Transcript Cost-utility analysis (CUA)

DRUG DATABASE AND PERFORMANCE INDICATORS FOR UTILIZATION MONITORING
Authors: Pongcharoensuk P1, Angsanant M2, Chantrakunopars P2, Phuthong P1, Kongsawat S1
Institution: 1Mahidol University; 2Ministry of Public Health, Thailand
Problem Statement: Drug expenditure accounts for 30% of health resources in Thailand. For health care
administrators, a standard drug database and management system are essential for utilization monitoring at
both national and facility levels.
Objective: To determine key performance indicators for drug utilization monitoring.
Design: Retrospective, time-series design.
Study Population: Twenty-six government hospitals (size range, 10–1,000 beds) with complete electronic drug
purchasing database for fiscal years 2000–2002, were purposively selected.
Intervention: Different drug codes (of similar drug products) from different hospitals were transformed into the
same standard drug code. Using the standardized drug codes, drug purchase data from all 26 hospitals were
consolidated to develop key performance indicators.
Outcome Measures: Total drug purchase (in baht), ratio of essential drugs (EDs) to nonessential drugs (NEDs),
and drug purchase by pharmacological group were used to monitor utilization trends. Unit price statistics of
generically equivalent drug products in 2002 were used as indicators for price monitoring of individual product.
Results: More than 100,000 drug purchase transactions were gathered. Results showed an increase in total drug
purchase from 817.8 million baht in 2000, to 908.3 million baht in 2001, and 978.4 million baht in 2002 (11.0%
growth in 2001 and 7.7% in 2002). Also increased was the ratio of NEDs to EDs—15:85 in 2000, 17:83 in
2001, and 18:82 in 2002. For pharmacological classification, anti-infectives ranked first in total purchases for
all three years but with a decreasing percentage, from 30.0% in 2000, to 26.4% in 2001, and 23.6% in 2002.
Cardiovascular drugs ranked second in total purchases but with an increasing percentage, from 11.4% in
2000, to 12.4% in 2001, and 13.7% in 2002. In unit drug price (baht) statistics for 2002 (listed in the following
order: number, average, standard deviation, minimum, maximum, p50, p75, p90), a wide variation was found
among both multisource and single-source drugs as follows: aledronate 10 mg tab—48, 48.99, 6.78, 45.86,
81.68, 48.15, 48.15, 48.15; ceftriaxone 1 g inj—22, 36.62, 22.74, 20.18, 124.12, 31.03, 37.98, 56.15; and
simvastatin 10 mg tab—66, 4.72, 1.92, 2.00, 8.92, 4.24, 5.35, 7.68.
Conclusion: Although not standardized, the drug database available in Thai hospitals can be a valuable resource
for utilization monitoring and system management to increase efficiency and effectiveness.
1
Introduction
Drugs accounts for 30% of health expenditures in
Thailand. Public hospitals are major health service
providers including pharmaceutical services.
In each hospital, drug database (such as purchase and
utilization records) has been kept accumulating for
many years. However, these drug databases have
limited utility for national policy-making due to the lack
of standard structure and code as well as inadequate
information technology in data management.
Hospital and health care administrators need to use these
available data as monitoring tools for decision making
and management at both national and facility levels.
2
Objective
To determine key performance indicators for drug
utilization monitoring:
1. Trend monitoring
– Total purchase, ratio ED:NED
– Total purchase by drug group, etc.
2. Price monitoring
– Unit price* statistics (mean, standard deviation,
min, max, percentile 50, 75, 90, 95) of generically
equivalent drug products.
*Unit price is net purchase price per unit (tablet, bottle).
3
Methodology
Study design: retrospective, time-series design.
Study Procedure
I. Construction of Standard Drug Code (25 digits)
– Type (Single* or combinations), 1 digit
– Drug Name, Paracetamol, 10 digits
– Dosage form, Tablet, 3 digits
– Strength, 500 mg., 4 digits
– Company, GPO, 4 digits
– Package size, 100 tab, 3 digits
*Only single chemical entity drug code is available for this study.
4
Methodology
II. Data collection
Sample: 26 public hospitals (10-1000 bed) were
puposesively selected. They had complete records
and were willing to provide a copy of Electronic Drug
Purchasing Database of the year 2000-2002.
III. Data transformation
Different drug codes (of the same drug product) from
different hospitals are changed into the same
standard codes.
IV. Data consolidation and analysis
All the transformed drug purchasing data from 26
hospitals were consolidated and then analyzed using
Microsoft Access, Excel and SQL.
5
Results: Key performance indicators
I. Total drug purchase (million baht) of
26 hospitals in 2000-2002.
----------------------------------------------------------Year
Purchase
%change
(adjusted*)
(adjusted*)
-----------------------------------------------------------
2000 817.8
2001 908.3 (898.4) 11.0 (9.9)
2002 978.4 (947.7)
7.7 (5.6)
---------------------------------------------------Notes: *adjusted for inflation of 1.1% in 2001 and 1.9% in 2002 (NESDB).
6
II. Drug Purchase as % of Total
Operating Expenses in 2002
------------------------------------------------------------Type of Hospital (bed) N Range % Avg.
------------------------------------------------------------Regional (500-1000)
7 15.6-29.4 20.7
General (200-500)
14 11.2-27.2
18.3
Community (10-90)
5 12.5-19.6 17.5
------------------------------------------------------------(The bigger the hospitals, the more drug used)
7
III. Total Purchase Ratio of Non-Essential
Drug (NED) and Essential Drug (ED)
--------------------------------------------Year
Ratio (NED:ED)
--------------------------------------------2000
15 : 85
2001
17 : 83
2002
18 : 82
--------------------------------------------8
IV. Ranking of Drug Purchase (%) by
Pharmacological Groups
------------------------------------------------------------Rank Group*
2002 2001 2000
------------------------------------------------------------1
3
4
5
6
8
10
Anti-bacterials
Anti-asthma
Anti-diabetics
Hypo-lipidemic
Anti-anaemic
NSAIDS
Cytostatic
21.6
4.0
3.9
3.3
3.3
2.5
2.2
23.8
4.4
3.3
2.8
2.7
2.6
1.7
26.6
3.9
3.0
2.4
2.1
2.0
1.4
Total (million)
824
779
730
------------------------------------------------------------------------------------------------------------------------*A total of 125 groups
9
V. Unit Purchase Price (baht) Statistics of
Generically Equivalent Products of 2002
--------------------------------------------------------------Drug
#
Min
Max
Avg
48 45.9 81.7
44 0.2 5.0
32 50.3 58.4
27 1.4 26.3
65 220 487
22 20.2 124
59
1.2 14.1
66
2.0 8.9
12 57.3 87.9
49.0
0.7
52.3
6.1
422
36.6
4.2
4.7
59.9
p50
p75
--------------------------------------------------------------Aledronate 10 mg tab
Alprazolam .25 mg tab
Atorvastatin 20 mg tab
Captopril 25 mg tab
Cefoperazone 1 g inj
Ceftriaxone 1 g inj
Ciprofloxacin 250 mg tab
Simvastatin 10 mg tab
Rosiglitazone 4 mg tab
48.2 48.2
0.5 0.8
50.3 52.3
7.5 7.5
486 486
31.0 38.0
3.8 5.4
4.2 5.4
57.3 57.3
---------------------------------------------------------------Notes: Italics are single-source drugs
10
Discussions
Lesson learned
An urgent need for data management of available hospital
database to be utilized for drug system management and
policy implementation.
Policy implications
Structure
- A central data warehouse under the Ministry of Public
Health, with capability for huge data management.
- An agency for determination and maintaining of standard
drug codes (possibly FDA?).
Process
- In addition of Quantity indicators of drug purchase as
shown, Quality indicators of drug use are necessary to
guarantee good outcomes for the patients and economic
outcomes for providers.
11
Conclusions
Though not standardized, hospital drug database
can be a valuable source for improving system
efficiency and effectiveness of drug use.
Recommendations
More research to be pursued:
– Quality of drug use and cost consequences of an
illness, particularly chronic diseases such as
diabetes or asthma.
– Quality of drug use and outcomes of treatment.
– Systematic management of drugs (selection,
procurement, utilization and outcomes).
This study is supported by the National Health Security Office.
12