Quality over quantity: reframing data woes in global health
When LifeNet International starts working with a new partner clinic in Uganda or Burundi, we know that we may soon see the clinics’ profits decreasing. It’s not because they have fewer patients using services, or because their expenses have risen. Rather, the decrease in revenue is due to drastically improved data collection that gives a more accurate depiction of the clinics’ financial story. Revenues appear to be decreasing at LifeNet International’s Uganda clinics. It is not because we helped our clinics to start providing services or medicines for free, or because we are charging them money for our programs. It is because we are teaching them how to keep their financial records more accurately.
LifeNet International provides monthly trainings on medical and management practices to clinics over a period of three or more years, and much of the curriculum focuses on improving record keeping. Our goals are to improve patient satisfaction, improve health outcomes, and improve clinic sustainability.
Within just two months, My Global Health Corps co-fellow, Prize Magezi, has been able to improve record keeping in clinics that have been making the same mistakes for years. One clinic accountant told her “The new tool that you introduced to us has actually opened my eyes to a loophole in our finance management system.”
Many of our donors understand sustainability. They often work in business and finance. As the monitoring and evaluation officer, it is my role to work with senior leadership to figure out how we can show our impact on clinic sustainability. The most obvious way to do this would be to show the change in profit over time. But with unreliable information about how the clinic was performing in the past, the changes we see before and after we start intervening misrepresent what is actually happening to the clinics profits. So what do I do?
This is not a problem that only affects my small non-profit organization, or a problem that only affects financial data. In my previous job, I worked on a project where we were asked to collect baseline data on school attendance in a handful of countries in Sub-Saharan Africa, but it was quickly evident that attendance was inconsistently reported and that daily attendance records often did not match head counts conducted the same day. In another setting, a friend of mine lamented to me that she was hired to do research at a clinic, but felt the data available was too questionable to ethically do any analysis before the state of the data was drastically improved. Data accuracy is a problem that plagues international development projects and evaluations everywhere.
Of course, the importance of data for international development has been widely addressed, and large scale reports about how big data can be used to make an impact have been published by the World Bank, the United Nations Sustainable Development Network, and the World Economic Forum, among others.
Data accuracy issues are not something that have been ignored either. At a higher scale, the Center for Global Development hosted an event last spring titled “Statistical Tragedy in Africa? Evaluating the Data Base for African Economic Development.” The event focused on the importance of data for measuring economic development and the continued accuracy limitations of the data that was available.
We know the great importance of data, and acknowledge the data accuracy issues that exist at every level of international development, from small projects to national data. Now, it is time for us to give improvements in data accuracy more time in the limelight. I believe we should start measuring data accuracy as an outcome. Right now, starting with poor data is often something that we see as a problem. We hide it in our limitations section as something we are trying to mitigate or overcome. We need to shift the conversation so that starting with bad data and improving it is in itself an accomplishment that is to be rewarded.