how Cleveland Clinic used TDABC to improve value

A time-driven activity-based costing pilot project reveals significant opportunities to improve processes of care and reduce costs related to heart-valve surgery.

As hospitals, health systems, and other providers prepare for the move to value-based payment, they face an increasing need to be able to determine the true costs of delivering care and service. Accurate cost data inform the allocation of scarce resources, help leaders understand profitability by service line, and better support clinical teams in driving efficiency and improving outcomes.

At Cleveland Clinic, a patient-encounter view of financial performance is used extensively to complement the organization's traditional operating statement view. Instead of reporting revenues and expenses for a functional area, it is possible to report the financial activity for any given patient population. We use this information to inform business plans, evaluate service lines or market areas, and examine the cost of care with the goal of reducing costs while maintaining or improving quality and outcomes.

Recently, Cleveland Clinic partnered with Robert Kaplan, PhD, the Marvin Bower Professor of Leadership Development (emeritus), and Derek Haas, senior project leader, Harvard Business School, to conduct a pilot project to explore the differences between time-driven activity-based costing (TDABC) and the relative value unit (RVU) costing system that our organization has used for 25 years in multiple hospitals and physician practices. The goal was to determine whether TDABC could improve the accuracy of cost information, provide additional insight into cost-reduction and value-improvement opportunities, and help drive improvements in clinical practice for two types of heart-valve procedures. The result: significant enhancements in clinical and administrative processes for the procedures studied and reductions in expense related to direct administrative and support processes for these procedures, such processes represent 6 to 7 percent of the total cost of an episode of care for patients undergoing heart-valve surgery.

The TDABC Pilot Project

Undertaking a TDABC pilot project offered Cleveland Clinic a valuable opportunity to simultaneously evaluate current costing methods, experiment with the TDABC approach, and compare the methodologies.

Hospitals and health systems employ several different cost-accounting methods to determine the cost of care provided. At Cleveland Clinic, we use an RVU-based method that allocates expenses from our general ledger and payroll systems to activity codes-both technical (hospital) and professional (physician)-in our cost-accounting system. These activity codes are largely derived from the billable items captured in our chargemaster, but they also include non-billable items that we track solely for costing purposes.

In this "top-down" allocation methodology, all costs of providing services are grouped by cost types and categories, and RVUs are used to allocate those costs across the units of service. RVUs are determined for labor by interviewing caregivers and establishing the average time or cost necessary to perform each unit of service by labor type (e.g., physicians, nurses, technologists) and other cost types. Labor RVUs are expressed in minutes, while RVUs for supplies are expressed in dollars and generally represent the acquisition cost. Overhead or indirect costs, such as facility or administrative costs, are also allocated to those same units of service.

Because the costs are directly allocated to the units of service, they are easily segregated between professional and technical codes and by type of service or supply (e.g., room and bed, imaging exams, sutures, drugs). The units of service and the associated costs-both direct and indirect-for a billing episode are aggregated to determine the cost of a patient encounter.

Our goal for the TDABC pilot was to experiment with the new TDABC methodology, compare the results achieved with our current results with RVU costing, evaluate the relative strengths and weaknesses of each methodology, and determine how to improve or change our costing methodology to help drive improvements in clinical practice.

The project team focused on a high-volume population of patients undergoing one of two types of valve surgery: mitral valve repair and aortic valve replacement. This population comprised a relatively high volume of patients with data readily available for the analysis, and the costing group already had a good understanding of the patient group from prior studies undertaken with Cleveland Clinic'sHeart & Vascular Institute. Most important, the program's clinical leaders already had demonstrated openness to changing clinical practice to improve patient value based on valid cost information. We utilized a hybrid approach of TDABC and our current costing methods, both in the interest of conducting the pilot in a timely manner and in recognition of the capacity constraints of the caregivers working on the project.

We focused our efforts where we felt we could gain the greatest value and insight from TDABC: process mapping and calculating the labor capacity cost rates for our Heart & Vascular Institute, surgical operations, and nursing institute caregivers. For supplies, implants, pharmaceuticals, lab tests, imaging, and other miscellaneous utilization, we deviated from the standard TDABC approach and used the results from our current cost accounting data.

We took this approach for two reasons. First, we wanted to manage the amount of time required to complete the pilot by eliminating the need to process map and calculate the capacity cost rates for imaging and lab services. Second, we have detailed historical utilization data for supplies, implants, and pharmaceuticals already available and costed in our current system. In addition, after evaluating the TDABC recommendation for calculating these types of costs, we determined our existing data are more accurate than the information we would receive from process-mapping sessions.

It is important to note here that the information provided by the Heart & Vascular Institute

The project team developed process maps for the two heart valve replacement surgeries, including both the clinical activities that had been extensively analyzed and the administrative processes that had not been studied in great detail. The complete cycle of care required ao distinct process maps, from the point of consent to surgery through preadmission activity, actual surgery, the fiveto seven-day inpatient stay, discharge, and, finally, postoperative follow-up.

When creating the process maps, we divided the clinical care process into small, related components. For example, the valve process was divided into eight components, including patient check-in, surgery, intensive care unit stay, step-down stay, and discharge. This approach ensured we were able to have the right caregivers in the room for our mapping sessions and that the group would be small enough to allow everyone an opportunity to participate. The mapping included the cost-accounting manager and analyst, the pilot project manager, and members of our continuous improvement team as well as OR nurses, floor nurses, nursing coordinators, scheduling coordinators, or patient education nurses, depending on the process to be mapped.

In each process-mapping session, the goal was to identify the five or six major components of the work, identify the tasks necessary to complete each component, and collect the information necessary to apply TDABC costing methodologies to the process. For each task, we recorded what type of caregiver performed the task (e.g., nurse, technician, coordinator, staff surgeon) and how many minutes, on average, the task took to complete. Interaction with the clinical caregivers provided invaluable information about the clinical workflows involved and the resources necessary to complete these workflows. Activities that had been unrecognized in our traditional approach, such as patient registration and education, were identified and discretely costed-giving us the ability to pinpoint previously unidentified opportunities for improvement.

After determining the time required for the work being performed, we were able to calculate the costs following standard TDABC methodology. An accurate calculation of capacity cost rates is one of the key success factors in this approach. The capacity cost rates for hourly employees used data directly from the existing cost-accounting system payroll feeds, since these feeds already captured actual hours and labor expenses. For staff physicians, we used several sources to obtain the necessary data, such as Medicare time studies and data from our cost-accounting system regarding time dedicated to clinical practice. (Note: We did not include the cost of malpractice insurance in the capacity-cost-rate calculation for staff physicians because that expense is accounted for in our allocation of indirect expenses. Reconstructing our allocation of indirect expenses for our TDABC pilot would not have been a valuable use of our time; it was possible for us to evaluate the benefits of TDABC without doing so.)

What We Learned

The large quantity of individual process steps and number of personnel involved during the patient's care cycle surprised our clinical and operational personnel. The project team quickly identified several patient and activity flows within existing processes that could be improved and made more efficient.

For example, surgical scheduling and preoperative testing required 43 distinct process steps involving 12 different resource types:

* Registered nurse for scheduling

* In-house nurse practitioner

* Surgeon

* TCI Program nurse or nurse practitioner for outpatient follow-up(a)

* Cardiologists

* Consulting services

* Fellow

* OR charge registered nurse

* OR charge patient assistant (PA)

* J -1 floor registered nurse

* Perfusionist

* Anesthesiologist

The amount of redundant activity recorded illustrated increased potential for overprocessing, unnecessary wait times, and unnecessary motion and transportation, all of which would be characterized as waste in a Lean improvement exercise. The team selected five distinct areas for possible improvement in this process alone:

* Reducing the no caregiver minutes required for cardiologists, nurse practitioners, surgeons, and TCI nurses to see patients

* Reducing the 90 caregiver minutes required for registered nurses, nurse practitioners, surgeons, and TCI nurses to review clinical information

* Reducing the 45 caregiver minutes required for cardiologists, nurse practitioners, and TCI nurses to order tests

* Reducing the 40 caregiver minutes required for nurse practitioners, surgeons, and TCI nurses to review test results

* Requiring no more than 10 minutes each for OR charge nurses, OR charge PAs, perfusionists, or anesthesiologists to perform staffing, room, or equipment analysis-for no more than a total of 3o minutes

In addition, the clinical team explored whether all caregivers were working to the top of their license or whether reassignments should be made to enable each staff member's time and talents to be utilized most effectively.

Administrative processes, such as scheduling, typically are not well-understood or delineated in traditional RVU cost accounting, which focuses on costing reimbursable procedures and tests. The TDABC analysis revealed that direct administrative and support processes for mitral valve repair and aortic valve replacement patients represent 6 to 7 percent of the total cost of an episode of care-a significant cost-reduction opportunity that remained hidden before Cleveland Clinic undertook the TDABC pilot. Through process mapping and capacity-cost-rate calculations, leaders gained a detailed look into process steps that could be consolidated, reduced, or performed with a lower-cost mix of personnel.

We also found that the TDABC method produced a cost that was about 10 percent lower for both procedures than the cost calculated using the RVU method. Further investigation revealed four key reasons for the variance.

Overstating costs associated with secondary procedures. On average, each heart-valve procedure resulted in two billed professional procedure codes by the cardiac surgery team. The presence of the secondary CPT code caused professional surgery costs to be overstated in our existing cost-accounting system because this system did not appropriately differentiate the reduced effort needed to complete a secondary procedure. In contrast, this reduced effort was clearly recognized using TDABC and allowed for the appropriate costing of heart-valve cases with secondary procedures.

Calculating the cost of unused capacity. Our RVU costing methodology allocates all expenses-including the cost of unused capacity-to our units of service, meaning that each unit effectively absorbs the cost of that unused capacity. TDABC allows us to clearly identify the cost of the unused capacity and to apply only the actual costs consumed by the activity, providing a more accurate result.

Determining equipment costs. Equipment-cost calculations were lower under TDABC because TDABC accounts only for the time the equipment is used, whereas the RVU method allocates all equipment expenses to activity codes.

Calculating time related to specific processes. Under RVU costing, certain activities were allocated a greater amount of time than what was identified via the process-map interviews.

Our overarching goal in the pilot was to evaluate the TDABC methodology to determine whether we could use it to improve Cleveland Clinic's cost-accounting accuracy and drive clinical practice improvement. Overall, our assessment of the TDABC methodology was positive, and we identified several significant findings.

The development of process maps for clinical processes offers finance professionals the chance to build and/or strengthen the working relationships with their clinical colleagues. The process maps enabled us to capture several important pieces of information that we had never clearly understood before: activity directly related to the clinical procedure that is not billed for, and the total number of people participating in the process. However, the interviews needed to gather the information required for TDABC were resource-intensive, requiring significant time commitments from many people.

We also believe that reliance on the interview process to determine implant, supply, and pharmaceutical utilization is risky because accurate recollection of this type of utilization for such a complex patient population would be a challenge even for experienced caregivers. Moreover, we often have multiple surgeons (nine for the cardiac pilot) performing these procedures, each with their own preferences and utilization patterns. To capture the variation among surgeons, we would have to interview them and their surgical teams individually, which could be timeand resource-prohibitive.

Finally, although we determined that TDABC allows for more accurate costing of a single episode of care based on the time spent providing care, it intentionally does not allocate the cost of unused capacity or waste, nor does it account for the cost of related activity, such as research or education. TDABC therefore cannot be used to calculate all of an organization's costs.

The Value of TDABC

At Cleveland Clinic, TDABC produced relatively small-but-significant differences in cost calculations in comparison with RVU costing. Although we do not believe that these results warrant a wholesale replacement of our current system, we did identify several changes Cleveland Clinic could make to improve the accuracy of our cost data.

For example, we found opportunities to more accurately identify unbilled activity to better allocate costs. One possibility is the creation of zero-charge items that more closely reflect the work being done, which would allow us to discretely capture and allocate costs.

Another area of opportunity lies in costing surgeon activity in the process. We are investigating the collection of procedure code modifiers that would allow us to more accurately cost codes billed as secondary procedures.

Overall, TDABC enabled us to gain further insight into our clinical processes and will help inform important future strategic initiatives, including the move toward bundled pricing and other forms of value-based payment. *

View an exhibit that maps out 43 care processes required lor heart-valve surgery at Cleveland Clinic and highlights areas the organization targeted lor potential improvement at hlma.org/cleveclinicTDABC.

AT A GLANCE

* Cleveland Clinic partnered with Harvard Business School to conduct a pilot project to explore the differences between time-driven activity-based costing (TDABC) and relative value unit costing.

* The goal was to determine whether TDABC could improve the accuracy of cost information and identify value-improvement opportunities for two types of heart-valve procedures.

* Using TDABC, leaders gained a detailed look into process steps that could be consolidated, reduced, or performed with a lower cost mix of personnel.

a. TCI stands lor 'To Come In' and refers to a program in which patients can arrive at Cleveland Clinic on the same day as their scheduled surgery.

Christopher J. Donovan, MBA, is executive director, fiscal services, Cleveland Clinic, Cleveland, and a member of HFMA's Northeast Ohio Chapter ([email protected]).

Mike Hopkins, MBA, is manager, decision support services, Cleveland Clinic, Cleveland ([email protected]).

Benjamin M. Kimmel is assistant director, fiscal services, Cleveland Clinic, Cleveland ([email protected]).

Stephanie Koberna, MBA, is financial analyst II, decision support services, Cleveland Clinic, Cleveland ([email protected]).

Carrie A. Montie is project manager, decision support services, Cleveland Clinic, Cleveland ([email protected]).