Current Projects and News
Some Recent News and Projects
On August 2016 I joined the Main Line Health System (MLHS) Center for Population Health Research at Lankenau Institute for Medical Research (LIMR), as Associate Director. The Center is a collaboration between the MLHS, LIMR and Thomas Jefferson University College of Population Health.
One of our first projects was the creation of a Population Health Dashboard (prototype). The main goal of this project is to find ways in which large, open datasets can be used to improve the design, implementation and evaluation of population health strategies by health systems. I believe tools like this dashboard can be used to improve decision making and collaborative efforts across partners involved in the production and maintenance of health beyond the hospital. Slides with a brief overview here.
A scientific review of the literature on Supervised Consumption Facilities for the City of Philadelphia. You can read the report here .
Cited by Vox
But they went further, developing models to quantify how many drug overdose deaths could be prevented and how much money could be saved with a supervised consumption facility in Philadelphia. They found that as many as 76 drug overdose deaths annually could be prevented, compared to the 907 people who died of an overdose in Philadelphia in 2016. And in terms of skin and soft tissue infections alone, the city would save as much as $1.8 million in hospitalization costs each year, according to the review.
Submission to the National Science Foundation Big Data Regional Innovation Hubs
Along with a multidisciplinary team, submitted an application to the Big Data Regional Innovation Hubs: Establishing Spokes to Advance Big Data Applications. You can read an abbreviated version of our submission (2017) here
As Jim Gray observed the first, second and third paradigms of science –empirical, analytical, and simulation –have successfully carried us to this point in history. The next wave of innovation and breakthroughs will come from data-driven science . Indeed, we are at the cusp of the fourth paradigm, increasingly grasping the power of data-intensive science and with this, its opportunities and challenges as we move away from hypothesis-driven to data-driven science. Among these challenges is the realization that algorithmic design and the data used for it may have profound societal implications across dimensions such as fair housing, economic opportunity and discrimination. As we enter this fourth paradigm it is vital to ask questions such as: are racial minorities less likely to find housing via algorithmic matching systems ? Does algorithmically-controlled personalization systematically restrict the information available to the economically disadvantaged ? Are content recommendations steering us away from decisions we would like to make? And, do online markets unfairly make goods more expensive for particular demographics or particular geographic locations ? In the context of health care, with algorithms that require highly dimensional patient data to make precise diagnoses and treatment options, we begin to ask whether populations in less participatory cultures be “information-poor” in this new system?
The intricate institutional and regulatory aspects of health care make it a particularly complex environment, which underscores the urgency to develop an open framework and tools to codify clinical practice guidelines, prediction and decision support algorithms, that decisively engage in multidisciplinary evidence generation and transparency and that extend into policy design and implementation in order to avert the potential harms from data and algorithmic bias
Some Recent Talks and Presentations
Design Thinking Workshop at NYU Data Science Center in partnership with Women in Data (2017) --Design Thinking & Data Science for Healthcare