The road to automated replenishment is not an easy one. However, through the integration of intelligence, technology, and expertise, effective demand forecasting in the health care supply chain is attainable.

Envision a scenario in which pressing the Enter key on a hospital computer
sets into motion a series of coordinated processes that reaches across the entire
health care supply chain – from the manufacturing plant, through transportation
and distribution networks, to the patient’s bedside.

Through a careful integration of the hospital system’s diverse data, including
information captured from clinical, financial, and operational sources, the
simple act of scheduling a routine medical procedure would activate an automated
system that:

  • Chooses products for the patient standardized to the (IC-9) procedure codes;
  • Assesses the need for backup contingency supplies and product options based
    on physician product preference and historical data;
  • Picks the supplies appropriate for that individual case, and groups those
    supplies with all other supplies needed for prep, recovery, and follow-up;
  • Verifies the latest contract price and ensures synchronized pricing accuracy
    among all supply chain stakeholders;
  • Determines whether there is a need to replenish supplies either at on-site
    stock locations or in the off-site warehouses of a distribution partner;
  • Aggregates the product replenishment requirements and automatically places
    an order; and
  • Communicates usage data to the manufacturer and other stakeholders for predicting
    future demand.

What Else?

These processes may well be the vision of the future for the health care supply
chain – a world in which medical and surgical supply purchases are uncompromisingly
driven by the demands of caregivers and their patients, but which also embraces
the most innovative and effective concepts of demand forecasting, inventory
postponement, and automated replenishment keeping supply chain costs under control.
It is a world where technology erases the lines of distinction between direct
purchases and distribution, sending an aggregated purchase order that automatically
selects the most cost-effective product combination and routing through the
supply chain, and then segregates and delivers the order to the appropriate
recipients.

 

It is a world where supply chain performance is measured end to end, analyzing
down to the SKU/location level, dramatically reducing the need for on-hand inventory
for scheduled medical procedures; and where health care information not normally
associated with the supply chain, such as aggregate patient data, medical research,
demographics, and the hospital’s own strategic objectives, is integrated to
extend demand predictability into the future.

This Future Is Within Reach

Today, there are three factors that make this vision of the future attainable:
the abundance of untapped information already available in the hospital setting;
a methodology of demand forecasting developed in other industries, but which
can be adapted to health care; and the recent growth and maturation of some
information technologies that enable an unprecedented level of integration of
data from diverse sources.

Integration of clinical data into the supply chain equation will revolutionize
the management process, enabling us to transform the existing yet often segregated
mix of information into actionable intelligence for health care decision makers.
Unique and customized application of some proven principles of supply chain
management from the retail industry, and some recent technological advances,
will gain health care supply chain efficiencies that we have been anticipating
for years. These efficiencies, in turn, will open the door to numerous corollary
benefits such as improved patient safety, more favorable clinical outcomes,
revenue enhancement, and an even more healthful work environment for clinicians.

Health Care Is Not Retail

This is not the first time the health care industry has considered applying
retail demand forecasting strategies. These concepts are well-proven, and have
been around since the late 1970s, when health care executives first used them
in efforts to standardize, manage, and package product in patient-ready (consumer)
configurations. In health care, those efforts met with some success. Proponents
found they could work well to support limited procedures and volume, but often
failed to cover contingencies clinicians frequently encounter at the pre-op,
procedural, and recovery phases.

Having incorporated these lessons into the lore of supply chain management,
we contend that our industry is now ready to develop its own set of principles
that draws upon the concepts of retail demand forecasting models, but which
fully accounts for the unique nature of health care – in which a failure to
meet the demands of consumers (patients and clinicians) can have dire consequences.

Advances in technology, historical trending, surgical scheduling, and clinical
information systems have given us the ability to manage usage more effectively.
The integration of supply chain management systems with point-of-use data entry,
and improvements in inventory management (such as the emerging RFID technology),
are enabling us to clear the way for the final and most important step in the
integration of health system data – linking the clinical pathways charted in
advanced patient data to the supply chain decision-making process.

Projecting Demand Accurately Is the Key

In the retail industry, one of the most common challenges facing executives
is a lack of reliable information for predicting what customers will buy. Retail
stores stock their shelves based on a combination of intuition, experience,
and market research that is often highly speculative. It is a risky proposition,
especially for the buyer responsible for filling warehouses that serve thousands
of retail outlets. The proliferation of clearance sales at the end of each buying
season attests to the odds for failure.

Yet, even with its inherent uncertainty, the retail industry has developed
a methodology of predicting demand that, to date, has worked better than in
other industries. Companies like Wal- Mart, Office Depot, Costco, and Dell have
been able to forecast with enough reliability to be able to significantly reduce
losses resulting from unsold or discounted inventory.

Health care executives are different – but not just because their decisions
have considerably more impact on the lives and well-being of the people they
serve.

Although hospital administrators often feel a lack of predictability, their
potential for success in determining what they will need to stock their supply
shelves may be substantially greater than in almost any other industry – at
least for certain subgroups of their total annual purchases. Where else but
in a hospital can a manager walk in and look at a schedule that tells the number
of customers and the potential product demand with relative accuracy for that
very day?

To date, efforts to automate the health care industry have been admirable.
Most hospitals now have automated systems in place to track diagnoses, medical
events, patient satisfaction, purchasing, accounts payable, reimbursements,
surgical team preferences, dispensing of pharmaceuticals, and more. Unlike the
retail environment, however, the principal challenge in hospital systems is
not a lack of reliable information to support decision making, but rather the
lack of visibility and integration of data that is already there – data that
is routinely gathered but stored in diverse information systems that rarely
communicate with each other in meaningful ways.

With so many data sources, and so many advances in technology and analytical
capabilities, there could be boundless possibilities for a health care institution
capable of organizing and using its information to gain business and clinical
advantages. Ultimately, the goal should be to create a health care supply chain
that is capable of keeping the clinicians supplied, and at the same time, limiting
on-hand inventory to what is necessary to address contingencies.

On the Road to Automated Replenishment

Hospitals are across the board in their level of sophistication in supply chain
management, ranging from facilities’ systems that still track physician preferences
on index cards, to systems that are managing the supply chain with advanced
methodologies such as activity-based costing, Internet-based analysis of purchasing
and utilization, and outsourcing the management of supply acquisition in high-dollar
clinical units.

Yet, further advances for even the most sophisticated institutions depend not
only on the application of new technological tools and importation of proven
methodologies from other industries, but also on the continued refinement of
their existing technologies that have shown their worth. We envision four key
mileposts on the road to the goal of automated replenishment:

  • Expansion of intelligence gathering and analytical tools to encompass not
    only the entire health care supply chain, but also to integrate clinical information;
  • Development of health care-specific demand forecasting methodologies;
  • Application of these new management methods to create a clinically driven
    demand chain supported by low on-hand inventory and automated replenishment;
    and
  • The application of this expanded supply chain intelligence for quality improvement
    and reimbursement optimization with the health system.

Figure 1. Potential Sources of Information to Support Effective Demand Forecasting
in a Health Care System

Expanding Proven Value

Experience has shown already that integrating and analyzing the data within
a single management system, such as hospital purchasing, can reap substantial
benefits. Internet-based mining of routine purchasing data, when properly analyzed,
can produce information that forms the basis for combining purchase orders,
standardizing products, reducing inventory, and improving contract compliance
– all proven strategies for reducing costs. Not only that, analysis of supply
chain activities has enabled the creation of models that more appropriately
determine actual supply costs. Integrating supply chain information with a hospital’s
clinical, financial, and human resources information systems offers an opportunity
to analyze relationships between supply utilization and other important measures
like labor costs, reimbursement levels, and patient outcomes. Such analysis
enables health care systems to identify the cost of care at the episode level,
detecting variations from the norm by clinician and procedure. Potential benefits
are immediate, including a reduction in the need for clinical employees to track
and manage inventory, and capturing supply consumption as a part of clinical
documentation.

Forecasting Demand

By integrating clinical information systems, the health care system can link
utilization to outcomes and provide more accurate clinical documentation, which
immediately offers the potential to improve charge capture and reduce reimbursement
denials. The hospital executive, or the hospital’s supply chain partner, has
reliable information to create a realistic bill of materials for procedures
from the preparation phase through recovery and follow-up.

In specific locations, such as the surgical suite, SKU- and location-specific
forecasting technology can be used to synchronize surgeon preference information,
and predict demand based on the hospital’s scheduling, patient demographics,
and even seasonal demands – with the consequent benefits of reduced inventory
levels in the OR, lower costs for case preparation, and improved fill rates
and service levels. In many cases, hospitals are likely to experience increases
in clinical satisfaction, productivity, and patient safety as well.

Because the risks associated with health care are unlike those of any other
industry, the quality of demand forecasting must be impeccable. The process
must:

  • Project demand by location and SKU and address long-term and short-term
    needs;
  • Support the hospital’s budgeting and planning process to account for introduction
    and expansion of medical/surgical programs; and
  • Address the importance of the critical lead times that are inherent and
    essential in the health care environment.

Automated Replenishment

In effect, this approach would convert the conventional medical/surgical supply
chain into a “demand chain,” in which information from many sources actually
drives the product flow. Originating at the patient’s bedside, the information
flows through the demand chain back to the production plant, where reliable
demand forecasts enable manufacturers to make intelligent production decisions.
The distributor uses the same information to reduce overstocks and product shortages,
and both the manufacturer and the distributor feed supply chain intelligence
back to the hospital in the form of an efficient product flow.

When the abundance of information in a health care system is properly analyzed
and applied to a well-tested demand response model, the risks associated with
automation can be reduced to an acceptable level in an environment that must
allow for emergencies, and cannot compromise patient safety. In this environment,
automated replenishment must also be impeccable – ensuring appropriate, accurate,
and on-time ordering tied directly to clinical demand, and incorporating processes
to:

  • Track and report exceptions;
  • Determine procedure-specific lead times for re-orders; and
  • Be self-correcting.

Improving Quality and Enhancing Revenue

Effective demand chain management produces tangible changes in the traditional
supply chain – most notably the cost savings from inventory reduction and postponement,
but there are many additional benefits from this approach of gathering information
from diverse sources within the health care system. The heightened visibility
of the demand chain enables hospitals to identify opportunities for improving
patient outcomes. It also allows better case documentation to support revenue
enhancement.

The Demand Chain Continuum

Despite many advances in supply chain management, too much of the medical and
surgical inventory in health care consists of slow-moving SKUs with a high risk
of obsolescence, expiration, damage, or recall. With adequate safeguards for
protecting patient identity, pressing the Enter key to schedule a particular
medical procedure can generate specific lists of medical/surgical supplies necessary
for every stage of a procedure – taking into account preferences of the clinician;
patient attributes such as age, weight, sex, medical conditions, and allergies;
and predictable external factors that influence the utilization of supplies.

Together, a fine-tuned forecasting system and automated replenishment process
can prepare the hospital for every scheduled procedure, reduce the costs for
such procedures through standardization, and help ensure favorable outcomes.
These systems also can help project the likelihood of unscheduled cases and
unexpected acuity issues, and determine the necessary product mix for backup
and support.