One of the main goals for a charitable organization is to help people around the globe to improve their lives. Charities invest in solutions to alleviate burdens on humanity such as AIDS, malaria, polio, tuberculosis, infectious diseases, hunger, and poverty.
Strategy decisions at charitable organizations are remarkably similar to commercial companies. Both businesses and charities start with a few basic questions to establish a strategy:
1. Where are we today?
2. Where do we want to be?
3. What do we need to do to get there?
4. How are we going to measure the results?
Let’s look at some data that help decision makers at charitable organizations answer these questions to profile investment opportunities and support the organization’s strategy.
The necessary data can be obtained from multitude of data providers, organizations that publish original survey data, or that aggregate, extrapolate and repackage data from other providers; many of these data sets are available on the Internet as a free service. For this exercise, a combination of data from the following providers gives us a good initial snapshot of health-related statistics:
To simplify access to their data, data providers may implement industry standard programmable interfaces (APIs) such as the Open Data API. Most data providers allow downloading data as Excel files, or will present the data in unstructured form. Bringing the data together from multiple providers and organizing it for analysis often is a non-trivial effort.
To prepare data for analysis, a charity may start with a few key data providers, combining data that covers one particular subject in depth with other data that covers a wider range of subjects. This combination would allow examining relationships between burdens (such as whether there is a relationship between health issues and poverty) and evaluate in-depth details of a burden (such as age and gender effects on pneumonia).
Four common data elements supplied by the majority of providers allow us to evaluate data across different providers – these are Indicator Name, Year, Country, and a numeric Indicator Value. Two additional metadata elements help users understand the meaning of the data (Indicator Definition) and how indicator’s values were derived (Methodology). This latter attribute is particularly important, because depending on the methodology, two indicators with the same name may have values that are significantly different – for example, there is a 50%+ difference between WHO’s and IHME’s estimates for malaria deaths in South East Asia (blogs.cgdev.org/globalhealth/2012/02/malariaestimate-sausages-by-who-and-ihme).
Once we combine the data from different providers, we can begin to examine the data and work through the questions posted in the beginning of this article.
It is important to note that without the insights provided by accurate data we run the risk of being deceived by our perceptions. For example, someone living in a developed country generally will not consider pneumonia to be a significant disease, given that the pneumonia vaccine has been available since 1977 and, if detected early, it can be treated with inexpensive antibiotics. Further, pneumonia does not generate as many headlines in the United States as do AIDS or malaria.
The data for pneumonia paints a very different picture, however, when considering the issue globally. According to WHO’s data, pneumonia is the leading cause of death in children worldwide, responsible for 18% of childhood deaths worldwide – more than AIDS, malaria and tuberculosis combined. In 2010, pneumonia killed an estimated 1.37 million children under the age of five, taking the life of a child every 23 seconds.
Similarly, those living in developed country may not consider malnutrition to be a serious problem, but the data shows that it is responsible for one-third of all child mortality and, according to the Lancet, malnutrition in the first two years is irreversible, resulting in poor health, lower education achievements, and even early death (www.concernusa.org/media/pdf/2011/06/1000_DAYS.pdf).