There is no room for complacency in the healthcare industry. Every passing minute is an opportunity for research that could save lives.
"I've always believed that the power of statistical methods lies not just in theory but in their ability to address problems that are happening in the real world. It's the possibility of saving and improving people's lives," Dr. Chen Yang remarked, encapsulating a philosophy that has driven his remarkable contributions to biostatistics and clinical research.
However, things are not always as clear as they may seem. In many cluster randomized trials (CRTs), only limited evidence, if any, is presented regarding the proper handling of subgroup analyses or overall methodological transparency.
For example, a stepped-wedge CRT (SW-CRT) designed to promote advanced care planning (ACP) among African-American patients can be criticized for its lack of racial disparity detection due to its focus on sample size calculation based on the overall treatment effect. This lack of transparency, if left unchecked, can greatly hinder future research effects in an otherwise critical field of study.
Breaking New Ground in Clinical Research
Dr. Yang's dedication to bridging the gap between theoretical frameworks and practical applications has set new benchmarks in the industry, transforming how data is utilized to improve health outcomes. One example is to evaluate the potential racial disparities for newly developed training programs for physicians. While the target intervention is designed for physicians, the feedback is collected from patients, leading to a CRT design for patients from different race groups.
To adequately power subgroup analyses in CRTs, particularly those with binary outcomes as the feedback, he led a project to develop a theoretical framework for power calculation in SW-CRTs, along with a user-friendly power calculation tool, which leads to 10% to 15% more accurate power. This tool is now used in ongoing NIH grant applications for research on improving communication between physicians and advanced cancer patients.
Dr. Yang's journey, marked by significant milestones, reflects his commitment to innovation and problem-solving. Holding a Ph.D. in statistics, an M.Sc. in statistics, and a B.Sc. in mathematics, he has transitioned from academia to the forefront of clinical research at the Icahn School of Medicine at Mount Sinai. His work in statistical methodology development has successfully advanced the field by providing invaluable tools for researchers worldwide.
One of Dr. Yang's notable contributions to his field is his 2024 paper, "Power Calculation for Detecting Interaction Effect in Cross‑Sectional Stepped‑Wedge Cluster Randomized Trials: An Important Tool for Disparity Research." This phenomenal tool allows researchers to monitor power calculations in real time, particularly for designing stepped-wedge CRTs tailored specifically to interaction effects similar to racial disparity. By applying the sample size recommended by Dr. Yang's method, these SW-CRTs are effectively protected from the risk of insufficient statistical power of significant discoveries or erroneous statistical outcomes.
From Theory to Practice: Real-World Impact
These advancements are vital in clinical trials, where understanding the heterogeneity of treatment effects can trigger more personalized and effective interventions. Dr. Yang and the team's proposed power calculation method was employed in a grant proposal aiming to study the impact of the clinical decision support system on improving the percentage of goal-of-care talk between oncologists and cancer patients identified as having a high risk of mortality.
In this trial, the goal-of-care talk is seen as a way for clinicians to make informed treatment recommendations–however, one that is unfortunately offered to a lesser degree for patients who belong to minority groups. The trial, designed in the required stepped-wedge format, included the participation of nine separate clinics over three consecutive six-week periods.
With Dr. Yang's power calculation tool, it was estimated that if each clinic had enrolled 200 patients, even with a 20% attrition rate, the statistical power for detecting a moderate level of racial disparity due to the intervention would be at least 86% accuracy in a scenario with 60% white patients, and 80% in a scenario with 70% white patients.
Advancing Finance Research Through Innovative Initiatives
Dr. Yang's innovative approach extends beyond clinical trials. His earlier work, "A Statistical Methodology for Assessing the Maximal Strength of Tail Dependence," published by Cambridge University Press in 2020, has also had expansive waves across the industry. This novel methodology helps estimate the likelihood of extreme co-movements between two random variables, a tool essential for risk management in finance and insurance. He has empowered researchers and practitioners to navigate complex data landscapes better by making these methods accessible and practical.
Existing methods for measuring tail dependence often underestimate extreme co-movements between dependent risks and may not align with the new paradigm of prudent risk management. To address this, Dr. Yang developed a new method that incorporated paths of maximal dependence between risks into measurements of extreme co-movements (measured monetarily), such as oil and gold prices or their investment returns.
However, these new measures currently lack empirical estimators and statistical inference theory, which hinders their practical application in promoting prudent risk management. This is another typical gap between theoretical results and practice. To bridge this gap, Dr. Yang proposed a procedure based on the average block minima estimator for these new measures. This procedure provides reasonable estimation using the observed actual data only, which avoids validating extra model assumptions. Moreover, the biases can be easily controlled by very few parameters.
The Broader Industry Context
Dr. Yang's proficiency extends beyond traditional statistical methods. His research encompasses various applications, utilizing techniques such as mathematical modeling, large-scale simulation, economic modeling, machine learning, and natural language processing.
His ability to quickly identify the most suitable analytic methods for various practical problems, even with limited background knowledge, sets him apart from many of his peers. This adaptability and problem-solving acumen have solidified his status as a key figure in translating complex mathematical frameworks into medical solutions.
Futurewise, industry forecasts predict continued growth and innovation in research methodologies. The demand for precise and adaptable statistical tools is expected to rise, driven by the increasing complexity of data and the need for personalized healthcare solutions.
Reflecting on the Journey, Looking to the Road Ahead
As Dr. Yang reflects on his journey, he remains focused on the future. "The ultimate goal of our research is to enhance the precision and applicability of statistical methods across various domains," he shares. "By continuously refining our approaches, we aim to provide researchers with tools that streamline processes to optional levels."
As the clinical research terrain continues to reshape and recolor, Dr. Yang's methods and insights will be key to setting new benchmarks.
Precision is a door in clinical research, and Dr. Yang is a turnkey.
