period. However, the study has some limitations. First, we did not ascertain thyroid cancer diagnoses, which limit our analysis of the association between iodine status and thyroid cancer. Second, although iodine intakes changed during the period 2004–2019, we are unsure of the specific time points of the iodine nutrition transition in these regions, which may impact the interpretation of the results. Third, we did not obtain information about the dietary patterns, smoking status, or alcohol consumption of participants in the baseline survey, which reduce our ability to explore some risk factors associated with thyroid disorders. Finally, the rate of loss to follow-up over the course of the study is substantial.

Conclusions

The incidence of thyroid disorders (except for thyroid nodules) stabilized or decreased among adults in the three regions from year 5 to year 15 of follow-up. Appropriate iodine fortification is safe and effective over the long term. Regardless of the previous iodine nutrition status, restoring the iodine nutrition status to appropriate levels will reduce the population's risk for thyroid disorders. Further research is required to explore the reasons for the observed increased incidence of thyroid nodules.

Footnotes

Acknowledgments

We thank the participants of this study. For continuous support, assistance, and cooperation, we thank the investigators of Zhangwu County Center for Disease Control and Prevention and Cangzhou Center for Disease Control and Prevention.

Authors' Contributions

Z.S. and W.T. were responsible for funding acquisition. W.T., Z.S., and E.N.P. were responsible for conceptualization. Yo.L. was responsible for formal analysis. Yo.L. did data curation. Z.S., Yu.L., Yo.L., H.W., D.T., X.T., W.C., X.S., J.L., J.G., Z.L., C.F., S.D., L.H., H.L., and W.T. were responsible for resources. Z.S., Yu.L., Yo.L., E.N.P., and W.T. wrote the original draft. All authors contributed to review and editing. Z.S., Yu.L., Yo.L., H.W., and W.T. had access to the raw data. All authors had access to all data in the study, and the corresponding author had final responsibility for the decision to submit for publication.

Author Disclosure Statement

The authors hereby confirm that no part of this article has been published or is under consideration for publication elsewhere. The authors have no potential conflicts of interest to declare.

Funding Information

This work was supported by the National Natural Science Foundation of China (Grant No. 81970682) and the NHC Key Laboratory of Thyroid Diseases Management (China Medical University, Grant No. 2019PT330001).

Supplementary Material

Supplementary Data

Supplementary Figure S1

Supplementary Table S1

Supplementary Table S2

References

Canaris

, Manowitz

, Mayor

, et al. The Colorado thyroid disease prevalence study. Arch Intern Med, 2000; 160:526–534; doi: 10.1001/archinte.160.4.526

Milanesi

, Brent

. Chapter 12: Iodine and thyroid hormone synthesis, metabolism, and action. In: Molecular, Genetic, and Nutritional Aspects of Major and Trace Minerals. (Collins JF. ed.) Academic Press: Boston, MA, USA; 2017; pp. 143–150.

International Council for Control of Iodine Deficiency Disorders; UNICEF; World Health Organization. Assessment of Iodine Deficiency Disorders and Monitoring Their Elimination: A Guide for Programme Managers, 2nd edition. World Health Organization: Geneva; 2001.

Gorstein

, Bagriansky

, Pearce

, et al. Estimating the health and economic benefits of universal salt iodization programs to correct iodine deficiency disorders. Thyroid, 2020; 30:1802–1809; doi: 10.1089/thy.2019.0719

Zimmermann

, Boelaert

. Iodine deficiency and thyroid disorders. Lancet Diabetes Endocrinol, 2015; 3:286–295; doi: 10.1016/S2213-8587(14)70225-6

Teng

, Shan

, Teng

, et al. Effect of iodine intake on thyroid diseases in China. N Engl J Med, 2006; 354:2783–2793; doi: 10.1056/NEJMoa054022

, Teng

, Ba

, et al. Efficacy and safety of long-term universal salt iodization on thyroid disorders: Epidemiological evidence from 31 provinces of mainland China. Thyroid, 2020; 30:568–579; doi: 10.1089/thy.2019.0067

Shan

, Chen

, Lian

, et al. Iodine status and prevalence of thyroid disorders after introduction of mandatory universal salt iodization for 16 years in China: A cross-sectional study in 10 cities. Thyroid, 2016; 26:1125–1130; doi: 10.1089/thy.2015.0613

, Shan

, Teng

. Effect of the transition from more than adequate iodine to adequate iodine on national changes in the prevalence of thyroid disorders: Repeat national cross-sectional surveys in China. Eur J Endocrinol, 2021; 186:115–122; doi: 10.1530/EJE-21-0975

10.

Petersen

, Knudsen

, Carlé

, et al. Increased incidence rate of hypothyroidism after iodine fortification in Denmark: A 20-year prospective population-based study. J Clin Endocrinol Metab, 2019; 104(5):1833–1840; doi: 10.1210/jc.2018-01993

11.

Bonofiglio

, Catalano

, Perri

, et al. Monitoring the effects of iodine prophylaxis in the adult population of southern Italy with deficient and sufficient iodine intake levels: A cross-sectional, epidemiological study. Br J Nutr, 2017; 117(1):170–175; doi: 10.1017/S0007114516004499

12.

Bjergved

, Jørgensen

, Perrild

, et al. Predictors of change in serum TSH after iodine fortification: An 11-year follow-up to the DanThyr study. J Clin Endocrinol Metab, 2012; 97(11):4022–4029; doi: 10.1210/jc.2012-2508

13.

Hong

, Stokes

, Otahal

, et al. Temporal trends in thyroid-stimulating hormone (TSH) and thyroid peroxidase antibody (ATPO) testing across two phases of iodine fortification in Tasmania (1995–2013). Clin Endocrinol (Oxf), 2017; 87(4):386–393; doi: 10.1111/cen.13371

14.

Sarfo-Kantanka

, Kyei

, Sarfo

, et al. Thyroid disorders in Central Ghana: The influence of 20 years of iodization. J Thyroid Res, 2017; 2017:7843972; doi: 10.1155/2017/7843972

15.

Choi

, Ock

, Kwon

, et al. Excessive iodine status among school-age children in Korea: A first report. Endocrinol Metab (Seoul), 2017; 32(3):370–374; doi: 10.3803/EnM.2017.32.3.370

16.

Teng

, Hu

, Teng

, et al. The study of thyroid diseases in a community not using iodized salt. Zhonghua Yu Fang Yi Xue Za Zhi, 2002; 36:176–179.

17.

, Teng

, Shan

, et al. Antithyroperoxidase and antithyroglobulin antibodies in a five-year follow-up survey of populations with different iodine intakes. J Clin Endocrinol Metab, 2008; 93(5):1751–1757; doi:10.1210/jc.2007-2368

18.

, Ding

, Han

, et al. Concentration-dependent differences in urinary iodine measurements between inductively coupled plasma mass spectrometry and the Sandell-Kolthoff method. Biol Trace Elem Res, 2021; 199:2489–2495; doi: 10.1007/s12011-020-02381-8

19.

Caldwell

, Maxwell

, Makhmudov

, et al. Use of inductively coupled plasma mass spectrometry to measure urinary iodine in NHANES 2000: Comparison with previous method. Clin Chem, 2003; 49:1019–1021; doi: 10.1373/49.6.1019

20.

Kratzsch

, Fiedler

, Leichtle

, et al. New reference intervals for thyrotropin and thyroid hormones based on National Academy of Clinical Biochemistry criteria and regular ultrasonography of the thyroid. Clin Chem, 2005; 51:1480–1486; doi: 10.1373/clinchem.2004.047399

21.

Hunt

, Bach

AJE

, Borg

, et al. The systematic bias of ingestible core temperature sensors requires a correction by linear regression. Front Physiol, 2017; 8:260; doi: 10.3389/fphys.2017.00260

22.

Laurberg

, Bülow Pedersen

, Knudsen

, et al. Environmental iodine intake affects the type of nonmalignant thyroid disease. Thyroid, 2001; 11:457–469; doi: 10.1089/105072501300176417

23.

Yip

, Fu

, Chen

, et al. 10 Years of health-care reform in China: Progress and gaps in Universal Health Coverage. Lancet, 2019; 394(10204):1192–1204; doi: 10.1016/S0140-6736(19)32136-1

24.

Bülow Pedersen

, Laurberg

, Knudsen

, et al. Increase in incidence of hyperthyroidism predominantly occurs in young people after iodine fortification of salt in Denmark. J Clin Endocrinol Metab, 2006; 91:3830–3834; doi: 10.1210/jc.2006-0652

25.

Laurberg

, Cerqueira

, Ovesen

, et al. Iodine intake as a determinant of thyroid disorders in populations. Best Pract Res Clin Endocrinol Metab, 2010; 24:13–27; doi: 10.1016/j.beem.2009.08.013

26.

Aghini-Lombardi

, Antonangeli

, Martino

, et al. The spectrum of thyroid disorders in an iodine-deficient community: The Pescopagano survey. J Clin Endocrinol Metab, 1999; 84:561–566; doi: 10.1210/jcem.84.2.5508

27.

Miranda

, Massom

, Catarino

, et al. Impact of nutritional iodine optimization on rates of thyroid hypoechogenicity and autoimmune thyroiditis: A cross-sectional, comparative study. Thyroid, 2015; 25(1):118–124; doi: 10.1089/thy.2014.0182

28.

Zaletel

, Gaberscek

, Pirnat

. Ten-year follow-up of thyroid epidemiology in Slovenia after increase in salt iodization. Croat Med J, 2011; 52:615–621; doi: 10.3325/cmj.2011.52.615