aka the Causal inference for Preventive Medicine and Epidemiology (CiPreme) Research LabThe goal of the CiPreme Lab is to advance the science of preventive medicine and epidemiology and contribute to the prevention of major chronic diseases through the application of innovative epidemiologic, econometric and causal inference methods, as well as computational modeling and simulation tools with the aim of translating research evidence into effective clinical, public health and policy interventions.
|
THE PROBLEMChronic diseases represent a huge burden in the U.S.Seven of the 10 leading causes of death are chronic diseases. These include heart disease, cancer, chronic lower respiratory diseases, stroke, Alzheimer disease, diabetes and kidney disease.
They are highly prevalent, represent a huge cost to the healthcare system and families yet they are mostly preventable. Rigorous study designs and causal inference methods are needed to better estimate causal effects and understand mechanisms. This could help identify, optimize and design effective health interventions to prevent these chronic diseases. |
Seven of the 10 leading causes of death are chronic diseases. Photo credit: cdc.gov |
THE TOOLSRigorous study designs and causal inference methods are needed!Examples of experimental designs include: 1) Randomized experiments: This is the most common approach and the gold standard for estimating causal effects. They include pragmatic and cluster randomized trials where individuals or groups are randomized to either the intervention or the control group and the mean outcome is compared between the two groups to determine the causal effect. 2) Quasi-experiments. These are sometimes known as instrument-based methods and leverage the quasi-randomness of a natural experiment such as a policy or event to evaluate its possible impact on a given outcome, typically at the local, state or country level. 3) Virtual experiments. These are hypothetical experiments that leverage existing data sources and published evidence as well as tools such as predictive analytics, computational modeling and simulation tools to evaluate the plausible impact of potentially achievable interventions at the individual and societal level. Other important tools Tools such as Robin's g-methods and marginal structural models as well as other methods such causal mediation analysis, causal survival analysis, machine learning and transportability methods are crucial in estimating causal effects (especially in the presence of time-varying confounding), quantifying heterogenous treatment effects and indirect effects and transporting effects from one study population to a target population. Additional important tools include: evidence synthesis and meta-analysis, machine learning, decision analysis and cost-effectiveness analysis. |
The Ladder of Causation from the Book of Why: The New Science of Cause and Effect by Pearl and Mackenzie.
|
THE HYPOTHESIZED MECHANISM AND SOLUTIONMost chronic diseases may share common pathways and may be preventable through lifestyle, metabolic and social interventionsOxidative stress, inflammation and endothelial dysfunction are thought to be common pathways to major chronic diseases including cardiovascular diseases, type 2 diabetes (and obesity), cancers and neuro-degenerative diseases such as Alzheimer's diseases and other dementias.
As such, common interventions may be used to prevent chronic diseases. These include lifestyle and metabolic modifications including but not limited to the following as outlined by the American Heart Association and termed "Life’s Essential 8™": diet, physical activity, nicotine exposure, sleep health, body weight, blood lipids, blood glucose, and blood pressure. In a like manner, the American College of Lifestyle Medicine (ACLM) identified six pillars of lifestyle medicine which includes a whole-food, plant-predominant eating pattern, physical activity, restorative sleep, stress management, avoidance of risky substances (e.g. tobacco, excessive alcohol consumption) and positive social connections. Lastly, social interventions such as the Special Supplemental Nutrition Program for Women, Infants, and Children (WIC) and Medicaid Expansion Program are examples of interventions that have strived to address modifiable social determinants of health and temporarily assist families in need. |
The American Heart Association (AHA)' Life Essential 8 |