Currently showing: Funding longer lives > Health/medicine

14 May 13 14:15

It has been suggested that transparency and patient data sharing in medical research can improve patient care leading to an increase in the number of lives saved and to novel and more effective treatments.

The difficulties arising from such a process though are not negligible. Researchers find it difficult to access medical information due to a number of factors including the large number of organizations involved in approving access, a lack of clarity around the law for accessing data and the complexity involved for using patient records.

A few months ago, The Guardian together with business analytics software SAS arranged a roundtable discussion inviting various medical professionals to discuss medical data sharing in NHS. Various views were expressed such as "remarkably difficult to share data in NHS", "people feel they are not allowed to share data", "in the drive to protect information we have gone too far the other way and created an obsession about not telling anyone anything", "patient confidentiality does not trump patient safety". Technology was another barrier which was raised by the meeting contributors.

The roundtable agreed that the large number of databases available in hospitals make the sharing of information with other systems a very challenging task and they all suggested that introducing standards for information sharing is vital.

In addition to that, lots of time and money is spent collecting irrelevant data as dataset is added to dataset thus leading to low quality information in the end.

A report from British Heart Foundation (BHF) calls Government to improve access to patient records and researchers. The attached video nicely presents the medical data sharing issue from both the researcher's and patient's point of view.

Could sharing your medical records help save lives?

A damning report from the British Heart Foundation (BHF) has revealed how UK red tape is strangling medical research that could save lives and is calling on ...

Category: Funding longer lives: Health/medicine


Brian Ivanovic - 15 May 2013, 3:53 p.m.

This seems like a very important question to get guidance on. As the technology to monitor our activities and our biometric information becomes less expensive the volume of personal information that will be generated will exponentially increase. Where is this information ending up and how will it be used? This data will give a clearer view of our day to day life vs. what we might like others to "think" we do. How will privacy laws treat this information that is often collected and summarized on the web? Even in a situation where collection of the information is unobtrusive will people want this information shared and used to assess their health habits? For those who do not want the infomaton shared will that decision be interpreted by others in a negative way?

Aspasia Angelakopoulou - 20 May 2013, 11:49 a.m.

I totally agree that in the future individually available data will dramatically increase. From a researcher's point of view that may look very exciting but challenging at the same time as data quality issues will also increase as well. Not sure how easy it will be to obtain high data quality standards when this will be changing so fast.

Regarding the individual willingness to share their personal data, I believe that as long as people better understand where and why their data will be used for they will be more willing to agree to its sharing (as the BHF video also suggests). Of course, there will always be this group of people who will be reluctant in having their data shared but I don't think that such a group should in any way be criticized.

Matt Singleton - 20 May 2013, 4:22 p.m.

A question for research experts such as yourselves ... now we have the computer capabilities to store and analyse enormous amounts of data, could there come a point when we have too much? As someone who has spent a lot of their career being confronted with data issues - mismatching databases, incomplete data, badly recorded data - there are a lot of challenges with the data we have already. But could we reach a point of saturation and what might that look like?

Brian Ivanovic - 22 May 2013, 3:41 p.m.

Is it too much information or is it that we have not yet figured out how to use this information to improve health? I don't know if our health improves that much moving from a weekly blood pressure reading over long periods of time to one collected daily, hourly or minute by minute. These will be questions that need to be researched to determine what is the "right" amount of information to collect to drive positive changes in health at the population level. We may have more and more enabling technologies to monitor things real time but that does not mean that everyone will benefit from that increased level of monitoring.

The other aspect of this discussion is how these exponentially growing volumes of information can be effectively utilized to improve health. In a paper published in 2012 by Crosson in the Annals of Family Medicine the quality of diabetes care was compared between practices using electronic health records (EHR) and those that were not. EHR practices are the ones that would first have the beginnings of the needed infrastructure to exploit the increasing amounts of data that could become available at the patient level. At this early stage of EHR rollout the Crosson et al found that the non-EHR practices delivered better care than the EHR ones!! I think this points out that acess to data in an of itself is not sufficient to drive postivie changes in health. You need to have developed systems to exploit this data in a way that will enable better health outcomes and train health care providers to use patient data effectively.

Aspasia Angelakopoulou - 22 May 2013, 4:04 p.m.

I agree that the biggest problem with Big data is how analysts can actually deal with these huge of volumes of data in order to address health related questions in the most efficient way. When we will have so much data in hand it will be difficult to use on-hand management tools or traditional data processing applications. There might be a point when even new software may need to be developed to deal with this exponential data increase. I guess most of the challenges will be related to storage, search, sharing, transfer, analysis and visualisation.

We are already dealing with so many problems when we are trying to analyze a dataset of thousands of people...Don't even want to imagine what problems I will have then!!!!

If you would like to leave a comment, please, log in.