Measuring up: what are outcomes and why are they important?

So why is the word "outcomes" creating a buzz in rehabilitation therapy these days? After all, everyone knows what outcomes are. Following her hip replacement, Mrs. Winterbottom goes into therapy, and after treatment she is able to walk. Great outcome. The nursing facility did its job, and Mrs. Winterbottom tells everyone what wonderful therapy she received. The story is warm and fuzzy and makes everyone feel good.

But real therapy outcomes are more than anecdotal stories and testimonials. Real therapy outcomes mean the benefit of therapy is actually measured on a national scale using hard data and benchmarked statistics over hundreds of thousands of patient records. Real outcomes are about reports customized to each patient and facility that relate measured functional gains to length of stay and the efficiency of therapy services from a variety of perspectives.

In other words, real therapy outcomes could reveal that Mrs. Winterbottom was rated at 0.5 for gait when she began therapy. After 21 days of treatment, she was rated at 3.0, which placed her above the national average for similar hip replacement patients. Further reports might reveal that the entire facility has a higher intensity of treatment (reflected in a higher percentage of the upper RUG categories) with shorter lengths of stay than the national average and that their patients are discharged to higher levels of independence and are much more likely to return home.

With this outcome information, the facility now can prove it did its job and justify the level of therapy provided. Mrs. Winterbottom is right on when she tells everyone what wonderful therapy she received.

"We are in an age of increasing responsibility to justify what it is we are doing," says Audrey L. Holland, PhD, Regents Professor Emeritus at the University of Arizona. "There is a need to gather evidence to suggest that treatment is beneficial and economically feasible. Outcomes measurement is beginning to make a difference."

Today's healthcare is all about gatekeepers--people who control access to the patient and/or to reimbursement. "Whether to the physician, the family, the federal government, or even the nursing staff, we must be able to prove our impact in some measurable way," says Bill Goulding, director of outcomes and appeals management for Aegis Therapies. "When we simply say that someone has increased his independence and ability to dress his upper body, that's meaningful to the patient, but it's not measurable. Gatekeepers need some justification to proceed with the plan of therapy."

How Do You Measure?

Outcomes objectively document the level of functional abilities assigned at the time of admission and the level of those abilities upon discharge. But how do you measure that level? "From a theoretical standpoint, any measure that is valid and reliable works," says Reg Warren, PhD, who, as principal at SeniorMetrix in Boston, has been involved in studies on the cost-to-outcomes relationship on a national level. "The point is to use that measure in a large enough scale so that compared data can be created between providers. If each therapy develops its own measure, the problem is that you can't go outside the use of that measure to compare your patients' outcomes with national or regional standards."

[ILLUSTRATION OMITTED]

For example, suppose Sunny Mountain Nursing Home has developed its own measurement scale. When our Mrs. Winterbottom tries to evaluate care at Sunny Mountain compared with care at Shady Forest Home in order to decide which facility to enter, the numbers mean nothing to her, especially if Shady Forest, too, has its own measurement scale.

Similarly, referral sources looking at the two facilities would have no idea how their patients would fare either by diagnosis, length of stay, or functional level at discharge. "One of the biggest problems with outcomes is that you must have a measure with interrater reliability," says Goulding. "If someone in Dubuque scores a patient at 2.5 level, someone in Boston should be able to evaluate the patient on the same scale and arrive at exactly the same score."

For that to happen, a scale must have reliability and validity. Several scales are currently in use, each with its own pros and cons. The Functional Independence Measure (FIM) is probably the most widely used scale in rehabilitation, certainly in acute rehab settings.

But according to Professor Holland, the FIM doesn't quite do the job for long-term care. "The Rehabilitation Outcomes Measure (ROM), marketed by Accu-Med Services (www.accu-med.com), is a better scale for nursing homes than the FIM," she says. "It is more refined, covers more areas, and asks more questions. It's much more specific in dealing with questions of language cognition, for example. And it will become an even better scale as it is more broadly used."

One advantage of the ROM, according to Mark Besch, vice-president of clinical services for Aegis, is that it is specific to skilled nursing facilities. "When balancing dollars and treatment worth, it's important to look at data from the same setting and from the same regulatory environment," he says. "The FIM scale is the largest overall, but it is not all post-PPS data and not all skilled nursing facility data."

Another advantage of the ROM is the ability to measure the abilities the patient will need after therapy discharge. For example, if patients are going home, it's important to measure their ability to do light housekeeping, prepare meals, and manage finances. If patients are remaining in the nursing home, the ability to bathe, dress, transfer, etc., is uppermost.

Mary Spooner, director of operations for Beverly Healthcare for Ohio, is sold on the ROM. "It's industry-specific, it captures the bulk of the diagnoses we are dealing with, and it drives quality," she says. All Beverly facilities use the ROM.

Aegis has built a data bank of more than one million outcomes records for its patients in skilled facilities that can compare ROM scores and data across many variables, including patient age, acuity (days post-onset), initial disability, ICD codes, general diagnostic group, length of stay, and discharge setting. "To my knowledge, it is the largest post-PPS outcomes database in the country," says Goulding.

Training and Certification

Of course, a scale means nothing unless those who administer it are thoroughly trained in its use. "There has to be some certification process," explains Besch, "so that people can prove that they clearly understand the scale and that everyone uses it in the same way."

For example, he notes that Aegis therapists go through three hours of intensive training in which they observe and rate how patients perform various functional tasks. After completing training, they take an examination to determine how well they understand the rating scale.

"The test is the same for everyone who uses ROM around the country," says Besch. "It is discipline-specific, so PT, OT, and Speech must each achieve a certain score for each discipline. It's all about making sure that when the therapist is presented with a clinical situation, he or she can understand how to translate that into the measure."

The Value of Outcomes to the Industry

Outcomes reports will play a vital role in upcoming changes in government regulation of healthcare. "Medicare Part B therapy caps are primarily aimed at guarding against overutilization, so outcomes data would be very beneficial for regulators," says Goulding. "Therapy caps also force efficiency, and that's what outcomes are all about. I'm hoping that by providing outcomes information, we can be proactive and give the government data that I'm not sure it has--that is, to tell them what it costs to provide nursing and therapy care for patients by diagnostic group."

In fact, Sen. Blanche Lincoln from Arkansas recently met with leaders of Aegis Therapies to ask for information to help with making decisions on therapy caps. "She wanted more than anecdotal testimonials on the effectiveness of therapy," reports Goulding. "She wanted outcomes information to back them up."

Outcomes data will also allow a nursing facility to predict the impact of reimbursement regulation on its bottom line by providing data on length of treatment, functional gains, cost of therapy, and cost to the home in terms of nursing labor costs. For example, the need to justify rehabilitation is more intense for Medicare Part B patients than for those Part A patients who have just been released from the hospital. "The government wants suggestions on payment structure, and one of the alternatives is for Medicare to pay X number of dollars for rehab for particular diagnoses," explains Goulding. "I don't know how a therapy company that doesn't have outcomes data will know if it can survive with that."

The Cost/Value Balance: The Future of Outcomes Data

The "Holy Grail" of outcomes data, according to Goulding, is to be able to prove that therapy is a good investment, that it reduces the burden of care and thus the financial burden on the healthcare system. He likens outcomes information to those old-time balance scales with a basket on each side of a fulcrum. "On one side is the cost, which is extremely measurable: for example, $3,000 worth of physical therapy," he says. "On the other side is the benefit or outcome. How much outcome balances the dollars? That's been hard to measure."

How much therapy a patient should receive, how long he/she should be in a particular post-acute setting, and what kind of functional change should be expected are topics that Dr. Warren explores for his healthcare-plan clients at SeniorMetrix. "Standards are now beginning to develop," he says. "It's important to have a uniform standard of functional change across therapies and across settings, and associate reimbursement with that measure. Quality and outcomes are disconnected in the current RUGs payment system. It is the linking of those two that will be important strategically for nursing homes in the future."

A step toward that goal is the comparison of therapy outcomes with MDS data. The goal is to see if nursing documents less labor, and therefore less labor cost, in caring for a patient because of therapy the patient received. For example, gains made in occupational therapy could be matched with nursing to see if the patient's ADL skills are getting better. Goulding notes that Aegis is beginning to compare its therapy outcomes with those of the Beverly Enterprises database. "There are many uses for outcomes," he says. "But the one that will really make a difference is the one that can prove that improvement in therapy equals reduction in cost or what is sometimes called 'burden of care.'"

What else is ahead for outcomes data? "I envision that outcomes reporting will eventually be part of the Medicare system," predicts Dr. Warren. "Medicare-risk lives are projected to triple in the next five years, and that means 12 million Medicare members will be under managed care plans. Most of them will require some form of outcomes associated with their payment."

And that means that the buzz about outcomes may become a very big noise indeed.

RELATED ARTICLE: How the ROM Scale Works

Using the Rehabilitation Outcomes Measure (ROM) scale, therapists score a patient's level of function in several areas at admission and at discharge, using a scale from 0.0 to 3.0 in half-point increments. A 0.0 means the patient is totally dependent upon staff, an extreme burden of care for the healthcare system. A 3.0 means the patient is totally independent and is a minimal burden of care. "The description of each level is very detailed, creating consistencies across the scale whether you are talking about bathing, stair climbing, or toilet hygiene," says Bill Goulding, director of outcomes and appeals management for Aegis Therapies. "That's why the rating can be standardized."