Developing an appropriate district health system performance assessment framework for Uganda
(2016)
Files
CKTashobyaThesisfinalMay102016.pdf
Restricted Access - Adobe PDF
- 4.17 MB
Details
- Authors
- Supervisors
- Abstract
- Introduction Health Systems Performance Assessment (HSPA) is still in its infancy, with most of the experiences only in the last three decades. Most of the documented experiences of HSPA involve high income countries (HICs), and focus on international or national levels. If at the subnational level, the experiences mostly pertain to the hospital level. Only a few HSPA frameworks have been developed and implemented in low income countries (LICs), and even fewer at the sub national health system level, for example the district level. Uganda has implemented the District League Table (DLT) since 2003. The Uganda DLT was developed to compare performance across districts, and includes input, process and output indicators, and a composite index used to rank districts’ overall performance. A number of achievements have been noted with the DLT including: the development of a database with a range of district variables; comparison of performance across districts; stimulation of interest among health system managers given the picture of variation in performance; and facilitation of discussion of health system-wide issues. However, a number of criticisms have also been raised against the DLT which include: comparison of districts with marked differences in context; the use of poor quality data to make definitive conclusions; the failure to identify factors behind observed performance; and limited use of findings for decision-making. Objectives The main objective of the study was to develop proposals for the adjustment of Uganda district HSPA. The study was also intended to contribute to the global HSPA literature with particular emphasis on LICs and the subnational (district) level. Methods Given the complexity of HSPA, and the limited theoretical development of the subject, especially in LICs, a multi-phase case study research strategy was used, using a mixed methods approach. In phase 1, desirable aspects of HSPA frameworks were extracted from the literature and used to construct a theoretical model of a HSPA framework. The model was validated by a Uganda-based expert group. Subsequently the theoretical model was used to review international HSPA frameworks. The theoretical model was thereafter referred to as a normative HSPA framework. In phase 2 mixed methods of research were used to carry out a comprehensive review of the DLT. Data from Key informant Interviews (KIIs) and document review was analysed using historical and policy analysis to tell the story of the development and implementation of the DLT. District HSPA data from the DLT database was analysed using Hierarchical Cluster Analysis (HCA) with the purpose of comparing the approach with league table rankings. The normative framework developed in phase 1 was used to carry out a comprehensive review of the DLT, using findings from the two previous approaches of phase 2 and from the KIIs. In phase 3, mixed methods of research were used to develop proposals for the adjustment of district HSPA in Uganda. HCA was applied on district HSPA data from the DLT and various government databases to explore relationships between different health system variables. Findings from this just mentioned sub study and from phases 1 and 2 were brought together to inform the development of draft proposals for the adjustment of Uganda district HSPA. The draft proposals were presented to a second Uganda-based expert group for validation and contextualisation. This led to the final proposals for the adjustment of district HSPA in Uganda. Results In phase 1 of the study, a normative framework for HSPA in a LIC like Uganda was produced, together with a number of lessons from international HSPA experiences. The framework is composed of seven attributes that cover content and contextual factors, and has been used in subsequent phases 2 and 3 of the study. A story of the development and implementation of the Uganda DLT emerged from phase 2 of the study, indicating a number of achievements but also some challenges and criticisms. The DLT was noted to have achieved some of its objectives, including the comparison of district performance, and to a lesser extent increased ownership of district performance by local government leaders. However, poor results were achieved in terms of determining factors behind observed district performance, instituting corrective measures and encouraging good practices. It was also noted that the DLT was better appreciated in the early years of its implementation. Changes in the Ugandan general context and in the health system since the late 2000s have not been supportive of health system-wide initiatives including HSPA models like the DLT. The study noted frequent change in power of the various actors at the district level, especially the Ministry of Health, donors, district managers (political, administrative, technical), and the community, which clearly has implications for district HSPA. The DLT approach to data analysis and presentation with the emphasis on ranking of all districts from the ‘best’ to the ‘worst’ performer was considered highly summarised, uninformative and not taking sufficiently into consideration relevant district contextual differences. HCA provides a compromise between league table rankings and the detail of individual district variable data. A comprehensive critique of the DLT along the lines of the normative HPSA framework was carried out to inform phase 3 of the study. Phase 3 of the study produced a set of recommendations for the adjustment of district HSPA in Uganda, structured along the lines (and attributes) of a normative HSPA framework. The process of development (adjustment) of Uganda district HSPA is meant to be participatory, particularly targeting local government managers. The adjusted framework explicitly relates to the Uganda health system conceptual model, also considering social determinants of health, and recognises the non-health care aspects of the health system as contributors to health system goals. The framework relates to the Uganda policy, as well as the organisational, epidemiological and societal context. Some of the aspects of the context that have been considered are the existence of multiple stakeholders in the district health system with varying power, and the frequent reforms that take place. Given the complexity and dynamism in many districts, the proposed framework puts emphasis on data collection, analysis and use of information for decision making at the district level. The objectives of the proposed framework are to support decision-making with emphasis on the district level, and to provide more accountability at all levels. The proposed framework has clear dimensions and sub-dimensions under which a number of indicators have been organised. The indicators have been chosen because of their strategic importance, feasibility and technical appropriateness. This study recommends that a unit should be put in place at the national level to support HSPA throughout the health system. Concerted efforts should be made to improve the availability and quality of data for district HSPA, beginning with data pertaining to the healthcare system and extending to the rest of the health system, more broadly defined (i.e. including the non-healthcare system and district context) in a strategic and phased manner. A schedule for data analysis and presentation has been proposed highlighting the level, focus, periodicity and analytical models to be utilised. The analytical models proposed include league table rankings together with HCA, social network analysis and performance trends analysis. The analytical models proposed can be used even as efforts are being made to improve the range and quality of data. Explicit mechanisms of change through which HSPA information is expected to lead to improvements in the health system have been proposed. These include benchmarking, peer learning, implementation of quality improvement initiatives and public reporting of HSPA information. The HSPA framework should be adaptable over time and applicable in a certain context, it is proposed. Reviews of the framework should be carried out every five years at the time of review of the health sector strategic plan. It should be feasible to implement the proposed framework for district HSPA in the Ugandan context today, since explicit attention went to relating the framework to this particular context. Some of the changes proposed involve improved clarity of what needs to be done in terms of district HSPA, by whom and when. A number of other changes are incremental, and can be achieved with minimal additional resources. However substantial resources are required to implement some of the other proposals, for example the establishment of a (national) unit responsible for HSPA, and improving data availability. In a context of limited resources and competing needs, decisions need to be taken on the level of priority HSPA should get. It is important to note that currently resources are being used for district HSPA; however this study has shown that the outputs of the current model of district HSPA are not being utilised all that much for decision making. Conclusion and Future Research Priorities A multi-phase case study strategy with mixed research methods was used for the development of an adjusted Uganda district HSPA framework. A number of lessons can be learned from this experience for Uganda, LICs, and globally, in regard to HSPA; and the methods that can be used for its study. In HSPA the content/design, processes of implementation of the framework, and the context and actors do matter. However it is possible for countries and organisations to learn from one another in regard to HSPA even when the context differs. This is possible if an approach of structured learning, as used in this study with the normative framework, is followed. The Uganda DLT was an innovation in the early days, as a form of subnational HSPA in a LIC. However, a number of things have changed since its launch in 2003: the Ugandan health system context, new developments in HSPA and information technology, among others. Other countries that have implemented HSPA frameworks for a long time have also had to adapt and learn. Proposals have been developed for the adjustment of district HSPA in Uganda in this study. It is recommended that the process to take this on board could begin with discussion of these ideas in a wider group of stakeholders in order to get their input. A lot remains unknown about HSPA globally, especially about HSPA in LICs, with the knowledge gap even bigger at the subnational level. A number of areas are proposed here for further research. There is a need to better understand the relationships between different dimensions of the health care system, and between the health care system and other aspects of the health system. A number of approaches and analytical models have been included in the proposals for a Uganda district HSPA framework, most of which have not been tested in HSPA in LICs so far – individually or in combination. It is proposed that the implementation of such approaches should be studied to see how they work out in the Ugandan health system context. Approaches include, among others, dual objectives for a HSPA framework and sharing of district HSPA information at the community level. Analytical models include HCA, social network analysis and performance trend analysis. It should be also interesting to study the evolution of the Uganda health system into – hopefully - a ‘learning organisation’.
- Affiliations
Citations
Tashobya, C. (2016). Developing an appropriate district health system performance assessment framework for Uganda. https://hdl.handle.net/2078.5/183254