Hospital Discharges Rise at Lucrative Times

How did data display help shed light on this problem? Explain.

Provide 3 areas in your companies where you might expect to see where data could be displayed that might better shed light on a problem / opportunity.

Wall Street Journal Feb 18, 2015 Hospital Discharges Rise at Lucrative Times Facilities Release Medicare Patients After Rules Trigger Higher Payments Christopher Weaver, Christopher.Weaver@wsj.com

http://www.wsj.com/articles/hospital-discharges-rise-at-lucrative-times-1424230201

Long-term-hospital companies time patients' discharges with an eye toward maximizing Medicare payments. WSJ's Christopher Weaver and Simon Constable discuss.

A Kindred Healthcare Inc. hospital in Houston discharged 79-year-old Ronald Beard to a nursing home after 23 days of treatment for complications of knee surgery.

The timing of his release didnt appear to correspond with any improvement in his condition, according to family members. But it did boost how much money the hospital got.

Kindred collected $35,887.79 from the federal Medicare agency for his stay, according to a billing document, the maximum amount it could earn for treating most patients with Mr. Beards condition.

If he had left just one day earlier, Kindred would have received only about $20,000 under Medicare rules. If he had stayed longer than the 23 days, the hospital likely wouldnt have received any additional Medicare money.

RELATED

Five Things to Know About Long-Term-Care Hospitals

Kindred declined to answer questions about Mr. Beards case, but said it seeks to provide quality care and comply with all laws and regulations.

Under Medicare rules, long-term acute-care hospitals like Kindreds typically receive smaller payments for what is considered a short stay, until a patient hits a threshold. After that threshold, payment jumps to a lump sum meant to cover the full course of long-term treatment.

That leaves a narrow window of maximum profitability in caring for patients at the nations about 435 long-term hospitals, which specialize in treating people with serious conditions who require prolonged care. General hospitals are paid under different rules.

A Wall Street Journal analysis found that many long-term-hospital companies discharge a disproportionate share of patients during that window when hospitals stand to make the most, a sign that financial incentives in the Medicare system may shape patient care.

Long-term-hospital executives sometimes pursued that goal for financial reasons rather than medical ones, say doctors, nurses and former long-term-hospital employees interviewed by the Journal.

The Journal analysis of claims Medicare paid from 2008 through 2013 found long-term hospitals discharged 25% of patients during the three days after crossing thresholds for higher, lump-sum payments. That is five times as many patients as were released the three days before the thresholds.

The Journal obtained records of all hospital claims paid by Medicare during those six years from the federal Centers for Medicare and Medicaid Services, for a fee. The data, which includes details of each individual hospitalization, provided the Journal with its broadest view yet of how American hospitals care for the tens of millions of seniors and disabled people in the Medicare system.

How Discharge Dates Affect Hospital Pay

The billing rules create a lucrative window for discharges... For most Medicare sepsis patients, for example, the larger lump-sum payments kicked in on day 20. Hospitals were more likely to discharge such patients during a three-day window beginning on that day.

Number of Medicare sepsis patients discharged, by length of stay, 2013

Notes: Analysis excludes patients who died during a hospitalization. Sepsis patients are those for whom hospitals received Medicare payment for Diagnosis-Related Group code 871, from Oct. 1, 2012, to Sept. 30, 2013. Source: Centers for Medicare and Medicaid Services. Source: Medicare claims data

Average, per-day Medicare payments to hospitals for treating sepsis patients, by length of stay, 2013

Notes: Analysis excludes patients who died during a hospitalization, and outlier payments for high-cost patients covered by a different system. Sepsis patients are those for whom hospitals received Medicare payment for Diagnosis-Related Group code 871, from Oct. 1, 2012, to Sept. 30, 2013. Source: Medicare claims data

Number of discharges by day for Medicare patients in all diagnosis categories, relative to threshold day, 2008 to 2013

Notes: Analysis excludes patients who died during a hospitalization. Includes long-term hospital claims that Medicare paid from Jan. 1, 2008, to Dec. 31, 2013. Source: Medicare claims data

Between mid-2011 and the end of 2013, the Kindred hospital that treated Mr. Beard discharged eight times as many Medicare patients on the day they reached their threshold as on the day before. In the days immediately after the lucrative three-day window, discharges plummeted. Kindred acquired the hospital, which has two campuses, in the summer of 2011.

Mr. Beard, a retired drilling-equipment salesman, was discharged from Kindreds facility on Nov. 12, 2011. His family says his condition had deteriorated at the hospital and they wish he had been released sooner.

In an email response to written questions, Sean R. Muldoon, chief medical officer of Kindreds hospital division, said the company strives for excellence of clinical outcomes and patient safety, and works to smooth care transitions and reduce patient costs.

The pattern of discharging patients at the most lucrative juncture is troubling and disturbing, says Tom Finucane, a doctor and professor at Johns Hopkins University School of Medicine, after learning of the Journals findings. The health-care system should serve the patients and try to improve their health, and any step away from that is a corruption.

Dr. Finucane and other medical experts say longer-than-necessary hospital stays increase risks for medical errors, infection and unnecessary care. Discharges that come too early can mean patients dont get care they need.

Medicare sets the lump-sum payment thresholds each year. They vary by patients diagnoses. The thresholds are set at five-sixth of the average length of stay for each diagnosis, based on earlier data. For patients being treated primarily for sepsis whom Medicare also deems to exhibit major complications, for instance, the threshold currently is 20 days; for the most common diagnosis for patients on ventilators, it is 27 days.

Payment thresholds

Former long-term-hospital executives say they sometimes called the threshold the normal low or five-sixth date, referring to the Medicare formula. The Journal interviewed 16 people who have worked at facilities operated by Kindred or rival for-profit system Select Medical Corp. in 10 states, including former hospital administrators, doctors and case managers who oversaw discharges. Those two publicly traded companies billed Medicare for 42% of all long-term-hospital claims it processed during the period the Journal studied.

MEDICARE UNMASKED

Part of a series examining payments in the roughly $600 billion Medicare system.

The former administrators say their corporate bosses exerted pressure to discharge as often as possible during the most lucrative days, rewarding managers who succeeded and questioning those who didnt.

Youd hear from the powers that be if your hospital was nothitting a pretty high percentage of your patients for Medicare soon after the payment threshold, says Karen Shammas, who was chief executive of a Kindred hospital in Peoria, Ariz., until late 2013, when she retired.

Ms. Shammas, like some other long-term-hospital administrators who were interviewed, described meetings in which hospital staffers would discuss plans for each patient at the facilityarmed with printouts from a computer tracking system that included, for each patient, the date at which reimbursement would shift to a higher, lump-sum payout.

Ms. Shammas says she never kept patients hospitalized for financial reasons if they were medically ready to leave.

Kindred declined to comment in detail on discharge patterns or corporate policies.

Former executives at both Kindred and Select say doctors, pressured by hospital administrators, sometimes ordered extra care or services intended in part to retain patients until they reached their thresholds, or discharged those who were costing the hospitals money regardless of whether their medical conditions had improved.

Former executives at hospitals run by each chain say their bonuses depended in part on maintaining a high share of patients discharged at or near the threshold dates to meet earnings goals.

In some cases, their bosses gave them specific targets for discharge rates during the most lucrative days, the former hospital executives say. When they missed their targets, some of the executives say, their bosses asked for explanations as to why individual patients werent released during the target window.

Select said in a written statement that its long-term hospitals discharge patients based on their medical condition and not on the Medicare reimbursement system and do not manipulate discharge timing based on financial considerations. The company said bonuses are based on overall financial performance, among other factors, and not the share of patients discharged near the threshold.

Selects corporate managers were very intense about managing that length of stay really effectively to maximize the profit potential for any particular patient, says Robert Marquardt, former CEO of a Select hospital in Fort Wayne, Ind.

If a patient was two days from the threshold, you were incentivized to see if you couldnt find a reason to keep them for two more days, says Mr. Marquardt, who left the company in December to work as a consultant.

Mr. Marquardt says he didnt believe the efforts caused harm. You might play the game a bit, but you would never put a patient at risk, he says.

Select said while it monitors discharge dates and other metrics and seeks to understand deviations from norms, it doesnt set discharge targets. It said efforts to prolong a patients stay or discharge a patient early would be a violation of Selects policies.

The Medicare data include nearly 860,000 claims the federal program paid for treatment at long-term-care hospitals during the six years through 2013. The data dont include claims for Medicare Advantage patients, who receive their benefits from private insurers. The Journals analysis of the data excluded patients who died during their stays because their deaths determined the durations of their stays.

Graphic: See Medicare Discharge Patterns For Long-Term-Care Hospitals In Your State

ENLARGE See discharge patterns for Medicare patients by searching our database of roughly 400 long-term acute-care hospitals across the U.S.

Three-day window

During the six years analyzed by the Journal, Kindred discharged 26% of patients during the three-day window and Select discharged 30% of its patients in the window.

Select said that if patients who died at hospitals were included in the calculation, its share of discharges during the three-day window would have been 26%. It said differences in rates between long-term hospitals are caused by factors such as variations in patients conditions.

Select also said that discharges should cluster in the period following the thresholds because those thresholds are based on averages, and because patients with similar diagnoses undergo similar treatment regimens. It said that government contractors audited 2,503 of its claims during the period the Journal studied, clearing more than 98% of them.

The potential cost to the taxpayer-funded Medicare program is significant. If all of the patients at long-term hospitals who were discharged during the three-day window were instead discharged one day before the threshold, the program would have saved nearly $2 billion over the six years ending in 2013, according to the Journals analysis.

The payment thresholds have come under scrutiny before. Last March, the congressional Medicare Payment Advisory Commission, criticized Medicares payments for long-term hospitals, saying data suggest discharge decisions [at long-term-care hospitals] are influenced at least as much by financial incentives as by clinical indicators.

Beginning later this year, Medicare is tightening the criteria for patients to be admitted to many long-term hospitals. It will pay long-term hospitals the higher lump-sum rates only after patients spend three or more days in intensive-care units at general hospitals, or when they are on ventilators.

EARLIER COVERAGE

Why Its So Hard to Fix Medicare Fraud (Dec. 25, 2014) Medicare Overbilling Probes Run Into Political Pressure (Dec. 11, 2014) Cancer-Care Giant Is Investigated Over Medicare Billing (Nov. 13, 2014) Doctors Cash In on Drug Tests for Seniors, and Medicare Pays the Bill (Nov. 10, 2014) A Fast-Growing Medical Lab Tests Anti-Kickback Law (Sept. 8, 2014) Walgreen Shakeup Followed Bad Projection (Aug. 19, 2014) How Agents Hunt for Fraud in Trove of Medicare Data (Aug. 14, 2014) Cancer Doctors Ring Up Big Medicare Bills for Tarnished Drug Procrit (June 19, 2014) Taxpayers Face Big Tab for Unusual Doctor Billings (June 9, 2014) Medicare Paid One Doctor More Than $20 Million in 2012 (April 9, 2014) Small Slice of Doctors Account for Big Chunk of Medicare Costs (April 9, 2014) Doctor-Pay Trove Shows Limits of Medicare Billing Data (April 9, 2014) Judge Ends 33-Year Injunction That Shielded Medicare Data on Doctors (May 31, 2013) Medicare Records Reveal Troubling Trail of Surgeries (March 29, 2011) Confidentiality Cloaks Medicare Abuse (Dec. 22, 2010) Top Spine Surgeons Reap Royalties, Medicare Bounty (Dec. 20, 2010) A Device to Kill Cancer, Lift Revenue (Dec. 7, 2010) Physician Panel Prescribes the Fees Paid by Medicare (Oct. 26, 2010)

Sean Cavanaugh, a deputy administrator of the Medicare agency, says: We work to ensure fair payment models are in place to pay entities for the level of care that is appropriate, medically necessary, and without intent on incentivizing care. He says officials anticipate the changes this year may address some of the issues.

For-profit companies such as Kindred and Select were more likely to discharge patients during the most-lucrative window than nonprofit competitors, the Journals analysis shows. Nonprofits discharged 16% of people during the window, compared with 27% at for-profits.

Some nonprofits, such as the New Britain, Conn., Hospital for Special Care, released more patients during the three days before they reached the thresholds than during the three days after.

Were never going to let the medical staff or the therapists know what the best financial day for discharge is, says John Votto, the hospitals CEO. I dont want to run a place where I have to worry that someone got discharged inappropriately.

In the case of Mr. Beard, the Houston patient, medical problems continued after his discharge. The Journal identified Mr. Beard through a Yelp.com review of the Kindred hospital posted by his granddaughter. A reporter examined billing records made available by Mr. Beards wife, Barbara.

Mr. Beard was admitted to Kindred Hospital Town and Country in Houston in late October 2011 after surgeons found the site of an earlier knee surgery had become infected with drug-resistant bacteria called MRSA. He was sent to the Kindred facility that Oct. 20 for a course of antibiotics, according to the records and Ms. Beard.

On his fourth day at the Kindred hospital, nurses administered the drug Remeron to treat sleeplessness. Mr. Beards wife says he had an allergy to that drugdocumented at the time on a wristband provided by another hospitaland he went into a coma for a time.

I wished then that I could take him somewhere else, says Ms. Beard, now 77 years old.

Over the next two weeks, Mr. Beards condition deteriorated as he received treatments from a dozen doctors. A Medicare document provided by his wife shows he received an hour and a half of critical care services on Nov. 9, three days before Kindred discharged him.

When he left the facility in a transport van on Nov. 12, bound for a nursing home, he complained of nausea, his wife says. The van driver called for an ambulance from a gas station. The ambulance took him to the emergency room at a general hospital in Katy, Texas.

Ms. Beard says doctors determined that, aside from low blood pressure, he was stable. She drove him to the nursing facility herself, but because Kindreds discharge papers had been left behind in the van, the nursing facility declined to accept him. He wound up back at the Katy hospital to begin an additional three-day hospitalization, where doctors performed tests to monitor an existing heart condition, the billing documents show.

Ms. Beard says she doesnt regret that her husband left Kindreds hospital when he did, despite the chaos of those days. I think if he had stayed at Kindred, he would have laid there and died, she says.

Mr. Beard had spent 23 days at the Kindred hospital. His diagnosis code, which the Journal confirmed by matching the payment amount against Medicares published rates for the Kindred hospital in late 2011, had a 23-day threshold for the full lump-sum payment, making it the most lucrative day to discharge Mr. Beard.

Medicare processed Kindreds full $35,887.79 payment that Dec. 19, the billing records show, less than two weeks before Mr. Beard died at home of heart arrhythmia.

HOW THE JOURNAL ANALYZED DISCHARGE DATA

The Wall Street Journal obtained all Medicare hospital claims from 2008 through 2013 from the federal Centers for Medicare and Medicaid Services, for a fee. The data included nearly 860,000 claims Medicare paid for long-term acute-care hospital stays.

The Journal analyzed patterns of discharges from those hospitals to determine how the duration of hospital stays corresponded to payment thresholds set by Medicare. The analysis excluded patients who died during their hospitalizationsabout 13% of the totalbecause their deaths determined the durations of their stays.

Under Medicare rules, payments to long-term hospitals jump from smaller payments for short stays to a larger lumpsum one when a patients stay reaches a certain duration. Those thresholds vary according to diagnosis. The Journal determined the threshold day for each claim based on its billing code and service date.

For each hospital, the Journal counted the number of patients discharged on the threshold day, and the number discharged on each day relative to the threshold day. If a patient was discharged the day before a threshold day, for example, it was counted as day -1. If a patient was discharged 10 days after the threshold, it was counted as day 10.

Lisa Schwartz contributed to this article.

Write to Christopher Weaver at christopher.weaver@wsj.com, Anna Wilde Mathews at anna.mathews@wsj.com and Tom McGinty at tom