Metrics guide PACS budget forecast model

Twitter icon
Facebook icon
LinkedIn icon
e-mail icon
Google icon

A PACS is not a one-time only purchase; new modalities, software applications and upgrades, expanding access to an existing system and increased storage requirements are among the many elements that a system administrator may be called on to implement.

“Hence, determining the correct amount of capital to reserve annually for the information technology infrastructure can be a difficult process for the administrator of a medical center,” wrote Steve G. Langer, PhD, associate professor of radiologic physics at the Mayo Clinic in Rochester, Minn.

Langer has developed a model for estimating infrastructure capital for the ongoing maintenance of PACS, which was published online before print in the Journal of Digital Imaging.

He came up with his model to guide the financial planning for the ongoing maintenance of PACS at his facility.

“Often a medical center’s PACS or department administrator is expected to track the annual expenses associate with running various information systems and infrastructure and predict the costs needed to maintain them in the future,” Langer wrote. “This can be a challenge.”

The team at Mayo came up with an approach that uses the known present-day adequacy of equipment, modalities and data storage along with the predicted data volume or procedure growth for the coming year to derive a future budget estimate.

“For any given performance requirement (storage, networking, computational power), if the growth rate of the performance/$ for that item is greater then the growth rate of the requirement, then for any budget cycle beginning with sufficient funds, a constant budget allocation will always be sufficient,” Langer wrote. “This is equivalent to stating that the slope of the requirement grows slower than the slope of the performance/$ graph for that resource.”

He noted that the model requires a few assumptions:

  • Data volumes can be predicted for one year, preferably two;
  • Networking requirements will scale with data volume;
  • Computational requirements will scale with data volume; and
  • Accurate data quantifying growth rate of the performance-to-dollar ratio for resources.

Once Langer assembled the metrics for his department, he used Microsoft Excel to determine the slope, via the SLOPE function in the application, from the table of values.

“Doing this provides one with two methods to determine budgets: either by comparing the slopes of the resource versus the need (from the tables) or by laying the two plots on the same graph and visually determining the relationship,” he wrote.

This method allows a department or system administrator to see one of three possibilities for their resource use against projected needs for a PACS, he noted.

According to Langer, if the slopes are the same one can budget constant dollars. If the need slope is less than the resource slope, one can reduce budgeted assets. If the need slope is greater than the resource slope, an increase in budget assets will be required.