Iodine nutrition, thyroid function, and thyroid autoimmunity four decades after Universal Salt Iodisation in India: a cross-sectional study among schoolchildren (ICMR-SALT-T 1)

Summary

Background

Epidemiological studies undertaken following the first and second decade after the introduction of Universal salt iodisation (USI) in 1986 had pointed towards mild persistence in goitre, iodine excess as measured by spot urinary iodine concentration (UIC) and increase in thyroid autoimmunity. We aim to look at the status of iodine nutrition, goitre prevalence, thyroid dysfunction, and autoimmunity in Indian school-children in the fourth decade of implementation of USI.

Methods

8903 school children aged 10–18 years from eight centres across India were assessed for goitre by physical examination, thyroid function (n = 5512), autoantibody status (n = 5726), iodine nutrition by spot urinary iodine concentration (n = 1853).

Findings

Overall goitre prevalence was 10.3% (95% CI 9.7%–11.0%), majority (98.6%) being grade I goitre. Female sex (aOR 2.1, 95% CI 1.7–2.5; p < 0.001), obesity (aOR 1.3, 95% CI 1.0–1.6; p = 0.03) and anti-TPO positivity (aOR 2.5, 95% CI 1.9–3.1; p < 0.001) were associated with higher odds of goitre. Subclinical hypothyroidism (SCH) was observed in 13% (95% CI 12.2%–14.0%) of children with female sex (aOR 0.8, 95% CI 0.7–0.9; p = 0.03) and age (>15 year vs. 10–15 year aOR 0.7, 95% CI 0.6–0.9; p = 0.001) associated with lower odds; and overweight (aOR 1.4, 95% CI 1.0–1.8; p = 0.03)/obesity (aOR 1.9, 95% CI 1.5–2.3; p < 0.001) and autoantibody positivity with higher odds. Anti-TPO positivity was seen in 10.4% (95% CI 9.5%–11.1%) with higher odds in females (aOR 2.1, 95% CI 1.7–2.5; p < 0.001). Socioeconomic status was not associated with goitre, SCH or autoantibody positivity. MUIC was 133.2 μg/L (IQR-90.62 μg/L −199.49 μg/L). More than 90% of the surveyed household reported use of iodised salt and 89% of the 1615 salt samples tested contained iodine more than 15 ppm.

Interpretation

Goitre prevalence in India is showing a steady decline, with no difference based on socio-economic background reported in previous studies, indicating positive long term impact of USI. High rates of thyroid autoimmunity and thyroid dysfunction need future follow up studies.

Funding

Grant from ICMR-New Delhi under ‘Investigator Initiated Research Proposal’ (IIRP) scheme.

Research in context.

Evidence before this study

We searched PubMed from inception up to August 2026 using the terms “Universal Salt Iodisation”, “Goitre”, “Thyroid”, “Urinary Iodine”, “India”, “School Children”, without language restrictions and identified relevant original and review articles. The Kangra Valley trial (Sooch and Ramalingaswami, 1956) established salt iodisation as effective against endemic goitre in India. Subsequent nationwide surveys documented goitre prevalence of 23·5% in 2003 and 17·6% in 2012 (Marwaha et al.), indicating a downward trend. Evidence from multiple studies suggested persistence of goitre related to micronutrient co-deficiencies and socioeconomic disparities. No nationwide survey on iodine nutrition and thyroid health in Indian school children had been conducted since 2012, leaving a critical evidence gap of over a decade.

Added value of this study

In a nationwide cross-sectional study across eight geographically diverse centres, we report a goitre prevalence of 10·3% in Indian school children, confirming a sustained decline since USI implementation. The socioeconomic gradient in goitre prevalence observed in earlier surveys has largely disappeared, likely attributable to the government’s Mid-Day Meal Scheme. Median urinary iodine concentration remained within the WHO-defined adequate range, affirming population-level iodine sufficiency. While anti-TPO antibody positivity appears to be rising, clinically significant thyroid dysfunction showed a settling trend compared with prior surveys. This is the first study to provide updated, nationally representative data on USI impact across multiple dimensions like goitre, urinary iodine, thyroid function, and autoimmunity in Indian school children.

Implications of all the available evidence

The sustained decline in goitre prevalence and absence of rising clinically significant thyroid dysfunction justify continued and strengthened USI implementation in India. The rising trend in thyroid autoimmunity requires longitudinal surveillance in future studies, as this cross-sectional design precludes causal inference. Periodic nationwide surveys remain essential to calibrate iodisation standards and sustain the public health gains achieved over four decades.

Introduction

Iodine has long been identified as an essential nutrient necessary for thyroid gland function with effects of its deficiency spanning from poor brain development and cognitive ability; reduced reproductive capacity to low work productivity and economic outcomes. Globally, salt iodisation was first introduced in the United States and Switzerland in the 1920s, followed by Canada, and several South American countries, with remarkable success. In India, the landmark Kangra Valley experiment of 1956¹ demonstrated the effectiveness of adding iodine to household salt in the control of goitre in the Himalayan belt. This led to the launch of National Goitre Control Program (NGCP) by the Government of India in 1962. The programme was revamped and renamed as National Iodine Deficiency Disorders Control Program (NIDDCP) in 1992 with Universal Salt Iodisation (USI) as its prime strategy. The wide variations reported in the iodine content of salt at the production and household consumption levels was attributed to several factors including the amount of iodine added, packaging and storage conditions and time taken to reach the household.

Total goitre rate (TGR) in school children has been used by WHO as an epidemiological parameter to define long term status of iodine nutrition, while median urinary iodine concentration (MUIC) serves as a short-term indicator of iodine nutrition.²

Our earlier nationwide surveys in school children in the first- and second-decade following implementation of USI revealed significant decline in goitre prevalence from 23.5% to 17.6%. Persistence of moderate level of goitre endemicity reflect poor iodine nutrition status, though additional confounders such as micronutrient deficiencies, environmental goitrogens and autoimmunity needs to be accounted for.3, 4, 5 We along with several other investigators from India had reported higher goitre prevalence in children from low socio-economic strata as compared to those from better socio-economic background6, 7, 8 suggesting possible role of nutritional deficiency and other environmental goitrogens on thyroid function.

Though, USI has been enforced since 1986, there are limited data evaluating its impact in Indian population; despite ample reports in literature suggesting increase in thyroid autoimmunity, thyroid dysfunction etc.9, 10, 11, 12 This nationwide epidemiological survey was undertaken under the aegis of the same investigators who had conducted the earlier two nationwide studies to assess iodine nutrition, goitre prevalence, thyroid dysfunction, and autoimmunity in Indian school-children in the fourth decade of implementation of USI.

Methods

This multi-centre cross-sectional study was conducted as part of ICMR–SALT-T 1 (Salt Iodisation, Autoimmunity & Long-Term Thyroid outcomes) among school-children aged 10–18 years from eight different regions of India (North-Srinagar; North-West–Chandigarh; North-east–Jalpaiguri; East–Bhubaneshwar; Central–Gwalior, West-Central–Pune; South-Trivandrum; South-East coast–Pondicherry). Study regions were purposively selected to represent diverse ecological and dietary contexts relevant to iodine exposure. The sample size calculated was 5571, using conventional prevalence estimation methods, considering goitre prevalence as the primary outcome and assuming a goitre prevalence of 17.6%,⁴ a 95% confidence level (Z = 1.96), and an absolute precision of ±1%.The details of sample calculation is given in Supplementary Index.

Thyroid function, autoantibody positivity and median urine iodine concentration were assessed as secondary outcomes. The details of sample size calculation is given in the supplement. The study was conducted from January 2024 to December 2025. The study flow has been depicted in Fig. 1. Schools in each region were selected based on permissions obtained from local government authorities and school principals, after intimation via mail, including the early responders. Consent forms in English/Hindi/regional language were distributed to all children in the age group 10–18 years. Permissions regarding conduct of the study were obtained at four levels: directors of school education and health services of the region; the respective school principals; written informed consent from the parents or legal guardians of the children; and a verbal assent from the participating children.

Fig. 1.

Flowchart depicting flow of patients.

Children were excluded from the study if they had any known systemic illness, including cardiac, hepatic, renal, or endocrine disorders; had received continuous medication for more than one month in the preceding six months; had participated in another investigational study within the previous three months; had family history of autoimmune thyroid disease; had a prior diagnosis of thyroid disease; were receiving levothyroxine or anti-thyroid medications; or had recent exposure to iodine through iodinated cough syrups or contrast-enhanced computed tomography within the previous six months.

During school visits, all enrolled children underwent a detailed clinical evaluation and assessment of goitre. Goitre evaluation was done by the principal investigators of each local site (paediatricians/endocrinologists), followed by repeat-examination by the Primary Principal Investigator (PD) and the Principal Project Supervisor (RKM). Discordant assessments were resolved by joint re-examination and if disagreement persisted, the child was excluded from goitre classification. Goitre was graded according to WHO/UNICEF/ICCIDD criteria.²

Venous blood samples (5–8 mL) were collected from participating children who consented for venipuncture, between 08:00 and 10:00 h. Serum and plasma samples were separated by centrifugation, stored in separate aliquots at −20 °C and transported by maintaining cold-chain to the central laboratory. Free triiodothyronine (FT3), free thyroxine (FT4), thyroid-stimulating hormone (TSH), and anti-thyroid peroxidase (anti-TPO) antibodies were measured from the plasma samples by electrochemiluminescence immunoassay (ECLIA) (Cobas e801 Germany, Roche Diagnostics GmbH, Mannheim, Germany), using calibrators and controls provided by the same manufacturer. The normal reference range and the intra and inter assay CVs of the assays were: TSH- 0.27 to 4.2 μIU/mL (3.4%, 11.2%), free T4(FT4)–11.9–21.6 pmol/L (4.6%, 7.5%) and free T3 (FT3)–2.0–4.4 pg/mL (5.3%, 5.3%) respectively. All parameters were included in the monthly external quality assurance services (EQAS) (Bio-Rad, USA) in which the central laboratory was a participant. Thyroid dysfunction was defined as the presence of subclinical or overt hypothyroidism or hyperthyroidism. Subclinical hypothyroidism (SCH) was defined as TSH > 4.2 μIU/mL with normal FT4 and overt hypothyroidism as TSH > 4.2 μIU/mL with low FT4. Subclinical thyrotoxicosis was defined as TSH < 0.3 μIU/mL with normal FT4 and FT3 and overt thyrotoxicosis as elevated FT3 and/or FT4 with TSH < 0.3 μIU/mL. Anti-TPO antibody levels ≥34 IU/mL were considered indicative of thyroid autoimmunity. Abnormal results were repeated to minimise errors.

Daily iodine intake was assessed by urinary iodine concentration (UIC) from spot urine sample which were obtained in a labelled specimen containers (∼5 mL) from at least one out of every ten participants, selected randomly. In addition, household salt samples (∼5 g) were collected from these same participants in a ziplocked, air and moisture proof polythene bag for analysis of iodine content. The analysis was done at a single centralised laboratory after transport in cold chain to ensure uniformity of measurements and quality control. Urinary iodine estimation was done by Sandell–Kolthoff Reaction (SKR) using a unique microplate reader at 405 nm with a specially designed sealing cassette to prevent loss of vapour and cross-contamination among wells.¹³ Ammonium persulfate digestion was performed in the microplate in an oven at 110 degrees C for 60 min following which, the mixture was transferred to a transparent microplate and SKR was performed at 25 degrees C for 30 min. The mean recovery of iodine added to urine was 98% (range: 89–109%). The theoretical detection limit, defined as 2 SD from the zero calibrator, was 0.11 μmol/L (14 μg/L). The mean intra and inter-assay CVs for samples with iodine concentrations of 0.30–3.15 μmol/L were <10%. The new method agreed well with the conventional chloric acid digestion method (n = 70; r = 0.991; y = 0.944x + 0.04; S(y|x) = 0.10) and with the inductively coupled plasma mass spectrometry method (n = 61; r = 0.979; y = 0.962x + 0.03; S(y|x) = 0.20) as per the Equip program of CDC Atlanta, USA thereby providing external quality assurance. Iodine content in salt samples was measured using iodometric titration and spectrophotometric evaluation methods, as described.¹⁴

Ethics statement

The study protocol was reviewed and approved by the Institutional Ethics Committees of the nodal centre (Post Graduate Institute of Medical Education and Research (PGIMER), Chandigarh- IEC-09/2023–2867) and all participating centres (approval of all participating centres included in Appendix). The study was conducted in accordance with the principles of the Declaration of Helsinki and National ethical guidelines for biomedical and health research involving human participants, ICMR, 2017. Written informed consent was obtained from parents or legal guardians, and verbal assent was obtained from all participating children prior to enrolment.

Statistical analysis

Continuous variables were assessed for normality using visual inspection of histograms and the Shapiro–Wilk test. Normally distributed variables are presented as mean ± standard deviation (SD), while non-normally distributed variables are presented as median with interquartile range (IQR). Categorical variables are presented as numbers and percentages. Goitre prevalence was used to indicate the magnitude of IDDs, and prevalence estimates are presented with 95% confidence intervals. The prevalence of goitre grades (WHO Grade 1 and Grade 2) was described by age group, sex and school type. To quantify variation in prevalence estimates across study sites, a meta-analysis of proportions was performed using R (version 4.3.1), with heterogeneity quantified using the I² statistic. Forest plots were generated to visually display site-specific prevalence estimates along with corresponding 95% CI. Binary logistic regression analysis was performed to identify factors associated with goitre, subclinical hypothyroidism (SCH), and thyroid peroxidase (TPO) positivity. Initially, univariable logistic regression analyses were conducted for each independent variable. Variables evaluated included sex, age group, school type, BMI category, goitre status, and TPO positivity, as appropriate for each model. These variables were then entered into multivariable logistic regression models to estimate adjusted odds ratios (aORs) with 95% confidence intervals (CIs). Study site was included as a fixed effect in all multivariable models to account for site-level variation. Overt hypothyroidism, overt hyperthyroidism, and subclinical hyperthyroidism were not subjected to logistic regression analysis due to the small number of cases; instead, differences in the prevalence of these conditions between goitrous and non-goitrous children were compared using the chi-square test, stratified by sex and school type. A sensitivity analysis was conducted with TPO positivity prevalence re-estimated using a high titre cut-off ≥102 IU/mL. Differences in UIC between goitrous and non-goitrous children were compared using the Mann–Whitney U test, given the non-normal distribution of UIC data. All Statistical analyses were performed using IBM SPSS Statistics for Windows, version 29.0 (IBM Corp., Armonk, NY, USA) and a two-sided p-value < 0.05 was considered statistically significant for all analyses. The study has adhered to Strengthening the Reporting of Observational Studies in Epidemiology (STROBE) guidelines.

Role of the funding source

The funder (Indian Council of Medical Research, New Delhi) had no role in study design; collection, analysis and interpretation of data; in the writing of the manuscript; or in the decision to submit the paper for publication. RKM, PD, MSVS, DB and LT had access to data and all authors take the responsibility to submit the paper for publication.

Results

Prevalence of goitre

A total of 8903 school children aged 10–18 years were included in the study, comprising 4603 boys and 4300 girls. The overall prevalence of goitre was 10.3% (95% CI 9.7%–11.0%) and WHO grade I goitres accounted for 98.6% of all detected goitres. Substantial heterogeneity was observed across study sites (I² = 80%), indicating considerable between-site variation in prevalence estimates (Fig. 2, Supplementary Fig. S1 and Table S4). In multivariable logistic regression analysis, female participants had significantly higher odds of goitre compared with males (aOR 2.1, 95% CI 1.7–2.5; p < 0.001). Underweight participants had lower odds of goitre than those with normal BMI (aOR 0.6, 95% CI 0.4–0.8; p < 0.001), whereas obesity was independently associated with higher odds of goitre (aOR 1.3, 95% CI 1.0–1.6; p = 0.03).

Fig. 2.

Goitre prevalence across various study centres. Goitre prevalence expressed in percentage (with 95% CI) by taking the number screened as denominator at each centre.

Participants with positive TPO antibodies had significantly greater odds of goitre compared with TPO-negative participants (aOR 2.5, 95% CI 1.9–3.1; p < 0.001). Age group and school type (fee-paying vs. non-fee-paying schools, used as a proxy of higher and lower socioeconomic strata respectively) were not significantly associated with goitre after adjustment.

Prevalence, age-wise and grade distribution of goitre among school-children is shown in Supplementary Table S1.

Thyroid function

Thyroid function assessment, FT3, FT4 and TSH was available, in 5512, who consented for venipuncture and whose samples were adequate for laboratory analysis. SCH was observed in 13% (95% CI 12.2%–14.0%) of children. In multivariable analysis, female sex was associated with lower odds of SCH compared to males (aOR 0.8, 95% CI 0.7–0.9; p = 0.03). Older age (>15 years) was also associated with reduced odds of SCH relative to the 10–15 year age group (aOR 0.7, 95% CI 0.6–0.9; p = 0.001) (Table 1).

Table 1.

Multivariable logistic regression showing factors predicting goitre.

Variable	Category	n/N	Prevalence (95% CI)	Unadjusted OR (95% CI)	p	Adjusted OR (95% CI)	p
Sex	Male	324/4603	7.0% (6.3–7.8)	Reference	–	Reference	–
	Female	596/4300	13.9% (12.9–14.9)	2.1 (1.8–2.5)	<0.001	2.1 (1.7–2.5)	<0.001
Age group (years)	10–15	675/6572	10.3% (9.6–11.0)	Reference	–	Reference	–
	>15	245/2331	10.5% (9.3–11.8)	1.02 (0.9–1.2)	0.74	0.9 (0.7–1.1)	0.31
School type	Non-fee paying	512/4052	10.5% (9.7–11.5)	Reference	–	Reference	–
	Fee-paying	408/4851	10.1% (9.2–11.0)	1.05 (0.9–1.2)	0.45	1.07 (0.8–1.3)	0.48
BMIa	Normal	566/5106	11.1% (10.3–12.0)	Reference	–	Reference	–
	Underweight	89/1481	6.0% (4.9–7.3)	0.5 (0.4–0.6)	<0.001	0.6 (0.4–0.8)	<0.001
	Overweight	106/946	11.2% (9.4–13.4)	1.01 (0.8–1.3)	0.91	1.1 (0.9–1.5)	0.37
	Obese	157/1284	12.2% (10.5–14.1)	1.1 (0.9–1.3)	0.24	1.3 (1.0–1.6)	0.03
TPOa positivity	Negative	502/5132	9.8% (9.0–10.6)	Reference	–	Reference	–
	Positive	139/594	23.4% (20.3–27.1)	2.8 (2.3–3.5)	<0.001	2.5 (1.9–3.1)	<0.001

Adjusted odds ratios were obtained from multivariable logistic regression models including study site as a fixed effect. n = number of people with Goitre; N = Total number of participants; OR: Odds ratio.

^aData not available for all participants.

The presence of goitre was independently associated with nearly twice the odds of SCH (aOR 1.9, 95% CI 1.5–2.3; p < 0.001). Socioeconomic status could not predict SCH (aOR 0.9, 95% CI 0.7–0.1; p = 0.11). With respect to BMI, overweight (aOR 1.4, 95% CI 1.0–1.8; p = 0.03) and obesity (aOR 1.9, 95% CI 1.5–2.3; p < 0.001) were independently associated with significantly higher odds of SCH, whereas underweight status was not (aOR- 1.1 95% CI 0.9–1.4; p = 0.36). TPO positivity was strongly and independently associated with SCH (aOR 1.9, 95% CI 1.5–2.5; p < 0.001) (Table 2).

Table 2.

Multivariable logistic regression showing factors predicting SCH.

Variable	Category	n/N	Prevalence (95% CI)	Unadjusted OR (95% CI)	p	Adjusted OR (95% CI)	p
Sex	Male	370/2688	13.8% (12.5–15.1)	Reference	–	Reference	–
	Female	349/2824	12.4% (11.2–13.7)	0.9 (0.8–1.0)	0.12	0.8 (0.7–0.9)	0.03
Age group (years)	10–15	578/4013	14.4% (13.3–15.5)	Reference	–	Reference	–
	>15	141/1499	9.4% (8.0–11.0)	0.6 (0.5–0.8)	<0.001	0.7 (0.6–0.9)	0.001
Goitre	Negative	598/4891	12.2% (11.3–13.1)	Reference	–	Reference	–
	Positive	121/621	19.5% (16.5–22.8)	1.7 (1.4–2.2)	<0.001	1.9 (1.5–2.3)	<0.001
School type	Non-fee paying	323/2521	12.8% (11.5–14.1)	Reference	–	Reference	–
	Fee-paying	396/2991	13.2% (12.0–14.5)	1.0 (0.9–1.2)	0.63	0.9 (0.7–1.0)	0.11
BMIa	Normal	382/3269	11.7% (10.6–12.8)	Reference	–	Reference	–
	Underweight	113/915	12.3% (10.3–14.6)	1.0 (0.9–1.3)	0.58	1.1 (0.9–1.4)	0.36
	Overweight	83/556	14.9% (12.2–18.1)	1.3 (1.0–1.7)	0.03	1.4 (1.0–1.8)	0.03
	Obese	135/722	18.7% (16.0–21.7)	1.7 (1.4–2.2)	<0.001	1.9 (1.5–2.3)	<0.001
TPO positivitya	Negative	609/4920	12.4% (11.5–13.3)	Reference	–	Reference	–
	Positive	109/570	19.1% (16.0–22.5)	1.7 (1.3–2.1)	<0.001	1.9 (1.5–2.5)	<0.001

Adjusted odds ratios were obtained from multivariable logistic regression models including study site as a fixed effect. n = number of people with SCH; N = Total number of participants; OR: Odds ratio.

^aData not available for all participants.

Overt hypothyroidism was observed in 1.4% ((5% CI 0.8%–2.7%) of goitrous and 0.3% (95% CI 0.2–0.5%)) of non-goitrous children. Both subclinical and overt hyperthyroidism were uncommon, with a prevalence of 0.04% and 0.2% in both goitrous and non-goitrous children respectively (Supplementary Table S2).

Thyroid autoimmunity

Anti-thyroid peroxidase (anti-TPO) antibody measurements were available for 5726 children. Overall anti-TPO positivity was 10.4% (95% CI 9.5%–11.1%) in the study population. On multivariable analysis, female sex was significantly associated with higher odds of TPO positivity compared to males (aOR 2.1, 95% CI 1.7–2.5; p < 0.001). The presence of goitre independently increased the odds of TPO positivity more than two-fold (aOR 2.5, 95% CI 2.0–3.1; p < 0.001).

Older age (>15 years) and socioeconomic status were not independently associated with TPO positivity after adjustment (p = 0.67 and p = 0.11, respectively) (Table 3).

Table 3.

Multivariable logistic regression showing factors predicting Anti-TPO positivity.

Variable	Category	n/N	Prevalence (95% CI)	Unadjusted OR (95% CI)	p	Adjusted OR (95% CI)	p
Sex	Male	194/2808	6.9% (5.9–7.8)	Reference	–	Reference	–
	Female	400/2918	13.7% (12.4–14.9)	2.2 (1.8–2.6)	<0.001	2.1 (1.7–2.5)	<0.001
Age group (years)	10–15	409/4140	9.9% (8.9–10.7)	Reference	–	Reference	–
	>15	185/1586	11.7% (10.1–13.3)	1.2 (1.0–1.5)	0.04	0.9 (0.8–1.2)	0.67
Goitre	Negative	454/5085	8.9% (8.1–9.7)	Reference	–	Reference	–
	Positive	139/641	21.7% (18.5–24.9)	2.8 (2.3–3.5)	<0.001	2.5 (2.0–3.1)	<0.001
School type	Non-fee paying	300/2614	11.5% (10.2–12.6)	Reference	–	Reference	–
	Fee-paying	294/3112	9.4% (8.4–10.4)	0.8 (0.7–0.9)	0.01	0.9 (0.7–1.0)	0.11
BMI∗	Normal	344/3379	10.2% (9.2–11.2)	Reference	–	Reference	–
	Underweight	81/956	8.5 (6.8–10.4)	0.8 (0.6–1.1)	0.1	0.8 (0.6–1.1)	0.16
	Overweight	68/581	11.7% (9.2–14.6)	1.2 (0.9–1.5)	0.26	1.1 (0.8–1.5)	0.42
	Obese	95/760	12.5% (10.2–15.1)	1.3 (0.9–1.6)	0.06	1.2 (0.9–1.5)	0.21

Adjusted odds ratios were obtained from multivariable logistic regression models including study site as a fixed effect. n = number of people with TPO positive; N = Total number of participants; OR: Odds ratio.

Sensitivity analysis was done with a cut off of ≥102 IU/mL and high titre anti-TPO positivity was seen in 2.7% (95% CI 2.3%–3.2%), 2.04% in <11 years, 2.6% in 11–15 years and 3.2% in >15 years. While comparing 11–18 year old children, the prevalence was 2.3% in 2012 and 2.8% in the current study (Table 4), which is not significantly different (p −0.129). Prevalence of anti-TPO positivity among goitrous and non-goitrous subsets is shown in Supplementary Table S3.

Table 4.

Temporal trends in iodine nutrition, thyroid outcomes, and thyroid autoimmunity in Indian schoolchildren across three nationwide surveys (2003, 2012, and 2025).

	2003	2012	2025
Goitre prevalence	23.5%	17.6%	10.5%
WHO grade 1	87.8%	96.7%	98.8%
WHO grade 2	12.2%	3.3%	1.2%
Median urinary iodine concentration
Overall	–		133.1 μg/L
Goitrous	320.3 μg/L	–	146.2 μg/L
Non-goitrous	283.5 μg/L	–	130.8 μg/L
Subclinical hypothyroidism
Goitrous	4.3%	10.3%	19.5%
Non-goitrous	2.36%	6.2%	12.0%
Overt hypothyroidism
Goitrous	0.89%	1.2%	1.5%
Non-goitrous	0%	0.3%	0.4%
Subclinical hyperthyroidism
Goitrous	0.98%	0.2%	0.0%
Non-goitrous	0%	0.2%	0.04%
Overt hyperthyroidism
Goitrous	0.13%	0.3%	0.8%
Non-goitrous	0%	0.1%	0.1%
Anti -TPO positivity
Goitrous	–	9%	21.5%
Non-goitrous	–	2.6%	8.8%
High-titre Anti -TPO positivity		2.3%	2.8%

Data shown are from three nationwide school-based surveys conducted in 2003, 2012, and 2025 including children in the age group 11–18 years. Total goitre prevalence and grading is based on WHO/UNICEF/ICCIDD criteria. Thyroid dysfunction definitions and anti-TPO cut-offs follow the respective survey protocols, and “–” indicates data not reported or not available in the corresponding survey.

Urinary iodine concentration

UIC was available for 1853 subjects. Median urinary iodine concentration (MUIC) was 133.2 μg/L with interquartile range (IQR) (90.62 μg/L −199.49 μg/L), and was not significantly different between the goitrous and non-goitrous children (146.09 μg/L (IQR-93.5 μg/L −200.2 μg/L) vs. 131.0 μg/L (IQR-90.5 μg/L −199.4 μg/L), p −0.12)).

Household salt iodisation

Household use of iodised salt was reported by 90% of the surveyed families. 1615 samples were analysed for iodine content and 89% contained iodine concentrations greater than 15 ppm.

Discussion

USI has been one of the most successful and cost-effective public health intervention. India has National Iodine Deficiency Disorders Control Program (NIDDCP) in place since 1992 and those children who are included in our study, up to the age of 18 years, born after introduction of USI is the ideal population to study it’s impact. The significant reduction in goitre rate from 23.5% to 17%–10.3% and goitre size in adolescent school children of India over the last 3 decades^3,4 suggest positive long-term impact of USI (Table 4), despite persistence of mild persistence of goitre. This residual goitre may be due to high prevalence of autoimmune thyroiditis; increasing rates of obesity¹⁵; low or excess iodine intake; micronutrient deficiencies; or exposure to environmental goitrogens. Various studies across the globe with iodine deficiency disorders have shown reduction in goitre prevalence ranging from 1.2% to 62.3% over 16–18 years after the intervention with iodised salt.¹⁶ WHO systematic review of iodised salt trials¹⁷ observed reduction in goitre prevalence in population studies. Gorstein et al.¹⁸ showed a fall in global TGR value from 13.2% to 3.2% between 1993 and 2019 by a regression model analysis. Zimmermann et al. observed a decrease in thyroid volume by 56% over 5 years after introduction of iodised salt in school children.¹⁹

Another interesting observation of high goitre prevalence in children from lower as compared to those from upper socio-economic strata was previously reported by us³ and several other investigators.6, 7, 8 It was suggested that micronutrient deficiencies other than iodine in poor children could play a role in thyroid metabolism and result in higher goitre prevalence. Several studies reported the role of iron, selenium and Zn and vitamin A in causation goitre in human studies.^5,20,21 The disappearance of difference in the present study may be the result of improved nutrition in poor children, provided mid-day meals as a government initiative; and improved reach of iodised salts to poor households. Though goitre can be argued to be just the tip of iceberg in relation to the wide spectrum of IDDs, these findings are still to be acknowledged as a significant progress, considering the fact that there is lack of nationwide data addressing the problem of iodine deficiency disorders.²²

The thyroid function evaluated in the present study noted an overall prevalence of SCH of 13%.

The prevalence of SCH in children was 4.3% in children aged 6–18 years,³ on follow up after the first decade (assayed by immunoradiometric assay technique (Immunocorp, Montreal, Canada), normal range of the kit being 0.3–4 μIU/mL and TSH cut off of 5 μIU/mL being used to define SCH), and 6.8% in children aged 5–18 years⁴ (assayed by immunoradiometric assay technique (Immunotech, Beckman Coulter analyser), with normal range of the assay 0.17–5.2 μIU/mL) after the second decade of USI in India. As the current study included children in the age group 10–18 years, these values are not truly comparable.

The prevalence of SCH observed in the present and previous studies is higher than that reported in other studies.23, 24, 25 Increasing iodine intake in populations has shown to be associated with increase in prevalence of SCH, primarily described in young and middle-aged adults.^25,26 Various explanations given for this phenomenon are an increase in thyroid autoimmunity, downregulation of thyroid function by iodine and reduction in pituitary type-2 deiodinase activity.²⁷ Lombardi et al.²⁸ reported a higher incidence of SCH after iodine supplementation in a study done in children younger than 15 years. An additional consideration is the transient changes in thyroid function seen at puberty and rising prevalence of childhood obesity. The prevalence of obesity in our study population was 14.5%, which is higher than the pooled prevalence of 8.4% from a recent meta-analysis.¹⁵ Whether SCH seen with a prevalence of 2.9%–14.1% among adolescent children in various studies29, 30, 31 is clinically relevant and warrants treatment is not known and need to be explored in further studies. A high rate of reversibility of SCH, seen in 41–73%^29,32 of the subjects, suggest that long-term large scale studies are needed to draw definite conclusions. Both structural and functional alterations in thyroid have been described in obese children, which is independent of autoimmune involvement.^33,34

Similar prevalence of SCH in boys and girls in the present study is no different from what was reported in our earlier study.⁴ Higher hypothyroidism prevalence in goitrous subjects than in non-goitrous subjects is in agreement with our earlier study⁴ and study carried out by Todd C H et al. in Zimbabwean school children.³⁵

Our study did not show any increase in prevalence of hyperthyroidism after three decades of salt iodisation, which is reassuring given concerns of hyperthyroidism on introducing iodine to iodine deficient populations.²⁷ In addition, long-term iodine sufficiency has shown to reduce development of autonomous nodules, as demonstrated in a 16-year follow-up study from Switzerland.³⁶

We noted a high prevalence (10.3%) of anti-TPO positivity. Similar prevalence of 9.7% was reported from China after 20 years of follow up.³⁷ There is significant increase in overall prevalence of autoimmunity as assessed by anti-TPO positivity from 4.4%⁴ two decades following USI to 10.3%. Pescopagano Survey in Italy 15 years after the USI introduction showed an increase in autoimmunity.²⁸ A study from Sri Lanka also noted a prevalence of elevated TPO Ab of 10.3%.²³ In contrast, no significant change was reported in population studies from Tasmania and North-East Germany.^38,39 Higher prevalence of thyroid antibodies was observed in goitrous than in non-goitrous subjects in our earlier studies also^3,4(Table 4). Several hypotheses has been proposed to explain rise in autoimmunity seen with iodine supplementation and include increased immunogenicity of thyroglobulin, increased MHC-II expression,⁴⁰ altered cytokine expression and modulation of the microbiota-gut-thyroid axis.^27,41 The higher occurrence of thyroid dysfunction in anti-TPO vs. anti-TPO negative subjects (21.4% vs. 11.2%, p < 0.001) could indicate probable association. Jayatissa et al.²³ reported strong association between high titre anti-TPO positivity and SCH, though the prevalence of SCH was much lower compared to that of anti-TPO positivity. Long term follow up studies show that thyroid function²³ and thyroid autoimmunity⁴² do settle down to baseline levels after iodine supplementation. Our finding of increase in thyroid autoimmunity and subclinical hypothyroidism is mirroring the findings of examination of iodine fortification programme in Denmark.⁴³

The MUIC of 133.4 μg/L (IQR-90.68–199.53) is indicative of adequate iodine nutritional status as per WHO. This indicates the impact of USI, given the situation that even developed countries like the UK and Australia⁴⁴ are facing the problem of iodine deficiency and reconsidering the need for salt iodisation programme.⁴⁵ This reinforces the need to continue USI in our country to sustain this achievement. USI and low salt intake in diet are not conflicting strategies and both should go hand-in-hand, while increasing awareness about the low iodine content of rock salt, which is in currently in vogue.

In the initial years following introduction of USI in our country, there was a tendency towards over-fortification of salt at the production level. Following the Revised Food Safety and Standards (Fortification of Foods) regulations released in 2019⁴⁶ introduced an upper limit for iodine concentration in salt both at production and consumer level at 30 ppm.

With 90% of the surveyed households reporting consumption of iodised salt and 89% of salt samples tested, having more than 15 ppm, it can be assumed that India has achieved reasonable success in the implementation of USI. This is a huge progress comparing 2008 UNICEF report that stated that only 51% of the nation’s households were consuming adequately iodised salt.⁴⁷

This study was school based and used non-probability sampling, which may limit national representativeness. Moreover, the schools in each region were selected based on permissions from local government and school authorities following a first-come, first-served approach, which could introduce a degree of bias. Unlike the previous two nationwide surveys which included children in the age group 6–18 years, the current study includes only children in the age group of 10–18 years, which limits comparisons. The study may be underpowered for the secondary outcomes, making it difficult to draw conclusions. ICC- based adjustment for multicentre study was not done, and this might potentially influence precision estimates. Its cross-sectional design precludes causal inference between iodine status, goitre, thyroid dysfunction, and autoimmunity. Blood samples were not available for all subjects for assessment of thyroid function and autoantibody status. Moreover, thyroid dysfunction was defined using kit reference ranges provided by the manufacturer, due to non-availability of established normative data in our population. Urinary iodine was measured using single spot samples, only in a subset of participants, reflecting recent intake and subject to intra-individual variability. Goitre assessment relied on clinical palpation without ultrasonographic confirmation, since conducting studies in the field using ultrasound is not permitted by the government. Potential confounders such as other dietary iodine sources, goitrogen exposure, micronutrient status, and pubertal effects were not comprehensively assessed.

In the fourth decade of USI, Indian school-children demonstrate overall adequate iodine nutrition and high household coverage of iodised salt. Despite this, mild residual endemic goitre persists almost entirely comprising WHO grade I goitre. High rates of thyroid autoimmunity and thyroid dysfunction need future follow up studies. USI, being a cost-effective, time-tested public health intervention needs continuation with quality control.

Contributors

PD, RKM and MMG conceptualised and designed the study. Methodology was finalised by PD, RKM, BKP, DB and NS. Funding acquisition was done by PD and RKM; and investigation was carried out by PD, RKM, MMG, NS and RSJ. Project administration was helped by PD, RKM, NS, SRM, MAG, KB, BKP, DN, JS, SSJ, PKJ, SBUS, and AN. All authors helped in supervision and validation of data. PD, RKM, DB, MSVS and LT curated the data and did formal analysis. Original draft was written RKM, PD, LT and AD; and reviewed and edited by PD, RKM, LT, SM, VJ, NS, BKP, MAG and Smed. PD, RKM and MMG had access to raw data and verified the data. PD and RKM take responsibility for the decision to submit for publication. The ICMR-SALT-T1 study group was involved in administration and supervision of this nation-wide survey at their respective centres.

Data sharing statement

All relevant data generated or analysed during this study are available from the corresponding author on reasonable request.

Editor note

The Lancet Group takes a neutral position with respect to territorial claims in published maps and institutional affiliations.

Declaration of interests

All authors have completed the ICMJE disclosure form and declare no conflicts of interest related to this work. All authors read and approved the final manuscript.