Patient-centred workload clustering in adults under 65 with long-term conditions from longitudinal primary care data.
6 affiliations
- 1Department of Population Health Sciences, King's College London, London, UK.
- 2School of Primary Care, Population Sciences and Medical Education, Faculty of Medicine, University of Southampton, Southampton, UK.
- 3School of Mathematical Sciences, University of Southampton, Southampton, UK.
- 4Department of Psychological Medicine, Institute of Psychiatry, Psychology and Neuroscience, King's College London, London, UK.
- 5Applied Research Collaboration Wessex, NIHR, Southampton, UK.
- 6University Hospital Southampton NHS Foundation Trust, NHS, Southampton, UK.
Abstract
Introduction
Although patient workload is increasingly recognised as a relevant dimension of the patient experience, it is not satisfactorily captured in current electronic health records. We set out to describe the distribution of patient workload patterns across long-term conditions and demographic groups using electronic health records annotated to reflect patient-centred workload derived from a qualitative evidence synthesis of patient experience.Methods
We used a predefined compendium of 63 workload indicators, derived from a previously published qualitative evidence synthesis of patient experience and mapped to electronic health record codes, classified into eight themes. We computed counts of these indicators in 765 930 patients born between 1956 and 1965 with two or more recorded long-term conditions, randomly sampled from the Clinical Practice Research Datalink. We then used an unsupervised learning approach to construct a typology of patients based on these counts.Results
Median age at first record was 31 years with 28 years of follow-up. The cohort was 49.4% (n=378 154) male and predominantly white (68.1%, n=521 859). Patients had a median of 15 workload indicators and 4 long-term conditions. Nine clusters were identified, with marked contrasts in workload intensity, long-term conditions, demographic profiles and size. While large clusters were less burdensome, smaller clusters showed more complex profiles, characterised by cardiometabolic or mental health long-term conditions, higher female-to-male ratios and greater ethnic diversity. The most burdensome cluster, characterised by a broad spectrum of rarer long-term conditions and dominated by indicators associated with investigation and treatment monitoring, appeared comparatively less complex.Conclusions
This unsupervised approach to patient workload revealed patient clusters with marked disparities in burden, long-term conditions, demographic characteristics and complexity, supporting future research directions aimed at better understanding and supporting patients with elevated workload demands.
Related
- Preventive governance for patient‐centred medical AI
- Advancing Patient‐Centred AI With Adaptive Machine Learning
- Effects of cognitive-motor dual-task walking exercise versus walking-only exercise on cognitive function and patient-centred outcomes in older adults with subjective brain fog: a randomised controlled trial.
- Feasibility and acceptability of a randomized controlled trial of a patient-centred transition program for stroke survivors and their caregivers—P-NAVISTROKE study
- Implementing a patient-centred and evidence-based intervention to reduce BEnzodiazepine and sedative-hypnotic use to improve patient SAFEty and quality of care (BE-SAFE): study protocol of a cluster randomised controlled trial
BibTeX
@article{delord2026patientcentred,
title = {{Patient-centred workload clustering in adults under 65 with long-term conditions from longitudinal primary care data.}},
author = {Marc Delord and Emilia Holland and Mozhdeh Shiranirad and Alex Dregan and Simon D S Fraser},
journal = {BMJ public health},
year = {2026},
month = oct,
doi = {10.1136/bmjph-2026-005763},
url = {https://europepmc.org/article/MED/42968904},
note = {CC BY},
}
RIS
TY - JOUR
TI - Patient-centred workload clustering in adults under 65 with long-term conditions from longitudinal primary care data.
AU - Marc Delord
AU - Emilia Holland
AU - Mozhdeh Shiranirad
AU - Alex Dregan
AU - Simon D S Fraser
PY - 2026
DA - 2026/10/30
JO - BMJ public health
DO - 10.1136/bmjph-2026-005763
UR - https://europepmc.org/article/MED/42968904
AB - Introduction Although patient workload is increasingly recognised as a relevant dimension of the patient experience, it is not satisfactorily captured in current electronic health records. We set out to describe the distribution of patient workload patterns across long-term conditions and demographic groups using electronic health records annotated to reflect patient-centred workload derived from a qualitative evidence synthesis of patient experience.Methods We used a predefined compendium of 63 workload indicators, derived from a previously published qualitative evidence synthesis of patient experience and mapped to electronic health record codes, classified into eight themes. We computed counts of these indicators in 765 930 patients born between 1956 and 1965 with two or more recorded long-term conditions, randomly sampled from the Clinical Practice Research Datalink. We then used an unsupervised learning approach to construct a typology of patients based on these counts.Results Median age at first record was 31 years with 28 years of follow-up. The cohort was 49.4% (n=378 154) male and predominantly white (68.1%, n=521 859). Patients had a median of 15 workload indicators and 4 long-term conditions. Nine clusters were identified, with marked contrasts in workload intensity, long-term conditions, demographic profiles and size. While large clusters were less burdensome, smaller clusters showed more complex profiles, characterised by cardiometabolic or mental health long-term conditions, higher female-to-male ratios and greater ethnic diversity. The most burdensome cluster, characterised by a broad spectrum of rarer long-term conditions and dominated by indicators associated with investigation and treatment monitoring, appeared comparatively less complex.Conclusions This unsupervised approach to patient workload revealed patient clusters with marked disparities in burden, long-term conditions, demographic characteristics and complexity, supporting future research directions aimed at better understanding and supporting patients with elevated workload demands.
SN - 2753-4294
ER -