Common Back-Pain Drug May Be Linked to Higher Dementia Risk, Large Study Finds

A common back-pain drug may be linked to higher dementia risk, according to a large observational study that has raised new questions about the long-term cognitive safety of gabapentin. The medication is widely prescribed for nerve-related pain and other chronic pain conditions, although its FDA-approved uses include postherpetic neuralgia and partial-onset seizures.

The headline deserves attention, but it also needs context. The study found an association, not proof that gabapentin causes dementia. Researchers could not randomly assign people to take the medication for years, and several factors—including chronic pain itself—may influence dementia risk.

That distinction has become even more important in 2026. A newer systematic review and meta-analysis pooled five observational studies involving more than 325,000 participants. It found modestly higher estimated odds of Alzheimer’s disease and related dementias among people using gabapentinoids. However, the authors rated the overall certainty of evidence as low and specifically said the available data cannot establish causation.

So, what did researchers actually find, how concerning are the numbers, and what should people taking gabapentin do with this information?

What the Large Gabapentin and Dementia Study Actually Found

Researchers Followed More Than 26,000 Matched Patients

The study that triggered much of the recent discussion appeared in Regional Anesthesia & Pain Medicine in 2025. Researchers used the TriNetX database to examine adults with chronic low-back pain whose medical records covered the period from 2004 through 2024.

After matching patients according to demographics, medical conditions, and pain medications, the researchers analyzed 26,416 adults. They excluded people who already had dementia, epilepsy, stroke, cancer, or previous gabapentin use before the study period.

The researchers then compared people prescribed gabapentin with similar patients who had chronic low-back pain but did not receive the drug.

Among patients with six or more gabapentin prescriptions, the study found a 29% higher relative incidence of dementia and an 85% higher relative incidence of mild cognitive impairment compared with matched patients who did not receive gabapentin.

Those numbers sound dramatic. However, relative risk does not tell you a person’s absolute chance of developing dementia. A 29% relative increase can represent a relatively small absolute difference when the underlying risk remains low.

Additionally, medical-record studies cannot perfectly capture every factor that separates people who receive a medication from people who do not.

That limitation matters because patients who require repeated treatment for chronic pain may differ from comparison patients in important ways that electronic health records cannot fully measure.

The Younger-Adult Finding Drew Particular Attention

One of the most striking findings involved adults between ages 18 and 64.

In that subgroup, gabapentin prescription was associated with roughly twice the relative risk of dementia and about 2.5 times the relative risk of mild cognitive impairment compared with matched patients who did not receive gabapentin.

Researchers also reported a pattern related to prescription frequency. Patients who received 12 or more prescriptions had higher recorded rates of dementia and mild cognitive impairment than patients receiving three to 11 prescriptions.

At first glance, that apparent dose-response pattern may strengthen concern. If greater exposure accompanies greater risk, researchers often consider that pattern worth investigating.

Still, prescription count does not equal a precise biological dose. A person with many prescriptions may have more severe pain, longer-lasting illness, more healthcare encounters, or other conditions that independently affect cognition.

Furthermore, dementia remains relatively uncommon in younger adults. Therefore, a large relative difference in a younger group does not necessarily mean that large numbers of younger gabapentin users will develop dementia.

The study authors themselves concluded that their findings support closer cognitive monitoring and further research into whether gabapentin plays a causal role. They did not establish that gabapentin directly causes dementia.

Why the Findings Do Not Prove Gabapentin Causes Dementia

Observational Studies Can Reveal Signals but Not Cause and Effect

A common back-pain drug may be linked to higher dementia risk, but the word “linked” carries most of the scientific weight.

The study was retrospective and observational. Researchers examined existing medical records rather than randomly assigning participants to gabapentin or another treatment.

That design allows scientists to study very large populations over long periods. However, it also leaves open the possibility of residual confounding.

For example, chronic pain itself may relate to dementia risk.

A 2026 systematic review examining chronic pain and dementia pooled data involving more than two million participants. It found that people with chronic pain had a higher estimated dementia risk than people without chronic pain.

That creates a difficult research problem.

People receiving repeated gabapentin prescriptions may have more persistent neuropathic symptoms, worse sleep, reduced physical activity, depression, anxiety, greater medication burden, or more severe underlying disease. Some of those factors may independently influence cognitive health.

Researchers can statistically adjust for known differences, and the 2025 study used propensity-score matching to do so. Nevertheless, no statistical method can guarantee that every meaningful difference between groups disappears.

Consequently, the study provides an important safety signal rather than definitive proof of harm.

Other Studies Have Produced Conflicting Results

The scientific literature does not point in only one direction.

A 2023 population-based study from Taiwan examined gabapentin or pregabalin exposure and reported a higher rate of dementia among exposed patients. The association appeared stronger with greater cumulative exposure.

However, another large study published in 2024 examined more than 200,000 adults aged 50 and older with chronic pain. That analysis did not find a statistically significant increase in dementia risk associated with gabapentin use, and researchers found no clear difference between lower and higher cumulative doses.

That conflict helps explain why the question remains unsettled.

The newest broad assessment adds useful perspective. A 2026 meta-analysis combined five observational studies involving 325,245 participants. Overall, gabapentinoid exposure was associated with approximately 28% higher estimated odds of Alzheimer’s disease and related dementias.

Yet the studies differed substantially from one another. Statistical heterogeneity reached more than 90%, which means their findings varied considerably.

More importantly, when researchers restricted the analysis to studies judged to have a low risk of bias, the association weakened and the confidence interval included the possibility of no increased risk. The authors therefore rated the certainty of evidence as low and emphasized that residual confounding and differences between patients prevent causal conclusions.

In other words, the evidence warrants further research and thoughtful prescribing, but it does not justify saying that gabapentin has been proven to cause dementia.

What Patients Should Know About Gabapentin Safety

Gabapentin Already Has Recognized Nervous-System Side Effects

Gabapentin is not a new medication. U.S. regulators first approved it in 1993.

Current FDA prescribing information lists gabapentin for postherpetic neuralgia in adults and as adjunctive therapy for partial-onset seizures. Doctors also prescribe it for other conditions in clinical practice, including several types of nerve-related pain.

The medication already has well-recognized effects on the nervous system.

According to current FDA labeling, common adverse reactions include dizziness and somnolence. The label also warns that sedation can impair driving or the operation of machinery.

Additionally, the FDA warns about respiratory depression when gabapentin is combined with central nervous system depressants such as opioids or when patients have underlying respiratory impairment.

These established effects are different from dementia.

Feeling sleepy, foggy, dizzy, or less attentive while taking a medication does not mean a person has developed a neurodegenerative disease. Likewise, temporary cognitive symptoms do not prove that permanent damage has occurred.

Still, the newer research gives clinicians another reason to pay attention when patients report persistent problems with memory, concentration, or thinking during long-term treatment.

Medication review becomes particularly useful when a person takes several drugs that can cause sedation or cognitive effects.

People Taking Gabapentin Should Not Stop It Suddenly

The new dementia research does not mean everyone taking gabapentin should discontinue it.

Abruptly stopping the medication can create problems of its own. Current FDA prescribing information warns that rapid discontinuation may increase seizure risk and can produce withdrawal symptoms or other adverse effects.

Therefore, anyone concerned about long-term gabapentin use should discuss the issue with the clinician who prescribed it rather than changing the dose independently.

A useful medication review can consider several questions.

Does gabapentin still meaningfully reduce the original pain or symptoms? Is the current dose still necessary? Has kidney function changed? Does the patient take opioids, sedatives, sleep medications, or other drugs that may increase sedation? Have new memory or concentration problems appeared?

Those questions allow clinicians to weigh benefit against risk for the individual patient.

The answer will not look the same for everyone.

Someone receiving clear relief from severe neuropathic pain may reach a different decision than someone who has taken gabapentin for years with little noticeable benefit.

Likewise, treatment alternatives depend on why the medication was prescribed in the first place. Physical therapy, exercise-based rehabilitation, topical treatments, other medications, psychological approaches to chronic pain, injections, or other interventions may help certain patients, but none provides a universal replacement.

The safest approach involves individualized medical review rather than a blanket decision based on one study.

How to Interpret Dementia-Risk Headlines More Carefully

Relative Risk Does Not Tell the Whole Story

A common back-pain drug may be linked to higher dementia risk, but a percentage increase can sound more frightening than the underlying numbers justify.

Researchers commonly report relative risk because it makes comparisons between groups easier.

Suppose a hypothetical condition affects 1 person in 1,000 in one group. A 30% relative increase would raise that figure to approximately 1.3 people per 1,000, not 301 people per 1,000.

That example does not represent the exact gabapentin data. Instead, it illustrates why readers need both relative and absolute risk whenever possible.

Age matters enormously as well.

Dementia risk rises sharply later in life. Therefore, a doubling of relative risk among younger adults can still correspond to a much smaller absolute number of cases than a modest relative increase among older adults.

Researchers also need to determine whether the younger-adult finding can be replicated in other populations.

The 2026 meta-analysis found that studies enrolling younger populations tended to produce larger risk estimates. However, the difference between age groups did not reach statistical significance.

Therefore, the observation remains interesting rather than definitive.

Future prospective research will need to examine medication duration, actual dosage, kidney function, pain severity, other medications, sleep, depression, physical activity, cardiovascular risks, and many other potential influences.

The Most Reasonable Message Is Caution, Not Panic

The evidence available now supports a middle position.

Ignoring the research would go too far because multiple observational datasets have produced a potential cognitive safety signal. The 2025 chronic low-back-pain study found higher recorded rates of dementia and mild cognitive impairment, and the 2026 meta-analysis found a modest overall association across observational studies.

However, declaring gabapentin a proven cause of dementia would also go beyond the evidence.

The newest meta-analysis explicitly described the certainty of evidence as low. Researchers noted substantial differences between studies and said confounding prevented causal inference.

Meanwhile, at least one large 2024 study of chronic-pain patients did not identify a significant dementia association.

Therefore, the most useful response involves appropriate monitoring and periodic review.

People who take gabapentin long term can ask their clinician whether the medication remains necessary and whether the current dose still provides meaningful benefit.

Clinicians can also monitor for cognitive complaints, sedation, balance problems, kidney-related dosing issues, and interactions with other medications.

Most importantly, patients should not interpret a news headline as an instruction to stop treatment.

Research changes clinical understanding gradually. One observational study can raise an important question. Repeated studies, better-designed comparisons, prospective research, and eventually stronger evidence determine how that question should change routine medical practice.

Frequently Asked Questions

Does gabapentin cause dementia?

Current research does not prove that gabapentin causes dementia. Several observational studies have found an association, while other research has not. A 2026 meta-analysis found modestly higher estimated odds of dementia among gabapentinoid users but rated the evidence as low certainty and stated that causal conclusions cannot be made.

How much higher was the risk in the 2025 study?

Among adults with chronic low-back pain who received six or more gabapentin prescriptions, researchers reported a 29% higher relative incidence of dementia and an 85% higher relative incidence of mild cognitive impairment compared with matched patients who did not receive gabapentin. These figures represent relative associations, not a prediction of an individual patient’s absolute risk.

Why did younger adults appear to have a larger association?

The 2025 study reported particularly large relative risk estimates among adults ages 18 to 64. Researchers have not established why. Differences in baseline risk, prescribing patterns, underlying illness, healthcare use, or unmeasured confounders may contribute. A 2026 meta-analysis also observed larger estimates in studies with younger populations, although the difference was not statistically conclusive.

Should I stop taking gabapentin because of this research?

Do not stop or rapidly reduce gabapentin solely because of these findings without discussing the change with the prescribing clinician. FDA labeling warns that abrupt or rapid discontinuation can increase seizure risk and may cause withdrawal-related adverse effects.

What symptoms should I discuss with my doctor?

Persistent memory problems, increasing confusion, unusual sleepiness, dizziness, concentration difficulties, balance problems, or significant changes in daily functioning deserve discussion with a healthcare professional. However, these symptoms have many possible causes, including other medications and medical conditions, so they do not automatically indicate dementia.

Conclusion

A common back-pain drug may be linked to higher dementia risk, and the emerging research on gabapentin deserves careful attention.

The 2025 study involving 26,416 matched adults with chronic low-back pain found higher recorded rates of dementia and mild cognitive impairment among people receiving repeated gabapentin prescriptions. The association appeared particularly large among adults younger than 65.

However, observational research cannot prove that gabapentin caused those outcomes.

Chronic pain itself may relate to dementia risk, and patients who receive repeated gabapentin prescriptions may differ from untreated patients in ways researchers cannot completely measure. Other studies have also produced conflicting findings.

The broader evidence now available in 2026 reflects that uncertainty. A meta-analysis of five observational studies and more than 325,000 participants found a modest overall association between gabapentinoid use and Alzheimer’s disease-related dementias, but researchers judged the evidence to be low certainty. When they focused on studies with lower risk of bias, the association became less conclusive.

For patients, the practical message is not panic and not abrupt discontinuation.

Instead, long-term gabapentin treatment deserves periodic review. Patients and clinicians can consider whether the medication still provides meaningful benefit, whether the dose remains appropriate, whether other sedating medications are involved, and whether new cognitive symptoms have appeared.

The research raises an important safety question. It does not yet deliver a final answer.