My research builds the framework of a data-fuelled learning system run by an inter-disciplinary team. The learning system aggregates and analyzes day-to-day experimentations as captured by clinical databases, where new knowledge is constantly extracted and propagated for quality improvement, and where practice is driven by outcomes, and less so by individual clinician knowledge base and experience and the local medical culture. Clinical databases provide a unique opportunity to evaluate both practice variation and the impact of diagnostic and treatment decisions on patient outcomes.
Critically ill patients are an ideal population for clinical database investigations because the clinical value of many treatments and interventions they receive is unproven, and high-quality data supporting or discouraging specific practices is relatively sparse. In addition, significant practice variation exists in the ICU; decisions are often based on clinician training and knowledge and local ICU culture. The Laboratory of Computational Physiology at MIT developed and maintains MIMIC, a public de-identified high-resolution database of patients admitted to the Beth Israel Deaconess Medical Center. I direct teams of clinicians (nurses, doctors, pharmacists) and scientists (database engineers, modelers, epidemiologists) who translate the day-to-day questions during rounds that have no clear answers in the current medical literature into study designs, perform the modeling and the analysis and publish their findings. The studies fall into the following broad categories: identification and interrogation of practice variation, predictive modeling of clinical outcomes within patient subsets and comparative effectiveness research on diagnostic tests and therapeutic interventions.
This learning system has two components: an electronic clinical database and an inter-disciplinary team that drives it. Unfortunately, the development of an electronic clinical database is a significant challenge in resource-poor settings where the benefits of quality and process improvement are even more compelling. To this end, I founded and lead Sana, a volunteer organization consisting of doctors, informaticians, engineers, public health experts, business entrepreneurs and social scientists with the goal of designing and implementing a cellphone-based information system to improve quality of care in resource-poor settings. Sana, hosted by the Computer Science and Artificial Intelligence Laboratory at MIT, developed and maintains an open-source cellphone-based software that allows capture and transmission of any type of medical data (e.g. text, images such as photo and ECG, audio such as lung sounds, video such as ultrasound) through cellular networks to a back-end electronic medical record system that a remote expert can access to provide real-time decision support to front-line community healthcare workers (CHW). The software allows embedding of decision trees, protocols and checklists to assist semi-trained CHW with triage, diagnosis and treatment. It is currently implemented in India for screening for oral cancer and heart disease. Pilot studies are underway in Brazil to screen for eye disorders that lead to blindness, in Greece to monitor diabetic foot ulcers and in the Philippines to diagnose and treat hypertension. Databases created by these implementations are envisioned to be used to develop population-specific predictive algorithms and image analysis, the ingredients to the team-based learning system that constantly seeks to improve processes and outcomes in healthcare delivery.