Hopkins inHealth investigators are devising analytic solutions to advance the delivery of high-quality, evidence-based, individualized patient care. As the amount of electronic health information generated by patients grows, sophisticated methodologies are needed to successfully translate this information into effective patient care tools. Through projects led by biostatisticians, clinical experts, and data scientists, Hopkins inHealth investigators are developing novel analytic tools focused on individualized patient care, learning health systems, and adaptive trial designs.
Individualized Patient Care: One of the challenges facing clinicians today is determining the optimal care plan for a patient while faced with the many treatment options often available and being sensitive to each patient’s individual health circumstances and preferences. To address this challenge, Hopkins inHealth investigators are designing and disseminating statistical tools that are better at characterizing each patient’s unique health status and predicting each patient’s outcome to a particular treatment. In one program, a Hopkins inHealth team is creating a statistical test to determine the pathogen most likely infecting a child diagnosed with pneumonia, greatly improving the chance for effective treatment. In another program, a clinical decision-support tool is being designed to predict the outcome of different prostate cancer treatments in individual patients, taking into account disease severity. The methodologies developed by these and other projects will be made available through open-source statistical software, ensuring their rapid dissemination and allowing clinicians and researchers to adapt the tools to their own needs. Visit the Learning Methodologies Core, inADM, and inCAS pages for more information on these and other projects dedicated to the individualization of patient care.
Learning Health Systems: Millions of healthcare visits occur annually in the United States, during which an unprecedented amount of health information is collected electronically. This information can be used to learn about the impact of clinical decisions on patient outcomes, improving subsequent care. To this end, several Hopkins inHealth projects are establishing learning healthcare systems, an innovative approach to healthcare delivery that allows for information generated within a healthcare system to be analyzed in real-time to support continuous improvement of patient care. The initial demonstration projects are focused on radiation oncology and interventional cardiology procedures. Visit the Oncospace and inCAR pages for more information.
Adaptive Trial Design: Though traditional randomized clinical trials provide useful information about the effectiveness of new treatments, many trials do not have the statistical power to ascertain how treatment outcome might differ by patient characteristics such as age or sex. Adaptive trials are ones in which clinical trial recruitment eligibility adapts over the course of the trial as information is learned about which patient characteristics are associated with treatment response. This novel trial design can lead to more personalized therapeutic options while increasing the efficiency of clinical trials. The Learning Methodologies Core is collaborating with clinical investigators to develop adaptive trial designs in the treatment of afflictions ranging from Alzheimer’s disease to HIV infection. Visit the Learning Methodologies Core page for more information.