Thyroid hormone replacement therapy is commonly prescribed when blood tests show a dip in thyroid hormone levels, but new research suggests it may not always be the best choice for older adults. 

The long-term study found that people over 65 who take thyroid hormone replacement therapy have a higher risk of death than their peers who don't. 

"The message here is to be cautious and conservative when prescribing levothyroxine [thyroid hormone replacement]," said study author Dr. Jennifer Mammen, an assistant professor of medicine at the Johns Hopkins University School of Medicine in Baltimore.

"Older adults go through a lot of changes — circadian rhythm changes, sleep changes, levels of chronic inflammation change — and lower thyroid levels may be an adaptation of age. It may be the body's way of cooling the engine down. If you give thyroid hormone, you may be overriding those changes," she said.

Mammen cautioned that the study shouldn't prompt anyone to immediately stop taking prescribed medication. Instead, she recommended that people talk with their doctor. 

"Someone who's been on thyroid hormone replacement since they were 20 because their thyroid failed, may need to stay on it. But, someone who started taking it at age 75 because their thyroid levels are a little different, maybe they don't need to be on it," she said. 

Dr. Kashif Munir, director of the Center for Diabetes and Endocrinology at the University of Maryland, said that patients who have clear low thyroid levels (hypothyroidism) should be given treatment. In fact, it can be dangerous not to replace thyroid hormone when it's truly needed. 

However, Munir said there are patients who may have some symptoms of hypothyroidism — such as feeling tired, cold or having difficulty concentrating — and have borderline test results (also called subclinical hypothyroidism). For this group, he said, "it may not be beneficial to treat these levels, and this study suggests it might be harmful." 

Mammen had been scheduled to present her findings at the Endocrine Society annual meeting, which was canceled due to coronavirus concerns. The study will be published in a special supplemental section of the Journal of the Endocrine Society.

The study included information from a U.S. National Institute on Aging Study. It had data on more than 1,000 people over almost two decades (2003-2018). Study participants were 65 and older. 

All of the participants had at least one thyroid-stimulating hormone (TSH) and one thyroxine (T4) measurement during the study. Thyroxine is the thyroid hormone that is replaced with thyroid hormone therapy. When thyroxine levels dip, the body usually responds by increasing thyroid-stimulating hormone levels. 

When the researchers compared the data year over year, they found that adults taking thyroid hormone had 60% higher odds of dying compared to people who weren't taking the treatment. 

Mammen said there are limitations to the study. For example, the researchers don't know why people were put on thyroid hormone, so they have no way of knowing if the study volunteers had significant hypothyroidism or borderline hypothyroidism. They also didn't know how long the volunteers had been taking thyroid hormone replacement therapy. 

Dr. Carol Chiung-Hui Peng, an internal medicine physician at the University of Maryland Medical Center in Baltimore, said, "This is a significant study that provides further evidence on a clinically relevant question," but added that definitive conclusions can't be drawn from one study. 

Peng recently conducted a meta-analysis with Munir and other colleagues. It was published in the Journal of Clinical Endocrinology and Metabolism. The analysis included 27 past studies on thyroid disease with more than a million patients over 60. 

They found that not treating clear cases of hypothyroidism increased the risk of dying from any cause. They also found that if people with subclinical hypothyroidism didn't receive treatment, their risk of dying didn't increase. 

While one or two studies may not change the way doctors prescribe medications, all three experts thought it would be useful for researchers to look at whether or not the range of what's considered a normal thyroid level should change based on a person's age.