Abstract
The study of decentralization in Fiji shows that increasing capacities is not necessarily related to increasing decision space of local officials, which is in contrast with earlier studies in Pakistan. Future studies should address the relationship among decision space, capacities, and health system performance.
Keywords: Decentralization of Health Systems, Decision Space, Capacities of Health Officials
The study of decentralization of health services in Fiji1 with an emphasis on the current initiatives in decentralization in one pilot region, Suva, is an excellent example of the use of my increasingly popular framework on the “decision space” approach to analyzing decentralization.2 This approach identifies a series of functions (financing, service delivery, human resources, governance) for which local authorities are granted some decision-making authority that can be “narrow” (little local choice), “moderate” (a range of choice but limited by central rules) or “wide” (little constraint on local choice). The Fiji study found that the decentralization process was relatively limited granting little choice over a few functions. In part because of the limited “decision space,” they introduced an interesting addition to my model with “no choice” or fully centralized as a new category.
The decision space approach has been used by Harvard T.H. Chan School of Public Health researchers to study decentralization in a number of countries including Colombia, Chile, Nicaragua, Morocco, India, Pakistan, Mongolia, Liberia, and South Africa.3 It has proven to be a useful framework to several other researchers, including most recently in Karnataka, India.4 It is interesting to note that the approach does not seem to depend on the size of the country and that it can be applied to large states in India as well as to a small island country like Fiji.
My approach also examines the relationship between “decision space” and the local capacities to exercise better choices. In studies in Pakistan with my colleague Andrew Mitchell, we found that those districts with more capacities tended to make more decisions within their formal decision space than did districts with less capacity.5 The Fiji study took this approach to identify capacity building as an issue related to decision space and unlike the situation in Pakistan they found that the government was increasing capacities but still restricting decision space. They found that although increased capacity was created at health centers in Suva and increased work load was achieved, very little “decision space” over key functions like human resources, budgets and financing accompanied these changes. The article shows how important it is to know how much authority and choice is allowed at local administrative levels and to consider the relationship between capacities and choices. The Fiji study suggests that the relationship found in Pakistan may not be universal and raises the question of whether just improving capacities without significantly increasing “decision space” might or might not result in improved performance.
The “decision space” approach was developed in order to provide more detailed guidance for policy-making about decentralization. Earlier literature tended to see decentralization as a single concept and did not distinguish the wide range of different ways decentralization can be implemented. It also tended to be ideological statements about the potential advantages or disadvantages of decentralization without demonstrating clear relationships between decentralization and actual performance. The decision space approach attempts to evaluate the effectiveness of different decision space configurations and to provide recommendations to design decentralization processes that will result in better health system performance.
In the Pakistan study, we identified some process indicators of health system performance and found some relationships suggesting that the combination of higher capacity and more use of decision space led to some improvements in these process indicators.6 The Fiji study does not examine the relationship between decision space and performance but that should be the subject of future research.
Ethical issues
Not applicable.
Competing interests
Author declares that he has no competing interests.
Author’s contribution
TJB is the single author of the paper.
Citation: Bossert TJ. Decision space and capacities in the decentralization of health services in Fiji: Comment on "Decentralisation of health services in Fiji: a decision space analysis." Int J Health Policy Manag. 2016;5(7):443–444. doi:10.15171/ijhpm.2016.51
References
- 1.Mohammed J, North N, Ashton T. Decentralisation of health services in Fiji: a decision space analysis. Int J Health Policy Manag. 2016;5(3):173–181. doi: 10.15171/ijhpm.2015.199. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 2.Bossert TJ. Analyzing the decentralization of health systems in developing countries: decision space, innovation and performance. Soc Sci Med. 1998;47(10):1513–1527. doi: 10.1016/S0277-9536(98)00234-2. [DOI] [PubMed] [Google Scholar]
- 3. Bossert TJ. Empirical studies of an approach to decentralization: “decision space” in decentralized health systems. In: Faguet JP, Pöschl C, eds. Is Decentralization Good for Development: Perspectives from Academics and Policy Makers. London: Oxford University Press; 2015:277-298.
- 4.Seshadri SR, Parab S, Kotte S, Latha N, Subbiah K. Decentralization and decision space in the health sector: a case study from Karnataka. Health Policy Plan. 2016;31(2):171–181. doi: 10.1093/heapol/czv034. [DOI] [PubMed] [Google Scholar]
- 5.Bossert TJ, Mitchell A. Health sector decentralization and local decision-making: decision space, institutional capacities and accountability in Pakistan. Soc Sci Med. 2011;72:39–48. doi: 10.1016/j.socscimed.2010.10.019. [DOI] [PubMed] [Google Scholar]
- 6.Bossert T, Mitchell A, Janjua M. Improving health system performance in a decentralized health system: capacity building in Pakistan. Health Systems and Reform. 2015;1(4):276–284. doi: 10.1080/23288604.2015.1056330. [DOI] [PubMed] [Google Scholar]