Study reveals potential drug links to Dementia risk

Analysis of millions of patient records suggests potential benefits of antimicrobials and harms of antipsychotics.

Study: Data-driven discovery of associations between prescription medications and dementia risk: a systematic review. Image credits: Orawan Pattarawimonchai/Shutterstock.com

From a recent study published in Alzheimer’s and dementiaa team of scientists systematically reviewed data-driven research on how prescription medications can influence dementia risk.

They analyzed the medical records of more than 130 million individuals and identified drug classes that may be associated with a higher or lower risk of dementia. These findings may help guide drug repurposing strategies and inform prevention efforts in dementia-related diseases.

Background

Dementia affects millions of people worldwide and is a significant global health problem, causing significant personal and economic burdens. Current treatments mainly target symptoms, but with limited success in altering disease progression.

Furthermore, efforts to develop effective disease-modifying therapies also face numerous challenges due to the complex mechanisms underlying dementia.

Progress in identifying potential pathogenic mechanisms, such as protein misfolding and inflammation, highlights potential therapeutic targets. Recently, repurposing existing drugs that have already been approved for use in other medical conditions offers a promising opportunity.

These medications may interact with dementia-related biological pathways, producing benefits beyond their original indications.

Although certain classes of drugs, including anti-inflammatory agents and vaccinations, have shown potential, findings remain inconsistent, necessitating systematic research to uncover consistent patterns that support effective prevention and therapeutic strategies.

About the study

The current systematic review used a comprehensive search strategy, focusing on studies that examined associations between medications prescribed for different diseases and dementia risk, using a data-driven approach. The team searched numerous databases, with data up to August 2023, without language restrictions.

Eligible studies were those that analyzed large data sets, such as electronic health records and administrative claims, using machine learning and statistical models to identify patterns. The studies included in the review all have a study population of adults diagnosed with dementia from all causes or subtypes, as defined by standardized criteria.

Specific attention was paid to medications prescribed to cohorts ranging from thousands to millions of participants, with follow-up durations of years. Data extraction involved variables such as sample size, predictors, outcomes, and analysis techniques.

However, methodologies varied widely among the 14 included studies. Techniques include logistic regression, random forests and other machine learning models. Some studies focused solely on medications, while others included broader characteristics such as demographics and medical history. Medication effects were assessed based on frequency of prescription, time of use, and associated risk of dementia. The review used tools to control bias and ensure high-quality assessment of all studies.

Key findings

The study found that specific drug classes were associated with changes in dementia risk. Medications such as antimicrobials, anti-inflammatory drugs and vaccines were associated with a reduced risk of dementia, while others, such as antipsychotics and certain diabetes medications, were associated with an increased risk of dementia.

The review found that certain medications were consistently associated with significant effects on dementia risk, both protective and detrimental. Antibiotics, antivirals and anti-inflammatory drugs were found to be protective, with risk ratios showing a notable reduction in the risk of developing dementia. These findings highlight the potential role of targeting inflammation and infections as part of dementia prevention strategies.

Similarly, vaccines for diseases such as hepatitis and typhoid showed associations with a reduced risk of dementia, suggesting that immune modulation may provide protective benefits against cognitive decline.

In contrast, the researchers discovered links between frequent use of antipsychotics and benzodiazepines and an increased risk of dementia. These results were consistent with previous evidence indicating the long-term neurological risks of these classes of drugs and underscored the need for careful evaluation of their use, especially in populations at risk for cognitive impairment.

Meanwhile, drugs that target vascular and metabolic pathways, including antihypertensives and statins, showed mixed results. The impact of these medications on dementia risk varied depending on the specific medications and demographic factors, indicating a need for more rigorous research to determine their role in dementia prevention.

The data-driven approach in the studies improved the identification of patterns across different data sets, while confirming previous findings and revealing new drug candidates. However, the associations varied depending on study design, medication classification systems, and population demographics, highlighting the need for targeted clinical trials to elucidate the mechanisms underlying these associations.

Conclusions

In summary, the reviewers reported on the complex interplay between medications prescribed for different health conditions or diseases and the risk of dementia, and revealed opportunities for drug repurposing. While certain medications showed promise in reducing the risk of dementia, the review found that other medications require caution because of possible adverse associations.

These findings highlighted the value of data-driven research in identifying therapeutic candidates and informing clinical decisions. The scientists believe that future work should focus on validating these associations through experimental studies and advancing our understanding of their underlying biological mechanisms.