Working Paper: NBER ID: w16900
Authors: Anirban Basu
Abstract: The United States aspires to use information from comparative effectiveness research (CER) to reduce waste and contain costs without instituting a formal rationing mechanism or compromising patient or physician autonomy with regard to treatment choices. With such ambitious goals, traditional combinations of research designs and analytical methods used in CER may lead to disappointing results. In this paper, I study how alternate regimes of comparative effectiveness information help shape the marginal benefits (demand) curve in the population and how such perceived demand curves impact decision-making at the individual patient level and welfare at the societal level. I highlight the need to individualize comparative effectiveness research in order to generate the true (normative) demand curve for treatments. I discuss methodological principles that guide research designs for such studies. Using an example of the comparative effect of substance abuse treatments on crime, I use novel econometric methods to salvage individualized information from an existing dataset.
Keywords: Comparative Effectiveness Research; Health Economics; Individualization
JEL Codes: C11; D61; I18
Edges that are evidenced by causal inference methods are in orange, and the rest are in light blue.
| Cause | Effect |
|---|---|
| CER information (Y10) | physician-patient dyad (I11) |
| physician-patient dyad (I11) | treatment choice behavior (D91) |
| treatment choice behavior (D91) | aggregate demand for treatments (E20) |
| traditional CER methodologies (C90) | inefficiencies (D61) |
| individualized CER (I11) | true marginal benefits curve (D61) |
| identification of individual-level treatment effects (C22) | significant heterogeneity (C21) |
| identification of individual-level treatment effects (C22) | improve decision-making processes (D91) |