Diabetes And Big Data

Could data hold the answer to treating diabetes?


Diabetes is a disease that affects millions of people worldwide. In the US alone, 29 million people or 9.3% of the total population suffer from the disease.

It can bring about premature death, limb amputations and heart attacks, and is more prevalent amongst non-caucasians throughout the US population. According to the CDC, ‘Non-Hispanic black, Hispanic, and American Indian/Alaska Native adults are about twice as likely to have diagnosed diabetes as non-Hispanic white adults’.

The treatment of diabetes generally consists of injections and frequent blood sugar checks, but more needs to be done in both preventative and clinical treatment.

Big Data may have the answer.

In the UK Outcomes Based Healthcare and Big Data Partnership have received funding to develop devices and data collection technology to help build a database to help with future monitoring and treatment for diabetics in the UK.

The idea behind the project is that it will not only collect data for better treatment, but with time, will even have predictive capabilities. This will mean that with the measurements it will be possible to establish potential future problems before they take place. It will allow clinicians to be pre-emptive to prevent future health problems or prepare adequately for a particular scenario.

It will not only use health data taken during regular readings. It will also include non-health data to supply a more complete picture of healthcare implications and the lifestyle choices that may accompany them.

The issue that many have with this kind of system is that maintaining anonymity of patients is often difficult. Simply by giving a postcode and a medical condition, it is highly likely that it would be possible to identify the individual.

As we have had suggested on the channels already, it would be possible to anonymize the data simply by analyzing it first before being used in modelling or otherwise. Case studies would need to be created from several different data points rather than just one individual and all identifiable data would need to be hidden.

It is the kind of data that many companies would want, mainly health insurance companies who would find this kind of information useful in amending their premiums.

Whether this particular programme will work in the long term where others have failed, is yet to be seen. However, if it does it could have a major impact on those suffering from this disease. 


Read next:

Maximizing the Organization: Turning Data Into a Competitive Advantage