Background In order to meet the demands of a growing and increasingly complex healthcare service, many hospitals have developed pharmaceutical assessment screening tools (PASTs) to prioritise workload.^{1 2} A lack of sensitivity across specialities is often cited as a drawback to PASTs, yet no neonatal or paediatric-specific tool has been developed.³

Following electronic patient record (EPR) implementation our hospitals use two PASTs concurrently:

• a historical pharmacist-assigned patient acuity level (PAL) determining frequency of review, ranging from level 1 (twice a week) to level 3 (daily).

• an EPR-assigned patient acuity score (E-PAS) which updates a cumulative total in real-time following the triggering of specific rules such as ‘overdue critical medicine’; this was designed as a single tool for use in neonatal, paediatric and adult patients.

A lack of outcome-based evidence for electronic scoring systems has led to a reluctance to use the E-PAS tool alone, especially in the absence of a neonatal or paediatric-specific tool.

Aims To evaluate the need for a standalone neonatal or paediatric-specific E-PAS tool by:

• comparing local PAL guidance with E-PAS rules to identify gaps.

• determining if there is a correlation between the PAL and E-PAS in neonatal and paediatric patients.

• establishing if the E-PAS is reflective of pharmaceutical acuity assignment by neonatal and paediatric pharmacists.

Method All neonatal and paediatric PAL criteria (n=126) were evaluated against the E-PAS tool to determine if they would appropriately trigger in all, most (>50%), some (<50%) or no patients. A single snapshot of PAL and E-PAS data from all current neonatal and paediatric inpatients (n=338) was collected utilising in-built EPR reporting functionality. 210 patients were excluded as no PAL was assigned. Descriptive statistical analysis of data was performed in Microsoft Excel.

Results Only 34.1% (43/126) of PAL criteria would appropriately trigger the E-PAS tool in most patients; this was significantly lower for intensive care areas (18.8% [6/32]) and more sensitive for level 1 (50% [11/22]) than level 3 (35.9% [28/78]) criteria. PAL criteria not reflected in the E-PAS tool related to specific diagnoses, therapeutic drug monitoring and pharmacist discretion.

The median E-PASs for each PAL were level 1 (2, n=36), level 2 (4, n=48) and level 3 (7, n=44); this demonstrated a strongly positive correlation (r=0.65) which reached significance (p<0.05). The distribution of E-PASs was wider for patients assigned level 3 (range 2–12, IQR = 4) than level 1 or level 2 (range 0–8, IQRs = 2.75 and 2 respectively).

Conclusion Positive correlation between the PAL and E-PAS provides initial reassurance that the current E-PAS tool may be sufficient to prioritise the most critical patients. However, the wide range of E-PASs reported for each PAL and gaps in the tool such as pharmacist discretion and relevance to critical care still suggest it does not yet truly reflect pharmaceutical acuity assignment by pharmacists. A significant proportion of patients were excluded for not having a PAL assigned; this is likely related to data capture timing in relation to EPR implementation and repeating this analysis after 6–12 months might be warranted.

References

1. Saxby KJ, Murdoch R, McGuinness J, Steinke DT, Williams SD. Pharmacists’ attitudes towards a pharmaceutical assessment screening tool to help prioritise pharmaceutical care in a UK hospital. European Journal of Hospital Pharmacy 2017;24:315–319.

2. Hickson RP, Steinke DT, Skitterall C, Williams SD. Evaluation of a pharmaceutical assessment screening tool to measure patient acuity and prioritise pharmaceutical care in a UK hospital. European Journal of Hospital Pharmacy 2017;24:74–79.

3. Abuzour AS, Hoad-Reddick G, Shahid M, et al. Patient prioritisation for hospital pharmacy services: current approaches in the UK. European Journal of Hospital Pharmacy 2021;28:e102–e108.