Stress & Hormones

Thyroid Function and Ageing:
When a Test Result Matters and When It Doesn't

Aevum Protocol7 min read

The thyroid gland sets the pace of the body's metabolism, and thyroid problems become more common with age, especially in women. But ageing also changes what "normal" looks like: TSH, the main thyroid screening test, naturally drifts upwards in older people. That creates a real risk of over-diagnosis. Many older adults are told they have "borderline" or subclinical hypothyroidism when their result is actually normal for their age. In a large trial of adults over 65 with mildly raised TSH, thyroid hormone tablets didn't improve symptoms or tiredness. At the same time, clearly abnormal thyroid function matters: a TSH of 10 or more is linked to a higher risk of heart disease, and a TSH that's too low, often from taking too much thyroid medication, is linked to about twice the risk of atrial fibrillation. This article explains how thyroid tests change with age, when results matter, and what that means in practice.

Key numbers

FindingDetail
Upper limit of normal TSH by age (US population)3.56 mIU/L at ages 20-29; 7.49 mIU/L at age 80+
Older people with TSH above 4.5 who are within their age-specific rangeAbout 70%
Levothyroxine for subclinical hypothyroidism in adults 65+ (TRUST trial, 737 people, average age 74)No improvement in thyroid-related symptoms or tiredness after 1 year
Subclinical hypothyroidism with TSH 10-19.9 (11 cohorts, 55,287 people)89% higher risk of coronary heart disease events (HR 1.89)
Subclinical hypothyroidism with TSH 4.5-6.9No increase in coronary heart disease events (HR 1.00)
Subclinical hyperthyroidism (low TSH) and atrial fibrillation (meta-analysis, 5 studies)About twice the risk (RR 1.99)

How the thyroid works

The thyroid, a small gland in the neck, makes two hormones, T4 and T3, that regulate metabolism, heart rate, body temperature, energy and many other functions. It's controlled by the pituitary gland, which releases TSH (thyroid-stimulating hormone):

Because of this feedback, TSH is a sensitive early indicator of thyroid problems, and it's usually the first test done. Doctors then check free T4 to see whether thyroid hormone itself is abnormal.

Key terms:

Risk factor table: thyroid findings in older adults

FindingAssociationTreatable / modifiable?Evidence strength
Mild TSH rise with ageUpper limit of normal roughly doubles by 80+; often normal for ageUsually doesn't need treatmentStrong (large population study)
Subclinical hypothyroidism, TSH below 10No increase in heart events at 4.5-6.9; treatment didn't improve symptoms in over-65sUsually monitor rather than treatStrong (large trial and pooled cohorts)
Subclinical hypothyroidism, TSH 10 or higherHigher risk of heart disease events (HR 1.89) and heart disease death (HR 1.58)Treatment often considered; benefit on heart outcomes unprovenModerate (pooled cohorts)
Low TSH (subclinical hyperthyroidism)About twice the risk of atrial fibrillationYes: often from too much thyroid medication; dose adjustmentModerate to strong (meta-analysis)
Overt hypothyroidism or hyperthyroidismWide-ranging effects on heart, bone, mood and metabolismYes: established treatmentsStrong

What the research shows

"Normal" TSH changes with age. A large US population study found that TSH shifts upwards with age. The upper limit of normal was 3.56 mIU/L in people aged 20-29, but 7.49 mIU/L in those aged 80 and over. The authors estimated that about 70% of older people with a TSH above 4.5 mIU/L were actually within the normal range for their age, and warned that subclinical hypothyroidism may be substantially over-diagnosed unless age-specific ranges are used (see How Hormones Change With Age).

Treating mild subclinical hypothyroidism in older adults didn't help. The TRUST trial randomised 737 adults aged 65 and over (average age 74) with subclinical hypothyroidism, a TSH of 4.6-19.99 mIU/L with normal free T4, to levothyroxine or placebo. After one year, there was no difference between groups in thyroid-related symptoms or tiredness. The researchers concluded that levothyroxine provided no apparent benefit in older people with subclinical hypothyroidism. Most participants had only mildly raised TSH, so the results apply most clearly to that group.

Two-panel bar graphic: in adults over 65 with mildly raised TSH, levothyroxine and placebo gave the same symptom scores and the same tiredness scores, with no difference after 1 year (Stott et al., New England Journal of Medicine, 2017, TRUST trial, 737 adults aged 65+)

Higher TSH levels are linked to heart disease. A 2010 analysis pooled 11 cohort studies with 55,287 people to look at subclinical hypothyroidism and heart disease. The risk depended on how high TSH was:

Overall deaths weren't increased. This is one reason many guidelines treat a TSH of 10 or more differently from milder rises. However, these are observational findings, and trials haven't shown that treatment reduces heart disease in this group.

Bar chart of coronary heart disease event risk in subclinical hypothyroidism by TSH level: hazard ratio 1.00 at TSH 4.5-6.9, 1.17 (not significant) at 7.0-9.9, and 1.89 at 10-19.9, compared with normal thyroid function (Rodondi et al., JAMA, 2010, 11 cohorts, 55,287 people)

Too much thyroid hormone is a risk too. A low TSH, which is often caused by taking a higher dose of thyroid medication than needed, carries its own risks. A 2024 meta-analysis found that subclinical hyperthyroidism was linked to about twice the risk of atrial fibrillation (RR 1.99), an irregular heart rhythm that raises stroke risk. Too much thyroid hormone can also speed up bone loss. For older adults on thyroid medication, getting the dose right, and not over-treating, is important.

What you can actually change

1. Interpret results in context. A mildly raised TSH in an older person is often normal for their age. Ask whether your result is outside the age-appropriate range and whether free T4 is normal.

2. Repeat before acting. TSH can fluctuate, and a single abnormal result is often repeated after a few weeks or months before any decision about treatment.

3. Don't treat numbers alone. For mildly raised TSH without clear symptoms, monitoring is often better than treatment, especially in older adults. Symptoms such as tiredness, weight gain or low mood have many other causes.

4. If you take thyroid medication, get your dose checked. Regular monitoring helps avoid both under- and over-treatment. Over-treatment raises the risk of atrial fibrillation and bone loss.

5. Know when to take it seriously. A TSH of 10 or more, a low free T4, a low TSH with a high free T4, or clear symptoms of an under- or overactive thyroid need proper medical assessment and often treatment.

Decision guide, thyroid results in older adults: mildly raised TSH with normal free T4 is often normal for age, so repeat and monitor; TSH 10 or higher, discuss treatment with your doctor; low TSH on thyroid tablets, the dose may be too high, so get it checked; abnormal free T4 or clear symptoms need proper assessment

Recommendations

Practical notes

Thyroid disease is common with age, but so is over-diagnosis. TSH naturally rises as people get older, and many "borderline" results are normal for age. In older adults with mildly raised TSH, thyroid hormone tablets haven't improved symptoms. Higher TSH levels, of 10 or more, are linked to heart disease and deserve closer attention, while too much thyroid hormone raises the risk of atrial fibrillation. The practical message is to interpret thyroid tests in the context of age, symptoms and repeat results, and to avoid both under- and over-treatment.

References
  1. Surks MI, Hollowell JG. Age-specific distribution of serum thyrotropin and antithyroid antibodies in the U.S. population: implications for the prevalence of subclinical hypothyroidism. Journal of Clinical Endocrinology & Metabolism, 2007;92(12):4575-4582.
  2. Stott DJ, et al. Thyroid Hormone Therapy for Older Adults with Subclinical Hypothyroidism (TRUST). New England Journal of Medicine, 2017;376(26):2534-2544.
  3. Rodondi N, et al. Subclinical hypothyroidism and the risk of coronary heart disease and mortality. JAMA, 2010;304(12):1365-1374.
  4. Singh, et al. Subclinical thyroid dysfunction and the risk of incident atrial fibrillation: a systematic review and meta-analysis. PLOS ONE, 2024.

Have you been told your thyroid
result is "borderline"?

Take the free Aevum Protocol assessment to see how your stress & hormones and 6 other longevity domains are performing — and get a personalised 90-day plan.

Take the Free Assessment →

Related articles