In addition to the theoretical advances described above, a major part of our current research program is to develop and implement statistical tools that can integrate the signals from many biomarkers into coherent, easy-to-interpret summary measures for use in many fields. Our goal is for our research findings to have a large and short-term impact on health and on research capabilities.

The integrated biomarker tools we are developing have the potential to upend the way clinicians use biomarker data to guide clinical decisions. For example, cholesterol may not be the best measure of what doctors are trying to measure when they look at cholesterol levels. Our tools may provide much more precise information about many aspects of the health state of patients: rather than separately looking at cholesterol, LDL, HDL, and triglycerides, a single lipid dysregulation score might be both easier to use and more precise, for example. Global dysregulation scores may aid in predicting clinical frailty. While not experts in Knowledge Translation (KT), we are partnering with clinicians and clinical data experts to develop both tools and their applications, and then to get clinicians to start using them.
The same tools that could aid doctors in clinical decision making may also prove to be important tools for clinical research. In particular, one of the main challenges of studying new pharmaceutical products is our inability to observe long-term side effects. If measures of physiological dysregulation are precise enough, they may provide a good window into the potential for long-term side effects of some medications without a need to actually follow patients for 20-30 years before regulatory approval. They might also show that some medications create systemic benefits beyond their intended therapeutic use.
As mentioned in the ecology and evolution section, our dysregulation measures are likely to serve as reasonable proxies for body condition of wild animals in ecological studies. We believe a similar principle can apply to lab animals as well, particularly in early testing of pharmaceutical products.
Economists, sociologists, epidemiologists, and demographers often seek reliable measures of health state in population surveys or cohort studies. Existing methods such as self-reported health are known to be biased or otherwise problematic. The dysregulation scores we propose are likely to provide an objective measure of individual health that can be obtained relatively cheaply reliably in such studies. In many cases, it can be calculated from existing data.
While not the core of our lab’s focus, we believe the approaches we are applying to aging and physiology generally can be profitably applied to some specific chronic diseases. For example, diabetes might best be understood as a disease of endocrine dysregulation, and the measures we propose might help illuminate the dynamics of this.