Epidemiology, University of North Carolina-Chapel Hill (2016)
Cohort Five Fellow
Professor/Director, Department of Epidemiology Injury Prevention Research Center
Director, National MCH Center for Child Death Review
Current Institutional Affiliation
Alaska Department of Health and Human Services: Program Manager, Alaska Birth Defects Registry/Fetal Alcohol Syndrome Surveillance Program
Areas of Expertise
Evidence-Based/Evidence-Informed Programs, Maternal Health, Prevention Science, Program Evaluation, Difference in Difference Modeling, Latent Class Analysis or Cluster Analysis, Longitudinal Data Analysis, Hierarchical Linear Modeling, Regression Modeling, Survival Analysis or Hazard Models, Structural Equation Models, Case Study Research, Adolescents and Young Adults, Infants and Toddlers, Racial/Ethnic Minority Groups, Child Death Review, System Dynamics, G-Formula
Jared Parrish is an epidemiology doctoral candidate (with a concentration in injury) at the University of North Carolina at Chapel Hill. His dissertation is building a longitudinal birth cohort through data linkages and multiple imputations, to quantify systematic error and elucidate weighted cumulative maltreatment risk profiles throughout the life course. Parallel to his education, Mr. Parrish is the senior scientist with the Maternal and Child Health Epidemiology Unit of the Alaska Division of Public Health. Mr. Parrish has primarily focused his research efforts on applying epidemiologic approaches to comprehensively quantify the incidence of child maltreatment through data linkages, cases ascertainment, and operationalizing definitions. He also participates on various committees concerned with applying a public health approach to maltreatment surveillance. Mr. Parrish holds a MS in Epidemiology and a BS in Community Health Science, and is a member of the International Epidemiological Association.
Quantifying Methodological Challenges in Maltreatment Research Through Novel Data Linkages, Outcome Classification, and Exploratory Risk-Set Modeling
Measurement errors often impact maltreatment studies but rarely receive adequate attention. The proposed study will implement a mixed-design strategy by combing the Alaska Pregnancy Risk Assessment Monitoring System with novel administrative records longitudinally. This study will specifically aim to: 1) quantify the amount of error in population based studies that cannot account for censoring, 2) assess the reliability of maltreatment classifications made by the child death review team using a sensitive tiered public health definition, and 3) create dynamic risk-set scoring using time-to-event and explore system dynamics solutions. Improving the validity of maltreatment research is extremely important to better understand the extent of the problem and target interventions. The proposed study will address three gaps that are un-validated, limited in scope, or currently operating under untested assumptions. By instituting population-based rigorous longitudinal epidemiologic studies at the state level, the gap between scientific inquiry and applied epidemiology can potentially be reduced.