Center for Large Data Research and Data Sharing in Rehabilitation

2019
Extending the generalizability of the Spinal Cord Injuries Model Systems to United States adults under rehabilitation due to traumatic spinal cord injuries

Navneet Kaur Baidwan, PhD
Post-doctoral Scholar, School of Health Professions
University of Alabama at Birmingham

Collaborator
James Graham, PhD, DC - Colorado State University

General Description
The overall goal of this study is to extend the generalizability of the estimates produced using the National Spinal Cord Injury Model Systems (SCIMS) database to United States adults in rehabilitation with traumatic SCI. This study will develop case-level weights for the SCIMS database using an auxiliary dataset created by combining Uniform Data System for Medical Rehabilitation (UDSMR) and eRehabData.

Specific Aims:

1. Develop weights accounting for differential selection and non-response while accounting for relevant variables, and apply an iterative fitting procedure to match characteristics between sample (SCIMS) and the auxiliary (UDSMR and eRehabData) data.
2. Provide overall representative estimates for those in inpatient rehabilitation due to SCI and to describe the patterns of socio-demographic and other lifestyle-related and behavioral variable that are included in SCIMS but not in the auxiliary dataset.

2019
Using Electronic Health Records to Examine Associations Among Stroke Severity, Social Risk Factors, Function-based Quality Measures, & Community Discharge from Inpatient Rehabilitation

Pamela Bosch, PhD, DPT
Associate Professor, Department of Physical Therapy
Northern Arizona University

Collaborator
Amol Karmarkar, PhD, MPH - Virginia Commonwealth

General Description
The long-term goal of this research is to preemptively identify high-risk, high-need patients with stroke for providing social support and timely home health interventions after discharge from IRF. The objective of this proposal is to examine the associations of NIHSS derived from hospitalization discharge records and function-related quality measures assessed at admission to an IRF. There are two underlying conceptual and clinical rationales of this investigation. First, function-related quality measures assessed at IRF admission can be a reasonable proxy for discharge NIHSS from acute hospitals. Second, the strength of associations between NIHSS and discharge to the community can be better determined after accounting for the mediating effect of social risk factors. Our proposed study will overcome limitations of prior research by using electronic health records (EHR) data from four IRFs capturing stroke in both younger (<65 years) and 65 and older patients.

Specific Aims:

1. Examine the association of NIHSS with the function-related quality measures derived from section GG self-care/mobility data elements
2. Examine the relative contribution of NIHSS and social risk factors along with function-related Quality Measures (GG self-care/mobility measures) to predict community discharge from IRFs

2019
Patient Factors Impacting Discharge to Community after Home Health for Individuals with and without Alzheimer’s Disease and Related Dementias

Sara Knox, PT, PhD, DPT
Assistant Professor, Department of Physical Therapy
MGH Institute of Health Professions

Collaborator
Kenneth J. Ottenbacher, PhD, OTR - UTMB

General Description
The objective of this proposed study is to examine the association between patient-level factors and successful discharge to community (DTC) for individuals with and without ADRD.

Specific Aims:

1. To determine if rates of successful DTC are different for individuals with and without ADRD.
2. To determine if associations between patient-level factors and successful DTC differ for individuals with and without ADRD.
3. To determine if dementia severity is associated with successful DTC for individuals diagnosed with ADRD.

2019
Trends and Predictors of Poor Discharge Outcomes following Inpatient Rehabilitation for Adults with Traumatic Brain Injury

Tolu Oyesanya, PhD, RN
Assistant Professor, Healthcare in Adult Populations Division
Duke University

Collaborator
Janet Prvu Bettger, ScD - Duke University

General Description
The goal of this project is to examine outcomes for adult patients with TBI who received IRF care between 2002 and 2018 using UDSMR data. Findings will inform interventions for transitional care planning.

Specific Aims:

1. To describe contemporary trends in discharge rehabilitation outcomes (discharge living setting and cognitive and motor function) for adult patients with TBI who received IRF care from 2002-2018.
2. To identify patient and facility characteristics that are associated with discharge rehabilitation outcomes.

2019
Pain phenotype, neuromuscular impairment, and development of disability among at-risk older adults: a secondary analysis of the Boston RISE Study

Eric Roseen, DC, MSc
Postdoctoral Research Fellow, Department of Family Medicine
Boston Medical Center

Collaborator
Kenneth J. Ottenbacher, PhD, OTR - UTMB

General Description
Musculoskeletal conditions are the leading cause of disability worldwide.1 Over 60% of adults over the age of 65 report at least one chronic pain condition, e.g., back pain and/or osteoarthritis.2,3 Americans are living longer lives, and the need for effective rehabilitative interventions for chronic musculoskeletal pain conditions is likely to grow. Studying the relationship between pain phenotypes and changes in function can provide novel insights into consequences and potential treatment of chronic pain. This project will use data from the Boston Rehabilitation Impairment Study of the Elderly (BostonRISE), a longitudinal study of 430 thoroughly-characterized older community-dwelling adults (mean age=77) recruited from primary care clinics and followed for functional outcomes over two years.

Specific Aims:

1. To assess whether baseline pain phenotype (e.g., number of pain sites, pain severity tertiles, etc.) is associated with changes in novel measures of neuromuscular attributes across baseline, one- and two years follow-up.
2. To assess whether baseline pain phenotype is associated with changes in patient-reported mobility across baseline, one- and two- years follow-up
3. To assess whether baseline pain phenotype is associated with changes in performance-based mobility across baseline, one- and two- years follow-up.

2019
Utilization of Outpatient Physical Therapy and the Impact on Opioid Use and Healthcare Utilization for Older Adults with Osteoarthritis

Cathy Schmidt, PT, PhD
Assistant Professor, Department of Physical Therapy
MGH Institute of Health Professions

Shweta Gore, PhD, DPT, GCS, CLT (Co-PI)
Assistant Professor, Department of Physical Therapy
MGH Institute of Health Professions

Collaborator
Kenneth J. Ottenbacher, PhD, OTR -Virginia Commonwealth University

General Description
The long-term goal of this research is to improve quality and access to non-opioid based rehabilitation for older adults with OA. The overall objective of this study is to explore the impact of a lost-cost, low medical risk intervention such as PT on healthcare outcomes in older adults with OA. Based on previous research in select cohorts of individuals with spinal conditions, our central hypothesis is that the utilization of PT will be associated with a decreased odds of opioid use and unplanned healthcare utilization. We will use the electronic health record (EHR) from Partners Healthcare to address the following specific aims.

Specific Aims:

1. Determine patient, provider and practice-level factors associated with receipt of outpatient PT at 30, 60 and 90 days for community-dwelling older adults with incident diagnosis of OA.
2. Evaluate the association of PT utilization (at 30, 60 and 90 days) on opioid use, planned healthcare utilization (diagnosing provider follow-up visits and elective hospitalizations) and unplanned healthcare utilization (emergent care and non-elective hospitalizations) for a period of 3 and 6 months following PT utilization.

2020
Using Medicare Data to Build Predictive Understanding of Elderly Abuse

Leila Wood, PhD
Assistant Professor
Center for Violence Prevention, Department of Obstetrics and Gynecology, University of Texas Medical Branch

Collaborator
Monique Pappadis, PhD, University of Texas Medical Branch

General Description

The project focuses on the assessment of elderly mistreatment in inpatient and outpatient Medicare recipients. Using Medicare data, we will build predictive models toward building a health care assessment tool for EM. We will explore the use of APS data to triangulate findings from Medicare data.

Specific Aims:

1. Use inpatient Medicare data to assess mortality, hospital readmission, and associated health conditions with ICD-10 codes for EM. We will produce two publications for peer review.
2. Use outpatient Medicare data to assess mortality, hospital readmission, and associated health conditions with ICD-10 codes for EM. We will produce publications for peer review.
3. Secondary analysis of APS data collected for a project on older survivors of domestic and sexual violence to assess for potential utility in project goals of assessment of EM.