Friedman Seminar: Postdoctoral Scholars

poster for post doc scholars seminar

This Friedman Seminar features six of our current postdoctoral scholars, presenting synopses of their varied research projects.

Alexandra Thorn


Alexandra Thorn holds a PhD in Biology from Tufts University, a Masters in Horticulture and Agronomy from the University of California, Davis, and a Bachelors in Biological Sciences from Wellesley College.  She interested in food security, sustainable agriculture, and regional resilience, and is currently working with Christian Peters on a project assessing the biological capacity to expand ruminant meat production in New York and New England.  Her general research interests include the application of biogeochemical modeling, spatial analysis, and stakeholder engagement to improve the sustainability and resilience of society.   Previous research includes the development of land cover scenarios for New Hampshireand assessment of the potential impact of future land cover on ecosystem services.


"Estimating Biological Capacity for Grass-based Ruminant Meat Production in New England and New York"

Meat production is a minor agricultural enterprise in the Northeastern United States. However, increasing interest in local food production and on-going challenges posed by drought and fire on Western rangelands give reason to believe that the importance of Northeastern grass-based ruminant meat could grow in the near future. The present study quantifies current and potential biological capacity for grass-based ruminant meat production in New York and New England –part of a project exploring supply chain barriers in the region. We summarized the spatial distribution of hay, pasture, and livestock production from the USDA Census of Agriculture, and then used land cover data from the Cropland Data Layer (CDL)to develop high resolution statistical models of land suitability for agriculture and grass production within the region. Regression tree analysis of CDL land cover categories as functions of soil and terrain properties show that the most important predictors of the location of cleared farmland for the date range 2010-2016 were the National Commodity Crop Productivity Indices (NCCPI) and the Non-Irrigated Capability Class (NICC). Eighty-two percent of current agricultural land cover is in map units with a dominant NICC of 1, 2, or 3. Roughly half of this land is perennial grassland (pasture and hay), which is skewed toward sites with lower NCCPI, more limited NICC, and higher slope. To estimate the potential productivity of pasture land area, we produced linear models of non-irrigated crop yield values for pasture, grass hay, and grass-legume hay from SSURGO against NICC classes and subclasses, drainage classes, slope, air temperature, and quintiles of depth to restrictive soil layers, using ordinary least squares linear regression.  We used these models to generate rasters of potential productivity, which we overlaid with our maps of the spatial distribution of pasture and hay, and aggregated to the county level.  The resulting estimates of potential county-level productivity were compared to livestock inventories from the USDA Census of Agriculture. We conclude that there is high biological potential to expand grass-based meat production in the region. Results will serve as input for a supply chain model of the grass-finished beef sector.

Yujin Lee


Yujin Lee is a postdoctoral fellow at the Friedman School of Nutrition Science & Policy at Tufts University.  She took the lead on a project to quantify the cardiovascular health and economic impacts of policy strategies to improve diet and reducethe incidence of CVD. Dr. Lee completed her PhD in Nutritional Sciences at Pennsylvania State University, her dissertation entitled “Effects of polyphenolic-rich dark chocolate and almonds on cardiovascular risk factors in overweight and obese adults”. She received her MS at Purdue University, and her BS degree in Nutrition Science at Yonsei University in South Korea.She has received several awards and honors and presented 6 abstracts in scientific meetings. She has authored 5 publications and is a co-author of a book chapter.  


"Cost-effectiveness of financial incentives for improving diet through Medicare and Medicaid" 

While economic incentives through health insurance are being considered to promote healthy behaviors, little is known about health or financial impacts of incentivizing diet, a leading risk factor for CVD. We estimated health and economic impacts of programs to incentivize healthful foods through Medicare and Medicaid over a 5, 10, and 20 y horizon.

Eleanor Shonkoff


Postdoctoral Fellow, ChildObesity180

Education: PhD, Preventive Medicine, Health Behavior Research, University of Southern California; MA, Social Psychology, Wake Forest University; BA Philosophy and Cognitive Science, University of Georgia

Previous experience: National Institutes of Health cancer prevention predoctoral fellow, Graduate teaching assistant, University of Southern CaliforniaResearch focus:parenting practices and child dietary intake


"Reliability and validity of digital images to assess plate waste in a restaurant setting"

Current methods for assessing child dietary consumption in restaurants are limited, particularly for large scale use. Digital images (DI) are valid and reliable for assessment in cafeterias, yet only preliminary evidence exists for restaurants. Given the significant increase in foods consumed away from home, an accurate and feasible method of determining consumption is critical. The objective of this study was to develop a robust digital imaging methodology and test reliability and validity compared to weighed plate waste (PW) for assessing consumption in the restaurant setting.Participants were parents who placed an order at one of 11 restaurants in two MA communities. Participants completed a survey and provided their child’s PW to be measured in grams with a food scale (OXO 1130800). For DI, PW was arranged on grid paper, details were annotated on the paper (e.g., packaging), and photographed from two angles. Two coders compared digital images to reference images and, without referring to weighed plate waste data, used a modified Comstock scale to estimate the percent of the original portion consumed. To calculate kcals consumed, grams consumed (DI and PW) were multiplied by kilocalories per gram (based on the restaurant’s nutrition information). Intraclass correlations, Spearman correlations, Wilcoxon signed rank tests, and Bland-Altman plots assessed correspondence between methods. Coders demonstrated acceptable reliability overall and within each meal component (e.g., entrées, side dishes. While mean differences were statistically different for some components, actual discrepancies were fewer than 6 kcals less than PW. Results demonstrate feasibility and reliability of digital images to assess child meal consumption in restaurant settings, including bothpurchased food and leftovers, with additional improvements to validity recommended for some meal components.

Heidi Lai


Heidi Lai is a Postdoctoral Fellow at the Friedman School of Nutrition Science & Policy at Tufts University. Her research focuses on objectively measured serial plasma phospholipid fatty acids and their longitudinal associations (including cumulative levels and change over time) with cardiometabolic health and healthy aging among the oldest old within the Cardiovascular Health Study in the United States, as well as globally within the Fatty Acids and Outcomes Research Consortium. Originally from the United Kingdom, Heidi completed her Ph.D. in Food Science and Nutrition, focusing on the dietary sources of (poly)phenols and the association with cardiovascular disease in the United Kingdom. and B.Sc. in Food Studies and Nutrition at the University of Leeds, UK. She had previously received an award for the Best Scientific Paper by A Young Scientist in 2016 from the Alpro Foundation in Belgium (Lai et al., 2015, Eur J Epidemiol; 30(9): p. 1035-48) and the Best Student Poster Prize 2015 from the Nutrition Society in the United Kingdom. Given thorough training in Food Biochemistry, Heidi is now focusing on strengthening nutritional epidemiology skills, by leading and collaborating in multiple projects within the Cardiovascular Health Study and Fatty Acids and Outcomes Research Consortium.


"Circulating n-3 polyunsaturated-fatty acids and the maintenance of healthy aging in older adults, the Cardiovascular Health Study"

Maintaining healthy aging (HA) is a crucial priority in older adults worldwide, givenglobal population aging, increased number of years living with disability, and the need for new treatments. Omega-3 polyunsaturated fatty acids (n3-PUFA) from seafood and plants exert favorable physiologic effects that could benefit HA. However, relationships between n3-PUFA and HA are not well-established, especially using serial biomarkers which provide highly objective measures.

Junxiu Liu


Junxiu (Juju) received her master of medicine from Peking Union Medical College, Beijing, China. She completed her PhD in epidemiology at the University of South Carolina Arnold School of Public Health. Her research interest is to understand the fetal origins of chronic diseases, life course epidemiology, and how comparative effective and cost-effective of diet policy intervention to improve diet and reduce the burden of chronic disease.


"Cost-Effectiveness of Financial Incentives and Disincentives for Improving Diet and Health through the Supplemental Nutrition Assistance Program"

The 2018 Farm Bill represents a major opportunity to reduce disparities in diet and health.  The largest component is the Supplemental Nutrition Assistance Program (SNAP), feeding 1 in 6 Americans.  Potential options include subsidizing fruits & vegetables(F&V), restricting sugar-sweetened beverages (SSBs), or implementing a broader food incentive/disincentive framework that preserves choice. Their comparative health impacts and cost-effectiveness are not established. Using a validated microsimulation model (CVD PREDICT), we estimated changes in CVD events, quality-adjusted life-years (QALYs), costs, and cost-effectiveness of 3 policy scenarios in SNAP adults: 1) 30% subsidy on F&V; 2) 30% F&V subsidy + SSB restriction; and 3) 30% subsidy on F&V, whole grains, nuts/seeds, seafood, plant-based oils, and 30% disincentive on SSBs, junk food, and processed meats. 

Graham R. Jeffries


Graham Jeffries is an agricultural data scientist and postdoctoral scholar working with Dr. Timothy Griffin. His interdisciplinary research unites remote sensing, agricultural modeling, and social science methods to leverage novel data streams for enhancing our understanding of sustainable and resilient agricultural systems. His postdoctoral work focuses on quantifying the capacity of Brazilian soy growers to adapt to shifts in climate variability driven by land use change (with support from the National Socio-Environmental Synthesis Center). He holds PhD and MS degrees from Tufts and a BA from Hampshire College.


"Satellite-based measurements of climate impacts on Brazilian soy production and farm management decisions"

A swelling stream of satellite imagery provides new opportunities to study the drivers and limitations of crop productivity, informing strategies for reducing yield gaps and improving the resilience of agriculture. In my lightning talk, I'll describe how I plan to use remote sensing, machine learning, and causal inference modeling to examine 1) climate variability impacts on Brazilian soy productivity and 2) the adaptive capacity of farmers to protect yields by shifting planting dates in response to altered wet season timings.


Wednesday, February 14, 2018 at 12:15 PM


Wednesday, February 14, 2018 at 1:15 PM


Behrakis Auditorium
The Jaharis Center
150 Harrison Avenue
Boston, MA 02111


Name: Jen Zahradnick
Phone: 617-636-3773


Boston Degree Programs
There is no content to display.

Whoops. Something is wrong here. You shouldn't being seeing this. Please try to load the page again.