Common bone-strengthening drugs were linked to fewer dementia diagnoses in Hong Kong

Osteoporosis, the thinning of bones with age, and dementia often go together in older people. Bisphosphonates are the most widely used osteoporosis drugs. They work by shutting down osteoclasts, the cells that break bone down. The common ones contain nitrogen (alendronate, ibandronate, risedronate and zoledronate) and block an enzyme in a chain of reactions that also produces cholesterol. Some studies have found more of the products of this pathway in the brains of people with Alzheimer's disease, and in mice the bisphosphonate alendronate improved memory and reduced brain inflammation. That raised the question of whether these bone drugs might also be linked to less dementia in people.

This study was published in 2025; it is getting attention now through a recent press release.

What they did

A team at the University of Hong Kong used CDARS, a database of all public hospitals and outpatient clinics in Hong Kong, which covers more than 80% of hospital admissions there. They found people aged 60 and over who were first diagnosed with osteoporosis or a fragility fracture (a break of the spine, hip, wrist or upper arm from a minor fall) between 2005 and 2020. People with cancer or with dementia already were left out. That left 121,492 people, with an average age of 77; 72.5% were women.

Many never received a bone drug: only 18% did by the end of 2020. The researchers made two comparisons:

• Bisphosphonate users vs untreated patients: 10,833 matched pairs.
• Bisphosphonate users vs users of other osteoporosis drugs (such as denosumab or teriparatide): 3,080 matched pairs.

Each user was matched with a similar person by age, sex, calendar year, fracture type, illnesses and recent medicines, at the same point after diagnosis. Dementia diagnoses in the first six months after matching were excluded, because the disease develops slowly and such early cases are unlikely to be related to the drug. People were followed for a median of about three and a half to four years, up to the end of 2021.

What they found

• Fewer dementia diagnoses with bisphosphonates. Compared with untreated patients, users had a 16% lower rate of Alzheimer's and related dementias. Compared with users of other bone drugs, the rate was 24% lower.
• In absolute terms: after five years, about 2 fewer people in every 100 treated with a bisphosphonate had been diagnosed with dementia. The team calculates that about 48 people would need to be treated for five years for one fewer case.
• Robust to two checks. Results were similar when people were dropped from the analysis once they switched osteoporosis drugs, and when the excluded early period was extended from six months to twelve.
• Subgroups. Against untreated patients, the link held in women, in men and in people with hip fractures. Against other bone drugs, it was clear only in women and in people diagnosed with osteoporosis. Other subgroups were small, and formal tests did not find that sex or fracture type changed the result.

What to keep in mind

• Observational. Patients were not assigned to a drug at random, so the authors note that "residual confounding" may remain.
• No memory tests at the start. The database had no measure of thinking skills when treatment began. People whose memory was already slipping may have been less likely to be offered a bone drug, and also more likely to be diagnosed with dementia later. The six-month delay was meant to reduce this, but cannot remove it entirely.
• Missing information. The records lacked bone density, physical function, education, social isolation, air pollution and over-the-counter supplements such as vitamin D and calcium.
• Diagnoses from records. Dementia was identified from hospital codes; when the team checked a sample, about 81% of coded cases were confirmed.
• One population. The patients were in Hong Kong's public health system. Earlier small studies of zoledronate did not find better scores on memory tests, though they were short and involved fewer than 130 people.

Why it matters

Bisphosphonates are cheap, familiar prescription drugs, already taken by many older people with weak bones. If they also lowered dementia risk, that would be a welcome extra. This study, with both untreated and actively treated comparison groups, found a consistent link. But only randomized trials can show whether the drugs themselves affect the brain. Bisphosphonates are prescription medicines with their own side effects, and the study offers no basis for taking them to prevent dementia.