summary: Researchers have developed MeAgainMeds.com, an AI-based tool that helps match patients with the most effective antidepressants based on their medical history. The free website aims to reduce the trial-and-error process of finding the right medication and improve patient outcomes.
By analysing data from millions of patients, AI can provide evidence-based recommendations that patients can discuss with their healthcare providers.
Key Facts:
- AI-powered matching: MeAgainMeds.com leverages AI to match patients with the most effective antidepressant medication based on their medical history.
- Large-scale data analysis: The tool analyzed data from more than 3.6 million patients and 10.2 million antidepressant prescriptions to make its recommendations.
- Patient-Centric: The website does not require any personally identifiable information and advises patients to consult with their health care provider about any changes to their medication.
sauce: George Mason University
Researchers at George Mason University's School of Public Health have harnessed the power of artificial intelligence (AI) analytical models to match patients' medical histories with the most effective antidepressants, helping them find relief more quickly.
The free website, MeAgainMeds.com, provides evidence-based recommendations to help clinicians and patients find the best antidepressant the first time.
“Many people with depression have to try multiple antidepressants before finding the right one that relieves their symptoms. Our website reduces the number of medications patients are asked to try.”
“The system recommends treatments to patients that have worked for at least 100 other patients with the exact same relevant medical history,” said Farooq Alemi, lead researcher and professor of health informatics at George Mason University's School of Public Health.
AI helped simplify the incredibly complex task of making thousands of guidelines easily accessible to patients and clinicians. The guidelines researchers created are complex due to the amount of clinical information related to prescribing antidepressants, but AI seamlessly simplifies this task.
MeAgainMeds.com has AI at its core, analyzing doctor or patient responses to anonymous medical history questions to determine the oral antidepressant that best suits your specific needs. The website does not ask for any personally identifiable information or prescribe medication changes. Patients are encouraged to speak with their healthcare provider about medication changes.
In 2018, the Centers for Disease Control and Prevention reported that more than 13% of adults were using antidepressants, and that number has only increased since the start of the pandemic and other epidemics in 2020. This website could help millions of people find relief faster.
Alemi and his team analyzed 3,678,082 patients who took 10,221,145 different antidepressants. The oral antidepressants analyzed were: amitriptyline, bupropion, citalopram, desvenlafaxine, doxepin, duloxetine, escitalopram, fluoxetine, mirtazapine, nortriptyline, paroxetine, sertraline, trazodone, and venlafaxine.
The researchers created 16,770 subgroups of at least 100 cases from the data, using information such as response to previous antidepressants, current medications, physical illness history, psychiatric illness history, primary procedures, and other information. Based on the subgroups and remission rates, the AI makes evidence-based medication recommendations.
“Matching patients to subgroups allows clinicians to prescribe the most effective medications to people with similar medical histories,” Alemi said. The researchers and the website encourage patients who use the site to share their information with their clinicians, who ultimately decide whether to prescribe the recommended medications.
Alemi and his team tested a prototype version of the site in 2023 and promoted it on social media. At the time, 1,500 patients used the website. Their goal is to continue improving the website and expanding the user base. The initial study was funded by the Commonwealth of Virginia and the Robert Wood Johnson Foundation.
In the latest in a series of papers on response to antidepressants, the researchers analyzed a subgroup of 2,467 patients who received psychological therapy.
Additional authors include Tulay G Soylu of Temple University, Mary Cannon and Conor McCandless of the Royal College of Surgeons in Dublin, Ireland.
About this AI and psychopharmacology research
author: Mary Cunningham
sauce: George Mason University
contact: Mary Cunningham – George Mason University
image: Image courtesy of Neuroscience News
Original Research: The access is closed.
“The Effectiveness of Antidepressants Combined with Psychotherapy” by Farrokh Alemi et al. Journal of Mental Health Policy and Economics
Abstract
The effectiveness of combining antidepressants with psychotherapy
background: Consensus guidelines for prescribing antidepressants recommend that clinicians should carefully tailor antidepressant choices to patients' medical histories, but do not provide specific advice on which antidepressant is best for a particular medical history.
Research objectives: This study provides empirically derived guidelines for prescribing antidepressants that fit the patient's medical history to severely depressed patients receiving psychotherapy.
Method: This retrospective observational cohort study analyzed a large insurance database of 3,678,082 patients. Data were obtained from US health care providers between January 1, 2001 and December 31, 2018. These patients received 10,221,145 antidepressant treatment sessions.
The study reports remission rates for the 14 most commonly prescribed solo antidepressants (amitriptyline, bupropion, citalopram, desvenlafaxine, doxepin, duloxetine, escitalopram, fluoxetine, mirtazapine, nortriptyline, paroxetine, sertraline, trazodone, and venlafaxine) and a category called “other” (any other antidepressant/antidepressant combination).
This study used robust LASSO regression analysis to identify factors influencing remission rates and clinicians' antidepressant choice. Selection bias in observational data was removed by stratification.
We combined the most important factors influencing remission with selection bias to organize the data into 16,770 subgroups of at least 100 cases. In this paper, we report on 2,467 subgroups of patients who received psychological treatment.
result: Large, statistically significant differences in remission rates were found within patient subgroups: remission rates for sertraline ranged from 4.5% to 77.86%, for fluoxetine from 2.86% to 77.78%, for venlafaxine from 5.07% to 76.44%, for bupropion from 0.5% to 64.63%, for desvenlafaxine from 1.59% to 75%, for duloxetine from 3.77% to 75%, for paroxetine from 6.48% to 68.79%, for escitalopram from 1.85% to 65%, and for citalopram from 4.67% to 76.23%.
Clearly, these medications are ideal for some subgroups of patients but not others. When patients are matched to subgroups, clinicians can prescribe the medication that will be most effective for the subgroup. Some medications (amitriptyline, doxepin, nortriptyline, trazodone) were not suitable as the sole antidepressant therapy for any subgroup because remission rates were always less than 11%.
Discussion: This study offers clinicians an opportunity to identify the best antidepressant for their patients before repeatedly trying antidepressants.
Impact on healthcare delivery and utilization: To help match patients to the most effective antidepressant, the study provides access to a free, non-profit decision support tool at http://MeAgainMeds.com .
Implications for health policy: Policymakers should evaluate how research findings can be used at the point of care through fragmented electronic medical records. Alternatively, policymakers could deploy AI systems that recommend antidepressants to patients online at home and prompt them to communicate the recommendation to their clinician at their next appointment.
Implications for future research: Future studies could investigate (i) the effectiveness of our recommendations in changing clinical practice, (ii) improving remission rates of depressive symptoms, and (iii) reducing costs of treatment. These studies need to be prospective yet pragmatic; randomized clinical trials are unlikely to address the large number of factors that influence remission.
