Jordi Blanch studied mathematics and ended up specialising in biostatistics “by chance”, after completing a master’s degree in applied mathematics. “At the time, I was working in a research group at UPC and had more free time than I do now, so I decided to enrol in a master’s degree in statistics to learn more about the field,” he explains. Since then, he has turned that interest into a career in health research. He completed a PhD in public health and is passionate about uncovering answers to people’s health problems from seemingly cold, hard data.
After spending three years at the Hospital del Mar Research Institute, he joined the Girona Research Support Unit (USR Girona) of IDIAPJGol in 2013 as a biostatistician and data manager. There, he provides methodological support throughout every stage of the research process, from helping design studies and selecting the most appropriate statistical methods to analysing and interpreting results.
Even so, he acknowledges that most of his work tends to take place towards the end of the research process, once the principal investigator has collected the data and needs help interpreting it. “If it’s a project led by our Unit, I’m more involved in the study design or protocol development. But when the project comes from outside, I often step in once everything has been collected and it’s time to analyse the data.”
A tool that serves research
Blanch sees biostatistics as a tool that serves research. “The point of biostatistics isn’t to develop statistical methods for their own sake, but to help answer research questions,” he says.
In his view, biostatisticians often play the role of keeping research grounded in reality. “We try to be the bridge between the ideas researchers have and the answers they ultimately obtain,” says Blanch. One of the most frustrating aspects of his work, he admits, is when the available data are simply not sufficient to answer the research question. “Researchers often ask us to do more than the data they’ve collected can actually support.”
To avoid this situation, he encourages researchers to seek statistical advice before starting their studies. “The role of a biostatistician isn’t limited to data analysis. We can add value at every stage of the research process, from study design through to publication. Even when writing a paper or report, our perspective is important. Quite often, you have to tell researchers, ‘No, you can’t make that claim based on the data you have.’”
Nevertheless, Blanch believes that research groups increasingly recognise the value of biostatisticians. “Today we’re well regarded and fully integrated into the research process, and that hasn’t always been the case,” he says.
Pharmacovigilance in cardiovascular health
In addition to supporting healthcare professionals conducting research through the Girona Research Support Unit, Jordi Blanch is a member of the Vascular Health Research Group, which focuses primarily on pharmacovigilance projects.
One of the group’s main research lines is evaluating the benefits and risks of statins in order to optimise prescribing according to cardiovascular risk. Another focuses on the relationship between cardiovascular risk and dementia. The group is also conducting studies to assess the effectiveness of new lipid-lowering drugs, for which, as Blanch points out, “there are very few studies, and even fewer based on real-world data.”
Among the group’s studies, Blanch highlights a paper published in the British Medical Journal in 2018, which concluded that statins were not associated with a significant reduction in cardiovascular disease or mortality among people over the age of 75 without diabetes. “That paper was highly influential and received a great deal of attention. In fact, it was selected as one of the best papers published that year,” he recalls.
SIDIAP opens new opportunities
One of the main sources of information Blanch works with is the SIDIAP database, which contains clinical information from almost everyone receiving primary care in Catalonia.
“Since 2013, I’ve seen SIDIAP grow from a database with only a handful of variables into one where you can now find almost everything,” he says. He adds that having access to this database has enabled his research group to participate in projects that would not have been possible otherwise. It has also “opened the door to many collaborations, both nationally and internationally.”
Optimistic about AI
Jordi Blanch enjoys his work. What does he find most rewarding? “Making sense of the data in front of you,” he replies. He also enjoys “trying new approaches” and questioning traditional ways of tackling problems in search of more efficient solutions.
He acknowledges that the field has changed dramatically. “When I completed my master’s degree twelve years ago, we were still being taught methods that were largely the same as those used in the nineteenth century. But over the last five or six years, especially since the COVID-19 pandemic, statistics has changed enormously, particularly in areas related to machine learning, which underpins generative artificial intelligence. We’ve moved from highly deterministic models to models that explicitly incorporate uncertainty, allowing us to capture much more of the variability in our data.”
When it comes to artificial intelligence, Blanch is optimistic. “I think it’s a useful tool if you know how to use it,” he says. The Girona Research Support Unit biostatistician notes that tools such as Claude and ChatGPT “give you ideas and perspectives you might not have considered,” while also stressing that “it’s important to remember that the ideas AI generates are based on existing information from elsewhere.”
He is also very cautious about the information shared with these tools. “We currently use web-based AI systems, which limits what we can upload. Quite often, when we’re writing papers or making specific queries, we’re handling sensitive information that we simply cannot share with AI platforms—at least not until we have a local AI solution within the Unit, something we’re gradually working towards.”
