Don’t be left behind – Data-driven cost accounting for health care


Improving cost accounting within hospitals is no longer a nice to have, but a critical component of meeting the increasing pressures on cost containment.

While many industries are using data-driven methodologies to conduct cost accounting, health care has been lagging far behind. Whether it’s due to high costs, change management challenges, low priority or other reasons, hospitals and health systems have put off revising the decades-old methodologies they use. Typical hospital cost management systems still use a combination of relative value units (RVU) and ratio of cost to charges (RCC) to allocate costs. While RVUs are an improved methodology over RCCs, they are still too subjective and generally not updated when processes are changed.

While this method sufficed in the past, in today’s environment — with growing amounts of data and clinical electronic medical record (EMR) systems — it’s simply not enough. Old methods and cost-accounting systems do not use the richness and depth of data available to provide an accurate view of costs. To continue on this outdated path is a prescription for failure.

With the passing of the Affordable Care Act (ACA) and other regulations, the move toward electronic health records, tracking of quality measures, and data collection has come to the forefront and will only increase. This shift provides a wealth of information that previously did not exist. Combining ACA with a changing reimbursement structure, hospitals and health systems need to adapt in order to compete.

The need for change
Improving cost accounting within hospitals is no longer a nice to have, but a critical component of meeting the increasing pressures on cost containment, changes in reimbursement practices and rates, and increasing awareness of patients on the cost of services across various providers. To keep up with the changes, we believe that cost accounting should be based on several underlying principles.

Keep direct costs direct: Averaging or allocating costs should only be used when a cost cannot be isolated as a direct cost. While many hospital models treat supplies and pharmacy as direct costs to a patient, there are many other costs that can be traced to specific departments, procedures and patients. The definition of direct costs should be expanded. For example, robotics equipment used for certain procedures should be allocated to only those procedures involving that equipment, not to all procedures performed in the operating room where the equipment is housed — or worse, excluded from costing entirely. Oral surgery could be performed in the same operating room as open heart surgery, but both procedures would not warrant use of the same robotics equipment.  

Understand cause and effect: Too little detail in a costing model will end up spreading costs to inappropriate targets, thereby under-costing the activity, procedure or other cost target. In the value-based health care market, understanding what causes costs to be incurred, the reaction of costs to changes in volume and understanding the appropriate cost targets will provide better data for analytics. Then you will have information to validate for hospital leaders what services and patients are profitable and where to expand and invest more capital.

Example of cause and effect: Knee replacement
We compare three costing methods to show which one will return the most value (see table). In this first example, the traditional costing method offers less cost differentiation between cases of the same type because it is based on charge-based formulas for supplies, drugs and implant costs. By not using actual cost drivers, this method doesn’t give us the complete and accurate picture on cost.

progressive costing method does a better job for cases where supplies, drugs and implants are a large part of case costs because it links actual costs to what was specifically used in that case. Although more accurate than traditional costing, the progressive costing method lacks labor detail, preventing you from accurately comparing cost by physician.

Neither of these two methods takes into consideration the intensity of the clinical resources expended or variation in physician practice patterns. For example, Dr. Bob takes five additional steps during heart bypass surgery that causes his surgical assistants and preoperative nurses to take 25 minutes longer in the Operating Room (OR) than Dr. Bill’s team and there is no difference in the quality of outcomes between Dr. Bob and Dr. Bill. Without that level of detail, we cannot cost compare the OR practice patterns of Dr. Bob to Dr. Bill. Another potential pitfall to the traditional or progressive method is lack of buy-in from physicians and clinical team members. Both may be quick to disregard or ignore a costing analysis if those methods lack labor and supply usage comparison data. The numbers won’t be of value to them.

The best approach is to use the advanced costing method. This method relates labor to actual minutes worked by level/type of staff instead of an allocation based on RVUs applied to charges. Operating room minutes are readily available from several systems. By applying the advanced costing method, you can directly assign patient treatment costs as a means to analyze procedures and service lines, identify physician practice variation, standardize clinical practice patterns, and support flexible budgeting and model payor contracts. By using more robust analytics, you can support decision-making in a rapidly changing environment. 

Knee replacement example

Use  the right cost driver: Expanding the number of drivers beyond RVUs for a model can enable more accurate costing. Using a driver that does not have a relationship with how the targets consume a cost pool could easily create inaccurate and potentially misleading results. Even when RVU collection is based on data other than a time or usage estimate, using appropriate raw data removes the need for collection and provides continual updates. See table below.

Right cost driver
Cost-accounting systems fall short if they can’t provide information to enable meaningful analytics on the services provided to a patient and their fixed and variable costs. With the wealth of data that is available today, a cost-accounting platform should be able to connect into clinical data sets such as inpatient/outpatient services provided, physician/nursing time, medications administered, supplies and equipment used, patient room, and ancillary department association (labs, imaging).

It should also link into:

  • Operational systems and data sets: scheduling, patient safety, satisfaction and quality
  • Financial systems and data sets: patient accounting, general ledger, payroll system, time and attendance system, and fixed asset system

It’s a long-term plan

Applying these outlined principles with the right information system gives hospitals valuable and actionable data to get them on the road to long-term financial stability. They will be better able to understand the cost of the full continuum of care and better follow reimbursement mechanisms such as value-based purchasing, bundled payments and shared savings.

Case study
Our client, a $10 billion integrated global health enterprise, wanted to implement a new costing methodology, where service costs were based on patient consumption of services. The existing methodology did not allow for detailed performance analysis by hospital department and service, or support comparisons across facilities, services, departments or physicians. Additionally, they wanted to create a standardized cost methodology across all hospitals, replace disparate costing systems, expand drivers used for costing and create data for other payor provider initiatives.

The Grant Thornton LLP team assisted with the implementation of Hyperion Profitability and Cost Management (HPCM). HPCM serves as an integrated system for profitability management, reporting and analytics. It provides business user-driven profitability application and modeling, a flexible allocations platform for any method (ABC, standard costing, etc.), support of both cost and revenue management, and a single point of maintenance for reduced administration cost. The team worked with the finance leadership in developing a pilot to develop the HPCM models and create an initial set of reports.

As a result of the new costing methodology, the enterprise has gained improved assignment of costs based on the actual drivers of expenses; a better understanding of the profitability of departments, patients and other relevant dimensions (hospital, physicians, etc.); automated generation of patient and service line profitability data collection and reports; standardized hospital costing; and a flexible model structure to allow for growth.

Read the full case study.
Getting started

Changing cost-accounting methods is a large undertaking and involves a substantial amount of change management efforts, in addition to the technical efforts. However, to get more valuable information to drive decision-making, there must be a starting point. 

  • Start with one department or one facility. This provides the opportunity to see the benefits as well as the challenges. A phased approach is easier on the project team and the employees who are involved in the change. To simplify this disruption, we recommend using the OR as a pilot area. For most hospitals and health systems, the OR is the primary engine that drives hospital profitability and resource utilization. Also, the OR has the ability to capture rich data specific to the patient, such as physicians, labor costs, surgical supplies, medical devices and equipment, pharmaceuticals, anesthesia, and other direct costs.

    To start, identify the consumption of costs by the procedures being performed in the OR and choose the most appropriate driver. Using a software that integrates well with your EMR system, patient accounting system and other systems to build a model will make it easier to understand your OR costs. Use this new methodology in your pilot area for a testing period. After the model and results are validated, the process can be rolled out across other departments in phased order, getting buy-in along the way of the new process.
  • Make a clean break from the old. Putting in new methodology and potentially new software takes time, budget, resources and effort. When a choice has been made to invest this time and effort, make sure you do not design the new system to mimic old methodologies.
  • Drive decision-making, not just improvement in cost accounting. A new cost-accounting method will help understand detailed costs, the cost drivers and direct costs. To perform meaningful analytics that you can use to transform your business, a strong reporting system is also necessary. With a transparent view and a single source of truth, the efforts of clinicians, operations and finance can be combined to drive informed decision-making. Data-driven methodology can:
    • Improve assignment of costs based on the actual drivers of expenses
    • Determine a better understanding of the profitability of departments, patients and other relevant dimensions (hospital, physician, service line, specialty, disease, etc.)
    • Generate automated patient and service line profitability data and reports
    • Support standardized costing
    • Provide flexible model structure to allow for growth

Modernize or fall behind

Accurate costing has been a missing component of the decision-making process for many health care organizations. The future success of health care providers and health systems hinges on moving away from traditional cost-accounting methods, using the right systems and taking advantage of the expanded amount of data now available to make informed decisions. As cost pressures grow, health care organizations will need to be able to use analytics to understand which areas of the organization are profitable and which are not. Along with the need to cut costs, there is growing expectation to improve quality and manage population health. Without better cost accounting, moving to assessment of services and making smart choices about growth strategy is not possible.