1
|
Zhu W, Tang H, Zhang H, Rajamohan HR, Huang SL, Ma X, Chaudhari A, Madaan D, Almahmoud E, Chopra S, Dodson JA, Brody AA, Masurkar AV, Razavian N. Predicting Risk of Alzheimer's Diseases and Related Dementias with AI Foundation Model on Electronic Health Records. MEDRXIV : THE PREPRINT SERVER FOR HEALTH SCIENCES 2024:2024.04.26.24306180. [PMID: 38712223 PMCID: PMC11071573 DOI: 10.1101/2024.04.26.24306180] [Citation(s) in RCA: 0] [Impact Index Per Article: 0] [Reference Citation Analysis] [Abstract] [Grants] [Track Full Text] [Subscribe] [Scholar Register] [Indexed: 05/08/2024]
Abstract
Early identification of Alzheimer's disease (AD) and AD-related dementias (ADRD) has high clinical significance, both because of the potential to slow decline through initiating FDA-approved therapies and managing modifiable risk factors, and to help persons living with dementia and their families to plan before cognitive loss makes doing so challenging. However, substantial racial and ethnic disparities in early diagnosis currently lead to additional inequities in care, urging accurate and inclusive risk assessment programs. In this study, we trained an artificial intelligence foundation model to represent the electronic health records (EHR) data with a vast cohort of 1.2 million patients within a large health system. Building upon this foundation EHR model, we developed a predictive Transformer model, named TRADE, capable of identifying risks for AD/ADRD and mild cognitive impairment (MCI), by analyzing the past sequential visit records. Amongst individuals 65 and older, our model was able to generate risk predictions for various future timeframes. On the held-out validation set, our model achieved an area under the receiver operating characteristic (AUROC) of 0.772 (95% CI: 0.770, 0.773) for identifying the AD/ADRD/MCI risks in 1 year, and AUROC of 0.735 (95% CI: 0.734, 0.736) in 5 years. The positive predictive values (PPV) in 5 years among individuals with top 1% and 5% highest estimated risks were 39.2% and 27.8%, respectively. These results demonstrate significant improvements upon the current EHR-based AD/ADRD/MCI risk assessment models, paving the way for better prognosis and management of AD/ADRD/MCI at scale.
Collapse
Affiliation(s)
- Weicheng Zhu
- NYU, Center for Data Science, New York, NY, 10001, USA
| | - Huanze Tang
- NYU, Center for Data Science, New York, NY, 10001, USA
| | - Hao Zhang
- NYU Grossman School of Medicine, Department of Population Health, New York, NY, 10016, USA
| | | | | | - Xinyue Ma
- NYU, Center for Data Science, New York, NY, 10001, USA
| | | | - Divyam Madaan
- NYU, Courant Institute of Mathematical Sciences, New York, NY, 10001, USA
| | - Elaf Almahmoud
- NYU, Courant Institute of Mathematical Sciences, New York, NY, 10001, USA
| | - Sumit Chopra
- NYU, Courant Institute of Mathematical Sciences, New York, NY, 10001, USA
- NYU Grossman School of Medicine, Department of Radiology, New York, NY, 10016, USA
| | - John A. Dodson
- NYU Grossman School of Medicine, Department of Population Health, New York, NY, 10016, USA
- NYU Grossman School of Medicine, Department of Medicine, New York, NY, 10016, USA
| | - Abraham A. Brody
- NYU Grossman School of Medicine, Department of Medicine, New York, NY, 10016, USA
- NYU Grossman School of Medicine, Rory Meyers College of Nursing, Hartford Institute for Geriatric Nursing, New York, NY, 10016, USA
| | - Arjun V. Masurkar
- NYU Grossman School of Medicine, Department of Neurology, New York, NY, 10016, USA
- NYU Grossman School of Medicine, Department of Neuroscience and Physiology, New York, NY, 10016, USA
- NYU Grossman School of Medicine, Neuroscience Institute, New York, NY, 10016, USA
| | - Narges Razavian
- NYU Grossman School of Medicine, Department of Population Health, New York, NY, 10016, USA
- NYU Grossman School of Medicine, Department of Radiology, New York, NY, 10016, USA
| |
Collapse
|
2
|
Liu H, Liang Q, Yang Y, Liu M, Zheng B, Sun S. Impact of mechanical ventilation on clinical outcomes in ICU-admitted Alzheimer's disease patients: a retrospective cohort study. Front Public Health 2024; 12:1368508. [PMID: 38601491 PMCID: PMC11004329 DOI: 10.3389/fpubh.2024.1368508] [Citation(s) in RCA: 0] [Impact Index Per Article: 0] [Reference Citation Analysis] [Abstract] [Key Words] [MESH Headings] [Track Full Text] [Figures] [Journal Information] [Subscribe] [Scholar Register] [Received: 01/10/2024] [Accepted: 03/18/2024] [Indexed: 04/12/2024] Open
Abstract
Background Alzheimer's disease (AD) is increasingly recognized as a pressing global public health issue, demanding urgent development of scientific AD management strategies. In recent years, the proportion of AD patients in Intensive Care Units (ICU) has been on the rise. Simultaneously, the use of mechanical ventilation (MV) is becoming more prevalent among this specific patient group. Considering the pathophysiological characteristics of AD, the application of MV in AD patients may lead to different outcomes. However, due to insufficient research data, the significant impact of MV on the prognosis of AD patients in the ICU remains unclear. Therefore, we conducted this study to comprehensively evaluate the potential influence of MV on the survival rate of AD patients in the ICU. Methods We obtained data from the MIMIC-IV database for patients diagnosed with AD. Using propensity score matching (PSM), we paired patients who received MV treatment with those who did not receive treatment. Next, we conducted Cox regression analysis to evaluate the association between MV and in-hospital mortality, 7-day mortality, 28-day mortality, 90-day mortality, 4-year mortality, length of hospital stay, and ICU stay. Results The data analysis involved a cohort of 641 AD patients spanning from 2008 to 2019, inclusive. Following a 1:2 propensity score matching (PSM) procedure, 300 patients were successfully paired, comprising 123 individuals who underwent MV treatment and 177 who did not. MV demonstrated an association with an elevated risk of in-hospital mortality (HR 5.782; 95% CI 2.981-11.216; p < 0.001), 7-day mortality (HR 6.353; 95% CI 3.014-13.392; p < 0.001), 28-day mortality (HR 3.210; 95% CI 1.977-5.210; p < 0.001), 90-day mortality (HR 2.334; 95% CI 1.537-3.544; p < 0.001), and 4-year mortality (HR 1.861; 95% CI 1.370-2.527; p < 0.001). Furthermore, it was associated with a prolonged length of ICU stay [3.6(2.2,5.8) vs. 2.2(1.6,3.7); p = 0.001]. In the subgroup analysis, we further confirmed the robustness of the results obtained from the overall population. Additionally, we observed a significant interaction (p-interaction <0.05) between age, admission type, aspirin use, statin use, and the use of MV. Conclusion In patients with AD who are receiving treatment in the ICU, the use of MV has been linked to higher short-term, medium-term, and long-term mortality rates, as well as prolong ICU stays. Therefore, it is crucial to break away from conventional thinking and meticulously consider both the medical condition and personal preferences of these vulnerable patients. Personalized treatment decisions, comprehensive communication between healthcare providers and patients, formulation of comprehensive treatment plans, and a focus on collaboration between the ICU and community organizations become imperative.
Collapse
Affiliation(s)
- Han Liu
- Institute for Global Health, University College London, London, United Kingdom
| | - Qun Liang
- First Affiliated Hospital of Heilongjiang University of Chinese Medicine, Harbin, China
| | - Yang Yang
- First Affiliated Hospital of Heilongjiang University of Chinese Medicine, Harbin, China
| | - Min Liu
- The Second Affiliated Hospital of Guangzhou Medical University, Guangzhou, China
| | - Boyang Zheng
- Heilongjiang University of Chinese Medicine, Harbin, China
| | - Shilin Sun
- The Second Affiliated Hospital of Guangzhou Medical University, Guangzhou, China
- Heilongjiang University of Chinese Medicine, Harbin, China
| |
Collapse
|
3
|
John London A, Karlawish J, Largent EA, Phillips Hey S, McCarthy EP. Algorithmic identification of persons with dementia for research recruitment: ethical considerations. Inform Health Soc Care 2024; 49:28-41. [PMID: 38196387 PMCID: PMC11001531 DOI: 10.1080/17538157.2023.2299881] [Citation(s) in RCA: 0] [Impact Index Per Article: 0] [Reference Citation Analysis] [Abstract] [Key Words] [MESH Headings] [Grants] [Track Full Text] [Journal Information] [Subscribe] [Scholar Register] [Indexed: 01/11/2024]
Abstract
Underdiagnosis, misdiagnosis, and patterns of social inequality that translate into unequal access to health systems all pose barriers to identifying and recruiting diverse and representative populations into research on Alzheimer's disease and Alzheimer's disease related dementias. In response, some have turned to algorithms to identify patients living with dementia using information that is associated with this condition but that is not as specific as a diagnosis. This paper explains six ethical issues associated with the use of such algorithms including the generation of new, sensitive, identifiable medical information for research purposes without participant consent, issues of justice and equity, risk, and ethical communication. It concludes with a discussion of strategies for addressing these issues and prompting valuable research.
Collapse
Affiliation(s)
- Alex John London
- Center for Ethics and Policy, Carnegie Mellon University, Pittsburgh, Pennsylvania, USA
| | - Jason Karlawish
- Department of Medical Ethics and Health Policy, University of Pennsylvania Perelman School of Medicine, Philadelphia, Pennsylvania, USA
- Department of Medicine, University of Pennsylvania Perelman School of Medicine, Philadelphia, Pennsylvania, USA
- Department of Neurology, University of Pennsylvania Perelman School of Medicine, Philadelphia, Pennsylvania, USA
| | - Emily A. Largent
- Department of Medical Ethics and Health Policy, University of Pennsylvania Perelman School of Medicine, Philadelphia, Pennsylvania, USA
| | | | - Ellen P. McCarthy
- Hinda and Arthur Marcus Institute for Aging Research, Hebrew SeniorLife, Boston, MA, USA
- Division of Gerontology, Department of Medicine, Beth Israel Deaconess Medical Center, Harvard Medical School, Boston, MA, USA
| |
Collapse
|
4
|
Dublin S, Greenwood-Hickman MA, Karliner L, Hsu C, Coley RY, Colemon L, Carrasco A, King D, Grace A, Lee SJ, Walsh JME, Barrett T, Broussard J, Singh U, Idu A, Yaffe K, Boustani M, Barnes DE. The electronic health record Risk of Alzheimer's and Dementia Assessment Rule (eRADAR) Brain Health Trial: Protocol for an embedded, pragmatic clinical trial of a low-cost dementia detection algorithm. Contemp Clin Trials 2023; 135:107356. [PMID: 37858616 DOI: 10.1016/j.cct.2023.107356] [Citation(s) in RCA: 0] [Impact Index Per Article: 0] [Reference Citation Analysis] [Abstract] [Key Words] [MESH Headings] [Grants] [Track Full Text] [Journal Information] [Subscribe] [Scholar Register] [Received: 03/22/2023] [Revised: 09/26/2023] [Accepted: 10/15/2023] [Indexed: 10/21/2023]
Abstract
BACKGROUND About half of people living with dementia have not received a diagnosis, delaying access to treatment, education, and support. We previously developed a tool, eRADAR, which uses information in the electronic health record (EHR) to identify patients who may have undiagnosed dementia. This paper provides the protocol for an embedded, pragmatic clinical trial (ePCT) implementing eRADAR in two healthcare systems to determine whether an intervention using eRADAR increases dementia diagnosis rates and to examine the benefits and harms experienced by patients and other stakeholders. METHODS We will conduct an ePCT within an integrated healthcare system and replicate it in an urban academic medical center. At primary care clinics serving about 27,000 patients age 65 and above, we will randomize primary care providers (PCPs) to have their patients with high eRADAR scores receive targeted outreach (intervention) or usual care. Intervention patients will be offered a "brain health" assessment visit with a clinical research interventionist mirroring existing roles within the healthcare systems. The interventionist will make follow-up recommendations to PCPs and offer support to newly-diagnosed patients. Patients with high eRADAR scores in both study arms will be followed to identify new diagnoses of dementia in the EHR (primary outcome). Secondary outcomes include healthcare utilization from the EHR and patient, family member and clinician satisfaction assessed through surveys and interviews. CONCLUSION If this pragmatic trial is successful, the eRADAR tool and intervention could be adopted by other healthcare systems, potentially improving dementia detection, patient care and quality of life.
Collapse
Affiliation(s)
- Sascha Dublin
- Kaiser Permanente Washington Health Research Institute, Seattle, WA, USA; Department of Epidemiology, University of Washington, Seattle, WA, USA; Kaiser Permanente Bernard Tyson School of Medicine, Pasadena, CA, USA.
| | | | - Leah Karliner
- University of California, San Francisco, San Francisco, CA, USA
| | - Clarissa Hsu
- Kaiser Permanente Washington Health Research Institute, Seattle, WA, USA
| | - R Yates Coley
- Kaiser Permanente Washington Health Research Institute, Seattle, WA, USA; Department of Biostatistics, University of Washington, Seattle, WA, USA
| | - Leonardo Colemon
- Kaiser Permanente Washington Health Research Institute, Seattle, WA, USA
| | - Anna Carrasco
- University of California, San Francisco, San Francisco, CA, USA
| | - Deborah King
- Kaiser Permanente Washington Health Research Institute, Seattle, WA, USA
| | | | - Sei J Lee
- University of California, San Francisco, San Francisco, CA, USA
| | | | - Tyler Barrett
- Kaiser Permanente Washington Health Research Institute, Seattle, WA, USA
| | - Jia Broussard
- University of California, San Francisco, San Francisco, CA, USA
| | - Umesh Singh
- Kaiser Permanente Washington Health Research Institute, Seattle, WA, USA
| | - Abisola Idu
- Kaiser Permanente Washington Health Research Institute, Seattle, WA, USA
| | - Kristine Yaffe
- University of California, San Francisco, San Francisco, CA, USA
| | - Malaz Boustani
- Indiana University School of Medicine, Indianapolis, IN, USA
| | | |
Collapse
|
5
|
Estiri H, Azhir A, Blacker DL, Ritchie CS, Patel CJ, Murphy SN. Temporal characterization of Alzheimer's Disease with sequences of clinical records. EBioMedicine 2023; 92:104629. [PMID: 37247495 DOI: 10.1016/j.ebiom.2023.104629] [Citation(s) in RCA: 0] [Impact Index Per Article: 0] [Reference Citation Analysis] [Abstract] [Key Words] [Track Full Text] [Journal Information] [Subscribe] [Scholar Register] [Received: 03/24/2023] [Revised: 05/05/2023] [Accepted: 05/10/2023] [Indexed: 05/31/2023] Open
Abstract
BACKGROUND Alzheimer's Disease (AD) is a complex clinical phenotype with unprecedented social and economic tolls on an ageing global population. Real-world data (RWD) from electronic health records (EHRs) offer opportunities to accelerate precision drug development and scale epidemiological research on AD. A precise characterization of AD cohorts is needed to address the noise abundant in RWD. METHODS We conducted a retrospective cohort study to develop and test computational models for AD cohort identification using clinical data from 8 Massachusetts healthcare systems. We mined temporal representations from EHR data using the transitive sequential pattern mining algorithm (tSPM) to train and validate our models. We then tested our models against a held-out test set from a review of medical records to adjudicate the presence of AD. We trained two classes of Machine Learning models, using Gradient Boosting Machine (GBM), to compare the utility of AD diagnosis records versus the tSPM temporal representations (comprising sequences of diagnosis and medication observations) from electronic medical records for characterizing AD cohorts. FINDINGS In a group of 4985 patients, we identified 219 tSPM temporal representations (i.e., transitive sequences) of medical records for constructing the best classification models. The models with sequential features improved AD classification by a magnitude of 3-16 percent over the use of AD diagnosis codes alone. The computed cohort included 663 patients, 35 of whom had no record of AD. Six groups of tSPM sequences were identified for characterizing the AD cohorts. INTERPRETATION We present sequential patterns of diagnosis and medication codes from electronic medical records, as digital markers of Alzheimer's Disease. Classification algorithms developed on sequential patterns can replace standard features from EHRs to enrich phenotype modelling. FUNDING National Institutes of Health: the National Institute on Aging (RF1AG074372) and the National Institute of Allergy and Infectious Diseases (R01AI165535).
Collapse
Affiliation(s)
- Hossein Estiri
- Department of Medicine, Massachusetts General Hospital, Boston, MA, USA.
| | - Alaleh Azhir
- Department of Medicine, Massachusetts General Hospital, Boston, MA, USA; Harvard Medical School, Harvard-MIT Program in Health Sciences and Technology, USA
| | - Deborah L Blacker
- Department of Psychiatry, Massachusetts General Hospital, Boston, MA, USA
| | | | - Chirag J Patel
- Department of Biomedical Informatics, Harvard Medical School, Boston, MA, USA
| | - Shawn N Murphy
- Department of Neurology, Massachusetts General Hospital, Boston, MA, USA
| |
Collapse
|