Font Size



PHDL Scientists Featured in The Economist

The statistics are staggering, and it is hard to overstate the scale of use and abuse of drugs in the U.S. A model developed by Dr. Graph showing the increase in opioid deaths.Hawre Jalal, faculty researcher at the Public Health Dynamics Laboratory (PHDL) at the University of Pittsburgh Graduate School of Public Health, demonstrates that the typical overdose victim is becoming younger and more urban, as reported by The Economist on October 26, 2017 in The shifting toll of America’s drug epidemic. The results (“Sub-epidemics within the Opioid Epidemic”, H. Jalal, J. Buchanich, L. Balmert, M. Roberts and D. Burke) presented by Dr. Jalal at the October 2017 Annual Society for Decision Making (SMDM) Meeting in Pittsburgh, PA, shows red alerts for U.S. drug overdose deaths per 100,000 population by age, demographics and drug type. The highest rates of prescription-opioid abuse can be found among midde-aged rural whites, including women. By contrast, both fentanyl and heroin users tend to be much younger, more likely to live in cities, somewhat more racially diverse and overwhelmingly male.

In another article, Forecasting the opioid epidemic, on October 28, 2017, The Economist raised the questions: When will the opioid epidemic peak? And how many will it kill? Dr. Donald Burke, dean of the Graduate School of Public Health, University of Pittsburgh, points out that the number of fatal drug overdoses has doubled every eight years for the past 37. Epidemiologists are frantically scrambling to go beyond simple best-guess estimates to dynamic models that can forecast addiction andGraph showing projected deaths from opioids. overdoses more accurately. Dr. Jalal and other scientists from the PHDL at the University of Pittsburgh are developing a dynamic transmission disease model of the opioid epidemic, matching data in the national drug-use survey to outcomes in mortality. These results (“A dynamic transmission disease model of the opioid epidemic”, D. Sinclair, H. Jalal, M. Roberts, D. Burke) presented by Dr. David Sinclair, PHDL post-doctoral researcher, at the 2017 Annual SMDM Meeting. predict that prescription opioid deaths will rise slowly to about 20,000 a year within the next five years, but heroin and fentanyl deaths will increase markedly to 72,000 per year by 2025.

PHDL Participates in SMDM Annual Meeting


The 39th Annual North American Meeting of the Society for Medical Decision Making (SMDM) was held in Pittsburgh, PA, October 22-25, 2017. The focus this year was on “Better Decisions Through Better Data Processes”. The Society for Medical Decision Making has a long history of developing methodologies which take advantage of complex data structures to enhance medical decision making and advance policy formation. The 2017 Annual Meeting explored themes to ensure the credibility and usability of the Society’s efforts and to promote their vision of an integrated approach to health care decision making, through wise use and thoughtful communication of data.

Several Public Health Dynamics Laboratory (PHDL) faculty and postdocs participated in teaching short courses, and oral and poster presentations. The PHDL also participated in an exhibitor’s booth to share information on many of the PHDL software tools, including a demo of FRED (Framework for Reconstructing Epidemiological Dynamics).

 David Sinclair SMDM 2017Hawre Jalal SMSM 2017

Click here to see our presenters

New PHDL Scientific Director

PyneSaumyadipta Pyne, PhD, joined the Public Health Dynamics Laboratory on October 1, 2017, as scientific director and will also serve as an associate professor in the Department of Biostatistics, both at the University of Pittsburgh Graduate School of Public Health.

Pyne was previously a professor at the Indian Institute of Public Health, Hyderabad. He formerly held the prestigious post of P.C. Mahalanobis Chair and professor and head of bioinformatics in the C.R. Rao Advanced Institute of Mathematics, Statistics and Computer Science. He is a former senior research fellow of the Indian Statistical Institute. Pyne is the founding chairman of the Computer Society of India’s Special Interest Group on Big Data Analytics and leads the Health Analytics Network.

Pyne received his doctorate from the State University of New York at Stony Brook, working simultaneously in the departments of computer science, and molecular genetics and microbiology. He conducted his postdoctoral research at Broad Institute of MIT and Harvard then worked at the Dana-Farber Cancer Institute, Harvard Medical School. His research interests include big data in life sciences and health informatics, computational statistics, and high-dimensional data modeling. He is actively engaged in promoting big data research and training activities.

CMU Wins CDC Competition Again

Roni RosenfeldFor three years now, Carnegie Mellon University's Delphi-Stat group, under the leadership of Roni Rosenfeld, PhD, has won the U.S. Centers for Disease Control and Prevention's competition in predicting week-to-week influenza activity levels in 10 regions of the nation, along with national totals. "We're gratified that our forecasting methods continue to perform as well as they do, but it's important to remember that epidemiological forecasting remains in its infancy", said Dr. Rosenfeld. The forecasting effort is expected to help health officials better plan for peak periods of infection, while providing the public with more accurate information to avoid getting the flu. Read more: CMU is the world's best at predicting influenza activity.

Modeling the Opioid Epidemic Workshop

Modeling the Opioid Epidemic WorkshopThe Public Health Dynamics Laboratory (PHDL) and the Clinical Translational Science Institute (CTSI), both at the University of Pittsburgh, co-sponsored a one-day national workshop on “Modeling the Opioid Epidemic: Insights and Potential Solutions from Computational Science” on October 2, 2017 in Pittsburgh. The workshop, the first in a series, was a great success. Over 140 researchers, clinicians, students and public health officials attended, representing more than a dozen academic institutions, federal agencies, community organizations, and health care systems. This workshop provided investigators and practitioners with a unique opportunity to learn more about the computational work being done at multiple scales in response to the opioid epidemic. Presentations and discussions featured speakers and panelists working at all levels and scales of computational modeling to examine the opioid crisis from the biological, to the individual human, to population-wide levels of science.

Modeling the Opioid Epidemic Workshop

To view recordings of the workshop presentations, click Read More or visit


Coming Soon: New Release of FRED

New Release!A new version of FRED (A Framework for Reconstructing Epidemiological Dynamics) will soon be ready for release. FRED is a tool for building epidemiological agent-based (individual-based) models and is designed to study how patterns of health conditions in defined populations vary over time. The new FRED will make population modeling easier. It is a unique tool for social science modeling and no computer programming is needed. A systems thinking approach is required to identify conditions of interest, their states, and the rules for changing states. FRED will simplify the workflow environment and manage the data produced by the simulation. To read more about the new FRED platform, click here.

You are here: Home News