Fee-for-service proves viable in academia

The radiology department at Leiden University Medical Center performs approximately 150,000 examinations each year. We developed and implemented a simple fee-for-service tool to monitor clinical output and relate this to required input.

Over the past 15 years, we have seen an increase in the number of clinical procedures, particularly the more complex ones such as CT, MRI, interventional radiology, and mammography. But while output was increasing, our budget-or input-remained stable. This situation was a potential disaster, threatening resources directed toward research and education. To meet this external pressure to boost clinical output, we had to increase our organizational effectiveness and work with sufficient budget.

Organizational effectiveness can be split into four categories:

• Operational effectiveness. Does the department run smoothly and meet minimum quality standards? Addressing this issue should take top priority.

• Cost-effectiveness. Does the department offer value for money?

• Commercial effectiveness. Is the department satisfying its customers?

• Strategic effectiveness. Is it forward-thinking, with plans for continuity and future developments?

During the late 1990s, we increased both the number of radiological procedures per 1000 outpatient visits and our tertiary referral rate faster than the average for the Netherlands. Commercial effectiveness proved to be poor, however. In August 2003, our patients were waiting over 20 weeks for elective MRI examinations, 10 weeks for elective body CT, and around eight weeks for myocardial scintigraphy.

The next phase of the process involved displaying output and relating this to production contracts or service level agreements. We summarized hundreds of procedural production costs into 82 codes. Past trends in demand for each procedure were used to gauge anticipated demand. These figures were compared with production costs to determine whether workload demands were achievable.

Real-time monitoring of production levels is an important part of this process. We prepared monthly reports on output in relation to contracted number of examinations and made these available on the HIS. Hospital managers could see the extent of rising demands made by referring physicians.

We then asked for more money to meet the increasing demand. We had to find a system that shows how changing output is related directly to changing input. Radiological reimbursement rates cannot be used for this in the Netherlands, because they are far lower than the real costs involved. Calculations need to be done on a procedure by procedure basis.

Performing radio-frequency ablation of a patient's liver tumor under general anesthetic may cost British Sterling 2000 in materials, 90 minutes' use of a CT room, 90 minutes for two technologists and one radiologist, and two days in a hospital bed. Costs associated with a neuroradiologic intervention to correct an arteriovenous malformation amount to British Sterling 2138 in materials, three hours time in an angiography suite, nine hours' time for three technologists, six hours' time for two radiologists, one outpatient appointment, and four days in a hospital bed. We forecast that we would perform 40 more RF liver tumor ablations in 2004 than in 2003, and 10 more AVM interventions. This breakdown of costs made it easier to see the financial impact.

The support of your hospital board and of management teams from requesting departments is vital when asking for budget alterations.

We now assess the actual production costs compared with contracted production costs every month on a department by department basis. If the change is more than 5% over the year, we send an invoice to that requesting department for the entire amount outstanding. Any excess in expenditure on interventional procedures is also invoiced annually.

We increased our production contracts in 2003, adding six full-time equivalent staff members, and increased our interventional budget by British Sterling 250,000. We raised production again in 2004 to reduce the waiting list. By summer 2004, we saw a marked reduction in waiting time for elective MRI, myocardial scintigraphy, and mammography. The number of elective CT procedures grew by 24%, though waiting time fell only slightly. This change reflects the growth in demand for CT.

We have tried to meet the needs of requesting departments and, where necessary, negotiated new contracts based on their demands. The situation requires continuous monitoring and adaptation as well as a good working relationship with requesting physicians.

PROF. BLOEM is chair of radiology and nuclear medicine at Leiden University Medical Center in the Netherlands. This column was adapted from his presentation at the 2004 EuroPACS/MIR meeting.