remain underrepresented within many research communities, especially in senior leadership.
Leadership and career development programmes are often offered as a solution within universities and research institutes. While these programmes can be inspirational and impactful, on their own they do not address the deeper, systemic barriers that shape who can access opportunities and progress.
Why structural change matters: The Gollum Effect
In academia, power is often shared asymmetrically among dominant and minority groups. To address this, we need to focus on the mechanisms that perpetuate these imbalances. Inclusion must go beyond inviting marginalised voices into existing systems, we need to transform how those systems work.
Recent research on the so-called Gollum Effect offers an interesting metaphor for understanding these dynamics. Like Tolkien’s character obsessively guarding the ring, researchers can become possessive over data, ideas, and resources. This creates a hypercompetitive culture, limits access to funding and opportunities and restricts open science and collaboration.
Importantly, this is not just about individual behaviour. It reflects the structures and norms within research systems that reward exclusivity over openness, disproportionately impacting marginalised groups and early-career researchers and disrupting scientific progress.
How does this apply in the context of health data science?
These dynamics are particularly visible in health data science, where access to data, tools and infrastructure is tightly controlled. So, how can we design systems that enable more equitable participation?
The researchers of the Gollum Effect identified potential solutions that support better collaboration and open science , several of which focus on infrastructure and data governance, such as:
- Establishing clear policies on data ownership and authorship to prevent territorial gatekeeping
- Mandating trustworthy open data, code sharing, and pre-registration for research projects to encourage and fund multi-institutional and interdisciplinary collaborations
- Interdisciplinary teamwork and recognising non-traditional research contributions such as technical support and data management
Team science
At HDR UK, team science is embedded from the outset. This means proactively valuing early career voices, recognising a diversity of skills, and engaging a range of partners across disciplines and sectors. This can be illustrated through the HxC: Healthier Science through Collaboration project which aims to drive system change by amplifying the voices of interdisciplinary scientists and technical professionals from diverse backgrounds.
Team science also underpins the Data Curation Skills for Sensitive Data project, which brings together people across roles and disciplines to build shared standards for high quality data curation in Trusted Research Environments. The project intentionally brings together technical specialists, governance professionals, operational staff, and researchers, recognising that expertise is distributed and often under-recognised within infrastructure environments. Its engagement strategy enables contributions from individuals in non-academic roles and from institutions of varying scale and resource. Findings from the project will inform training pathways that are accessible, scalable, and responsive to diverse career trajectories, helping to strengthen workforce inclusion and progression across the sector.
Inclusive infrastructure
Developments within health data science offer a unique opportunity to design technical infrastructure that actively reduces exclusion. But inclusion doesn’t happen by default, it requires intentional design. Prioritising accessibility, usability, and trustworthiness can help ensure that trusted research environments are fair and usable in practice, reducing barriers to entry and enabling a broader range of researchers to participate.
Our technical infrastructure programmes – including the Health Data Research Gateway, the Cohort Discovery Service, and the Safe People Registry – are designed to make datasets easier to discover and support scalable and trustworthy access to sensitive data, supporting more equitable access to health data for research.
Similarly, in collaboration with the UK Health Data Research Alliance, we have developed the SafeGUARDs framework. This practical tool supports trustworthy decision-making and makes data access decisions more transparent and easier to understand for everyone, including researchers, data custodians and the public.
More broadly, building trust in health data science means addressing deeper, systemic sources of mistrust stemming from historical misuse of data, unethical practices, misinformation biased assumptions about minoritised groups and unequal experiences within the healthcare system.
Find out more…
Embedding equity into health data ecosystems – through inclusive infrastructure, equitable policies, and collaborative research cultures – creates the conditions for innovation and better science. You can explore the work mentioned in this blog post further:
