Thursday, October 30th, 2014
Increasingly in the health sector, scorecards are being used to visually represent statistical data. They are used at varying levels: to compare countries within a region, to compare sub-national districts within a country, and at a more local facility level. Essentially, scorecards are a way of depicting data attractively and accessibly, but many players have taken them a step further as a tool to (a) support policy making, and/or (b) improve accountability. By communicating data more effectively and transparently, the scorecards themselves and the process undertaken to produce them provide a platform for dialogue and participation around the data. Decisions can be thus taken based on this data, scores monitored, and decision-makers held to account.
MannionDaniels has completed an assessment for FP2020 on scorecards, assessing existing health scorecards and interviewing designers and end-users to gain from their experiences. The MD team consisted of Rolla Khadduri (Technical Team Lead), Clea Knight (Public Health Specialist) and Dr. Poonam Thapa (Associate, Rights Specialist).
Some general positive conclusions regarding scorecards emerged from this assessment:
(+) Scorecards can be an excellent way of presenting a limited amount of data succinctly and convincingly for policy-makers to make resourcing and policy decisions
(+) Having scorecards can, over time, improve the quality of data collected
(+) Scorecards can be a platform for improved data utilisation, if used within an existing platform of data review and decision-making. These could be District Health Management Meetings, or Joint Annual Review processes for example.
(+) The use of scorecards can improve service delivery at facility level if presented at sub-national level with specific facility information such as enablers / gap-analysis
However, there were commonly-held cautions amongst about the following issues:
(*) For scorecards to be used, country-ownership is paramount. Technical support from outside organisations are encouraged, but only when scorecards are asked for and needed by national governments first.
(*) Similarly, indicators should be chosen based on specific country contexts of what are priorities in the country and what data is available
(*) Countries are being asked to collect a multitude of indicators, and use a multiplicity of tools already. This can be counter-productive, so the decision to proceed with developing a scorecard needs to be taken in consideration of other data collection/representation national contexts. Principles of harmonisation and aid effectiveness call for increased integration with existing initiatives so as to reduce the burden on national governments.
(*) Developing, designing and maintaining scorecards takes a great deal of time and effort (up to one year).
For presentation, the key message is: The simpler, the better
(+) Do keep the data simple
(+) Do use a traffic light colour system (red, amber, green) for easy visual representation
(+) Do use info-graphics and icons, instead of text
(-) Don’t use indicators where there are lots of blank values
(-) Don’t try to include too much data on one scorecard
(-) Don’t use complicated graphs that need lengthy explanation
Other summary findings from the assessment include:
High-quality scorecards depict both values and also change in values over time. Having a change in value allows deeper analysis of gaps and bottle-necks, which can instigate action. The process of data collection, and the process of acting upon the data, is more important to data use than the data itself.
Starting with the right approach to scorecards is important from the beginning of the process, emphasising supportive tracking and transparent monitoring as the aims of using scorecards.
Integrating a rights-based approach is key from the start of the process. At a macro-level, disaggregating data by age, marital status, wealth quintile, education, region, ethnicity and urban/rural is useful to visualise needy populations and determine priority setting. At facility-level, including indicators on quality and data from private facilities would increase representation of services that more of the population use.