| Theoretical biostatisticians
lead methodology development and study design
efforts for clinical and economic analysis and
address measurement issues in natural history
studies and bias adjustment in observational
studies. Experience includes employing Bayesian
techniques to profile health care providers;
applying the structural equation modeling technique
to address the direct and indirect effects of
hemoglobin level on medical costs in ESRD patients;
using modeled rates to improve the stability
of standardized incident and prevalent rates;
and developing Bayesian hierarchical spatial
models to smooth event rates, costs, and other
variables, thus revealing geographic patterns.
Theoretical biostatistical staff members use
the instrumental variable approach to adjust
the biased effect of diabetic care on medical
outcomes within the aged population, and apply
the Marginal Structure Model technique to incorporate
the historical relationship between EPO dose
and hemoglobin level in order to arrive at valid
estimates of EPO effect on medical outcomes for
ESRD patients.
Our strength in biostatistical methods and epidemiological
techniques rests on our expertise with very large
databases
and our ability to address the complex issues
related to study bias. We study incidence and
prevalence rates, analyze morbidity and mortality
outcomes, and interpret findings from large and
merged data sets.
Members of our group conduct and coordinate health
services research, including large observational
studies and multicenter clinical trials, and
we provide expert assistance on clinical and
epidemiological study design. We evaluate quality
of care, assess pharmaceuticals and delivery
systems, analyze economic burdens, and identify
trends in chronic diseases.
|
