Abstract
The medical expenditure for outpatients under the Civil Servant Medical Benefit Scheme’s direct
billing system continued to be very high in 2009-2010 due to the use of drugs outside the National Lists of
Essential Medicines. This study estimated potential cost-savings using the aggregate drug utilization data
from 28 large public hospitals under two scenarios of reimbursement: 1) the lowest price of generic in each
drug group, and 2) median price of each drug. If the reimbursement rate was set most restrictively based
on the minimum price of the essential drug with an equivalent Anatomical, Therapeutic, and Chemical
(ATC) code 1-3 (eg. the lowest price of simvastatin for antilipids), the expenditure for antilipids,
antiosteoarthritis (non-steroidal anti-inflammatory drug (NSAID), cyclo-oxygenase (COX-2) inhibitor, and
symptomatic slow-acting drugs for osteoarthritis (SYSADOA)), and antisecretory/antiulcers (histamine-
2 receptor antagonist (H-2RA) and proton pump inhibitor (PPI)) in 2010 would be reduced by 72.8, 85.7,
and 86.5%, respectively. If each individual hospital was allowed the drug reimbursement as is but no
higher than a ceiling determined as the median of price per prescription for the drug with the equivalent
ATC 1-5, the expenditure for antilipids and antiosteoarthritis would be reduced by 13.4 and 13.3%, respectively.
For 20 drug groups with relatively high expenditure, the reimbursement ceiling could reduce the
expenditure by 14.0%. The least restrictive reimbursement condition using a fixed rate referent to the
median price of drugs with equivalent ATC 1-5 from all hospitals could reduce the expenditure for the 20
drug groups by 2.9%. Therefore, to contain expenditure using a fixed rate of reference price based on
drugs with an equivalent ATC 1-5 would be the least restrictive policy option.
