Document text
Principal Investigator: HONG YU
Organization: UNIVERSITY OF MASSACHUSETTS LOWELL
Fiscal Year: 2024
Award: $782,282
Funding agency: National Institute on Aging
Social and behavioral determinants of health and Alzheimer’s Disease: Cohort study of the US military
veteran population
Alzheimer’s Disease (AD) affects an estimated 5.8 million US adults. Veterans are particularly susceptible to
AD due to demographic, clinical, and economic factors. Social determinants of health are the conditions in
which people are born, live, work, and age. Adverse social determinants of health include job loss and financial
and food insecurity. Together with behavioral health factors (e.g., smoking and substance use) and mental
health, adverse social and behavioral determinants of health (SBDH) contribute to adverse health outcomes.
Associations between SBDH and AD have been noted, but most studies used structured electronic health
record (EHR) or survey data. SBDH are not routinely added to structured EHR. Natural language processing
(NLP) approaches can be developed to automatically extract SBDH and their attributes. This application
responds to PAR-22-093 and NOT-AG-18-047. The specific aims are:
Aim 1: Establish NLP-enriched case definitions of adverse SBDH and AD-related information (e.g., signs and
symptoms of cognitive decline), and examine their incidences by first chart-reviewing ~10,000 EHR notes (e.g.,
primary care, neurology, psychiatric, and social work notes) and then developing and evaluating sophisticated
NLP systems for automatically capturing SBDH and AD-related information.
Aim 2: Using NLP enriched SBDH as independent variables from a nested case-control design, we will
analyze the associations between adverse SBDH and incident AD. We will also evaluate how the associations
vary by age, sex, race/ethnicity. We will compare results using NLP-enriched SBDH vs. using structured data
(only) SBDH. Hypothesis 1: Patients with adverse SBDH have substantially higher AD risk, after adjusting for
potential covariables (e.g., patient-specific demographic and clinical factors). Hypothesis 2: The effects of
adverse SBDH on AD risk vary by age, sex and race/ethnicity, after adjusting for covariables (e.g., patient-
specific clinical factors). Hypothesis 3: The effects of adverse SBDH on incident AD are likely cumulative and
duration-dependent, with more and longer adverse SBDH leading to higher AD risk.
Aim 3: Early AD diagnosis may prevent or delay AD development through intervention efforts on SBDH.34
Cognitive decline occurs 4-8 years prior to AD diagnosis.35 We will study whether inclusion of NLP-enriched
adverse SBDH and AD-related information helps early AD diagnosis. We will use three types of predictive
models: statistical regression, traditional machine learning, and innovative deep learning models.
Terms: <21+ years old><65 and older><65 or older><65 years of age and older><65 years of age or more><65 years of age or older><65+ years><65+ years old><> 65 years><AD dementia><AD prevention><Address><Adult><Adult Human><Affect><Age><Aged 65 and Over><Alzheimer Type Dementia><Alzheimer disease dementia><Alzheimer disease prevention><Alzheimer prevention><Alzheimer risk factor><Alzheimer sclerosis><Alzheimer syndrome><Alzheimer's><Alzheimer's Disease><Alzheimer's diagnosis><Alzheimer's disease diagnosis><Alzheimer's disease risk><Alzheimers Dementia><Amentia><Behavioral><Black American><Brain Trauma><Cause of Death><Clinic><Clinical><Cognitive Disturbance><Cognitive Impairment><Cognitive decline><Cognitive function abnormal><Cohort Studies><Concurrent Studies><Data><Dementia><Development><Diagnostic Findings><Disturbance in cognition><Dose><Early Diagnosis><Economic Factors><Economical Factors><Electronic Health Record><Ethnic Origin><Ethnicity><Exposure to><Financial Hardship><Generations><Health><Health Care Costs><Health Care Systems><Health Costs><Health Surveys><Healthcare><Healthcare Costs><Healthcare Systems><High Prevalence><Homelessness><Impaired cognition><Incidence><Informatics><Infrastructure><Integrated Health Care Systems><Intervention><Intervention Strategies><Investigators><Job loss><Learning><Machine Learning><Maintenance><Measures><Mental Health><Mental Hygiene><Modeling><Natural Language Processing><Neurology><Non-Hispanic><Nonhispanic><Not Hispanic or Latino><Observation research><Observation study><Observational Study><Observational research><Outcome><PTSD><Patient outcome><Patient-Centered Outcomes><Patient-Focused Outcomes><Patients><Persons><Population><Post-Traumatic Neuroses><Post-Traumatic Stress Disorders><Posttraumatic Neuroses><Predisposition><Primary Care><Primary Senile Degenerative Dementia><Process><Psychological Health><Public Policy><Race><Races><Regression Analyses><Regression Analysis><Regression Diagnostics><Reporting><Research><Research Personnel><Research Resources><Researchers><Resources><Retrieval><Risk><Risk Reduction><Secure><Signs and Symptoms><Smoking><Social Service><Social Work><Social isolation><Statistical Regression><Stress><Structure><Susceptibility><System><Time><Traumatic Brain Injury><United States Department of Veterans Affairs><United States Veterans Administration><Update><Veterans><Veterans Administration><Veterans Affairs><Veterans Health Administration><Veterans Health Affairs><Vulnerable Populations><Woman><Work><above age 65><adulthood><after age 65><age 65 and greater><age 65 and older><age 65 or older><age > 65><age of 65 years onward><aged 65 and greater><aged 65+><aged ≥65><ages><alzheimer risk><behavioral health><case control><case-controlled><cognitive dysfunction><cognitive loss><computer based prediction><cost><cost effective><deep learning based model><deep learning model><dementia risk><design><designing><developmental><digital repositories><early detection><electronic health care record><electronic health medical record><electronic health plan record><electronic health registry><electronic medical health record><electronic repositories><electronic structure><ethnic minority><experience><financial adversity><financial burden><financial distress><financial insecurity><financial strain><financial stress><food insecurity><health care><health determinants><high risk group><high risk individual><high risk people><high risk population><homeless><human old age (65+)><improved><innovate><innovation><innovative><integrated health system><integrated healthcare systems><integrated system of care><intersectionalities><intersectionality><interventional strategy><machine based learning><malleable risk><men><military veteran><modifiable risk><natural language understanding><old age><older adult><older adulthood><opiate overdose><opiate related overdose><opioid drug overdose><opioid induced overdose><opioid intoxication><opioid medication overdose><opioid overdose><opioid poisoning><opioid related overdose><opioid toxicity><over 65 years><patient oriented outcomes><population health><post-trauma stress disorder><posttrauma stress disorder><predictive modeling><prevent><preventing><primary degenerative dementia><racial><racial background><racial minority><racial origin><reduce risk><reduce risks><reduce that risk><reduce the risk><reduce these risks><reduces risk><reduces the risk><reducing risk><reducing the risk><risk factor for dementia><risk for dementia><risk-reducing><senile dementia of the Alzheimer type><sex><social><social health determinants><structured data><substance use><substance using><traumatic brain damage><traumatic neurosis><unhoused><veteran population><virtual><vulnerable group><vulnerable individual><vulnerable people><≥65 years>