Using burden of disease information for health planning in developing countries: the experience from Uganda

https://doi.org/10.1016/S0277-9536(02)00246-0Get rights and content

Abstract

Given the growing interest in both the use of evidence in planning and in using the burden of disease measure (BOD) and cost-effectiveness analysis, we explored health planners’ perception of the usefulness of the BOD in priority setting and planning in developing countries, using Uganda as an example.

An exploratory qualitative approach involving in-depth interviews with key policy makers in health at district and national levels was employed. Interviews were supplemented with a review of relevant documents. Analysis involved identification of key concepts from the interviews. Concepts were grouped into categories, namely, the appeal of quantitative data, data limitations, opaque methodology, planning as a political process and opportunity costs. These form the basis of this article.

We found that the BOD study results have been used as the basis for the national health policy and in defining the contents of the national essential health care package. The quantification and ranking of disease burden is appreciated by politicians and used for advocacy, resource mobilization and re-allocation. The results have also provided information for priority setting and strategic planning. Limitations to its use included poor understanding of the methodology, poor quality of data in-puts, low involvement of stakeholders, inability of the methodology to capture key non-economic issues, and the costs of carrying out the study. There is commitment, by Ugandan planners to using evidence in priority setting. Since this was an exploratory study, there is a need for more studies in developing countries to document their experiences with the use of evidence, and specifically, the BOD approach in planning and priority setting. Such information would contribute to further synthesis of the approach.

Introduction

There is growing interest in the use of evidence in planning and priority setting world-wide (Niessen, Grijseels, & Rutten, 2000). However, priority setting in developing countries is difficult and full of uncertainties, due to lack of dependable evidence, analytical methods for identifying priority options, and coherent processes for decision making (Bryant, 2000).

The burden of disease (BOD) approach, whose results were first published in 1993, has been proposed for use in health planning and priority setting (Murray & Lopez, 1996a; World Bank, 1993). Disability adjusted life years (DALYs), a measure of BOD, incorporates estimates of morbidity and mortality by cause, incidence, average age of onset, duration, degree of disability, and time lost due to premature mortality. Social preferences about the value of future health and the value of a healthy year of life lived at different ages are also incorporated (Murray & Lopez, 1997).

Leading causes of DALYs lost can be used to identify priorities for research. Since it can also be used in cost-effectiveness analysis, DALYs are of practical use to policy makers in identifying priority interventions (Jamison, Saxenian, & Bergevin, 1995; Murray, 1996; Murray & Lopez, 1996b).

According to Jamison (1993) and Lozano (1997), additional uses of BOD results include providing criteria for time allocation in training, pointing out the information gaps in health information systems, identifying otherwise neglected health problems, and enhancing informed debates on the social values that influence resource allocation in health. The BOD methodology is also thought to force maximum use of available information, favour systematization of information for some disease conditions, and to allow for participation of numerous groups of researchers and decision makers (Jamison, 1993; Murray & Lopez, 1997).

Uganda participated in the East Africa region BOD study and cost-effectiveness analysis carried out in 1994. This was done to transfer to the country teams familiarity with all the requirements, assumptions, and the process used in this approach. It was envisaged that as a consequence, with some additional support, countries would be assisted to develop an ability to reproduce this analysis on their own and to use it in planning (Hutchinson, 1996; Kamugisha, Katumba, Makumbi, Mugarura, & Albright, 1994). Participants were mainly ministry of health officials.

Due to lack of good morbidity data, discounted life years lost (YLL)—without age or disability weighting—were used in the cost-effectiveness analysis, instead of DALYs. Three types of intervention were considered for each cause of YLLs lost: preventive, community, and curative. The most cost-effective types of intervention for the identified leading causes of YLLs were selected to comprise the national essential health care package (Hutchinson, 1996) (Box 1).

According to the 1999 health policy,

…The minimum package will comprise interventions that address the major causes of the BOD and shall be the cardinal reference in determining the allocation of public funds and other essential inputs. Government will allocate the greater proportion of its budget to the package in such a manner that health spending gradually matches the magnitude of priorities within the BOD… (MOH, 1999a).

Uganda also carried out a pilot BOD study in 13 districts in 1996 in order to capture epidemiological differences between districts and to train district planners in the methodology. District results were similar to the national BOD results, but with slight differences in the ranking (MOH, 1996). However, each participating district used its own local information to estimate the cost values for the cost-effectiveness analysis.

The BOD and cost-effectiveness approach seem to fulfil the requirements for effective priority setting; namely, provision of information about costs and outcomes, and clarity of objectives (Klein & Williams, 2000). While there has been an increasing acceptance of the approach internationally (Niessen et al., 2000), there have also been reservations. These have been in the areas of: the quality of the data used, especially in the case of developing countries (Coleman, 1998; Cooper, Osotimehin, Kaufman, & Forrester, 1998); the values incorporated in age—and disability—weighting (Arnesen & Nord, 1999; Paalman, Bekedam, Hawken, & Nyheim, 1998); and the lack of contextualization of health problems (Sayers & Fliedner, 1997; Ustun et al., 1999).

Given the theoretical and ethical concerns that have already been expressed about DALYs (Anand & Hanson, 1995; Barker & Green, 1996; Williams, 1996), and given the relative novelty of the approach, there is a need to assess its use for further development of the methodology (Murray, 1996). The purpose of the current study was to explore health planners’ perception of the usefulness of the BOD measure in priority setting and planning at national and district levels in Uganda, and to draw conclusions for wider application.

Section snippets

Materials and methods

A qualitative exploratory approach was employed. This involved in-depth key informant interviews and document review. Informants included policy makers and health planners at national, district, sub-county, and village levels. All had either participated in the BOD studies or were involved in health planning at their level (see Table 1, Table 2).

Respondents at national level identified the study district, Mukono. The district was among the first to be decentralized, and also participated in the

Results

We found that the BOD information has been used, to some extent, in health planning at district and national levels.

Discussion

A qualitative approach was employed in this investigation because it is appropriate for exploratory studies and captures a social world of “lived experience” that facilitates deeper understanding of phenomena (Denzin & Lincoln, 1994; Kvåle, 1996). Reliability in the present study was catered for by triangulation of sources of data. Triangulation in addition, adds rigor, breadth and depth to an investigation (Flick, 1992, Kapiriri & Norheim, 2002; Robson, 1997). Personal experience, given the

Conclusions and recommendations

Authorities in Uganda have attempted to use BOD data (excluding morbidity, i.e. YLLs) to form the basis for national health policy and the essential health care package. It has also been used to guide, according to the national policy, resource reallocation within the health sector. Additional areas of application include strategic planning, identification of health priorities and cost-effective interventions, and the revitalization of policy makers’ interest in evidence-based policymaking.

Acknowledgements

We thank our respondents from Mukono District and the Ministry of Health, Uganda. We also thank Dr. Elizabeth Ekochu for assisting in the data collection, and the following people for their earlier comments on this manuscript: Professor Gunnar Kvåle, Merete Underland, Trude Margrete Arnesen and the anonymous reviewers.

References (43)

  • D Chandramohan et al.

    Verbal autopsies for adult deathsIssues in their development and validation

    International Journal of Epidemiology

    (1994)
  • A.J Coale et al.

    Models of mortality and age composition

  • R Coleman

    Disease burden in Sub-Saharan Africa

    Lancet

    (1998)
  • N.K Denzin et al.

    Entering the field of qualitative research

  • U Flick

    Triangulation revisitedStrategy of validation or alternative?

    Journal for the Theory of Social Behaviour

    (1992)
  • L.W Green et al.

    Health promotion planning an educational and environmental approach

    (1991)
  • Hutchinson, P. (1996). Assessing the results of the burden of disease/cost-effectiveness analysis in the 13 pilot...
  • D.T Jamison

    Disease controls priorities in developing countriesAn overview

  • D.T Jamison et al.

    Investing in health wiselyThe role of needs-based technology assessment

    International Journal of Technology Assessment in Health Care

    (1995)
  • Jeppson, A. (1999). Application of the burden of disease and cost-effectiveness analysis as an instrument for district...
  • A Jeppson

    Financial priorities under decentralization in Uganda

    Health Policy and Planning

    (2001)
  • Cited by (66)

    • How have systematic priority setting approaches influenced policy making? A synthesis of the current literature

      2017, Health Policy
      Citation Excerpt :

      Kapiriri et al. reported findings from a Uganda study where national and district level decision makers perceived the approach favourably by virtue of its being systematic and evidence based. Furthermore, the numerical ranking of the priorities was found to be handy when advocating for more resources from politicians [95]. This paper also reports that the BOD/CEA supported decision makers to focus on the key health problems, and to maximize the use of their resources.

    • International Perspectives on Resource Allocation

      2016, International Encyclopedia of Public Health
    • How effective has the essential health package been in improving priority setting in low income countries?

      2013, Social Science and Medicine
      Citation Excerpt :

      Lastly, the EHP approach, by focusing on technical methods and not the process, assumes that PS is technical. This runs contrary to the evidence that PS is a political process (Jamison, 2006; Kapiriri et al., 2003). Despite these criticisms, the EHP has continued to attract interest from national and international stakeholders, who have supported its use in several LICs (Birn et al., 2009; WHO, 2008; Xingzhu, 2003).

    View all citing articles on Scopus
    View full text