Assessing the costs and cost-effectiveness of rapidscale up for the maternal, neonatal, and child health: - PDF

Please download to get full document.

View again

of 22
All materials on our website are shared by users. If you have any questions about copyright issues, please report us to resolve them. We are always happy to assist you.
Information Report

Food & Beverages


Views: 7 | Pages: 22

Extension: PDF | Download: 0

Related documents
Assessing the costs and cost-effectiveness of rapidscale up for the maternal, neonatal, and child health: The economic component of the impact evaluation strategy for the Catalytic Initiative Benjamin
Assessing the costs and cost-effectiveness of rapidscale up for the maternal, neonatal, and child health: The economic component of the impact evaluation strategy for the Catalytic Initiative Benjamin Johns, Marjorie Opumi-Akuamoa, Damian Walker The Institute for International Programs Department of International Health The Johns Hopkins University Bloomberg School of Public Health Draft Version 2 Not for citation or public release 1 Table of Contents Figures, tables, and boxes... 2 List of Abbreviations... 3 Acknowledgements Introduction Specific objectives of the cost and economic component of the impact evaluation Methods A. Conceptual framework for assessing the cost of rapid scale-up B. Framing the analysis C. Classification of costs D. Study design E. Allocation methods F. Uncertainty analysis G. Data sources, collection methods, and data collection tools Data analysis and outcomes Limitations List of recommendations References Glossary of terms Appendix 1: Estimating incremental costs Appendix 2: Example of how these methods are being applied in Malawi to estimate cost and cost-effectiveness Appendix 3: Example of tables for reporting results Figures, tables, and boxes Figure 1: The step-wise approach to impact evaluation... 7 Figure 2: Conceptual framework for assessing the cost of rapid scale-up Box 1: Data envelopment analysis & stochastic frontier analysis Table 1: Types of data potentially collected by site and method Table 2: Types process, intermediate, and outcome indicators and data needed Table 3: Summary of main recommendations List of Abbreviations BMGF CI CCM CHAs CMH DALYs DEA DHAs GDP GNI HFS IC ICER IIP IMCI JHU LiST MCE MDGs MNCH NGO OOP RMM SFA TC UN UNFPA WHO WHO-CHOICE Bill and Melinda Gates Foundation The Catalytic Initiative to Save One Million Lives Community Case Management Community Health Agents Commission on Macroeconomics and Health Disability Adjusted Life Years Data Envelopment Analysis District Health Accounts Gross Domestic Product Gross National Income Health Facility Survey Incremental Costs Incremental Cost-Effectiveness Ratio The Institute for International Programs Integrated Management of Childhood Illness The Johns Hopkins University Lives Saved Tool Multi-Country Evaluation Millennium Development Goals Maternal, Neonatal, and Child Health Non-Governmental Organization Out-Of-Pocket payments Rapid Mortality Monitoring Stochastic Frontier Analysis Total Costs United Nations United Nations Population Fund World Health Organization WHO s Choosing Interventions that are Cost-Effective programme 3 Acknowledgements The authors would like to acknowledge the expert feedback provided to earlier versions of this document by David Bishai (The Johns Hopkins University), Lesong Conteh (The London School of Tropical Health and Hygiene), Dan Chisholm (WHO), Dean Jamison (University of Washington), David Newlands (Aberdeen University), and Virginia Wiseman (The London School of Tropical Health and Hygiene). Further comments were provided by members of the Institute for International Programs and their research partners. While these comments have improved the document, all responsibility for the content and any errors remain the responsibility of the authors. 4 1. Introduction Most low-income countries are making slow progress in addressing child and maternal mortality too slow to achieve Millennium Development Goals (MDGs) 4 and 5 by Governments, donors and development agencies are responding to the situation by redoubling efforts to stimulate and support child survival activities in countries, particularly in Africa. The Catalytic Initiative to Save a Million Lives (CI) is a central part of the movement to accelerate progress towards achieving MDGs 4 and 5. In late 2006, the Bill and Melinda Gates Foundation (BMGF) approved a grant to the World Health Organization to work with other United Nations (UN) agencies to provide proof of concept that accelerating coverage with interventions of proven efficacy can be achieved rapidly and will lead to reductions of at least 25% in under-five mortality. Although WHO is the grantee, each UN agency is responsible as lead partner for one country. WHO is the lead partner in Malawi, UNFPA in Mozambique, and UNICEF is the focal agency for the project in Burkina Faso, leading and coordinating support provided by other UN agencies. This grant to WHO is considered part of the larger Catalytic Initiative. This project includes an independent prospective evaluation led by the Institute for International Programs (IIP) at The Johns Hopkins University (IIP-JHU) in collaboration with research institutions in each collaborating country. The overall objective of the impact evaluation is to serve as a proof of concept that it is possible to rapidly intensify effective interventions targeting maternal, neonatal and under-five mortality. Other objectives of the impact evaluations include: 1. Measure the additional number of children saved through the accelerated approach; 2. Measure the cost per child life saved in accelerated approach districts relative to the other districts of the country; 3. Supply the MOH and partners with information on the effectiveness of the implementation of the project. Objective two utilises the methods and tools of economic evaluation, which are rooted in the fundamental problem by which economists characterize decision-making: making choices between alternatives in the context of scarce resources. Within the scope of national and international public health, these choices are often framed in the debate as to which interventions should have priority. Economic evaluation attempts to identify ways in which scarce resources can be employed efficiently. Efficiency has two principal meanings in this context. First, there is technical (or operational) efficiency, which concentrates on maximizing the achievement of a given objective within a given budget doings this right. Second, there is allocative efficiency, which is a broader concept as it focuses on choosing the optimal mix of interventions for a given level of expenditure optimal in the sense that 5 they maximize health gains doing the right things. When allocative efficiency is being analysed, as in the case of cost-effectiveness, technical efficiency is often adjusted to be equalized in each of the comparison branches [1]. This document outlines generic recommendations for use in the economic evaluation of rapid-scale up of MNCH programmes as a component of the larger impact evaluation. It is intended to be a guide for estimating costs across country studies to ensure consistency in data collection, analysis, and reporting; it is not anticipated that all countries will desire to answer all the questions potentially addressed by economic evaluations, or that all of the proposed data collection activities will be undertaken in all settings. Rather, this document serves to propose possible activities and analyses; decisions as to which components are appropriate should be made in each country during the evaluation design process (see Appendix 2 for an example). While the extent to which the results from economic evaluations can be generalized is questionable [2], the collection of data in a standardized way is necessary if any comparisons across settings are to be undertaken. Thus, as with other components of the impact evaluation, it is recommended that countryspecific evaluations adhere to the recommendations set forth in this document to the extent possible. The IIP evaluation approach is based on a step-wise evaluation philosophy (see figure 1), with specific questions adapted to the country-level situation [3,4]. To answer the questions posed within this approach, a number of cotemporaneous studies will be undertaken, which may include: 1. Careful documentation of programme activities in both control and rapid scale-up districts (step 1 through 3); 2. Measuring the quality of case management for common childhood illnesses provided at health centres and/or by CHAs (step 2); 3. Prospective evaluation of the impact of the Rapid Scale-Up (steps 3 through 5); 4. Rapid mortality monitoring (RMM) (steps 3 through 5); 5. Collection of vital events by community members such as community health agents (CHAs) (steps 3 through 5); 6. Costs and cost-effectiveness evaluations (step 6). Additionally, and of note to readers of this document, a separate component on assessing and evaluating the equity implications of the whole programme are included in the IIP evaluation. It is anticipated that the economic component will contribute to and supplement the equity analyses. 6 Figure 1: The step-wise approach to impact evaluation Implications for costeffectiveness Is program good value for money? Is it sustainable? Is there an impact on health and nutrition? Have adequate levels of effective coverage been reached in the population? Are these services being used by the population? Measuring effectiveness Utilization; quality adjustments Demand side costs Are quality services being provided? at health facility level? at community level? Are the interventions and plans for delivery technically sound and appropriate for the epidemiological and health system context? Supply-side costs Identify activities included in the programme Source: [3,4] Details of each of these components of the impact evaluation are available elsewhere [ref or refs]. While the last step directly incorporates the output of an economic evaluation in the overall evaluation philosophy, the step-wise approach can also be adapted to shape the questions asked and data collected within an economic evaluation. For example, at the first step, collecting data on the budgetary allocations for planned activities can help to identify what activities will be included in the rapid scale-up for which cost data need to be calculated and also to establish (at a preliminary level) the adequacy of the budget in relation to the planned activities. The second step reflects the need to collect cost of activities from a supply-side perspective. Step 2 represents the minimal step required for an economic evaluation, but a full economic evaluation should assess costs from the perspective of society (including both supply- and demand-side perspectives) and not just the health care system or supply-side [5,6]. Thus, costs to patients are assessed in conjunction with step 3. Data from steps 2 and 3 can be used to answer questions of technical efficiency e.g., are services being delivered for a minimal amount of inputs? (or, are the maximum number of services being delivered for a given set of inputs?) [7,8]. Data from steps 4 and 5, combined with the costs from steps 2 and 3, then answers the questions posed in step 6, which assesses questions of allocative efficiency are the proposed activities (e.g., those encompassed under the rapid-scale up) providing good value for money as compared to not doing these activities or compared to other possible activities? This step necessitates the use of either process indicators (at a minimum) such as utilization rates [Step 4], or (more ideally) health outcome measures such as deaths 7 averted, life years saved, disability adjusted life years (DALYs), or the like [Step 5]. A further step, needed for a complete economic evaluation, is to assess the non-health impact of health interventions, which may include the impact of improved health on people s productivity, educational attainment, etc. [9] [other refs]. This document is intended first for economist working in the IIP evaluation, other staff at IIP working on the evaluation, and their partners. It is secondarily intended for distribution among other economist conducting economic evaluation of large-scale health programmes. Finally, it is intended as background reading for people working in public health in developing countries seeking to gain an understanding of the principals and methods used in the economic evaluation of large scale programmes. As such, it covers a range of suggestions, some standard to most, if not all, economic evaluations and some more specific to the issues related to economic evaluations of large-scale programmes in developing countries. More advanced analysis topics are not covered in detail in this document; it is likely that these issues will be addressed by analysts that have the training to apply the appropriate techniques, but the specifics of the techniques will change based on the data available. 8 2. Specific objectives of the cost and economic component of the impact evaluation The economic component of the larger impact evaluation is intended to assess the cost of the interventions aimed at reducing MNC mortality. This component of the analysis has two specific and interrelated objectives: 1. Determine the cost of the activities funded by the CI. The specific activities employed under the CI will vary by country, and the specific activities for which costs are estimated will be identified in conjunction with the broader evaluation. 2. To estimate the cost and cost-effectiveness of providing the MNCH prevention and care services the CI activities are seeking to influence at both the facility- and community-level. Understanding the cost-effectiveness necessitates the use of effectiveness data from the broader evaluation. Other objectives can be added as per country needs. Some secondary analyses that can be conducted with the data that needs to be collected in order to meet these two objectives are suggested in sections 5 and 6 below. 9 3. Methods 3A. Conceptual framework for assessing the cost of rapid scale-up Figure 2 presents a conceptual diagram of the total costs involved in the rapid scale-up of MNCH interventions. Figure 2: Conceptual framework for assessing the cost of rapid scale-up TC 2 Total Costs E D C Incremental costs TC 1 B A Time Start of programme Target coverage / saturation achieved The x-axis in the diagram represents time from before the start of the rapid scale-up programme until target levels of coverage (or mortality reduction) have been achieved (or until the end of the evaluation). The y-axis represents the annualized, total costs of MNCH activities, with higher levels of costs being associated with higher levels of total costs. There are five cost categories represented in the framework: Area A: The rectangular region above the X-axis represents the total amount of resources used for MNC health services and preventive action at the beginning of the observation period. Area B: Represents the change in total costs normally associated with secular trends in most health systems 1. The change in total costs can be attributed to changing population, inflationary pressures, and the introduction of new technologies, among other factors. In 1 Area B could also be thought of as the change in area A over time. 10 this diagram, secular trends are seen as increasing total costs (as is normally, but not universally, the case); this does not conceptually imply that average or marginal costs are necessarily increasing. Nor is there any particular reason to believe that secular trends will affect total costs in a linear way as depicted in the diagram (this is done here for simplicity s sake). Note that secular trends will affect the total costs of the MNCH programme both before and after the introduction rapid scale-up. Area C: Represents the start-up costs associated with the rapid scale-up; these costs are specific for the rapid scale-up programme and thus are incremental to the costs represented in Areas A and B. In economic terms, the start-up costs will be annualized over the lifetime of the programme and thus are shown in the figure as constant over time, although this again may not necessarily be the case. The size of area C to some extent depends on the analytic horizon of the analysis. Area D: Represents the operational programme-level costs associated with the rapid scale-up; these costs are specific for the rapid scale-up programme and thus are incremental to the costs represented in Areas A and B. Again, there is no pre-determined pattern for how these costs will change over time; for convenience they have been depicted here as increasing with the expansion of rapid scale-up activities. Area E: Represents the increased costs due to the increased patient load under the rapid scale-up programme. Again, as rapid scale-up increases coverage, total costs will invariable increase, but there is no a priori reason to believe that unit costs will remain constant, increase, or decrease. Area E must be assessed as part of the economic component of the impact evaluation. It should be noted that any changes in unit costs due to programme scale-up will also, likely, affect the unit costs associated with areas A and B. The costs associated with areas C and D may be estimated using four different methods: 1. Documentation approach: Through careful documentation of activities and correctly identifying which costs are unique to the rapid scale-up and allocation of the costs for activities shared by rapid scale-up and other activities; 2. First difference approach: The costs for C and D may be estimated by calculating programme costs at baseline and subtracting these baseline costs from the total programme costs observed after implementation. This approach ignores secular trends; 3. Aggregate difference approach: Subtracting the programme costs in the control districts from the programme costs in the rapid scale-up districts after rapid scaleup has occurred [1]. This assumes that the baseline costs in the control and rapidscale up areas are the same. 4. Difference-in-difference approach: Methods 2 and 3 can be combined to control for secular trends. In this methods, programme costs are collected before and after rapid scale-up in both the control and implementation districts, and the difference in costs in the control districts is interpreted as the secular trend in programme 11 costs, and taken out of the difference in costs between the two time points in the rapid scale-up districts [1]. It should be noted that the assumption that secular trends are similar in the comparison and rapid scale-up districts is dubious. Even if districts are randomized, the likely small number of districts available for randomization and interference by other, non-ci programmes (an interference which usually cannot be considered random) likely mean that randomization alone may not ensure comparability [10]. Further, randomization (by itself) does not tell us why secular trends are as observed, and thus does not contribute to an understanding of the causes of heterogeneity in secular trends [11]; the generalizeability of cost data depends critically on understanding the particular activities and settings the data reflect and how these are similar to or different from other settings 2. Thus, for this analysis, it is recommended that method 1 be used in combination with one of the other methods (method 4 is the preferred) in order to assess which costs are attributable to rapid scale-up. Similar methods should be used to estimate Area E, but further adjustments may be necessary. Having time series cost data tracked by activity may allow for the calculation of total, average, and incremental cost per recipient in both the scale-up and control areas, since the change in the utilization, population, and the unit costs will be possible. Using the full change (comparing baseline to full implementation) in unit costs in the rapid scale-up areas would assume that all changes in the unit costs of service delivery and the utilization rate are attributable to the rapid scale-up, which ignores the influence of secular trends (See Appendix 1 for more details). Thus, secular trends may be controlled for using comparison areas (i.e., using the aggregate or difference-indifference approaches described above). However, the collection of time series cost data is obvio
We Need Your Support
Thank you for visiting our website and your interest in our free products and services. We are nonprofit website to share and download documents. To the running of this website, we need your help to support us.

Thanks to everyone for your continued support.

No, Thanks

We need your sign to support Project to invent "SMART AND CONTROLLABLE REFLECTIVE BALLOONS" to cover the Sun and Save Our Earth.

More details...

Sign Now!

We are very appreciated for your Prompt Action!