Low Yield of Thyroid-Function Tests in Adult Hospitalized Patients — A Retrospective Analysis

Background In the US, serum thyroid-stimulating hormone (TSH) and thyroxine measurements are the fourth- and tenth-commonest laboratory tests ordered, respectively. Diagnosis of thyroid disorder requires clinical suspicion supported by laboratory values. However, in the setting of acute illness, both the clinical and laboratory pictures can be confounded. Objective To study clinical outcomes and derangement patterns of inpatient thyroid-function tests. Design This retrospective study was conducted at an academic center on admissions aged ≥18 years and TSH tests performed over a 1-year period. Admissions with active pregnancy and/or prior thyroid-related diagnosis were excluded. Main Outcomes Clinical outcomes were divided based on new diagnosis of thyroid-related disorder, newly prescribed thyroxine replacement, or antithyroid drugs/ endocrinology referrals, or both. In order to analyze the derangement patterns of abnormal TSH, only the results of the first test ordered were considered (as some admissions had multiple TSH tests ordered). Results A total of 7,204 admissions aged ≥18 years had TSH tests done. Of these, 1,912 were excluded. Of the 5,292 admissions with no prior thyroid disorder or active pregnancy, 183 (3.46%) were assigned a new diagnosis of thyroid-related disorder, 54 (1.02%) received treatment/referral, and 46 (0.87%) received both a new diagnosis and treatment/referral. Based on the TSH results (reference range 0.42-4.0 mIU/L) of the 5,292 admissions, 4,312 (81.5%) and 980 (18.5%) admissions were flagged normal and abnormal, respectively. Of the 980 admissions with one or more abnormal TSH results, 21 (2.14%) had first ordered TSH 10 mIU/L. Conclusion There is low value in testing inpatients for thyroid disorders, and testing comes at significant expense to the health-care system.