OIG identifies 46 hospitals for DRG 475 upcoding

As part of its crackdown on upcoding, OIG has started identifying hospitals that it believes are upcoding to DRG 475, which encompasses respiratory diseases for patients who need continuous mechanical ventilation.

The good news is that only a small percentage of hospitals — 46 institutions out of 3,714 — in OIG’s sample appear to be upcoding. The hospitals were not identified. The bad news is that because DRG 475 is one of the pricier diagnostic groups, the federal fraud hounds are likely to pay special attention to it. An earlier OIG analysis found that upcoding lower DRGs to DRG 475 cost Medicare about $10,000 per case. All 46 hospitals have been referred to the agency’s Office of Investigations.

The study looked at claims from between 1993 and 1996. At the institutions in the sample, DRG 475 discharges made up more than 1.5% of all discharges, and the proportion of DRG 475 discharges to total discharges blossomed by more than 100% between 1993 and 1996.

The 46 hospitals flagged for having abnormal billing patterns had DRG 475 rates that soared 160% over three years, compared to a national average of 23%. The proportion of total discharges bloomed 159%, or 2.12% of total discharges, compared to a national increase of 16% or .85% of all discharges. That should give you a good clue as to the baseline numbers OIG and carriers will be using in their audits.

OIG estimates that these problem hospitals were overpaid as much as $11.5 million in 1996, or 31% of their total DRG 475 reimbursement. An earlier HCFA study estimated that upcoding accounted for 7% of DRG 475 claims, for overpayments of $67 million.

OIG admits it can’t say for certain that upcoding occurred in each case. "We recognize that only record reviews by trained professionals will establish if incorrect coding has occurred at the 46 hospitals identified," it concedes in the conclusion of the report. As in previous reports, OIG recommends that HCFA routinely analyze hospital billing data for upcoding.