Working Paper: NBER ID: w12623
Authors: Christopher C. Afendulis; Daniel P. Kessler
Abstract: What are the important tradeoffs in consulting a single expert for both diagnosis and treatment? On one hand, an integrated diagnostician may have the incentive to recommend treatments that are not in the buyer's best interests. On the other hand, joint production of diagnosis and treatment by an integrated diagnostician may be more efficient. We examine an important special case of this problem: the costs and health outcomes of elderly Medicare beneficiaries with coronary artery disease. We compare the empirical consequences of diagnosis by an "integrated" cardiologist -- one who can provide surgical treatment -- to the consequences of diagnosis by a non-integrated cardiologist. Diagnosis by an integrated cardiologist leads, on net, to higher health spending but similar health outcomes. The net effect contains three components: reduced spending and improved outcomes from better allocation of patients to surgical treatment options; increased spending conditional on treatment option; and worse outcomes from poorer provision of non-surgical care. We conclude that accounting more completely for doctors' incentives to refer patients in setting reimbursements, or in the alternative, allowing doctors more freedom to make and receive payments for referrals, could reduce spending and improve quality.
Keywords: health care; diagnosis; treatment; medicare; cardiologists
JEL Codes: I1
Edges that are evidenced by causal inference methods are in orange, and the rest are in light blue.
Cause | Effect |
---|---|
Diagnosis by an integrated cardiologist (I11) | Higher health spending (H51) |
Diagnosis by an integrated cardiologist (I11) | Similar health outcomes (I14) |
Redirection from bypass surgery to angioplasty (G33) | Reduction of spending (H61) |
Redirection to more costly treatments (H23) | Increase in spending (H59) |
Diagnosis by an integrated cardiologist (I11) | Increased mortality rates for nonsurgical patients (I12) |