All Content

Admitting to a Readmit Problem

Admitting to a Readmit Problem

We have a friendly disagreement within our hospitalist group. Some of our physicians believe we should track readmission rates. They believe it is a marker of quality. Others do not. What do you think?

Richard Mackiewicz, MD, New York, NY

Dr. Hospitalist responds:

Policymakers certainly are thinking about hospital readmission rates these days. Hospital readmissions sometimes can indicate poor care or poor coordination of care. Most hospitalist programs do not track readmission rates…but maybe they should.

I have a feeling payers, such as Medicare, will implement policies in the future that will force hospitals and hospitalists to closely monitor readmission rates. Why do I think that? Because, aside from poor care, unnecessary readmissions cost the system money—lots of money. What if I told you 17% of your patients are readmitted to a hospital within 30 days? Not high enough? How about 31%?

I admit my hospitalist program doesn’t track readmission rates. I have no clue what percentage of our patients get readmitted within 30 days. But MedPAC does. A recent MedPAC analysis of 2005 Medicare Provider Analysis and Review data found 6.2% of patients discharged from hospitals are readmitted within seven days. This percentage grows to 11.3% at 15 days and 17.6% at 30 days. That 17.6% translates to roughly $15 billion in Medicare spending.

Data for patients with end-stage renal disease (ESRD) are even more staggering. Hospitalized ESRD patients are readmitted within seven days at a rate of 11.2%. Within 15 days, that becomes 20.4%. Within 30 days, 31.6% of patients with ESRD are readmitted to the hospital.

Surprised at the high numbers? I was. It’s not just patients of this type. Some of my patients get readmitted for reasons that have nothing to do with previous admissions. How can we prevent that? MedPAC ran numbers with only “potentially preventable hospital readmission rates.” The readmission rates for all comers were 5.2% at seven days, 8.8% at 15 days and 13.3% at 30 days. This translated to $5, $8, and $12 billion dollars, respectively, in potentially unnecessary spending of Medicare dollars.

If unnecessary hospital readmissions are so bad, why haven’t hospitals and hospitalists placed a bigger emphasis on preventing them? There are several reasons. One is a lack of awareness of the problem, but the main reason likely is lack of financial incentive to do so.

Most hospitals receive Medicare payment regardless of readmissions. In some states, CMS contractors and quality improvement organizations aggressively have denied payment for readmissions within 30 days, but these are the exceptions, not the rules. In many parts of the country, hospitals have no financial incentive to reduce readmissions unless they can fill the unused beds with more “profitable” patients.

Under the case-based DRG payment model, Medicare actually rewards hospitals for shorter lengths of stay. Hospitals have developed systems to encourage providers to discharge patients as quickly as possible. In fact, many hospitals even look at physicians’ inpatient length of stay as a measure of performance. From the physician perspective, why not discharge the patient as quickly as medically appropriate? The hospital commends you for doing so and if the patient is readmitted, you get to bill a higher admission code rather than a lower-paying subsequent day visit code. More admission and discharge billing means more money.

So how will policymakers address the issue of unnecessary hospital readmissions? Simple. They’ll restructure the compensation model. Medicare addressed the problem of hospital-acquired infections by not paying for them. Hospitals reacted by implementing measures to minimize and prevent the development of these complications. MedPAC has suggested Medicare disclose the risk-adjusted readmission rates for all hospitals and determine benchmark readmission rates for certain conditions (e.g., heart failure, COPD exacerbations, and CABG). Hospitals would receive payment based on how close they come to these benchmarks.

Comment on this Article

Your email address will not be published. Required fields are marked *