Few discredit the many advancements that could be made when centralizing all known medical records. However many right remark that there is an inherent risk involved. Moreover this data that they would willingly divulge has enormous value and can be used for both good and bad. The misuse of medical data deserves its own post so for now it suffices to say that in order to reduce the likelihood of such an occurrence steps must be taken. The anonymous records should not be able to be traced back to their subject.
Is it even possible to make medical records unspecific enough not to identify an individual but so that they contain the necessary data to prove valuable to researchers. Clearly a cardboard box into which you can insert a form or two will not suffice. An on-line database with the relevant info? It’s only a matter of time before something gains information they are not privy to. This is the main problem with centralization of data. When it is combined with centralization(s) of power, there is no telling how much harm will be done. However noble the idea is to provide doctors worldwide with the facts necessary to cure their patients, an small amount of people will have much to gain by procuring this confidential info and using it for their themselves.
So that rules out entries that contain name and contact information. What about date of birth? Okay that’s probably too much of a give away. Occupation? Age? DNA sequence?
This is big data we’re talking about. The kind of big data that will be able to identify which doctor treated a patient based on a phrase he likes to use or the time of day that he uploaded the form. Try as we may to conceal personal and confidential things, big data will find them. It’s just a matter of time. Which begs the question of course; is it even possible to store any data at all without compromising the data that was of interest in the first place?
No one seems to have a decent answer here. Either they put their faith in complex encryption mechanisms which in turn hands over the keys to people who keep and guard the data. Or they simply do not believe data can be centralized at all. I kind of like the idea of using hybrid solutions. Solutions that do not keep one long file containing a complete medical history. Rather they could contain a group of more or less related data and nothing more. True, patterns would still be found, but it would be harder, much harder to confirm that they belong to person X. Especially if there is no guarantee that there is a file you are looking for. Do you have an out-of-the-box idea that could revolutionize the storage and standardization of medical data?
Andrew
http://sciencenordic.com/your-identity-not-safe-anonymous-data
Parallel Computing - Harder Better Faster Stronger 
Yes I agree that when the information put in the global such as Internet it will be not 100% secured. The most importance it is, the more attached to that information.
The only thing that the programmer or system administrator have to do is try to protect their system as they can. In my own I think only the information that store locally can be secured. Or if we want to make it secured we have to store only the reference to actual data. I mean we can store only reference number as Id in global and refer to the actual data in locally. And what do you think the way how to make your system secured?
LikeLike
The main objective of centralization is to provide researchers and doctors with this data. This means access to data that is not local. I will admit I cannot see a clear way forward. As you have clearly said it will never be completely secure.
LikeLike