This dissertation uses proprietary drug screening data, illicit drug prices from the DEA STRIDE database, and national and local macroeconomic variables to measure the price responsiveness and treatment effectiveness of methadone maintenance therapy (MMT) on patients in a voluntary, methadone-treatment program in a rural Alabama county. This is done using conventional, myopic, and rational models of demand. The demand for illicit drugs is found to be sensitive to national drug prices as estimated from the Drug Enforcement Agency's System to Retrieve Drug Evidence (STRIDE), length of time in treatment, previous consumption, and the local unemployment rate. An important innovation in this paper is the use of temperature data from coca-plant-growing regions as an instrument for the cocaine prices taken from the DEA STRIDE database. Use of this instrument yields estimation results more in line with the predictions one obtains from economic theory. The estimation results imply that methadone maintenance therapy is a substitute for all illicit drugs under analysis. The implied price elasticity of cocaine use by MMT patients ranges from -0.0003 to -0.01284 and the unemployment elasticity of cocaine use is -0.00385.