Projects Delivered

The Group A Streptococcus (GAS) is a bacteria that causes skin and respiratory tract infections. Invasive GAS infection occurs when the bacteria invades part of the body where it is not usually found, like the blood, lungs, deep muscle and fat tissue. The infections are diagnosed by growing the bacteria in the laboratory. This process can take several days and is affected by the prescription of antibiotics to the patient. In January 2023, NHS Lothian introduced a PCR test to improve the diagnosis of patients with invasive GAS. Using the DataLoch curated databases, we aim at investigating the impact of this new test in patient's health by comparing with the traditional testing pathway.

After a brain scan, many people without any symptoms of a stroke learn they have scan appearances similar to those of a stroke (‘covert cerebrovascular disease’). People with covert cerebrovascular disease have a higher risk of future stroke or dementia, but no treatments are known to reduce this risk. In part, this is because it is difficult to find and ask enough people with covert cerebrovascular disease to take part in trials of new treatments. 

We have developed new computer methods that can look at massive numbers of brain scan reports. We think these new computer methods could find and invite a large number of people affected by covert cerebrovascular disease to studies to test new treatments. We will test whether this approach is feasible in this project and in later work test if it is acceptable to people with Covert cerebrovascular disease (CCD).
If this method is feasible and acceptable, our future work will use brain scans collected in normal clinical practice and invite people to join a study of an existing blood pressure treatment in NHS Lothian. 

In Scotland, and across the world, the increase in lifespans is leading to an increase in the number of people living with long-term healthcare conditions. This results in reduced quality of life and is a challenge for healthcare systems worldwide which have traditionally been set up to treat single conditions.
    
In an acute care setting, being able to identify key long-term conditions will quickly provide clinicians with the background knowledge on patients, allow them to make time-critical decisions in a better informed manner, and improve the outcomes for patients .

Within the DataLoch secure environment we will build a solution which helps clinicians to identify patients with multiple long-term conditions rapidly, apply artificial intelligence to create personalised risk stratifications, better visualise their data, then integrate this with guidelines to support clinical decision making following an emergency department attendance. This solution will be risk assessed before release from this environment.

In partnership with NHS Lothian and Health Innovation South East Scotland (HISES) we will then evaluate how the dashboard and AI risk stratification impacts clinical care in a real-world environment.

A first infection with the varicella zoster virus manifests as chickenpox, after which the virus remains “hibernated” in nerve cells in the body and can break out occasionally to cause shingles. The virus has recently been implicated in a variety of health outcomes related to inflammation of blood vessels, such as heart attacks and strokes. Understanding whether this virus really is involved in the causal pathway that leads to such health outcomes could be very useful to our understanding of these diseases, and ultimately how to treat and prevent them. In addition, because there is an effective vaccine to reduce the chance of getting shingles, it is important to find out if the vaccine can prevent some of these adverse health outcomes. If so, promoting uptake of the vaccine, and potentially providing it already at younger ages or as additional “booster shots”, could be an effective tool for preventing these adverse health outcomes. We plan to take advantage of the way in which the shingles vaccine was rolled out in Scotland to help us control for the fact that usually those who get vaccinated are different in various health-related characteristics compared to those who do not get vaccinated. Using our approach, we will be able to contribute to the research evidence on the varicella zoster virus in a new and important way because we will be able to provide estimates of the shingles vaccine’s effects that are less vulnerable to bias than most other existing studies as we will be in a position to more convincingly compare population groups that are truly comparable to each other. In addition, we will be able to provide a rigorous evaluation of the effect of vaccine on shingles prevention when the vaccine is rolled out in the routine health system as opposed to in a highly controlled research setting.

Chest pain is a serious symptom that may indicate heart disease. It is a common reason for people to attend hospital or their GP. Specialist cardiac services have been designed to review those with suspicious symptoms and start important treatments if necessary. These are known as Rapid Access Chest Pain Clinics (RACPC). However, there has been no systematic review of referral patterns and outcomes after RACPC assessment. Not every patient with chest pain is referred by their GP, and not every referral is triaged to RACPC. We suspect that opportunities to prevent heart attacks may be being missed. To investigate this, we will use linked healthcare data to evaluate outcomes in patients with recent onset chest pain, over the last decade in NHS Lothian. We will review how investigations, treatments, and outcomes (such as heart attacks and deaths due to heart disease) have varied between those reviewed and not reviewed in specialist services. Our aim is to find improvement opportunities and identify the types of referrals that should be prioritised. 

We aim to understand and compare the impact of different respiratory infections in hospitalised patients of all ages, specifically human metapneumovirus (hMPV), human parainfluenza virus (PIV), respiratory syncytial virus (RSV), and influenza, on people seeking healthcare at NHS Lothian between 2010 and 2022. We will estimate, for each of the virus groups, the severity of disease, requirement of different levels of medical care and the clinical outcomes. This will allow a comparison of the requirement of different levels of medical care, and clinical outcomes with those in each of the viral groups.  We will examine data separately for different age groups and underlying health condition status.

The findings will help healthcare providers allocate their resources more effectively, develop strategies to prevent and manage these infections, and guide the development of vaccines for hMPV and vaccination policies for specific groups.

Frailty and multimorbidity are terms used to describe people at risk of worsening health, increasing disability and death. These are often older people who have multiple health problems and frequently present to health and social care services at a point of crisis. However, identifying these vulnerable people earlier may allow extra support and assessment to prevent such crises. There are multiple ways of identifying multi-morbid people, but our current record systems for primary care (GP practices), secondary care (hospitals) and social care are not well connected.

In this project, we will bring data together data from GP and hospital records to provide a combined view of patients’ health and care information to health and social care partnerships in the Lothian region. These groups are responsible for services that impact the health and wellbeing of the population. No individual level patient data will be shared as part of this project.

Sepsis is a term used for severe infection which affects multiple body systems. Suspected sepsis is a common presentation to the Emergency Department and must be treated immediately as it can be life threatening. One of the challenges in improving care for people who present with suspected sepsis is around figuring out who has the condition, and who has an infection which is less severe, and who has a different condition entirely. Currently, we do not have a good way of identifying people who present with sepsis using routinely collected data. This project hopes to compare different ways of using data that is collected during the time people are in hospital. If we can accurately find out patients who present with sepsis to the Emergency Department, we will be able to better understand the disease, and also be able to plan future research projects where we can compare tests or treatments which can improve how we care for those with sepsis.