Transcript the impact of decentralization on health services in

THE IMPACT OF DECENTRALIZATION ON
HEALTH SERVICES IN UGANDA: A LOOK AT
FACILITY UTILIZATION, PRESCRIBING, AND
AVAILABILITY OF ESSENTIAL DRUGS
Ogwal-Okeng JW, Anokbonggo WW,
Obua C,Ross-Degnan D, Aupont O.
International Conference on Improving
Use of Medicines
Chiang Mia, Thailand
April 1, 2004
INRUD Uganda and Makerere University
Project supported as part of the Joint Initiative on Improving
Use of Medicine under a grant from ARCH
Abstract
Problem Statement: Uganda started implementing a structural adjustment program, which resulted in
the decentralisation of health services to district level, in July 1994. It was intended to improve the quality
of health services and pharmaceutical supplies in the hospitals, with resultant increases in the level of
utilization of health services.
Objective: To assess the impact of this policy on facility utilization; prescribing patterns for ARI,
diarrhoea, and malaria; and on availability of essential drugs in hospitals in Uganda.
Design: Mixed evaluation study design using both qualitative and quantitative methods. With time series
analysis of indicators before and after the policy.
Setting and Population: The study was done in two district hospitals in northern Uganda. Patient records
in the hospitals were analyzed retrospectively. In-depth interviews were conducted with politicians; openended questionnaires and focus group discussions were administered to patients and health workers
respectively.
Outcome Measures: Facility utilization was evaluated by average monthly attendance in the outpatient
department and pediatric ward admissions. Prescribing indicator outcomes were: average number of drugs
prescribed per case; % drugs given by injections; % antibiotics prescribed; % chloroquine tablets for
malaria; % chloroquine injections prescribed for malaria; and % anti-diarrheal drugs prescribed for noninfective diarrhoea. Availability was assessed as average percent availability of nine essential drugs and
number of drugs whose availability was more than 40 %.
Results: There was a general increase in patient attendance in both hospitals, although the initial increase
later declined in Apac district. Drug availability was erratic and not always adequate in both districts. This
was much better in Lira district, where funding for drug procurement was more accessible. Prescribing
patterns varied, with improvement in some indicators, while other indicators showed no change or even
worsened.
Conclusions: The decentralisation policy increased health facility utilization. All stakeholders considered
the policy to be good. However, it has so far failed to solve drug shortages, inefficient utilization of
resources, and general low morale among hospital staff. Staff should be retrained and better remunerated
in order to cope with the implementation of the policy. Local politicians should clearly understand their
roles and responsibility under the new policy. Efficient utilization of funds at all levels of the district
administrative structures should be ensured.
Study Funding: Applied Research in Child Health (ARCH) through a grant from the United States Agency
for International Development (USAID)
Introduction
•
Decentralization advocated to improve access to health and
family planning services, and quality and efficiency of public
sector.
•
Uganda adopted a structural adjustment program (SAP) in
1994 in two phases. Lira was in the first phase that
implemented the policy in 1994 and Apac started in 1995.
•
These changes took the level of decision-making nearer to the
people and involved the community and their local leaders in
their health matters.
•
This study examined the impact of the policy on monthly
health facility utilization; availability of essential drugs and
prescribing patterns for ARI, diarrhea, and malaria in two
district hospitals in Uganda.
General Objective
To assess the impact of
decentralization policy on quality of
health care services in Uganda based
on levels of facility utilization,
prescribing behavior and stock levels
of essential drugs
Specific Objectives



To assess the average monthly patient attendances in
the out patient, maternity and pediatric units in Apac
and Lira District Hospitals before and after introduction
of decentralization policy in Uganda.
To assess the stock levels of some essential drugs in
Apac and Lira Hospitals before and after introduction of
decentralization policy in Uganda.
To assess the prescription patterns for non-complicated
malaria, non-dysenteric diarrhoea and acute respiratory
infections (ARI) in Apac and Lira Hospitals before and
after introduction of decentralization policy in Uganda.
Methods
Design and setting
•
Study conducted in Lira and Apac district hospitals
•
Retrospective study design covering 2 years before and 5 years
after policy implementation
•
Mixed evaluation methods using both quantitative qualitative
approaches.
Analyses of facility utilization records, Phamacy stocks, and
prescriptions.
Key informant interviews and focus group with 351
Stakeholders
Data sources
•
Facility utilization: Monthly attendance records of study hospitals
from 1992 – 1999
•
Drug availability: Mid-month hospital drug availability for specific
tracer drugs during 1992 - 1999
•
Prescribing Patterns: Analysis of 5040 patients records
(prescriptions).
Variations in Facility Utilization
Figure I. Facility Utilization
8000
7000
Lira
6000
Apac
Attendence
5000
Lira
Apac
4000
3000
2000
1000
0
92
93
94
95
96
Year
97
98
99
Results
Mid-month availability of tracer drugs before and after decentralization
Tracer Drugs
Before
LIRA 1992
1993
Chloroquine tablets 5,250 154,833
After decentralization *
1994
1995
1996
1997
1998
1999
41,000
58,750 34,833
53,333
43,750
65,667
Cotrimoxazole tablets 3,250
1,667
20,750
18,500 13,375
14,334
29,000
26,750
Mebendazole tablets 2,500
6,583
13,750
18,417 10,083
31,667
6,750
11,000
Ferrous sulfate tablets 29,333
9,667
54,417
12,750 7,166
0
42,083
29,333
Aspirin tablets 68,750 68,333
66,879
72,083 93,250
57,000
47,250
74,167
ORS sachets 2,833
1,179
1,417
0
4,100
878
0
15,858
10,964
26,583
25,167
Ampicillin capsules
0
683
0
18,833 9,167
Results
Mid-month availability of tracer drugs before and after decentralization
APAC 1993
1994
Chloroquine tablets 19,583 17,167
1996
1997
1998
1999
N/A
44,667
95,750 44,417
13,583
11,843
N/A
7,342
40,833
58,333 37,750
28,667
17,167
N/A
17,917
24,000
13,583 16,167
9,500
6,583
N/A
Ferrous sulfate tablets 11,333 64,833
47,167
70,583 15,417
34,917
6,583
N/A
Aspirin tablets 11,917 50,583
61,417
77,000 72,500
48,833
30,500
N/A
ORS sachets
1,410
4,332
N/A
22,917
21,988
N/A
Cotrimoxazole tablets
500
Mebendazole tablets 9,333
Ampicillin capsules
*
1995
800
1,783
2,062
0
5,900
41,083
433
283
41,583 28,417
Decentralization started in 1994 in Lira, and in 1995 in APAC
Results
Prescribing Patterns for Malaria
Before
After
National
decentralization
Standard
Lira Hospital
1992 1993 1994 1995 1996
% chloroquine tablets
% chloroquine
injections
Average # drugs
83.3
41.7
90.6
35.5
79.4
64.7
60.9
78.3
44.4
64.3
100
< 15
1.6
1.9
1.9
1.4
1.1
2.0
Apac Hospital
1993 1994 1995
1996
1997
% chloroquine tablets
% chloroquine
injections
Average # drugs
28.1
68.8
36.7
73.3
48.6
62.9
33.3
78.8
31.3
70.8
100
< 15
1.9
1.7
1.8
1.7
1.9
2.0
Summary
•
•
•
•
•
•
Increased facility utilization observed with
decentralization policy
Drug shortages persisted, and even worsened in some
facilities
Drugs stocks increased but quantities remained
inadequate
Prescribing behavior remained poor fater decentralization
Health staff had mow morale and lacked of training
opportunities
Poor relationship between health staff and political
leaders hampered successful implementation of
decentralization policy