Discover PlayDecide. Download games, prepare, play. GET STARTED

eTRIKS: The Value of Medical Research Data and its Reuse

Choose your language

PlayDecide games may be available in multiple languages

Play the game

Download, prepare, discuss & collect results.

SIGN INRegister

Across Europe health researchers are collecting more and more information about diseases and treatments and how they impact individual patients. Often data is stored in incompatible, inaccessible systems that make it difficult or even impossible to reuse the information to improve patient outcomes. The challenges to using the data to their full value may be both practical and ethical.

Author / translator Paul Peeters

Across Europe health researchers are collecting more and more information about diseases and treatments and how they impact individual patients. Often data is stored in incompatible, inaccessible systems that make it difficult or even impossible to reuse the information to improve patient outcomes. The challenges to using the data to their full value may be both practical and ethical.

Datasets that are in a standardised format that makes combining them with other datasets are more valuable. They are more valuable because they can be reused as part of studies. By combining datasets researchers aim to answer scientific questions which are not possible to answer with smaller individual datasets. Poor adoption of standards, data privacy regulations, organisational policies, and a general lack of appreciation that data has value beyond its original use are all barriers that limit the value society gets from medical research data.

Created 4 March 2016
Last edited 10 October 2018
Topics Ethics, Health, Science

Policy positions

Policy position 1

Research data should only be used in the project that the individual originally consented to.

Policy position 2

Individuals own their research data, trust them to share. New uses of research data should require re-consent.

Policy position 3

In some instances research data can be shared when there is adequate protecion of personal data.

Policy position 4

It should be compulsory to share all research data.

Story cards

Story card - no image

I take data, study it and use it to help develop high quality treatment.

High quality data includes data that I can re-use. This means that the data I have can be integrated with any new datasets, so that new insights can be identified.

I am concerned that data may be too difficult to integrate or that I may have to analyze datasets with missing data.

Translational Researcher
Story card - no image

I don’t trust everything that is being done within medical research.

I would like to see that what’s being done, is done in a transparent way. If this is not the case I may take steps to make sure it happens.

Patient Advocate/NGO
Story card - no image

I want to see my data used well. I want to make sure that what I have consented to is what is happening.

I also want to know that if I don’t benefit from the data, at least others will. I want to have an impact on future treatment.

Story card - no image

Where and how will the money be most effectively used? What diseases and initiatives need the most attention? How can I best improve the success of research? Who do I need to know? What`s the most up to date information?

Lots of advocates for particular causes contact me. I need to be able to the right decisions.

Policy maker
Story card - no image

I see a lot of effort in cleaning databases. I see a lot of value in standards as they will limit the amount of data that is not comparable and minimize the transformational work.

I spend most of my time querying people on incorrect data, “There will always be manual work”.

Data Scientist
Story card - no image

I am concerned about public health and that the company I work for is successful. This means I may compete with other pharmaceutical for the same data.

I would like others to share data and am prepared to share data that is not/no longer commercially sensitive.

Pharmaceutical Researcher


Medical innovation depends upon access to medical research data

Research advances require large datasets that are costly to generate and can take a long time to collect. Most clinical decision-making is still based more on opinion than data. A fair statement would be that medical innovation is dependent upon having large amounts of data available. There is enormous cost and time saving potential in re-using existing data.
The benefit to society is so large that it should be a requirement to share data.

Patients want to be aware of how their data is being used

When someone consents to a study it is for a specific purpose, i.e. to research a specific disease. Re-using the data for other purposes does not respect this requirement. Many existing datasets would lose their value if patients had to be consented for a new type of analysis. 2015 EU regulation stipulates that new users of medical research data have to gain patient permission, unless there is a specific exclusion to the regulation permitted.

Data protection is not to the level it should be

In some industries, e.g. the telecom industry, a high level of encryption of data is used. Medical research data has varying levels of security. Medical research data risk to be compromised. Higher levels of data security could be costly and difficult to implement for legacy datasets. The risks related to data loss and disclosure is often understated. It may be impossible to eliminate risk, so a balance of benefit vs risk is required.

Risk of ‘research parasites’

To generate a dataset a group of researchers invest time and resources to collect data and assure quality. That group of researchers then wants to get value out of their effort and being the first to publish. Editors in the NEJM stated that by making data open there is a risk of having ‘research parasites’ which are researchers who have done nothing to collect the data but then analyze and publish before those working hard to collect the data.

Data hoarding

Data is often controlled by individuals or organizations who may or may not own the data. Data hoarding occurs when access to data blocked by someone in control of the data. They can effectively block access to the data by others limiting who is able to use the data for research purposes. This is often done to protect the investment they made in collecting and processing the data. They want to be first to publish the findings from a set of data.

Lack of use of data standards limits the value of medical research datasets

At best 10% of the world’s research data has been collected and structured using standards. The amount of effort needed to judge the quality of the dataset and the amount effort needed to combine datasets or even re-use them is increased markedly when standards are not used. Often standards are not used because it takes effort to do in the beginning. Researchers are focused on getting a study done and not thinking about a datasets future value.

There is an under-appreciation of the patient perspective on medical research data re-use

Some surveys are revealing that up to 70 -80% of patients in research studies would be willing to have their data shared. Yet regulations and privacy protection measures seem to disregard the unique nature of the medical research data that is being made available by patients because they want to help themselves and others with similar conditions. Patients share data to achieve a greater degree of health and well-being.

Mobile health and personal health monitoring increases the sensitive of medical research data

The implications of the flood of data from new technology will impact on the use and re-use of medical research data. Personal health monitoring provides a rich set of data that is highly personal.

Limiting the use and re-use of this data threatens to stifle the disruptive innovation that is now developing. Yet a major breach of this type of data could also limit the innovation potential of this technology as well.

Lack of understanding of the basics about data

Many patients do not understand the basics about data and how it is used. This raises questions as to how ‘informed’ they are in consenting to the use of their data. This extends down to understanding how their data is protected. There may be misconceptions about the level of risk for the re-use of their data. Many people share sensitive personal data on a daily basis on social media platforms. Medical research data is often less sensitive data.

Data Anonymisation is a myth

Many patients require that their medical research data be anonymised, but is data anonymisation possible? There is a growing belief that data anonymisation is in fact impossible.

What is medical research Data?

Medical data is the information recorded during a medical research study. This can include measurements, personal information, and the results of scientific tests.

Data is typically processed and analysed leading to different levels of data.
Level 1 – Raw, unprocessed data
Level 2 – Processed Data (standardised, curated, normalised)
Level 3 – Modified data through analysis and validated
Level 4 – Integrated datasets

What happens to medical research data?

Data is stored in databases so that it can be processed and analyzed by researchers. Data is analysed using statistical and analytical methods. This can be done on focused or local level however more recent approaches to medical research require the integration of multiple types of data that is best done in a centralized database.

Data sharing

Once carried out on a set of data, the data remains valuable for use in other research studies. Existing data can be useful to test an idea prior to going through the cost and time of collecting new data. This can be useful in deciding which new therapies are worth testing.

We now have the ability to collect and process much more data than before. Sharing of data is becoming more common as a means of getting the most out of collected data.

Data protection

Data must be protected. Health data involves aspects such as data confidentiality and data integrity.
The process to access and make changes to data must be well controlled.

Data security entails anonymisation at different levels to minimize the chance of identifying patients.

Applications used to analyse data also have to be validated for use.

Informed consent

This is the means by which an individual chooses to accept the conditions of a medical research study. Participants are typically provided with details of all aspects of the study.

This is the principle that individuals should be aware of the risks they will be exposed to by participating in a study.

Data Standards

Standardization makes datasets more useful to others. They make it easier easier to integrate, analyze, process and check that the data is of sufficient quality and has not been compromised.

Standards facilitate the re-use & sharing of data. The FDA is mandating that standards are applied to data submitted in the application for approval of a new therapy. This saves time and money as well as improves the quality of the submitted data.

Data analysis

Data analysis is the process of examining data and making sense of this data by applying statistics to determine if there are trends and markers that relate to disease and health. These trends or markers may indicate opportunities for new therapies.

Data analysis can also be used to provide evidence for the effectiveness of a test or medical therapy.

Personal Data

These are pieces of information associated to data that can be used to identify an individual.

These markers uniquely identify an individual either directly, for example, by name, address or phone number; Or indirectly by combining several of them for example; date of birth and postal code in small or middle size cities.

Genetics is a recent addition to the list of identifiers.

Data reuse

Processing already existing research data for a purpose not originally planned for.
For example, health data collected to conduct a clinical trial on a given intervention, which are subsequently used to compare the results of another intervention, or a study by another company.

Who owns medical research data?

In the process of collecting, processing & analyzing a dataset many parties have added value to the data and could have the right to protect their intellectual input.
Right to ownership may vary based upon what type of data is under consideration. Personal data such as your name is more clearly the ownership of an individual, whereas with an analysis report carried out on a dataset it is more uncertain as who owns or can control access to it.

Register to download vote results of this PlayDecide game.Register