Formal ROI Analysis Is Rare

Howard J. Anderson,
Executive Editor
Health Data Management Magazine, December 1, 2007

When El Centro (Calif.) Regional Medical Center contemplated investing in an emergency department information system, the hospital's chief nursing officer and other executives demanded a precise quantification of the potential return on investment.

"That was a first for us," says John Gaede, director of information services at the 165-bed hospital. "Now our board has demanded that we do the ROI exercise for all other systems we buy."

Indeed, it's extremely rare for hospitals and clinics to set ROI benchmarks before acquiring software and then actually measure the application's performance against those expectations, many consultants say.

"It has something to do with the culture of health care and hospitals," laments Bruce Eckert, senior manager at Beacon Partners, a Weymouth, Mass.-based consulting firm. "The not-for-profit culture is to avoid creating the business case and having the accountability that a for-profit business might."

Another consultant, however, says that situation is beginning to change.

"Hospitals know they have to invest in clinical information systems to improve the quality of care, but measuring the actual results has dragged behind the recognition that they have to put the system in," says Nancy Chapman, a partner at ACS Healthcare Solutions, a Dearborn, Mich.-based consulting firm. "Hospitals, however, are beginning to recognize that they have to demonstrate how a software investment will improve care."

With a clinical system ranging in price from $2 million for a small hospital to $15 million or more for a large facility, executives are beginning to demand quantifiable evidence of improvement in patient care, Chapman says.

"Some of our clients make a very good business case for a software investment, but then don't do as good a job checking whether the benefits were achieved," says Barbara Cox, senior principal and senior researcher at the Center for Health Information, the research unit of Noblis Inc., a Falls Church, Md.-based consulting firm. "I can't think of a client that's done the entire process very well."

Health care organizations attempting to create ROI benchmarks should form a team of clinicians, I.T. experts and others to tackle the challenging task, Cox advises.

"It takes time to put the right team together to figure out the goal of the project and what type of approaches are needed to create a baseline of measurable information," she says. "It takes a little bit of effort and some research to determine the benefits that can be achieved and determine what approach to take to measuring them."

Cox stresses that CIOs should not take the lead role in measuring ROI. "Justifying the value of an investment should be the primary responsibility of the executive sponsor in the part of the organization that will benefit from the investment," she suggests.

The committee establishing benchmarks first has to understand the goals of the organization before making a business case for the investment, Cox stresses. And the business case should change over time as the goals of the organization change, she adds.

Once software is implemented, the performance against the benchmarks should be measured regularly. "A lot of organizations don't have the energy after an implementation to go back and study whether they have indeed achieved the anticipated savings," she says.

In addition, many organizations also fail to provide an adequate budget for the post-implementation analysis, she notes.

Cox and other consultants acknowledge that setting ROI benchmarks for clinical systems is far more challenging than for financial systems. But many factors, including medication errors reduction and nurse productivity improvement, can be measured.

How Complex?

Setting up ROI expectations can range from hiring a management engineer to do "time and motion" studies of staff productivity to simply setting expectations for such process improvements as reducing chart pulls.

When he was a hospital CIO, Eckert of Beacon Partners hired an engineer to conduct a time and motion study of the medication administration process, identifying the resources required. Two months after implementing a new clinical system, the engineer did a follow-up study to demonstrate how the software made the hospital more efficient.

Hiring an engineer to conduct a sophisticated study can prove costly; Eckert's hospital spent $40,000. But if an organization is spending 20 times that figure on software and sees large potential for ROI, the investment is worthwhile, he contends.

For those not interested in hiring an engineer, Beacon Partners provides simpler ROI checklists that its clients can use to measure potential benefits. For example, hospitals can measure reductions in support cost for a new system vs. an old one; declines in power consumption; savings in space devoted to paper files; and cuts in labor costs associated with tracking down paper records.

Eckert advises provider organizations to focus on an easily measurable "global indicator," such as full-time equivalent employees per occupied bed or percentage of time nurses spend on administrative tasks.

"The amount of time that nurses spend treating patients, if they don't have a clinical information system, is typically 40%," says Chapman at ACS. "The rest of their time is spent on manual documentation. So a hospital could set a goal of using software to reduce the amount of time in documentation by 15 minutes per day per nurse."

Such an improvement could enable a hospital to free up nurses' time for treating more patients, enabling the organization to open up more beds and, ultimately, generate more revenue, she notes.

Chapman argues that the ideal way to measure ROI on clinical systems is to measure improvement in clinical outcomes, which includes reductions in medical errors.

"Hospitals don't implement clinical systems to save money," the consultant stresses. "They do it to improve outcomes and gain the ability to treat more patients more effectively with fewer medication errors."

Breaking New Ground

Like many hospitals, El Centro Regional Medical Center had never conducted a formal ROI study before acquiring software. But when it considered the purchase of an emergency department system, things changed.

The hospital's new chief nursing officer was skeptical of the value of the system and demanded proof of ROI, recalls Gaede, the director of information services.

So he crunched some numbers and came up with some easily measurable benchmarks that indicated the necessary hardware and software, which cost $650,000, would pay for itself within four years. The board approved the purchase of software from MEDHOST Inc., Addison, Texas, and was pleased when the payback came in only six months.

Major sources of payback included an increase in billings per patient as well as slashing dictation costs from $400,000 per year to virtually nothing, Gaede says.

The emergency department, on average, was using paper records to justify E&M codes averaging 2.8. Evaluation and Management codes are the modifiers to CPT codes that indicate the complexity of a case on a scale of 1 to 5. Today, these codes average 4.0 because the software more accurately captures all procedures.

Before installing the software, billing per patient averaged $284; by August 2007, that figure climbed to $341, not including price increases, Gaede says.

As a result, the hospital has seen a $2.4 million increase in billings per year, Gaede says.

The MEDHOST software uses the online documentation to generate billing codes. Coders then review the codes for accuracy and to fine-tune the methodology. Soon, the ED system will deliver charges directly to the hospital's billing system, where they'll be checked.

After the success with measuring ROI for the emergency department system, Gaede suggested to the board that a similar exercise be required for all future I.T. investments. He acknowledges that measuring ROI for other clinical systems may prove much more difficult. Next on the agenda is an electronic records system for ambulatory care settings.

Clinical Exercise

CentraState Healthcare System, Freehold, N.J, already is tackling the difficult task of measuring ROI for an enterprisewide clinical information system.

The integrated delivery system, which owns one 271-bed hospital, has implemented the Sorian clinical information system from Siemens Medical Solutions Inc., Malvern, Pa. Before installing the software, the system selection team created benchmarks for achieving goals in four areas: patient safety and quality; nursing and physician productivity; and operational efficiency, explains Indranil "Neal" Ganguly, vice president and CIO.

Among these goals are: Reducing avoidable medical errors; trimming nurse overtime as a result of after-hours documentation tasks; increasing physicians' use of online order sets; and reducing hospital-acquired pressure ulcers.

Since implementing many components of the system in January 2006, the hospital has seen a 42% drop in avoidable medication errors, the CIO says. Sorian's workflow engine helps reduce the number of errors by providing nurses with the appropriate dose of certain drugs based on lab values displayed in the clinical system, Ganguly explains.

To minimize the occurrence of pressure ulcers, the organization is using the workflow engine to flag nurses when an at-risk patient needs to be turned in their bed. If the nurse does not acknowledge that the task has been completed, the system sends reminders and ultimately alerts the nurse manager, the CIO explains. So far, that effort has resulted in a 10% decrease in pressure ulcers, Ganguly adds.

"Measuring of results is becoming institutionalized and ongoing as we develop each new workflow," he says.

Getting Some Help

Faced with the need to complete an ROI exercise, some organizations turn to consultants for help.

MeritCare Health System in Fargo, N.D., got outside help in setting expectations for implementation of an electronic health records system supplemented by a document management system. Hiring a consultant helped the project's backers win the support of senior executives for the project by adding an extra layer of credibility to forecasts, says Caryn Hewitt, R.N., executive partner for health information management.

The consultant's forecast predicted the investment in hardware and software would pay for itself in about four years.

In one year, as a result of implementing document management from Hyland Software Inc., Westlake, Ohio, for use at 12 clinics, the organization trimmed 18 full-time equivalents who had formerly been assigned to chase down paper documents, Hewitt says. The benchmark had been 22 FTEs reassigned within three years, so that ROI goal clearly is within reach, she adds.

MeritCare is saving money by either leaving these positions vacant as staffers depart or by reassigning the FTEs to other duties.

A wide variety of paper records that are not part of the EHR, such as signed patient consent forms, are now scanned using the Hyland software and stored in the records system, from GE Healthcare, Waukesha, Wis.

Measuring ROI doesn't have to be a complex process nor require hiring a consultant, some physician group practices are discovering.

For example, West Ashley Family Medicine, a five-physician group in Charleston, S.C., is focusing on measuring the impact of electronic records on its reimbursement and staff size, says John Gastright, M.D., who owns the practice.

Since implementing a records system from Columbia, S.C.-based HealthPort, formerly Companion Technologies, the practice has increased its typical bill for patient visits by $20 because of more precise coding of services provided, he says. The records system automatically suggests codes based on the online documentation. The physicians then have the option of accepting or rejecting those codes.

By using the records system, the group practice estimates it has avoided hiring about five staff members to shuffle paper, Gastright adds.

Beyond the financial benefits, the practice has achieved other efficiency gains, such as no longer losing paper charts and more easily writing prescriptions on computers, the physician adds.

Similarly, physicians at Mt. Kisco (N.Y.) Medical Group are seeing an increase in revenue thanks to practice management and electronic records software from Misys Healthcare Systems, Raleigh, N.C.

The practice estimated when it implemented the software that its 145 physicians would each be able to see one to two more patients a day because of less time spent documenting care, says Abe Levy, M.D., medical director and chief quality officer. That goal was quickly achieved, he says.

"The most important lesson we've learned is that it's important to think of improving physician efficiency as the goal rather than simply cutting costs or looking to save money by avoiding transcription," Levy says.

The practice's physicians continue to dictate notes, which are entered into the electronic records by transcribers, because they find this fits best in their workflow, he says. "A lot of groups struggling with electronic records adoption have tried to save on transcription costs, but that has made their doctors less efficient, and then the doctors rebel," he contends.

Sidebar

Experts advise health care organizations to carefully measure the return on investment in software. But for some doctors, the measure that really matters is the effect on their lifestyle.

"I'm not a stereotypical, overworked, frazzled physician," says Jeffrey Cooper, M.D. "My day would be much more difficult with paper charts."

Cooper, a solo practitioner in Duluth, Ga., who recently hired a partner, uses records software from JMJ Technologies Inc., Raleigh, N.C.

Simply put, the pediatrician invested in the software and related hardware to save time. "The doctor is the most expensive employee in the practice, so anything that improves my productivity is probably a good investment," he says.

Cooper uses the software to generate reports to measure the return on investment. For example, he now has an average turnaround time for patients of 32 minutes, including waiting time and exam. Before installing the software, turnaround took more than an hour. "The application makes us so efficient that I can see more patients," he says.

Before installing the software, Cooper saw about six patients per hour; now, he sees eight or more.

Cooper's very pleased with the increases in his revenue and profits. But equally important, he says, is that "I get to go home at a reasonable hour."

Thanks to computers, his office is quieter and less cluttered, he adds. "Being in an office with a stack of charts would just bump up my blood pressure."

For more information on related topics, visit the following channels:

• Revenue Cycle Management
• Hospitals