Do NSAIDs Cause Osteoporosis? The Impact on Bone Health

Regular NSAID use does not appear to cause osteoporosis in the way that, say, long-term corticosteroid therapy does. Multiple observational studies of older adults have found that daily NSAID users actually tend to have slightly higher bone mineral density than non-users, which is the opposite of what you’d expect from a drug that weakens bones. The real concern with NSAIDs and bone health lies elsewhere: in their ability to slow fracture healing, to potentially interfere with osteoporosis medications, and in a biological mechanism that is genuinely double-edged. The story turns out to be far more nuanced than a simple yes or no.

How NSAIDs Interact With Bone Biology

To understand why the evidence is so contradictory, you need to know one thing about how bones maintain themselves. Bone is constantly being broken down and rebuilt in a process called remodeling. A molecule called prostaglandin E2 (PGE2) plays a central role in this cycle. PGE2 stimulates both the cells that tear down old bone and the cells that lay down new bone. It does this through different receptor subtypes, and one receptor in particular, called EP4, appears to be the main driver of new bone formation. In mouse studies, activating that receptor restored bone mass lost from immobilization or estrogen deficiency.¹

NSAIDs work by blocking cyclooxygenase (COX) enzymes, which are required to produce prostaglandins including PGE2. That means NSAIDs suppress both the bone-destroying and the bone-building sides of remodeling simultaneously. Studies in mice genetically lacking the COX-2 enzyme found that fracture healing was severely delayed, with a high rate of non-union, and that adding PGE2 back to bone marrow cell cultures completely rescued the defect in bone formation.² This dual role of PGE2 is why researchers have raised concerns about NSAIDs and bone remodeling during conditions like osteoarthritis, where the skeleton is already under stress.³

The biology, in other words, gives you a plausible reason to worry. But biological plausibility and what actually happens in living humans are not always the same thing, and this is one of those cases where the lab bench and the clinic tell different stories.

What Bone Density Studies Show

If NSAIDs were meaningfully eroding bone, you’d expect people who take them regularly to have lower bone mineral density (BMD). Most observational studies have found the opposite. A study of older women in Rancho Bernardo, California, found that those using propionic acid NSAIDs (a class that includes ibuprofen and naproxen) had higher BMD at all five skeletal sites measured, and statistically higher BMD at the lumbar spine, femoral neck, and total hip. The effect persisted even after excluding women with osteoarthritis and adjusting for other factors. Women who combined these NSAIDs with estrogen therapy had the highest BMD of all, suggesting an additive benefit.⁴

Another study found that current use of COX-2 selective NSAIDs combined with aspirin was associated with roughly 4-5% higher BMD at the whole body and total hip.⁵ A larger study following women for a decade through the Danish Osteoporosis Prevention Study found no meaningful difference in BMD changes between NSAID users and non-users, though there was a small trend toward higher spinal BMD gain in those taking NSAIDs.⁶ And a study of daily aspirin and NSAID users among older women found a multiply adjusted BMD increase of about 1-3% at the hip and spine.

So the density data, taken at face value, are reassuring and in some cases mildly positive. This does not mean NSAIDs are bone-building drugs. It more likely reflects a mix of biology (suppressing bone resorption slightly outweighing suppressed bone formation in steady-state remodeling) and the fact that people who take NSAIDs tend to be more physically active, since the drugs help them manage pain from conditions like arthritis.

The Fracture Risk Puzzle

Density is only part of the picture. What people actually care about is whether they’ll break a bone. Here, the evidence gets murkier. The Danish study mentioned above found that NSAID-exposed women had about 44% more fractures after adjustment, even though their BMD was not lower.⁶ That sounds alarming until you ask why someone is taking NSAIDs in the first place. People who use these drugs regularly often have painful musculoskeletal conditions that themselves increase fall risk. Pain, stiffness, impaired balance, and reduced mobility all raise the odds of a fracture independent of bone quality.

A large Dutch pharmacy-record study involving over 280,000 fractures concluded that while raw NSAID use was associated with roughly 47% more non-vertebral fractures, this likely reflected the underlying conditions rather than a drug effect on bone. There was no meaningful difference between regular and occasional users, and the specific NSAID used didn’t matter much. The researchers concluded the results were “not supportive of clinically significant effects of NSAIDs on bone metabolism.”⁷ Another study that separated different NSAIDs found substantial variation: ibuprofen was associated with about double the fracture risk at low doses, while celecoxib showed no increase at all. The authors noted that falls, rather than bone weakening, could explain much of the increased risk with certain drugs.⁸

This is a pattern that shows up repeatedly in NSAID research and in observational drug studies generally. The people who take the drug are not identical to the people who don’t, and the reasons they take it may matter more than the drug itself. Researchers call this confounding by indication, and it makes it very difficult to draw clean causal conclusions from fracture-risk data.

Where NSAIDs Clearly Affect Bone: Fracture Healing

The strongest and most clinically relevant concern about NSAIDs and bone health is not about causing osteoporosis. It’s about what happens when you take them after a fracture. A meta-analysis pooling data from over 645,000 patients found that NSAID use around the time of a fracture roughly doubled the odds of adverse bone healing events in adults.⁹ That sounds decisive, but a closer look at the data reveals that duration of use matters enormously.

A separate meta-analysis of randomized controlled trials found that short-term NSAID use (under two weeks) was not significantly associated with non-union, while long-term use carried over five times the odds of non-union compared to controls.¹⁰ A practice management guideline from the Eastern Association for the Surgery of Trauma reinforced this, finding that acute NSAID use under 30 days showed no association with non-union across 18 different bones studied.¹¹

The practical takeaway is straightforward: a few days or even a couple of weeks of ibuprofen or naproxen after a broken bone is probably fine. Months of continuous NSAID use while a fracture is trying to heal is a different matter. If you’ve had a fracture and your doctor recommends something other than NSAIDs for ongoing pain, this is likely why.

Spinal Fusion and Surgical Bone Healing

The same duration-dependent pattern appears in spinal fusion surgery, where new bone must grow to fuse vertebrae together. In the early 2000s, the spine surgery literature broadly warned against postoperative NSAID use. But a cross-disciplinary review found that nearly all human studies published after 2005 showed that short-term use (under two weeks) had no effect on fusion rates. The dose dependency seen with two-week courses was not present when NSAIDs were used for just 48 hours after surgery.¹²

A systematic review of human trials reached a similar conclusion: the available evidence did not show strong support for the idea that NSAIDs used for post-fracture or post-fusion pain therapy lead to increased non-union. The researchers noted, however, that the data were so varied and conflicting that no firm clinical recommendation could be made in either direction.¹³ This is an area where surgeons often have strong individual preferences, and those preferences sometimes outpace the evidence.

COX-2 Selective Drugs May Be the Bigger Concern

Not all NSAIDs are created equal when it comes to bone. The COX-2 enzyme is the one most directly involved in bone repair and formation, as the mouse knockout studies showed. That raises a logical question: do drugs that selectively block COX-2 (like celecoxib) interfere with bone healing more than nonselective NSAIDs (like ibuprofen or naproxen, which block both COX-1 and COX-2)?

Some evidence says yes. A study examining fracture non-union rates found an association between COX-2 inhibitors and non-union, but not with nonselective NSAIDs. The authors noted this was at least partially consistent with animal studies showing worse fracture healing outcomes with COX-2 selective drugs.¹⁴ This creates an ironic situation: COX-2 inhibitors were developed partly to be gentler on the stomach than older NSAIDs, but they may be harder on healing bone. The fracture-risk data, on the other hand, showed celecoxib associated with lower fracture risk than ibuprofen, which means the fracture-healing concern is distinct from the fracture-risk question.

The Animal-to-Human Translation Problem

A large reason the NSAID-bone debate remains unresolved is that much of the most alarming evidence comes from animal models, and translating rodent bone biology to humans is unreliable in this area. A systematic review and meta-analysis of animal studies confirmed negative effects of NSAIDs on bone healing in rodents but emphasized that the clinical literature in humans is thin and contradictory. Some human retrospective studies confirmed negative effects after hip fracture, while others showed no effect, and one prospective trial actually found beneficial effects among patients with wrist fractures treated with casting.¹⁵

A critical analysis of the entire body of literature put it bluntly: the data across animal, in vitro, and human studies present “great diversity,” with even studies using identical parameters producing opposing results.¹⁶ Rodent fracture healing is faster and involves slightly different cellular dynamics than in humans, and drug doses used in animal experiments often exceed what people actually take. This doesn’t mean the animal data are worthless, but it does mean that someone who reads a mouse study and concludes “ibuprofen will ruin my bones” is leaping across a wide evidentiary gap.

NSAIDs May Blunt Osteoporosis Medications

Perhaps the most practically important finding for people already being treated for osteoporosis is that NSAIDs may interfere with bisphosphonates, the most commonly prescribed class of osteoporosis drugs. An exploratory analysis from a randomized trial of clodronate (a bisphosphonate) found that the drug’s ability to reduce fracture risk was largely negated in women who were also taking NSAIDs. Among women not using NSAIDs, clodronate reduced osteoporotic fracture risk by about 29%. Among NSAID users, there was essentially no benefit. BMD loss at the hip and femoral neck was also significantly greater in women taking both clodronate and NSAIDs compared to those on clodronate alone.¹⁷

This was a post hoc analysis from a single trial, which means it needs to be confirmed in studies designed specifically to test this interaction. But the mechanism is biologically plausible: bisphosphonates work partly by being incorporated into bone during active remodeling, and if NSAIDs suppress that remodeling, there may be less active bone surface for the bisphosphonate to reach. If you’re on a bisphosphonate for osteoporosis and regularly using NSAIDs for arthritis pain, this is a finding worth discussing with your doctor.

NSAIDs and Vitamin D

Another interaction that has gotten less attention involves vitamin D. A study of adults with multiple health conditions found that NSAID users were significantly less likely to reach adequate vitamin D levels after 12 months of supplementation compared to non-users. The researchers noted that chronic inflammation or pharmacological interference with vitamin D pathways could explain the limited benefit. Since vitamin D is critical for calcium absorption and bone mineralization, this is another indirect pathway through which chronic NSAID use could affect bone health, even if the drugs aren’t directly weakening bone tissue.

Aspirin as a Special Case

Aspirin is technically an NSAID, but it behaves differently from others in several ways, including its effects on bone. A review of dose-dependent effects found that low-dose aspirin (the kind taken daily for cardiovascular protection) likely benefits bone mass by stimulating bone-building cells and inhibiting bone-resorbing cells through a mechanism that doesn’t even depend on the COX pathway. Higher-dose aspirin and other NSAIDs have more ambiguous effects because they suppress both sides of bone remodeling simultaneously.¹⁸

A cross-sectional study of patients with type 2 diabetes found that low-dose aspirin users had slightly (though not statistically significantly) higher BMD at the spine, femoral neck, and Ward’s triangle compared to non-users.¹⁹ The numbers were modest and didn’t reach statistical significance, but the trend was consistently in the protective direction. If you’re taking a daily baby aspirin for your heart, there’s no reason to think it’s hurting your bones, and some reason to think it may be mildly helpful.

Postmenopausal Women and Ibuprofen

Postmenopausal women are the population most concerned about osteoporosis, and the NSAID evidence in this group is particularly interesting. A randomized controlled trial of low-dose ibuprofen combined with resistance training in postmenopausal women found no effect on BMD at the lumbar spine, femoral neck, or total body. However, there was a potentially meaningful finding at Ward’s region of the femoral neck, a clinically vulnerable area: the control group lost about 6% of BMD there, while the group receiving ibuprofen and stretching gained about 3%.²⁰ The researchers concluded that ibuprofen may have some beneficial bone effects for this group.

Separately, an animal study found that flufenamic acid, a less commonly used NSAID, inhibited bone-resorbing cells and protected against estrogen-deficiency bone loss in mice.²¹ Whether this translates to humans remains to be seen, but it reinforces the idea that the relationship between NSAIDs and bone is not simply destructive. Some of these drugs, at some doses, may actually support bone in people who are losing it due to estrogen deficiency.

When You Take an NSAID May Matter

An unexpected wrinkle in the NSAID-bone story comes from chronobiology. A study in mice found that giving an NSAID during the animals’ active phase (equivalent to daytime for humans) resulted in significantly better fracture healing than giving the same drug during the resting phase. Mice treated during their active phase had larger fracture calluses, greater mineralized bone area, and fewer bone-resorbing osteoclasts.²² This is a mouse study, so the direct translation to humans remains uncertain, but it opens the door to the possibility that the timing of NSAID doses, not just their total amount, could influence bone outcomes. It also helps explain why the same drug at the same dose can produce different results in different studies where dosing schedules vary.

If this finding holds in human research, it would mean that taking your ibuprofen in the morning rather than before bed could make a real difference to how a healing fracture responds. For now, this is early-stage science, but it’s the kind of practical detail that could eventually change clinical recommendations.