This site is intended for health professionals only

Sepsis prediction tool helps GPs identify patients needing hospital treatment

Sepsis prediction tool helps GPs identify patients needing hospital treatment

A simple tool for predicting sepsis in the community could help GPs decide which patients may need admitting to hospital, research suggests.

It is the first study to look at a risk score for patients with suspected severe infections at home, say the researchers, who found it had similar levels of accuracy to the NEWS risk score that is routinely used in hospitals.

While not intended to replace clinical judgment, the predictive model, could help support decision making, reduce variation and improve collaboration between primary and secondary care, the British Journal of General Practice reported.

The prediction tool uses age, temperature, blood pressure, heart rate, oxygen saturation and mental status. Researchers also tested biomarkers such as lactate, C-reactive protein and procalcitonin but this did not help improve predictions.

It was tested in 357 adults in the Netherlands who had contacted a primary care out-of-hours service where infection was suspected, including in cases of fever, confusion and general deterioration.

Just over half (56%) of patients were referred directly to the emergency department of which the vast majority (94%) were then admitted to hospital.

In all, 151 patients (42%) had sepsis according to later analysis of records by an expert panel.

Of nine factors looked at, six were included in the final model which were aged >65 years; temperature >38°C; systolic blood pressure ≤110 mmHg; respiratory rate; peripheral oxygen saturation ≤95%; and mental status, each of which scored one point.

The model was shown to accurately predict sepsis with a C-statistic value of 0.76-0.83 which was comparable to outcomes with NEWS, the researchers noted.

One of the benefits of the model is that it is easy to use in practice and uses clinical signs that GPs can easily ascertain in daily practice, they added.

But more research will be needed on how best to use the score in practice to improve the management of patients with possible sepsis.

‘Though the difference between empirical clinical assessment by the GP and performance of the present model was modest, it can help support clinicians during the busy daily routine, reduce variation in the quality of primary care, and improve collaboration between primary and secondary care for this potentially life-threatening condition.

‘The model is not intended to overrule the GP’s overall judgement but rather to inform the GP on the probability of the sepsis outcome. The GP can subsequently use this information to decide whether or not to refer the patient to hospital,’ the researchers concluded.


Visit Pulse Reference for details on 140 symptoms, including easily searchable symptoms and categories, offering you a free platform to check symptoms and receive potential diagnoses during consultations.


Please note, only GPs are permitted to add comments to articles

Malcolm Kendrick 22 April, 2022 9:23 am

‘Though the difference between empirical clinical assessment by the GP and performance of the present model was modest…’

In short, thier model was no better than clinical judgement. However, as is always the way of such things, we will inevitably end up with a rigid six point socring model for Sepsis which will then determine admission to hospital. Woe betide any patient who dies not score highly enough. Woe betide any GP trying to over-ride the score and use their own judgement.

David Jarvis 22 April, 2022 11:15 am

I always find these scores seem to match my clinical judgement. Read into that what you wish.

Stephen Fowler 22 April, 2022 2:24 pm

Medicine by numbers – why not use laymen who can fill in the score, would be much cheaper

David jenkins 23 April, 2022 12:03 am

bollox – if you suspect sepsis, admit. don’t waste time putting a number on someone who is that ill !