A Novel Method to Identify the Start and End of the Winter Surge in Demand for Pediatric Intensive Care in Real Time

Using a genetic algorithm to solve a non-linear location allocation problem for specialised children’s ambulances in England and Wales

Since 1997, special paediatric intensive care retrieval teams (PICRTs) based in 11 locations across England and Wales have been used to transport sick children from district general hospitals to one of 24 paediatric intensive care units. We develop a location allocation optimisation framework to help inform decisions on the optimal number of locations for each PICRT, where those locations should be, which local hospital each location serves and how many teams should station each location. Our framework allows for stochastic journey times, differential weights for each journey leg and incorporates queuing theory by considering the time spent waiting for a PICRT to become available. We examine the average waiting time and the average time to bedside under different number of operational PICRT stations, different number of teams per station and different levels of demand. We show that consolidating the teams into fewer stations for higher availability leads to better performance.

Modelling the allocation of paediatric intensive care retrieval teams in England and Wales

BACKGROUND

Following centralisation of UK paediatric intensive care units in 1997, specialist paediatric intensive care retrieval teams (PICRTs) were established to transport critically ill children from district general hospitals (DGHs). The current location and catchment area of PICRTs covering England and Wales are based on historical referral patterns. National quality standards specify that PICRTs should reach the patient bedside within 3 hours of accepting a referral.

OBJECTIVE

To determine what proportion of demand for PICRT services in England and Wales can be reached within 3 hours and to explore the potential coverage impact of more stringent 'time to bedside' standards.

METHODS

We used mathematical location-allocation methods to: (1) determine the optimal allocation of DGHs to current PICRT locations to minimise road journey time and calculated the proportion of demand reachable within 3 hours, 2 hours, 90 min, 75 min and 1 hour and (2) explore the impact of changing the number and location of PICRTs on demand coverage for the different time thresholds.

RESULTS

For current (and optimal) location of 11 PICRTs, 98% (98%) of demand is reachable within 3 hours; 86% (91%) within 2 hours; 59% (69%) within 90 min; 33% (39%) within 75 min; and 20% (20%) within 1 hour. Five hospitals were not reachable within 3 hours. For the 3-hour standard, eight optimally located PICRT locations had similar coverage as the current 11 locations.

CONCLUSIONS

If new evidence supports reduction in the time to bedside standard, many more hospitals will not be adequately covered. Location-allocation optimisation is a powerful technique for supporting evidence-based service configuration.

Prospective paediatric intensive care registry in Latvia: one year outcomes

Background

In Latvia, there is a single eight-bed paediatric intensive care unit (PICU) where all critically ill children are admitted. A recent retrospective audit of the outcomes of paediatric critical care in this unit revealed a high number of unplanned extubations and excess crude mortality. In 2017, our centre joined the UK and Ireland based Paediatric Intensive Care Audit Network (PICANet) as a pilot project to investigate the feasibility of developing a paediatric critical care registry in Latvia and in the Baltic states.

Methods

Riga Stradins University Ethics Committee approved the study. Anonymized data on all patients admitted to our unit from 1 June, 2017 to 31 May 2018 were prospectively entered onto the PICANet database.

Results

A total of 774 PICU admissions were analysed; 45% of admissions were elective. The median age was 59 months (IQR: 14-149). The highest admission rate was on Wednesdays representing the flow of elective surgical patients. The median length of stay was 0.95 days (IQR: 0.79-1.98). Twenty-five percent required respiratory support. The expected number of deaths estimated using the Paediatric Index of Mortality 3 (PIM 3) 15.16; 15 patients (1.94%) died resulting in Standartized Mortality Ratio (SMR) of 0.99 (95% CI 0.57-1.60). The emergency readmission rate within 48 hours after PICU discharge was 0.9%. There were 1.8 unplanned extubations per 100 invasive ventilation days. Other paediatric intensive care audit networks reported similar adjusted mortality rates but lower rates of unplanned extubations. Thirty days after PICU discharge, 653 (84.36%) patients were alive and outside hospital, 98 (12.66%) were inpatients, six (0.78%) had died, two (0.26%) were lost to the follow-up. We observed a marked peak of infant emergency respiratory admissions in February.

Conclusions

This project explored the possibility of prospective paediatric critical care audit in Latvia by joining an established international network. This allowed direct comparison of outcomes between the countries. Excess mortality was not observed during one-year data collection period, however a high rate of unplanned extubations was revealed. The results allowed a better planning of elective patient flow by spreading elective cases over the week to avoid "rush hours".

Use of power-law analysis to predict abuse or diversion of prescribed medications: proof-of-concept mathematical exploration

OBJECTIVE

To conduct a proof-of-concept study comparing Lorenz-curve analysis (LCA) with power-law (exponential function) analysis (PLA), by applying segmented regression modeling to 1-year prescription claims data for three medications-alprazolam, opioids, and gabapentin-to predict abuse and/or diversion using power-law zone (PLZ) classification.

RESULTS

In 1-year baseline observation, patients classified into the top PLZ groups (PLGs) were demographically and diagnostically similar to those in Lorenz-1 (top 1% of utilizers) and Lorenz-25 (top 25%). For prediction of follow-up (6-month post-baseline) Lorenz-1 use of alprazolam and opioids (i.e., potential abuse/diversion), PLA had somewhat lower sensitivity compared with LCA (83.5-95.4% vs. 99.5-99.9%, respectively) but better specificity (98.2-98.8% vs. 75.5%) and much better positive predictive value (PPV; 34.5-45.3% vs. 4.0-4.6%). Of top-PLG alprazolam- and opioid-treated patients, respectively, 20.7 and 9.9% developed incident (new) Lorenz-1 in followup, compared with < 3% of Lorenz-25 patients. For gabapentin, neither PLA nor LCA predicted incident Lorenz-1 (PPV = 0.0-1.4%). For all three medications, PLA sensitivity for follow-up hospitalization was < 5%, but specificity was better for PLA (97.3-99.2%) than for LCA (74.3-75.4%). PLA better identified patients at risk of future controlled substance abuse/diversion than did LCA, but the technique needs refinement before widespread use.