The trusted source for
healthcare information and
Complexity criteria helps center with budgeting
Goals are to improve workflow, efficiency
Many clinical trials offices struggle with determining an accurate estimate of costs for research services, sometimes falling short of comfortable margin. In an effort to improve this process, a Detroit clinical trials office has developed complexity criteria and productivity standards.
"We pick studies because we think they’re good for our patients, but we can’t afford to go in the hole," says Carolyn Schmidt, RN, MSHA, OCN, administrative director of Josephine Ford Cancer Center in the clinical trials office at Henry Ford Hospital in Detroit.
Several years ago, Schmidt and Tiffany Pearce, BS, RHIA, financial grant manager specialist at the cancer center, decided it was necessary to develop standards that would help them get a better handle on study costs.
"We decided to develop a complexity criteria to assign protocols either high, moderate, or low complexity, depending on what the work involved," Pearce says.
"We held brainstorming sessions within the office and with employees, but it mostly was trial and error and based on Carol’s experience," Pearce says.
High, medium, or low
Three components went into the complexity criteria, including administrative, data management, and case management, Schmidt notes.
"Our challenge was we needed to start billing departments for services offered through our centralized office," she says. "Either we could have people keep minute-by-minute tabs of their time, which is not a fun thing to do, or we could develop criteria from each of the three services."
So the same three ratings of high, moderate, and low were applied to each service, Schmidt says.
Pearce and Schmidt met with staff and asked people what makes a protocol highly complex for them.
For example, the administrative staff might say that pharmaceutical trials are more complex because they typically require that the sponsor review and approve the consent document; whereas, it’s acceptable to use the institution’s consent language in most cooperative group trials, Pearce explains.
Staff also noted that nursing protocols requiring multiple registrations, such as a leukemia trial, tend to be higher complexity than a protocol where there are not multiple registrations, she says.
Study coordinators find that phase I studies are more complex because they require more one-on-one monitoring than phase III studies, Pearce adds.
For data management and case management staff, a trial in which there are eight or more cycles of chemotherapy treatment have a high complexity, while a prevention study in which patients receive hormones and no chemotherapy is a low-complexity study, she says.
Developing the complexity criteria was only the first step since this is an evolving process, Schmidt notes.
The institution’s plans are to centralize clinical trials services so the office will handle more than cancer clinical trials, and this will require an entirely new look at complexity, she says.
"As we look at the specificity of each of those new protocols, we’ll ask, What is it about those protocols that we could label low, moderate, or high complexity?’" Schmidt says.
It’s possible the final complexity criteria will be specific to the type of study conducted, with oncology trials having their own criteria and infectious disease trials having a separate set of criteria, she notes.
"Also, we looked at time factors," Schmidt says.
So if a study uses drugs and will require pharmacology support or if radiology oncology department consultation will be necessary, then more time will be required and the cost is increased, she adds.
The clinical trials office is set up as its own business that charges the hematology/oncology department for the cost of clinical trials, and this cost now is based partly on the complexity factor, Pearce notes.
"It helps us to negotiate budgets with the external sponsor, and it gives us an idea of how much money is coming into our office to pay salaries," she says.
"Initially, things worked very well, but we’re finding that as time goes on the predictability of accrual for clinical trials is difficult," Pearce adds. "We don’t feel as though we’re right on target yet, so we’re adjusting as we go — it’s a very long process."
Schmidt and Pearce also developed productivity standards as a second major step in the process.
"We looked at all tasks performed in each of three areas," Pearce says. "For example, with data management, we have protocol registration, long-term follow-up, and sometimes materials handling."
The goal was to see how many low, moderate, and high complexity activities a person could do in one year if this was all he or she had to do, Schmidt says.
For each of those three areas, they assigned tasks performed and spoke with personnel to develop an idea of how long it takes to do each of those tasks, Pearce explains.
"Sometimes the answers were varied," she says. "I’d talk to five data people and ask how long it takes to do a certain task and I’d get different answers."
Another common finding was that staff tended to think it would take longer to do a particular task than what management thought it would take, Pearce adds.
"We came up with exact numbers, averaged out the numbers, and added time, and — to our knowledge — came up with the maximum number of hours a each task would take," she says. "But the numbers are very liberal because some people work more slowly than others."
Also, some areas require very intense attention, such as care for cancer patients, Schmidt notes.
"A cancer patient may call the case manager for a zillion things that may be unrelated to the protocol because the case manager in oncology often is their primary contact person," she explains. "So it does take a lot of time to work with these patients; and from a purely practical perspective, it’s hard to judge how many interruptions or phone calls or emergency room admissions you’ll have to deal with — these are very sick people."
But it was only in the nursing area that productivity numbers needed significant adjustment, Schmidt says.
Once task hours were determined, Pearce and Schmidt developed a set of standards based on the major task that each employee performs.
For instance, for the administrative staff, the longest task is the IRB application; and with data management, the longest task is completing the case report forms for new patients, Pearce notes.
"With nurses their biggest task is managing patients on a protocol," she says. "Then we set a standard of if all a person did, for example, was new IRB applications, how many could they do in a year?"
That number was divided by 2,080 hours per year. So if the IRB application required 40 hours, the number 40 was divided into 2,080 to come up with the answer that a person who did nothing but these applications could do 52 of these in a year provided the person was working full time and not busy with other tasks, Pearce explains.
"We used the biggest task as our standard for a year, so the employee would have to hit 52 productivity points in a year and receive productivity points prorated on time spent on other tasks, as well," she says. "We looked at the time it takes to do an IRB renewal, maybe three hours, and that would equal 693 per year, so the weight of that task would be adjusted to go into the 52 standard."
The productivity standards are integrated into the office’s software so every time a report or task is completed, it’s automatically put into the report, Pearce adds.
"If the office seems overwhelmed with work, I’ll monitor it weekly to see if I can shift work around," she says. "It might be that someone is way above their productivity level and someone else is below."
This process gives administrators a clear idea of when it’s necessary to hire additional staff or when the office is overstaffed and people need to be shifted to other assignments, Pearce notes.
"It can be used to talk with people about their productivity," she says.
For example, the productivity numbers could be added to employee report cards and referred to in annual evaluations, Schmidt says.
The chief benefit is that it assists in predicting trial cost and department overhead, Schmidt and Pearce say.
"Because we have identified how long it takes to do each task, we’ve identified an average salary for personnel in each of those groups, and we’ve figured out how much it costs to run a trial based on that," Pearce says. "We’re still tweaking it because we’re still not right on as far as productivity should be."