Answers are grounded in this paper and cite their sources.
Background: Coffee taken together with levothyroxine (LT4) tablets reduces LT4 absorption in small pharmacokinetic and clinical studies. Whether habitual coffee intake is associated with poorer thyroid hormone replacement in the general population of LT4 users is unknown. Methods: We analyzed adults aged 20 years or older in the US National Health and Nutrition Examination Survey (NHANES) 2007-2012 who had thyroid function tests and a reliable day-1 24-hour dietary recall. LT4 use was identified from the prescription medication file. Coffee intake (cups/day; 1 cup = 240 g) and the clock time of the first coffee were derived from individual food records. The primary analysis was the design-based, survey-weighted linear regression of log-transformed thyrotropin (TSH) on coffee cups per day among LT4 users, adjusted for age, sex, survey cycle, race/ethnicity, body mass index, serum cotinine, proton pump inhibitor use, use of drugs that bind LT4, and dietary fiber, calcium and iron intake. All other analyses were exploratory. Results: Of 8,232 eligible adults, 531 used LT4 (78% women; median age 66 years); 16.8% had TSH >4.5 mIU/L and 15.6% had TSH <0.45 mIU/L. In the primary analysis (n = 522), each additional cup of coffee was associated with a 5.7% lower TSH (beta = -0.059; 95% CI -0.139 to 0.022; p = 0.15). Coffee intake was not associated with TSH >4.5 mIU/L (odds ratio per cup 0.99; 95% CI 0.77 to 1.28). In exploratory analyses, LT4 users who drank coffee before 10:00 had lower TSH than those who did not (-33%; p = 0.047), and among coffee drinkers the association of the time of first coffee with TSH differed between LT4 users and non-users (interaction p = 0.018). These exploratory associations were in the direction opposite to impaired absorption and were attenuated without survey weights. Conclusions: In this nationally representative cross-sectional sample, habitual coffee intake and earlier coffee drinking were not associated with higher TSH or more frequent under-replacement among LT4 users. The findings do not support a population-level adverse effect of coffee on LT4 replacement, but cannot exclude effects that are compensated by dose titration.
No takes yet. Share an insight, caveat, or question.
Kenji Ohe (2026) studied this question.
Synapse has enriched 8 closely related papers on similar clinical questions. Consider them for comparative context: