Prevalence and Incidence of Thyroid Dysfunction in Type 1 Diabetes, Type 2 Diabetes and Latent Autoimmune Diabetes of Adults

The Fremantle Diabetes Study Phase II

Kirsten E. Peters; Stephen A. Paul Chubb; David G. Bruce; Wendy A. Davis; Timothy M. E. Davis


Clin Endocrinol. 2020;92(4):373-382. 

In This Article

Abstract and Introduction


Objective: Since the results of published studies assessing thyroid dysfunction complicating diabetes have been variable in quality, inconsistent and may not reflect contemporary clinical care, the aim of this study was to determine its prevalence and incidence in a large, well-characterized, representative cohort.

Design: Community-based, longitudinal, observational study.

Patients: A total of 1617 participants from the Fremantle Diabetes Study Phase II (FDS2), including 130 (8.0%) with type 1 diabetes, 1408 (87.1%) with type 2 diabetes, and 79 (4.9%) with latent autoimmune diabetes of adults (LADA).

Measurements: Serum thyrotropin (TSH) and free thyroxine (FT4) at baseline between 2008 and 2011 and in those attending Year 4 follow-up.

Results: The prevalence of known thyroid disease (ascertained from baseline self-reported thyroid medication use or hospitalization data) was 11.7% (189/1617). Of the remaining 1428 participants, 5.1% (73/1428) had biochemical evidence of subclinical hypothyroidism, 1.1% (15/1428) overt hypothyroidism, 0.1% (2/1428) subclinical hyperthyroidism and 0.2% (3/1428) overt hyperthyroidism, representing an overall baseline prevalence of thyroid disease of 17.4% (282/1617). During 5694 patient-years of follow-up, 25 (3.0%) of the 844 with a normal baseline TSH and follow-up data developed known thyroid disease. Of the remaining 819, 3.4% developed subclinical hypothyroidism, 0.2% overt hypothyroidism and 0.5% subclinical hyperthyroidism. There were no statistically significant differences in the prevalence or incidence of thyroid dysfunction by diabetes type.

Conclusions: Thyroid dysfunction, known or detected through screening, is common in diabetes. These data suggest the need for periodic clinical and biochemical screening for thyroid disease in all types of diabetes.


A possible link between diabetes and thyroid dysfunction has long been recognized.[1] A variety of studies have suggested that thyroid disease is more prevalent in both type 1 and type 2 diabetes than in the general population,[2–6] but several used clinic-based samples, the characteristics of control groups have not always been provided, and clinical management (including recommendations for screening for thyroid dysfunction) has changed since some were conducted. A recent large population-based study found the prevalence of thyroid disease to be increased in Norwegians with type 1, but not type 2, diabetes.[7] This finding needs validation but suggests that characterization of diabetes type, which can be inaccurate in administrative database studies,[8,9] is important in establishing the risk of thyroid dysfunction complicating diabetes.

Thyroid dysfunction detected by screening using serum thyrotropin (TSH) and free thyroxine (FT4) is usually classified as subclinical hypothyroidism/hyperthyroidism if only the TSH is outside the reference interval, and as overt primary hypothyroidism/hyperthyroidism if the TSH and FT4 are concordantly abnormal. An increasing baseline TSH from within the reference interval was associated with an increased risk of overt hypothyroidism over 20 years in the Whickham study.[10] However, we[11] and others[12–14] have found that subclinical hypothyroidism can resolve without treatment over several years in 30%-50% of cases. While thyroid antibody positivity is another risk factor for incident overt hypothyroidism,[10] both the TSH (particularly when mildly raised[12]) and thyroid antibody positivity have low specificity for predicting subsequent hypothyroidism in an individual. One guideline recommends thyroxine replacement if the TSH is >10 mU/L, or if the TSH is raised but <10 mU/L, and symptoms of hypothyroidism are presented.[15] However, a recent meta-analysis of thyroxine therapy trials for subclinical hypothyroidism demonstrated no benefit for quality of life and thyroid-related symptoms,[16] which questions the need for treatment in this group.

The American Diabetes Association (ADA) recommends universal screening for thyroid dysfunction in type 1 diabetes.[17] While the ADA previously recommended that, for type 2 diabetes, screening should target females aged >50 years and those with dyslipidaemia,[17] current ADA guidelines[18] and those of the UK National Institute for Health and Care Excellence[19] do not recommend thyroid function monitoring in type 2 diabetes. Similarly, guidelines from Italy do not support screening for thyroid dysfunction in type 2 diabetes,[20] but a Welsh guideline does recommend screening with TSH and thyroid antibodies for all diabetes patients at diagnosis, with limited follow-up testing in those with type 2 diabetes.[21] The need for more real-world data to better inform screening recommendations in type 2 diabetes has, however, been highlighted.[22]

We have determined the baseline prevalence of known and undiagnosed thyroid disease in participants with well-characterized diabetes from the Fremantle Diabetes Study Phase II (FDS2). We also investigated the incidence of new thyroid dysfunction over the ensuing 4 years. We hypothesized that, despite contemporary approaches to detecting thyroid dysfunction, there remain people with diabetes and undiagnosed overt thyroid disease who should be considered for treatment.