Dependent older patients may cost hospitals more

Hospital costs were 23 percent higher for older patients who required help with daily activities such as bathing or eating than for older patients who could perform these activities for themselves, according to a study of patients at an Ohio hospital that was conducted by researchers from the San Francisco VA Medical Center (SFVAMC.) This is the first study to examine the differences in hospital costs for dependent and independent older patients, the researchers report.

“Functional status is extremely important for geriatric patients. The population is aging and a lot of people have functional disabilities, yet no one had done a study taking into account how functional impairments in older patients affect hospital costs,” says the study’s lead author, Kenneth Chuang, MD, a Veterans Affairs National Quality Scholar at SFVAMC. “While most people understand that Medicare patients who are sicker or less functionally independent are going to end up costing more, they probably don’t realize that hospitals usually don’t get paid more for their care.” The study appears in the December Journal of the American Geriatrics Society.

Researchers need to conduct such studies at other hospitals, Chuang says. If the findings are similar, he says, it would mean that under the current Medicare payment system there is an economic incentive for hospitals to discourage admission of dependent older patients, or to quickly discharge them. “If these findings hold true at other institutions,” Chuang says, “it will be important for Medicare to address these financial inequities.”

Chuang and his team based their analysis on a study conducted at the University Hospitals of Cleveland between 1993 and 1997. In this earlier study, investigators evaluated whether interventions such as physical therapy could decrease the length of stay and improve outcomes for older inpatients. They also tracked costs for each patient using the hospital’s cost management information system, a proprietary cost accounting system that assigns direct and indirect costs to individual patients based on their use of hospital resources.

Chuang’s team examined records for 1,612 patients aged 70 or older and compared each patient’s assigned costs with an evaluation of his or her functional status upon admission. Evaluations consisted of asking patients whether they could perform certain routine activities without requiring assistance two weeks before admission to the hospital and at the time of admission to the hospital. Patients were rated as dependent if they needed help with one or more of the following tasks: bathing, dressing, eating, toileting, or transferring from a bed to a chair.

The analysis showed that the patients’ mean age was 81, two out of three were women, and 12 percent were admitted to the hospital from a nursing home. Sixty-four percent were dependent in one or more of the five tasks on admission to the hospital. On average, dependent patients were three years older than independent patients. Nearly all patients admitted from nursing homes were dependent.

Chuang and his team determined that, overall, patients who were dependent in one or more daily activities incurred costs of $5,300 during their hospitalizations, while independent patients incurred $4,060 in costs: a difference of 31 percent. When the team adjusted these figures to take into account the difference in profiles of dependent and independent patients, they still found that dependent patients cost the hospital 23 percent more than independent patients.

The team found that most of the differences in cost could be attributed to the fact that dependent patients had longer average stays than independent patients: 7.7 days versus 5.7 days. Yet even after accounting for length of stay, dependent patients cost the hospital 6 percent more than independent patients.

“There were big differences here,” Chuang says. “And the costs were not made up by Medicare reimbursement. If these findings hold true in other hospitals, they suggest that there are incentives for hospitals to discriminate against patients with lower functional status.”

Medicare designs its reimbursement system to pay providers fairly and to ensure that no groups of patients are financial winners or losers for hospitals. Payments to hospitals are based on a patient’s diagnosis at the time of discharge, largely independent of the length of stay. Thus, if two patients have the same diagnosis—pneumonia, for example—a hospital will receive the same payment for a 95-year-old Medicare patient who is an inpatient for eight days, as for a 72-year-old who is an inpatient for four days. This system of payment was instituted in 1983 in an effort to prevent needless delays in discharging patients. Under the old Medicare formula, payment was based on a patient’s length of stay, a system that provided an easily abused financial incentive for hospitals to retain patients as long as possible, Chuang says.

When it was first put into effect, the new payment system raised the concern that hospitals might be systematically underpaid for frail older patients; that is, older patients who generally lack strength and are particularly susceptible to disease and infirmities. But a study published in 1987 found that there was little difference in the amounts hospitals charged for patients under the Medicare system, regardless of factors such as a patient’s age, income or history of previous hospitalizations, or whether the patient had been living in a nursing home at the time of admission.

Chuang’s study differs from the 1987 study and most other investigations of the Medicare payment systems in two ways. First, the research team used clinically relevant information gathered at the time of admission for each patient. Most other studies, including the 1987 study, have been based on administrative data such as age, permanent disability and income. Second, Chuang based his work on actual hospital costs, rather than on the amounts that hospitals charge for patients. “Most previous studies have looked at charges because it’s hard to measure costs accurately,” Chuang says. “Since costs are usually assigned to departments rather than individual patients, in order to do a study like this, we had to have a specific system that could attribute costs to patients.”

Other studies that have evaluated Medicare reimbursement systems have examined factors such as a patient’s age, sex and income. In these studies, costs have been found to vary by about 3 to 10 percent between groups. “No other study has found cost differences as high as those we found between dependent and independent patients,” Chuang says. “While cost differences below about 10 percent could end up balancing out between groups, that’s not the case with differences as high as those we found. I was really surprised by the magnitude of our results.”

Other investigators in the study include senior author Seth Landefeld, MD, associate chief of staff for geriatrics at SFVAMC, professor and chief of geriatrics at University of California, San Francisco, and Senior Scholar in the Department of Veterans Affairs National Quality Scholars Program; Kenneth Covinsky, MD, MPH, staff physician, Division of Geriatrics, SFVAMC and associate professor of medicine, UCSF; Laura Sands, PhD, School of Nursing and Center for Aging and the Life Course, Purdue University, West Lafayette, IN; Richard Fortinsky, PhD, Center on Aging, University of Connecticut Health Center, Farmington, CT; and Robert M. Palmer, MD, MPH, Section of Geriatric Medicine, Cleveland Clinic Foundation, Cleveland, OH.

The study was supported by grants from the National Institute on Aging and the John Hartford Foundation. Kenneth Chuang was supported by the VA National Quality Scholars Fellowship program. Dr. Covinsky was supported in part by an independent investigator award from the Agency for Healthcare Research and Quality and a Paul Beeson Faculty Scholarship in Aging Research from the American Federation for Aging Research.