Source: United Kingdom – Executive Government & Departments

July 10, 2025

An observational study published in Regional Anesthesia & Pain Medicine looks at gabapentin prescription for chronic pain and the risk of dementia and cognitive impairment. 

Prof Ian Maidment, Professor in Clinical Pharmacy, Aston University, said:

“This study found an association between gabapentin and dementia. It was an observation study and therefore conclusions about causality cannot be drawn. Furthermore, the research did not control for length of treatment or dose of gabapentin. Other similar recent studies have failed to find a link. Therefore, overall the jury is out on whether gabapentin causes dementia.”

Prof Martin Prince, Professor of Epidemiological Psychiatry, Institute of Psychiatry, Psychology & Neuroscience, King’s College London, said:

“This is an interesting pharmaco-epidemiological study, using a retrospective (historical) cohort design, and reporting a significant increased risk of dementia incidence among those prescribed gabapentin for chronic low back pain. The authors are right to stress that they are reporting an association, and not necessarily a causal link. As I will discuss, confounding and reverse causality are tenable explanations for the observed effects. The strengths of the study include a large sample size, a long surveillance period, a state of the art propensity matching on a large number of potential confounding factors, and additional confounders controlled for in the analysis.

“There are some issues apparent with the research.

1. The most significant is that the title of the paper (which refers only to chronic low back pain) appears to be misleading. Closer reading of the research methods reveals that the inclusion criteria were actually “chronic pain (ICD- 10- CM G89.29), chronic pain syndrome (ICD- 10- CM G89.4), lumbar radiculopathy (ICD- 10- CM M54.16), or chronic low back pain (ICD- 10- CM M54.5)”. This is a much broader group, and would include, among other conditions, post-herpetic neuralgia and painful diabetic and other peripheral neuropathies, which are particularly common indications for prescription of gabapentin. There is an inconclusive literature linking shingles episodes to an increased risk of dementia (and suggesting that the Shingrix vaccine may lower the risk), and recent research using the UK biobank linking multisite chronic pain with the incidence of dementia and hippocampal atrophy (1). None of this body of potentially relevant research was referenced or discussed in the current paper. Confounding by indication therefore remains a distinct possibility – the condition that gabapentin is treating, rather than the drug itself being responsible for the increased dementia risk. Of note is that use of gabapentin would likely be reserved for those with more severe pain, and therefore a more severe underlying condition. While the investigators clearly sought to limit the potential for confounding by indication (for example by excluding from consideration individuals prescribed gabapentin for epilepsy), their efforts are likely to have been only partially successful. Reverse causality must always be considered in dementia cohort studies given the 20 years or more interval between the earliest detectable signs of Alzheimer’s Disease (from neuroimaging and blood biomarkers) and clinical onset. Those diagnosed with dementia would not, strictly speaking, have been dementia-free at cohort inception. It is possible that the CNS effects of Alzheimer’s disease modulate pain processing and appreciation, leading to more complaints of more severe pain, at multiple sites. Hence that Alzheimer’s disease caused the pain, and, ultimately the Gabapentin prescription, not vice versa. Or that there is an underlying common cause, for example inflammation, that is driving both the neurodegeneration and the neuropathic pain.
2. I could not understand why mild cognitive impairment, frontotemporal dementia and dementia with Lewy Bodies were listed as factors that were propensity matched at baseline, when the onset of mild cognitive impairment and all cause dementia were outcomes of interest and hence presumably excluded at baseline? It isn’t very clearly explained. It is possible that those with MCI at baseline were left in when assessing dementia as an outcome, but excluded when assessing MCI as an outcome. But leaving FTD and DLB cases in at baseline (with the implicit assumption that they could be considered as remaining at risk for developing AD or vascular dementia), seems to be an odd approach.  
3. Since, apparently, separate diagnostic codes for Alzheimer’s Disease and Vascular dementia were available, I am surprised that no attempt was made to explore whether the association with gabapentin prescription was similar or different across the two sub-types. The relationship of gabapentin use to both AD polygenic risk scores, and AD-specific blood biomarkers would also be another area for future research.”

1. W. Zhao, L. Zhao, X. Chang, X. Lu, & Y. Tu, Elevated dementia risk, cognitive decline, and hippocampal atrophy in multisite chronic pain, Proc. Natl. Acad. Sci. U.S.A. 120 (9) e2215192120, https://doi.org/10.1073/pnas.2215192120 (2023).

Prof Tara Spires-Jones, Director of the Centre for Discovery Brain Sciences at the University of Edinburgh, Group Leader in the UK Dementia Research Institute, and Past President of the British Neuroscience Association said:

“This study by Eghrari and colleagues examined medical records from over 24,000 people in the US and found that prescription of the medication gabapentin for chronic pain was associated with a higher risk of developing dementia.  While authors used statistical methods to try and account for other risk factors, this type of study cannot prove that gabapentin was the cause of increased dementia risk.  One very important factor that was not examined in this study is levels of physical activity.  People with chronic pain requiring gabapentin may have been less physically active, which is a known risk factor for developing dementia.”

Prof Sir John Hardy, Group Leader at the UK Dementia Research Institute at UCL, said:

“While this is interesting, one has to worry that these types of findings are artefactual and result (for example) from a marginal acute effect on cognitive performance rather than effects on the underlying disease.”

Dr Leah Mursaleen, Head of Clinical Research at Alzheimer’s Research UK, said:

“Research shows that nearly half (45%) of dementia cases could be prevented or delayed if 14 health and lifestyle risk factors are addressed by people and society. At the moment, there’s not enough evidence to suggest pain medications are linked to higher dementia risk, but this research gives us interesting insights.

“This large observational study looked at health records of over 26,000 people in the US diagnosed with chronic lower back pain and who were prescribed gabapentin within a 10-year period. They found gabapentin prescription was associated with an increased risk of dementia and mild cognitive impairment, especially in people under the age of 65.

“Some of the strengths of this research was the large sample size and some dementia risk factors were considered, such as age and high blood pressure.

“However, this study only shows an association between gabapentin prescriptions and mild cognitive impairment or dementia, so we do not know if the medication is directly causing the higher risk. Gabapentin dosage wasn’t recorded, and there was no information on how long people were on the medication. 

“Because this study only used health records of people with chronic pain, we cannot rule out other factors that might be influencing the findings. And previous studies looking at people prescribed gabapentin for other conditions like seizures, didn’t show a link between the medication and higher dementia risk.

“Managing chronic pain is very important and if anyone has any concerns about the medication they are receiving, they should speak to their doctor.”

‘Risk of dementia following gabapentin prescription in chronic low back pain patients’ by Nafis B Eghrari et al. was published in Regional Anesthesia & Pain Medicine at 23:30 UK time on Thursday 10 July. 

DOI: 10.1136/rapm-2025- 106577

Declared interests

Prof Ian Maidment: No declarations of interest

Prof Martin Prince: No conflicts of interest to report

Prof Tara Spires-Jones: I have no conflicts with this study but have received payments for consulting, scientific talks, or collaborative research over the past 10 years from AbbVie, Sanofi, Merck, Scottish Brain Sciences, Jay Therapeutics, Cognition Therapeutics, Ono, and Eisai. I am also Charity trustee for the British Neuroscience Association and the Guarantors of Brain and serve as scientific advisor to several charities and non-profit institutions.

Prof Sir John Hardy: Have consulted for Eisai

For all other experts, no reply to our request for DOIs was received.