Term of Award

Fall 2019

Degree Name

Doctor of Public Health in Biostatistics (Dr.P.H.)

Document Type and Release Option

Dissertation (open access)

Copyright Statement / License for Reuse

Creative Commons License
This work is licensed under a Creative Commons Attribution 4.0 License.


Jiann-Ping Hsu College of Public Health

Committee Chair

Karl Peace

Committee Member 1

Lili Yu

Committee Member 2

Xinyan Zhang


In Clinical Practice, combination drug therapy has become common in treating many disease conditions. The purpose of these combinations is often to ensure optimal efficacy and to reduce adverse effects that may arise from monotherapy. Clinical trials have also been conducted to ensure efficacy and safety of these combinations before they are introduced into the market. However, adverse effects still occur with combination therapies. The objective of this study is to (1) To determine a region of optimum doses of Drug A and Drug B in combination while focusing on efficacy alone (2) To determine a region of optimum doses of Drug A and Drug B while focusing on efficacy incorporating important safety constraints. Using Hypertension as the disease of interest, the primary efficacy endpoint is the change in diastolic blood pressure from baseline to the end of treatment at 8 weeks, and the adverse effect is edema. Drug A in doses of 0mg, 2.5mg, 5mg, 10mg and Drug B in doses of 0mg, 20mg, 40mg, 80mg were used in combination in the treatment of hypertension I and II. Response Surface Methodology (RSM) was used to find the doses of Drug A and Drug B which maximized efficacy, and a Probit-Normal Model was used to model the probability of the occurrence of edema. Results showed an unconstrained optimal decrease of 21mmhg corresponding to 12mg of A and 48mg of B. The point of minimum risk of the probability of edema occurring coincided with 2.5mg of Drug A and 20mg of Drug B (0.006%). The region of minimal edema incidence lay around the combination of 2.5mg of Drug A and 20-40mg of Drug B. A combination of 10mg of Drug A and no dose of Drug B showed the highest probability of edema. At a desired target of zero probability of edema occurrence, the constrained optimal value was about 15mmhg. Response Surface Methodology (RSM) has been applied in many disciplines; constraint optimization methods are relatively novel in their application to health and clinical dosing. Application of this method will require formulation of constraints that are appropriate for the disease of interest, and that is also clinically appropriate.

Research Data and Supplementary Material