Clinical Informatics and Data Science

Dr. Maggie Horn’s research agenda focuses on understanding the multifactorial experience in patients with musculoskeletal pain by leveraging secondary data from claims, administrative, clinical outcomes, and registry data sources. She has a particular interest in developing sustainable and scalable processes to collect patient-reported outcome measures and clinical data through the EHR to undertake research and quality improvement projects and develop data analytics platforms. 

Dr. Horn is scientifically trained as a Rehabilitation Scientist and Epidemiologist and clinically trained as a Physical Therapist. Her current research and collaborations are focused on leveraging clinically collected data from the EHR to
1) Develop predictive models to optimize care and treatments options for patients with lumbar spinal stenosis,

2) Investigate the role of psychosocial factors and comorbidity to inform health care delivery strategies to prevent the chronification of MSK injuries in military populations,

3) Determine the prevalence of persistent pain and poor outcomes (i.e, reduced physical function, reliance on pharmacological management of pain, psychological comorbidity) among patients undergoing total knee, hip, and shoulder arthroplasty (TKA, THA, and TSA, respectively) and 4) Implement and measure the effectiveness of the implementation of clinical practice guidelines for physical therapist treatment of neck pain. The over-arching goal of Dr. Horn’s research is to understand how patients with musculoskeletal pain utilize healthcare systems, how care pathways affect downstream utilization of health services and disease states, and ultimately find better ways to deliver healthcare in this population by leveraging technology and knowledge translation.   

Dr. Horn is passionate about mentoring students with interests in data science, data analytics, data visualization, implementation of clinical practice guidelines and evidence-based practice as well as the development of systematic reviews on key topics related to data analytics and care delivery in orthopedics and rehabilitation. 
 

Current Research Studies and Collaborations

Lumbar Stenosis Prognostic Subgroups for Personalizing Care and Treatment (PROSPECTS)
Funding Source: R01 R01AG069891 (PI: Rundell, S; Site PI, Goode, A)
Role: Co-investigator

The overall goal of this project is to identify a phenotype of older adult patients that may benefit from nonsurgical care for lumbar spine stenosis.


Development of Persistent Musculoskeletal Pain in the Military: The Prediction of Outcomes, Utilization, and Readiness (POUR) Cohort
Funding Source: DOD PRMPR (MIRROR 21) (PI: Rhon, D)
Role: Co-investigator

The objective of this proposal is to use an innovative predictive framework consisting of psychosocial factors and comorbidity to validate algorithms that inform health care delivery strategies to prevent the chronification of MSK injuries


Total Joint Arthroplasty Learning Health Unit  
Funding Source: Duke Health System (PI: Bolognesi, M.)         
Role: Co-investigator                                                                        

The purpose of the development of the Joint Arthroplasty Learning Health Unit to improve outcomes after total knee, hip, and shoulder arthroplasty (TKA, THA, and TSA, respectively) in the Duke Health System. The goal of this project is to identify patient, provider, system, and environmental factors that contribute to poor outcomes in this population with the goal of implementing care redesign and clinical prediction tools within the EHR to improve patient outcomes and cost-effectiveness.


Duke Department of Physical Therapy and Occupational Therapy 
Clinical Leads: Marissa Carvalho, PT, DPT (PI), Katie Pruka, PT, DPT, Michelle Ramirez, PT, DPT, Zachary Stearns, PT, DPT
Role: Research Collaborator

The goal of the collaboration between the investigators is to use a shared knowledge base and informatics resources to 1) Administer and develop a standardized process to track patient-reported outcome measures (PROMs) within the EHR to measure outcomes from physical therapy intervention,
2) Implement the use of clinical practice guidelines by optimizing EHR documentation, providing training, and measuring the patient progress and outcomes after implementation and
3) Determine the relationship between psychosocial factors, clinical outcomes, and treatment in patients seeking physical therapy using novel tools.