Predicting Patient Acuity

Nursing workloads could use a fix

Can Accurate Predictions Be Made Using a Mathematical Model?

Predicting patient acuity would be a powerful tool for any hospital, especially for nurses. But is it possible? Can estimations and predictions for the care of patients be produced from healthcare data?

That’s the questions researchers of a recent study wanted to answer. They created a mathematical model for the automated assignment of patient acuity scores and then looked at data from 23,528 electronic patient records.

The results showed that is was possible to obtain accurate predictions about acuity scores for the coming day based on the assigned scores and nursing notes from the previous day.

According to the study, making same-day predictions leads to even better results, as access to the nursing notes for the same day boosts the predictive performance. In addition, textual nursing notes allow for more accurate predictions than previous acuity scores.

The best results appear to be achieved by combining both of these sources, according to the researchers.


To evaluate the effectiveness of patient acuity scores, the researchers relied on the patient classification systems (PCS), which assesses and classifies patients according to their activity, their need of care, as well as the nursing activities that are needed to fulfill those care needs during a specific time period. Nurses managers have used PCS to organize the care process and allocate the necessary resources.

OtherPCS systems are available, but because the researchers were based in Finland, they used one commonly applied in that country — the Oulu Patient Classification (OPCq), which has proven to be a reliable measure. It consists of six nursing care subsections: (1) planning and coordination of care; (2) breathing blood circulation and symptoms of disease; (3) nutrition and medication; (4) personal hygiene and excretion; (5) activity, movement, sleep and rest; and (6) teaching, guidance during care and follow-up care, and emotional support. Each subsection is dreaded by the nurse daily on a scale of A to D, with being 1 point. The higher the score, the more demanding the need of care. Patients are then categorized into five different levels.

The data analyzed consisted of 23,528 patients with any type of heart problem that were admitted to a university hospital between 2005 and 2009. The patients had to have stayed in the hospital longer than one day.

The researchers then applied a mathematical model to automatically assign patient acuity scores on the data set, and then used more mathematical models to predict a patient’s acuity.


The results showed that the texts and previous scores contain important and complementary information, with the best results being achieved by combining both sources.

Also, having access to notes from the same day that the prediction is made also further boost predictive accuracy.

The researchers concluded a reliable patient acuity predicting system would decrease a nurses’ workload, freeing them for patient care. They said the availability of this kind of predicting tool would also transform patient classification into a real-time process instead of a monthly reporting task.

ASSIGN for Nurses, a MedAptus patient assignment software program that matches the right patient with the right nurse, has a built-in patient acuity tool that automates scoring.