Send message to Authors

Short abstract:

The paper highlights the conflict between evidence-based medicine, data-driven medicine and practice-based medicine (held by GPs) developed in Italy during the Covid-19 outbreak. Multiple meanings of ‘empirical evidence’ and the downgrading of GPs’ practical knowledge and experience are analyzed.

Long abstract:

Since February 2020, strategies aimed at containing and managing the Covid-19 outbreak have been developed by European countries. Among these measures, the possibility of an early treatment of the disease has been considered of fundamental importance, both for curing the disease and governing the outbreak. Despite their potential, early (practice-based) therapies were neglected in Italy and the debate around them gave rise to a strong conflict between their proponents and opponents, to the point that some of the former (mainly General Practitioners) organized a properly political movement in order to promote the integration of early home therapies in the official health protocols.

Unlike, many scientists and doctors argued the necessity of waiting for the outcome of the randomized (evidence-based) studies; which, however, arrived after several months. Meanwhile, medicine has moved mostly on the basis of the quantitative data available, sometimes collected with different and not very transparent criteria.

Hence, a harsh conflict developed between several GPs (supporters of the practice-based approach) and scientists and doctors accustomed to following evidence-based approaches and forced to rely on data-driven approaches.

The paper highlights this epistemological conflict among these 3 different visions of medicine and science, the hierarchy of scientific knowledge and professional skills, the downgrading of GPs’ practical knowledge and experience, the multiple meanings of the concept of ‘empirical evidence’.

Particularly, health protocols will be considered as socio-technical objects embedded in a vast range of cultural, political and economic factors that contributed to the general resistance towards practice-based therapies into national guidelines.

Send message to Authors

Long abstract:

What is the relationship between evidence-based and data-driven health care? In this paper, we address this question from a sociological perspective, exploring the epistemic, institutional and political processes that bring each of these types of health care to bear. We draw on the contrast between two empirical cases studies: reproductive medicine, an area where there is increased use of big, user-led data to obtain regulatory approval and capture markets, and evidence-synthesis communities, and how they have responded to the challenge of big data in the last decade to develop new analytics and review methodologies. We propose a model to understand these transformations that focuses on the shifting configuration between market, standards and democratic legitimacy in contemporary health care. Specifically, we argue that data-driven health care poses challenges to the effective balancing between commercial technology development, standardisation and patient-centeredness in medicine, whereby existing standardisation configurations are challenged and/or adapted in view of rapid market expansion and the participatory promises of user-led data. In doing so, we highlight where tensions arise between evidence-based-medicine (EBM) imperatives and data-driven health care and locate productive convergence zones. We conclude with a discussion of potential implications for the future of EBM in the age of big data.

Send message to Author

Long abstract:

Real-world data (RWD) refer to routinely collected data relating to patients’ health status and the delivery of health care originating from a range of sources other than traditional clinical trials. Examples of RWD include electronic health records, patient registries or personal information collected from wearable devices. RWD are often used in particular epistemic projects, the pharmaceutical industry relying more and more on observational data such as health records instead of experimental data one would gain as part of Randomized Controlled Trials (RCT) to learn about the effectiveness and safety of medicines. In such instances, RWD are framed as an alternative form of evidence, which can supplement or even replace RCTs. This shift greatly impacts what we consider ‘good’ science and objectivity in research. In this paper, I begin mapping out what counts as RWD and how these data are used in order to understand how they shape the knowledge produced about the world. Crucially, I explore how the use of RWD challenges the established hierarchy of evidence within healthcare, and discuss its implications for evidence-based medicine.

Send message to Author

Long abstract:

Ranging from electronic health records, mhealth and sensors to more sophisticated advancements in AI, genomics, and Big Data, EU policymakers have sought to find ways to translate digital health technologies (DHT) into ‘trustworthy’ tools for healthcare. Despite several initiatives, one of its most prominent translational tensions resides around the evidence standards that are used by regulatory agencies to assess the quality and safety and reimbursement-potential of medical Digital Health Technologies. According to developers, industry, and researchers alike, the struggle for these technologies to navigate these regulatory approvals arises from the fact that these policy frameworks still rely on evidence standards that mismatch with the dynamic nature of these technologies, hindering their translation into ‘trustworthy’ tools for healthcare. This misfit ties into a far wider debate that has been incubating for years in the world of drug approvals on the move from Evidence-based Medicine towards Real-world evidence (RWE) for assessing the quality and safety, and reimbursement of biomedical innovations. By investigating three ‘regulatory experimentations’ to develop RWE as governing tools for DHT (e.g., In Silico Modeling, Digital Clinical Trials, and RWE reimbursement frameworks), this paper builds on STS, innovation-, and RRI studies to investigate the social dimensions of RWE as a new mode of governing for DHT: how does this governing technique come into being? Which type of knowledges are produced, and by whom? and how does this reconfigure our understanding of ‘medical’ expertise, -data, and patients?

Send message to Authors

Long abstract:

Evidence-based medicine (EBM) relies on the ‘hierarchy of evidence’, which ranks different studies according to their methodological rigour, to assess the quality of evidence. Systematic reviews and RCTs are favoured, whereas experience-based knowledge and case studies are valued least. EBM thus evaluates evidence following a frequentist logic, i.e. drawing statistical inferences from frequent events. Despite criticism for being too narrow, inflexible and undervaluing many types of knowledge, it remains the dominant approach in EBM.

Large volumes of data, especially so-called ‘real-world data’, together with advanced data analytics present new opportunities to generate insights that are inaccessible to EBM’s core methodologies. Computational and AI-based methods are employed to process these data by identifying patterns and correlations that supposedly support the prevention, diagnosis and (personalised) treatment of diseases.

Despite the diversity of data, these AI-based technologies often operate within a frequentist framework similar to EBM, where statistical probabilities are determined by observed frequencies. AI-based analyses of ‘big data’ are thus said to reproduce prevailing modes of knowledge production in EBM that prioritize quantitative data and statistics.

Drawing on AI-based analyses of patient experiences shared online to inform evidence-based guideline development, we empirically scrutinise the types of reasonings elicited by AI analyses. While RCTs follow a generalised one-in-many frequency principle, AI methods enact a different, diverse frequentism. With AI methods, we obtain many different individual experiences, a large “gathering of exceptions”. Understanding these specific modes of reasoning illuminates the kind of knowledge that can be generated by AI and its relation to EBM.

Send message to Authors

Long abstract:

Evidence-based medicine (EBM) is a dominant paradigm organizing the production and translation of knowledge in public health. Within this approach epidemiological evidence generated in randomized-controlled trials (RCTs) provides the ‘gold standard’ for public health interventions in at-risk target groups (Timmermans & Berg 2003). In recent years big data analytics have come to be heralded as offering a radically novel approach to public health research and practice. Variably known as ‘personalized prevention’ or ‘precision public health’ this approach uses aggregate data from multiple domains to arrive at personalized risk-profiles for disease and intervention outcomes (Hoeyer 2019). Proponents have argued that this data-driven public health obviates the need for key tenets of EBM, including sampling quality, hypothesis-generation, and causal inference. Traditional public health scientists have, however, been skeptical of these claims, arguing for the continued relevance of EBM criteria and emphasizing the need for mutual collaboration (Mooney & Pejaver 2018). In spite of this on-going discussion, few studies have looked at the ‘trading zones’ (Gallison 1997) that appear in research collaborations between these two approaches to public health. In this study we use the notion of ‘repertoires’ put forward by Ankeny & Leonelli (2016) to uncover how researchers deal with tensions between data-driven and evidence-based repertoires of public health research, asking how they craft convergence between the two approaches. Our analysis is grounded in an ethnographic case study of a large-scale research consortium engaged in modelling early childhood developmental risk, bringing together data-scientists, public health researchers and professionals in childhood preventative care.

Send message to Author

Long abstract:

On-line tools and Apps have transformed our society. Today, millions of people are using on-line tools to diagnose themselves with various psychiatric conditions; often receiving a diagnosis by e-mail: a 'mailbox diagnosis'. The developmental lifecycle of psychiatric diagnostic categories themselves, how new diagnoses are created and become established via EBM, is the topic of my presentation.

On-line platforms have led to a proliferation of new ‘seedling’ diagnostic categories. They are nurtured by varying networks of associated research, industry, and lay support, which support more traditional EBM- like approaches (epidemiology studies) to validate seedling diagnoses. This presentation will start by reviewing which seedling diagnoses currently have a platform. They include more established diagnostic categories “mild cognitive impairment” (mild disorientation, forgetting names, dates: touted as forerunner to dementia); less established, e.g., “screen attention disorder” (limited attention caused by over-use of social media/screens), and new child diagnoses e.g., “pathological demand avoidance” (rigid resistance to action after being told what to do) and many more.

My aim is to introduce a 'life cycle' model of diagnostic category showing how both data harvesting and EBM interact to establish new diagnosis through a model that engenders fluid interdisciplinary thinking in the study of the social causes, construction, and implications of diagnostic categories.