Successful Implementation of Vendor-Managed Inventory at a Major Teaching Hospital

1
Successful Implementation of Vendor-Managed
Inventory at a Major Teaching Hospital
Tsui M,1 Wilson DI, 2,3 Merry H, 4 Phulwani K, 1
Dooley MJ1,5 1) Pharmacy Dept, Bayside Health,
Victoria, 2) Dept of Civil Environmental
Engineering, University of Melbourne, Victoria,
3) Master Research Aust P/L, Victoria, 4)
Clifford Hallam Healthcare, Australia 5)
Department of Pharmacy Practice, Victorian
College of Pharmacy, Monash University, Victoria.
Results continued 4. Fill rates Initial fill
rates were lower than the target of 97, but
gradually increased to be above 98 after 4
months. Although the wholesaler has met
the fill rate target, it is being compromised by
its supplier, the manufacturers. This has
resulted in the wholesaler having to keep a
higher than necessary stock holding as buffer
stock in order to meet its own customer needs.
Results 1. Staffing Since December 2005, all
imprest locations are checked and restocked by
CH2 staff. As a result, 3.5 EFT imprest-related
staff have been re-deployed to clinical support
duties. 2. Stock Holding Stock holding initially
jumped unexpectedly during Christmas New Year
but reduced to 2.5M after 5 months, compared to
the expected holding of 2.9M, based on patient
throughput projection. The reduction was
maintained, and in June 2007 stock holding was
2.4M. 3. Customer service The first
survey conducted in July 2006, found that the
average score given by customers for the overall
service was 5 neither poor nor excellent.
However, when asked to compare services before
VMI and after, the average score was below 4,
which was unexpected. Changes were therefore made
to the service delivery from July to November
2006, and a second survey conducted in January
2007 indicated that perceptions of service had
improved significantly.

• Introduction
• Traditional pharmacy medication imprest services
  require significant local labour resources and
  stock holding. The concept of Vendor Managed
  Inventory (VMI) used extensively in the business
  sector may provide efficiency gains if a
  wholesaler manages the logistics of medication
  imprest in a hospital pharmacy.
• Objective
• To implement VMI for medication imprest services
  in a major teaching hospital
• To reduce number of hospital staff required to
  provide imprest services
• To reduce stock holding
• To improve customer service
• To maintain fill rates at or above 97
• Method
• The Alfred, a 400-bed, acute tertiary referral
  hospital, introduced Ward Box, Clifford Hallams
  (CH2) version of VMI, to 40 imprest locations, in
  December 2005. In this model, CH2 staff would
  check imprests, order from their warehouse, pick
  non-CH2 stock from Pharmacy, and rotate stock.
  They would also check expiry dates in the ward
  imprests.
• An analysis of pharmacy stock holding was
  conducted over a 15 month period (9 months prior
  and 5 months after implementation). Actual stock
  holding was compared to projected stock holding
  estimated from hospital activity.
• To measure customer satisfaction, a
  self-administered questionnaire was distributed
  to hospital staff in July 2006. The questions
  assessed standard customer satisfaction elements.

Discussion The results show that the VMI project
has achieved the objectives. Despite an initial
surge of stock holding, the one-off savings from
stock reduction was maintained in June 2007, 18
months after implementation. In terms of fill
rates, random audits are still being conducted,
and the results fed back to CH2 on monthly basis.
Conclusions A comprehensive Vendor Managed
Inventory system was successfully implemented at
a major teaching hospital. All key objectives
have been met. This is the first site in
Australia that such a system has been
implemented. However, it was found that
manufacturers have lagged behind in fill rates
compared to wholesalers. Improved data
interchange between wholesalers and manufacturers
will in theory reduce stock holding across the
entire supply chain.