Abstract:
A large literature documents the wide variety in medical treatments in the United States. However, in the absence of administrative data, similar variations have received less attention for primary care and certainly less attention in low and middle income countries. We use a novel methodology that combines audit studies with direct tests of health care providers’ knowledge to address this gap. We show that “avoidable” care, defined as care that was unlikely to improve the underlying condition, accounts for as much as 75% of all expenditures in our rural Indian setting. These costs arise both due to the use of unnecessary medicines and under-evaluation that leads to incorrect treatment for the underlying condition. Using the tests of knowledge, we show that the majority of avoidable care occurs because doctors truly “believe” that these medicines are necessary—and this is more so for fully trained and qualified providers. On the other hand, under-evaluation whereby the doctor knew to give the correct treatment but did not when faced with the same patient, appears to reflect a lack of effort and accounts for 45% of avoidable care. We discuss the implication of these results for regulation and policy and pose further questions for theoretical and empirical research.