Laird Harrison

March 16, 2017

SAN DIEGO — In patients undergoing knee replacement, comorbidities do not necessarily mean higher costs, according to results from a new study.

This complicates efforts to account for risk factors that could be used to adjust the bundled payments that the Centers for Medicare and Medicaid Services (CMS) is beginning to use as a replacement for fee for service, said Noelle Larson, MD, from the Mayo Clinic in Rochester, Minnesota.

"We did find higher cost with Charlson comorbidities, but the big drivers of cost were individual payment plans, hospital contracts, and individual states," Dr Larson said here at the American Academy of Orthopaedic Surgeons 2017 Annual Meeting.

Because fee-for-service reimbursement could create an incentive for overtreatment, CMS has been asking medical centers to accept lump payments for "episodes of care," such as total knee replacement, as reported by Medscape Medical News. The payment covers all care provided from admission to 90 days after discharge.

The concern is that we start rationing care and not providing procedures to high-risk patients.

But orthopaedic surgeons have argued that payments should be higher for patients at higher risk for complications.

"The concern is that we start rationing care and not providing procedures to high-risk patients," session moderator David Fisher, MD, an orthopaedic surgeon from Indianapolis, told Medscape Medical News.

To see how such payments might be structured, Dr Larson and her colleagues used data from Optum Labs, a research company with data on 100 million patients covered by insurance, either private or Medicare Advantage plans.

The researchers looked at 38,889 patients younger than 65 years who were covered for the 12 months before and 90 days after total knee replacement. Mean hospital stay was 2.8 days.

The median total 90-day cost was $35,798.80. For the one-quarter of cases that were least expensive, the median cost was $27,637.60. For the one-quarter of cases that were most expensive, the median cost was $47,263.20.

Patients were categorized according to Charlson comorbidity index score, which is calculated on the basis of the number of diagnosed conditions, such as heart disease, AIDS, and cancer. The 90-day payments for knee replacement were calculated for both individual comorbidities and Charlson score.

There was a significant increase in cost for patients with obesity, atrial fibrillation, history of myocardial infarction, dementia, renal disease, HIV, and malignancy. For patients with private insurance, obesity added $1135 to the cost, heart failure added $2998, and diabetes added $434.

Although the cost was lower when the Charlson score was 0 than when it was 4+ ($35,300 vs $38,800), that $3500 difference is a relatively small proportion of overall costs, Dr Larson reported.

"We were surprised that there was not much variation with the Charlson comorbidities," she said, "so we started looking for other factors."

She and her colleagues found that costs were twice as high for patients covered by private insurance as for those covered by Medicare Advantage, regardless of the number of comorbidities.

Longer hospital stays, discharge to a skilled nursing facility, and complications all increased cost. But overall, posthospital care made up just 14% of the total cost among privately insured patients and just 18% among those covered by Medicare.

This means that cost-reduction strategies will have to focus on what happens in the hospital, said Dr Larson.

Out-of-pocket costs covered by the patient were also not a determining factor in overall cost.

The strongest driver of cost is, I think, individual hospital contracts.

What seemed to matter more was geography. Cases with similar complication rates cost more than $50,000 in some states and less than $30,000 in others.

"We were all shocked when CMS came out and there was no disease severity or risk stratification," Dr Larson explained. "We've been working with CMS to try to develop risk stratification. Unfortunately, we can't do any better, so we've been looking for strong drivers of high-cost patients. The strongest driver of cost is, I think, individual hospital contracts."

Given this finding, "it's hard to argue with CMS, saying you need risk stratification," she concluded.

Saving Money With Bundled Payments

Even without stratification, bundled payments can benefit medical centers, said Brian Curtin, MD, from OrthoCarolina in Charlotte, North Carolina. He presented a comparison of 4757 bundled-payment reimbursements and 8415 fee-for-service reimbursements received by OrthoCarolina.

Bundled payment "is finally becoming profitable," he told Medscape Medical News.

The bundled-care program saved $2700 per case, reduced skilled nursing facility admissions by 6%, and reduced home health usage by 6%. In addition, the readmission rate dropped from 12% to 10%.

Still, he said, some medical centers will begin to deny care to patients if CMS doesn't come up with a risk-stratification program.

Already, OrthoCarolina will not perform knee replacements on patients who smoke or who have a body mass index, albumin level, or hemoglobin A1c level that is cause for concern, Dr Curtin reported. Instead, physicians refer these patients to other providers who can help with issues such as weight loss or smoking cessation.

Despite what Dr Larson's team found, Dr Fisher said he believes the data warrant risk stratification in CMS bundled payments.

"We know there are comorbidity factors, and CMS has not taken them into account," he told Medscape Medical News.

Dr Larson, Dr Curtin, and Dr Fisher have disclosed no relevant financial relationships.

American Academy of Orthopaedic Surgeons (AAOS) 2017 Annual Meeting: Abstract 237. Presented March 15, 2017.


Comments on Medscape are moderated and should be professional in tone and on topic. You must declare any conflicts of interest related to your comments and responses. Please see our Commenting Guide for further information. We reserve the right to remove posts at our sole discretion.