Our latest update takes a close look at some of the work we do at DataLoch beyond preparing data extracts for NHS service-improvement projects or novel research explorations. There’s a lot that goes on in the background…
Connect with Us if you want to find out more.
Curating a reliable overview of lifestyle data
It is widely recognised that lifestyle factors – smoking, diet, exercise – are associated with preventable ill health and are socioeconomically related, meaning they contribute to widening health inequalities. To support research incorporating these factors, we need to provide high-quality data. When we focused on alcohol consumption data, we discovered that answering the apparently simple question of “how much alcohol do people drink?” is a lot more complex than we anticipated…
Marta’s Perspective: making it easier for researchers to explore lifestyle impacts on health
Resolving the challenges of GP coding
Coded data allows details of GP visits, test results, symptoms, and other information to be easily recorded and shared for approved purposes (like research) without the need to read complete typed-in patient notes. However, a key complication is the diversity of computer systems and individual clinical preferences in recording codes, which requires significant attention from our team before the data can be made available for authorised projects.
Ally’s Perspective: opportunities and challenges in primary care data
Early explorations in making free-text available for research
Health data research is driven by coded data. However, about 70-80% of the value of medical records is locked away in clinical (free-text) notes. Upon a foundation of public involvement, we are exploring how the various types of privacy risks contained within clinical notes could be consistently mitigated, thereby harnessing the full potential of this information for future health innovations.
Arlene’s Perspective: the untapped potential of clinical free-text
Thoughts from our Public Reference Group
Patient and public involvement and engagement (PPIE) is crucially important for research. It ensures that the value of research – tangible benefits for society such as improved care services – remains the key focus, while researchers benefit from the perspectives of those with genuine lived experiences. Our Public Reference Group shares their thoughts on the potential of PPIE for research and how researchers and public participants can access supportive resources.
Public Reference Group Perspective: the importance of PPIE
Improving understanding of later life through data
As more people live longer lives, there is an increased need to develop improved services to support those experiencing health or social challenges. One way to approach the problem is a better use of the data hosted within health and social care systems. However, with so many factors to consider, we have to use a collaborative approach to define the key starting points. In November 2023, we started our journey through the first Ageing Data Workshop.
Arlene’s Perspective: enabling the value of ageing data in research
Our SARA collaboration: developing semi-automated prototypes
Current approaches to data linkage and management are usually locally developed and manual in nature. The SARA (Semi-Automated Risk Assessment) project was led by Arlene Casey, DataLoch’s Natural Language Processing expert, and saw substantial progress in developing prototype dashboards that would support efforts in data provenance (i.e. data ingestion, processing, and linking) and privacy risk assessment for free-text.
Unveiling the Impact of the SARA Project
New research we have enabled: the relationship between socioeconomic factors and COVID-19
Led by the NHS Lothian Infection Service, researchers found that the risk of deaths due to the COVID-19 pandemic were doubled for those with specific indicators of deprivation. In particular, there was an increased risk for those living in areas with a) lower levels of average income and/or b) higher numbers of hospitalisations due to alcohol consumption.
Therefore, for future allocation of resources, there may be greater benefit in looking at specific measures of local deprivation - such as income, education level, crime rates, access to amenities - rather than the overall deprivation level, particularly in areas where overall affluence may mask pockets of deprivation.
View the open-access article
