Has your performance as a materials manager ever been
measured using metrics that seem unrelated to your efforts to control
supply costs? Have you had to defend why your supply costs are higher
than another hospital within your health system, despite the fact that
your purchasing practices are very similar, and the performance
benchmark was adjusted for case mix?
Surprisingly, most supply expense metrics used by
hospitals nationwide fail to accurately factor in the kinds of patients
the hospital treats, and the supply needs dictated by those patients.
It’s no wonder materials professionals find themselves on the defensive.
Current methods: Incomplete at best, misleading at worst
Your organization may measure supply costs as a
percentage of revenue or of total expense, or per adjusted patient days or per adjusted discharge. Used consistently,
these measurements can track performance over time within your
organization ("internal benchmarking"). However, these metrics provide
limited information and cannot be used to compare your facility with
another – even within the same health system ("external benchmarking").
Flaws exist with each of these popular metrics,
especially when used for external benchmarking. These measurements
assume that all hospitals and health systems perform the same cases – in
type and proportion. This is obviously not realistic.
Some hospitals admit a high percentage of their patients
for supply-intensive procedures such as spinal fusion, pacemakers/AICDs
or joint replacements. Others see a higher proportion of their patients
admitted for normal vaginal deliveries, pneumonia or congestive heart
failure – far less "supplies-intense" admissions. Logic dictates that
the supply expense will be relative to patient types. Unfortunately,
current supply metrics may put unlike facilities in the same "peer
group" using unrelated measurements.
Furthermore, measuring supply expense as a percentage of
adjusted patient days, for example, may actually be counterproductive. A
hospital with an inefficient, higher average length of stay (ALOS) will
appear to be more supply-efficient, because supply costs are spread out
over a higher ALOS in the metric. This creates an inaccurate snapshot of
supply chain efficiency.
Case Mix Index adjustment is not the solution
One common mechanism used in an attempt to level the
playing field is the Case Mix Index (CMI). CMI was originally created to
"weight" patient admissions for reimbursement for the entire scope of
patient care. It was never intended as a predictor of that slice of
patient care cost related to supplies.
Here’s one striking example of how CMI doesn’t work to
predict supply expense: DRG 520 (a spinal fusion surgery patient) has a
case weight of 1.63. DRG 121 (treatment of a heart attack patient) has a
case weight of 1.62.
DRG 520 involves setting up an operating room for a
procedure that includes the implantation of expensive hardware plus the
usual O.R. supply costs, in a multi-day length of stay.
DRG 121, on the other hand, includes at most a
diagnostic cath procedure (no stent), in a multi-day LOS. DRG 121
averages only $942 in supply costs. So while these two DRGs share a
virtually identical Case Weight, their expected use of supplies differs
by a factor of five.
Time for a new supply performance metric
To enable credible external benchmarking, a new supply
cost metric must factor in a facility’s unique patient mix, and then
account for the expected average supply costs of those patients.
Expected average supply costs will be more reliable if they are derived
from analysis of a large number of facilities.
Consider the following example. Aspen Healthcare Metrics
compared the supply intensity of a five-hospital system with the average
supply intensity of a sample of 1,400 acute care hospitals nationwide.
We analyzed the actual acquisition costs for all medical/surgical
implants, pharmaceuticals and miscellaneous supplies for all 540
inpatient DRGs.
In our analysis, the IDN had a CMI 9.1 percent higher
than the national average. However, when standardized supply cost data
was applied, the difference between the national average and the health
system was a whopping 47 percent higher. Materials managers were in the
crosshairs.1
But further analysis revealed why this health system’s
supply costs were 47 percent higher than the national average, although
its CMI was just 9 percent higher: The health system treated more
patients in certain supply-intensive DRGs and service lines than the
national average.
For example, this health system has 33 percent more
total joint procedures, 48 percent more pacemakers/AICDs , 123 percent
more lumbar fusion cases, and 229 percent more drug-eluting stent
admissions than the national average. These procedures are some of the
most intensive supply consumers. Ultimately, further analysis quantified
that supply cost management was in fact quite efficient when patient mix
was accounted for.
So before supply chain managers are penalized unfairly,
their organizations should explore more reasonable and rigorous metrics
of supply chain efficiency. Ultimately, it all gets down to patient
care.
