Working Paper: NBER ID: w22873
Authors: Gautam Gowrisankaran; Keith A. Joiner; Jianjing Lin
Abstract: A recent literature finds that hospitals “upcode” when doing so increases revenues, suggesting that incomplete information creates substantial distortions. However, reporting complete information is itself costly. We examine the impact of both revenues and coding costs on hospital billing practices for Medicare inpatients. Following the literature, we investigate the fraction of patients top coded as the revenues from top coding vary. We then examine how this fraction changes following Medicare reforms—which increased the requirements and complexity to justify top codes—interacted with hospital electronic medical record adoption—which may decrease coding costs. We find evidence that coding costs drive top coding behavior.
Keywords: hospitals; payment incentives; coding behavior; Medicare; electronic medical records
JEL Codes: H51; I11; I13; O33
Edges that are evidenced by causal inference methods are in orange, and the rest are in light blue.
| Cause | Effect |
|---|---|
| Increases in extra revenue from top coding (H27) | Increases in hospitals reporting top codes (I11) |
| Increases in revenues (H29) | Reported top codes (Y10) |
| EMR adoption (I18) | Reported top codes post-reform (I18) |
| EMR adoption (I18) | Better documentation (Y50) |
| EMR adoption (I18) | More never events reported (Y10) |
| EMR adoption (I18) | Top coding behavior for medical and surgical DRGs (I11) |
| EMR adoption (I18) | Top codes for medical DRGs (I18) |
| EMR adoption (I18) | Top codes for surgical DRGs (I11) |