As part of our ongoing efforts to expand the opportunities for innovative research, we have incorporated further datasets within our repository covering a variety of specialist areas: National Early Warning Score 2 (NEWS2), 4AT delirium questionnaire scores, weight measurements, radiology orders, and TrakCare appointments.
National Early Warning Score (NEWS) 2
The National Early Warning Score (NEWS) 2 dataset collates assessments made by clinicians in Emergency Departments and other acute care settings.
A NEWS score is based on observation data that is routinely recorded when patients are treated in hospital. Six key measurements are captured: respiratory rate (breaths per minute); blood-oxygen level; body temperature; blood pressure; pulse rate; and level of consciousness (focused on apparent confusion or agitation).
In clinical practice, scores related to each measurement are combined to give the overall NEWS figure, which supports decisions on how patients should be treated. For example, a score of 7 or more sees the patient being assessed as high-risk, which prompts emergency intervention by critical care specialists. NEWS is regarded as an important assessment for patients suspected as having sepsis and other rapidly progressing conditions.
“Making routinely collected NEWS2 observations available for research enables characterisation of a patient’s physiological status and how it changes over time, beyond isolated snapshots. Combining NEWS2 with other datasets permits the rigorous evaluation of new approaches to assessing illness severity and offers the opportunity to develop more timely interventions that greatly improve patient care.”
Ewen Harrison, Professor of Surgery and Data Science, University of Edinburgh
Delirium assessment: 4AT questionnaire
4AT is a brief questionnaire used by clinicians if they suspect delirium: a sudden change in the brain that causes confusion and difficulties in maintaining attention. The questionnaire captures:
1. the clinician’s assessment of how alert the patient is;
2. how well the patient remembers their age, date of birth, current location, and current year;
3. how well the patient can say the months of the year in reverse order; and
4. an assessment of recent changes in alertness / mental sharpness.
The 4AT score does not diagnose delirium directly, but indicates if the clinical team needs to pay further attention and assess whether their patient is experiencing a delirium episode.
“The 4AT is a brief, well-validated tool for detecting delirium and cognitive impairment and is already used in routine care. Having 4AT scores in DataLoch means researchers can study delirium across large patient populations and examine its associations with outcomes over time.”
Alasdair MacLullich, Professor of Geriatric Medicine and Consultant Physician, University of Edinburgh
Inclusion of weight measurements
Our Observations table is a collection of lifestyle-related data that draws together primary and secondary care data to improve reliability. This dataset already includes alcohol consumption, body mass index, smoking status, and frailty scores. We have now added weight data to improve the data available for research exploring lifestyle factors that can impact health and wellbeing.
Radiology Orders
This dataset captures the basic information around requests for imaging a patient, including the timing and urgency of the request. Specific details include when the request was made and completed, the level of urgency for the procedure, as well as the type of image, such as X-ray, MRI scan or other technique. Furthermore, information is captured about which body parts should be imaged.
However, the actual scans and indicated interpretations are not part of this dataset.
TrakCare Appointments
Key information about outpatient – and some emergency and inpatient – appointments is found in this dataset, which covers hospital or clinic visits, home visits, and telephone consultations. It provides comprehensive details about the hospitals and specialist departments involved alongside referral dates. Additionally, the dataset offers insights into appointment characteristics, such as whether the appointment is an initial consultation or a follow-up, as well as stating whether the patient attended the scheduled appointment.
Summary of the data we host
