Diabetes is a manageable condition, but many patients have poor outcomes. IDF Europe wanted to uncover gaps in clinical practice and gain insight into care from the patient and healthcare-professional perspective.
IDF Europe harnessed cognitive analytics technology from IBM to review medical literature and analyze social media data, supporting the publication of a comprehensive study of diabetes care across Europe.
50% fasterliterature review—from 8 months to 4—via automatic identification of relevant articles
Explorespatients’ experiences of living with diabetes through social media analysis
Providesdetailed evidence to help healthcare organizations improve diabetes care
Business challenge story
Identifying gaps in diabetes care
Diabetes poses one of the biggest healthcare challenges of the 21st century. The disease can lead to serious health problems, and have a significant impact on quality of life for patients and their families.
With 58 million people living with diabetes across Europe, the financial and societal impacts of the disease are also a major concern for governments and healthcare providers. In 2017, diabetes was responsible for over 9 percent of total adult health expenditure, and caused over 477,715 deaths in Europe alone.
The most tragic aspect of these statistics is that in many cases, these outcomes could have been avoided. Although most types of diabetes are incurable, the disease is comparatively well-understood, and medical science has developed numerous strategies to mitigate and control it. The problem is that physicians and people living with diabetes often lack the information, resources or support they need to manage the condition effectively.
To take a deeper dive into the reasons why many people with diabetes do not receive the level of care they need, IDF Europe embarked on a major research project, which would ultimately result in a publication called “Integrating diabetes evidence into practice: challenges and opportunities to bridge the gaps”.
Dr. Niti Pall, Chair Elect of IDF Europe, explains: “There’s a huge gap between the amount of diabetes research that gets done in hospitals and universities, and the reality that people with diabetes see when they go to get treatment. We wanted to bridge that gap by exploring the reasons why standards of care vary so widely between different countries, and highlighting opportunities to improve diabetes education and standardize practices.”
Cognitive technology powers innovative research
As part of the “Integrating diabetes evidence into practice” project, IDF Europe was keen to explore new technologies that could augment its traditional research methods. When an opportunity arose to work with IBM, the IDF Europe research team quickly grasped the potential of the collaboration.
Maria Stella de Sabata, Regional Manager at IDF Europe, comments: “IBM was able to bring new technology and an extra level of data science expertise, while we helped the IBM team gain greater insight into how NGOs contribute to the understanding and delivery of healthcare. Working together on this project was a win-win.”
Like most research projects, the first phase involved a review of all the recent literature on diabetes. Looking at articles published between 2000 and 2017, the team counted 7,000 that were potentially relevant. Sorting through all these articles by hand would have been extremely time-consuming, so the team decided to automate the process using IBM® Watson® Explorer instead.
Working closely with IBM experts, the researchers imported the articles into Watson Explorer in PDF format, defined a dictionary of keywords to search for, and provided examples of the right places to look for information in each document. For example, the Watson solution learned to read the abstract for information about the scope of each article, and also annotate the references section to identify other papers that might be useful for the project.
Cristian Andriciuc, the project coordinator for IDF Europe, recalls: “Watson Explorer was a tremendous help in finding the most important sections of relevant articles, and presenting them to our researchers for deeper analysis. As a result, we were able to complete the literature review in just four months—it would probably have taken twice as long using traditional methods.”
The second phase of the project involved a survey of IDF Europe’s member organizations, examining the standards of diabetes care and levels of diabetes education in different countries and regions. This was also combined with data from other sources, such as information on the costs of diabetes care provided by the World Health Organization. IBM data scientists helped the research team aggregate and analyze these different data sets, and visualize the results.
Dr. Niti Pall comments: “IBM helped us make the survey data approachable to a general audience, which is critical because our research isn’t just for doctors and politicians—it’s for people with diabetes too. We did this study because we wanted to focus on the people who are living with the condition, and understand their perspective.”
In the third phase of the project, the team placed an even stronger emphasis on understanding the everyday experience of people with diabetes, by analyzing social media data. Using IBM Watson Analytics® for Social Media, the researchers were able to capture public posts from platforms such as Twitter and Facebook, anonymize them, and analyze the topics discussed and sentiments expressed.
Dr. Niti Pall comments: “The wonderful thing about social media is that the data is completely unfiltered. In particular, younger people will communicate about their diabetes on social media in a much more open way than if we asked them to fill in a survey or speak to their doctor.”
As a result, the research team was able to identify a number of new topics that were clearly important to people with diabetes. The team was also able to gain insight by analyzing social media metadata. For example, the timestamps on each post helped the team understand the times of day when people with diabetes most often seek help or advice from social media.
“Understandably, people tend to turn to social media outside of working hours, when traditional health services may not be available,” says Maria Stella de Sabata. “If health systems can find ways to provide reliable information 24/7, it could avoid situations where people put themselves at risk by following incorrect advice from the internet.”
Facilitating better care
Collaborating with IBM helped IDF Europe publish the results of the “Integrating diabetes evidence into practice” study within just a few months. In particular, the use of IBM Watson Explorer to accelerate the literature review stage had a significant impact on the speed of research.
Cristian Andriciuc says: “Watson Explorer is a great tool for researchers who need to search through large amounts of medical literature—it cut the time we spent on literature review by around 50 percent.”
Maria Stella de Sabata adds: “Using Watson Analytics for Social Media to mine social media data has attracted a lot of attention—it shows that our study goes beyond the traditional academic approach, and highlights the real impact of diabetes on people’s lives.”
The conclusions of the study map out the huge variations in delivery and quality of care for diabetes in different countries across Europe. In some countries, diabetes is not seen as a priority, or recommended care models are not adopted, for a variety of reasons. In others, effective guidelines are in place, but clinician shortages or a lack of training prevent good practices from being adopted. Finally, even in countries where sufficient resources are in place, diabetes treatment may still fall short if healthcare systems fail to communicate with patients effectively.
Dr. Niti Pall explains: “The study provides objective evidence that will help healthcare organizations put the latest diabetes research into practice. Above all, it emphasizes the importance of physicians and nurses paying attention to the patient’s experience, not just their medical condition. Putting patients at the center is the key to boosting adherence to therapy and reducing the burden of diabetes on patients, their families, and society as a whole.”
She concludes: “This ground-breaking project with IBM shows how innovative research methods can open up new perspectives on diabetes care, giving governments and healthcare organizations the insight they need to help people with diabetes live fuller, healthier lives.”
International Diabetes Federation Europe
IDF Europe is the European chapter of the International Diabetes Federation (IDF). As an umbrella organization representing 71 national diabetes organizations in 46 countries, IDF Europe is a diverse and inclusive multicultural network that represents both people living with diabetes and healthcare professionals. The organization aims to influence policy, increase public awareness and encourage health improvement, as well as promote the exchange of best practice and high-quality information about diabetes throughout the European region.
Take the next step
IBM Analytics offers one of the world's deepest and broadest analytics platform, domain and industry solutions that deliver new value to businesses, governments and individuals. For more information about how IBM Analytics helps to transform industries and professions with data, visit ibm.com/analytics. Follow us on Twitter at @IBMAnalytics, on our blog at ibmbigdatahub.com and join the conversation #IBMAnalytics.