60 Participants Needed

Reinforcement Learning for Aging

(REINFORCE-EHR Trial)

JL
Overseen ByJulie Lauffenburger, PharmD, PhD
Age: 18+
Sex: Any
Trial Phase: Academic
Sponsor: Brigham and Women's Hospital
No Placebo GroupAll trial participants will receive the active study treatment (no placebo)

Trial Summary

What is the purpose of this trial?

The overall goal of the proposed research is to refine and adapt and perform efficacy testing of a novel reinforcement learning-based approach to personalizing EHR-based tools for PCPs on deprescribing of high-risk medications for older adults. The trial will be conducted at Atrius Health, an integrated delivery network in Massachusetts, and will intervene upon primary care providers. The investigators will conduct a cluster randomized trial using reinforcement learning to adapt electronic health record (EHR) tools for deprescribing high-risk medications versus usual care. 60 PCPs will be randomized (i.e., 30 each to the reinforcement learning intervention and usual care \[no EHR tool\] in each arm) to the trial and follow them for approximately 30 weeks. The primary outcome will be discontinuation or ordering a dose taper for the high-risk medications for eligible patients by included primary care providers, using EHR data at Atrius. The primary hypothesis is that the personalized intervention using reinforcement learning will improve deprescribing compared with usual care.

Will I have to stop taking my current medications?

The trial does not specify if participants must stop taking their current medications. However, it focuses on helping doctors reduce or stop high-risk medications for older adults, so your doctor might discuss changes to your medication.

What data supports the effectiveness of the treatment Reinforcement Learning for aging?

Research shows that machine learning models, which include reinforcement learning, can help detect and predict frailty in older people, potentially improving health outcomes as people age.12345

Is Reinforcement Learning for Aging safe for humans?

The research articles provided do not contain specific safety data on Reinforcement Learning for Aging. They discuss general risks of adverse drug reactions in older adults, but do not address the safety of this specific treatment.678910

How does the treatment Reinforcement Learning for Aging differ from other treatments for aging?

Reinforcement Learning for Aging is unique because it uses a learning-based approach to adapt behaviors based on rewards, which may help older adults improve their ability to adjust to new situations despite age-related declines in explicit knowledge and attention. Unlike traditional treatments that might focus on medication or physical interventions, this approach leverages cognitive processes to enhance adaptation and decision-making.1112131415

Eligibility Criteria

This trial is for primary care providers (PCPs) at Atrius Health in Massachusetts. It's focused on helping them to safely reduce or stop prescribing high-risk medications for older adults. The PCPs will be part of a study that uses a new method within their electronic health records.

Inclusion Criteria

Intervention targets primary care providers (including physicians and PCP-designated nurse practitioners and physician assistants) at Atrius Health
I am 65 or older and have been prescribed 90 or more high-risk pills in the last 6 months.

Exclusion Criteria

Prior randomization to intervention arm in the prior NUDGE-EHR trial
Not a primary care provider at Atrius Health

Timeline

Screening

Participants are screened for eligibility to participate in the trial

2-4 weeks

Treatment

Primary care providers are randomized to either a reinforcement learning intervention or usual care for deprescribing high-risk medications

30 weeks

Follow-up

Participants are monitored for safety and effectiveness after treatment

4 weeks

Treatment Details

Interventions

  • Reinforcement Learning
Trial OverviewThe study tests a reinforcement learning-based tool integrated into EHR systems against usual care without the tool. Sixty PCPs will be randomly assigned to either use the new tool or continue with standard practices, and they'll participate for about 30 weeks.
Participant Groups
2Treatment groups
Experimental Treatment
Active Control
Group I: Reinforcement learning interventionExperimental Treatment1 Intervention
The intervention is a reinforcement learning program that personalizes EHR-based tools for PCPs to promote deprescribing high-risk medications over follow-up. The reinforcement learning intervention selects a tool for each provider based on an algorithm from an inventory of EHR tools and chooses tools that are predicted to motivate action for the individual provider. The effectiveness of each tool will be assessed on a selected interval based on whether a deprescribing action is taken by PCPs for eligible patients. The algorithm is trained to maximize these actions over time.
Group II: Usual careActive Control1 Intervention
No EHR-based tools provided beyond those used in regular clinical practice.

Find a Clinic Near You

Who Is Running the Clinical Trial?

Brigham and Women's Hospital

Lead Sponsor

Trials
1,694
Recruited
14,790,000+

National Institute on Aging (NIA)

Collaborator

Trials
1,841
Recruited
28,150,000+

Atrius Health

Collaborator

Trials
3
Recruited
1,500+

Findings from Research

Aging significantly increases the risk of liver-related adverse drug reactions (L-ADR), with a 33% increase in relative risk for every 10-year increase in age, based on an analysis of 64,702 reports from the China National ADR Monitoring System.
The study identified that certain drug categories, particularly antiarrhythmic, antilipemic, and antihypertensive medications, have a high correlation with L-ADR in older adults, highlighting the need for careful risk management when prescribing these drugs to this age group.
Age-Associated Risk of Liver-Related Adverse Drug Reactions.Han, YZ., Guo, YM., Xiong, P., et al.[2022]
Current drug safety systems often fail to identify certain adverse drug reactions (ADRs) that may mimic age-related chronic diseases, especially in older populations.
To better detect these 'silent' ADRs that can develop from long-term drug use, a new systematic search strategy is needed alongside existing clinical trials and reporting systems.
The out-of-focus bias in drug surveillance.Gnädinger, M., Mellinghoff, HU.[2021]

References

A multi-center, randomized, controlled trial to assess the efficacy of optimization of drug prescribing in an elderly population, at 18 months of follow-up, in the evolution of functional autonomy: the OPTIM study protocol. [2022]
General guidelines for the management of older patients with cancer. [2022]
Unsupervised learning of aging principles from longitudinal data. [2022]
Machine learning approaches for frailty detection, prediction and classification in elderly people: A systematic review. [2023]
Measuring and modeling interventions in aging. [2020]
Age-Associated Risk of Liver-Related Adverse Drug Reactions. [2022]
The out-of-focus bias in drug surveillance. [2021]
[Prevention of adverse events through renal dosage adjustment in institutionalized elders]. [2010]
Predicting and detecting adverse drug reactions in old age: challenges and opportunities. [2013]
10.United Statespubmed.ncbi.nlm.nih.gov
On the perspective of an aging population and its potential impact on drug attrition and pre-clinical cardiovascular safety assessment. [2022]
Age-related variations of visuo-motor adaptation beyond explicit knowledge. [2021]
12.United Statespubmed.ncbi.nlm.nih.gov
The effects of aging on the interaction between reinforcement learning and attention. [2019]
Forgetting Enhances Episodic Control With Structured Memories. [2022]
14.United Statespubmed.ncbi.nlm.nih.gov
Intact Reinforcement Learning But Impaired Attentional Control During Multidimensional Probabilistic Learning in Older Adults. [2020]
Reinforcement learning: Computational theory and biological mechanisms. [2021]