Predictive Ability of Emergency Medical Priority Dispatch System Protocols Should Be Assessed at the Atomic Level of the Determinant Code

Increasing emergency number utilisation is not driven by low-acuity calls: an observational study of 1.5 million emergency calls (2018-2021) from Berlin

BACKGROUND

The Emergency Medical Service (EMS) in Germany is increasingly challenged by strongly rising demand. Speculations about a greater utilisation for minor cases have led to intensive media coverage, but empirical evidence is lacking. We investigated the development of low-acuity calls from 2018 to 2021 in the federal state of Berlin and its correlations with sociodemographic characteristics.

METHODS

We analysed over 1.5 million call documentations including medical dispatch codes, age, location and time using descriptive and inferential statistics and multivariate binary logistic regression. We defined a code list to classify low-acuity calls and merged the dataset with sociodemographic indicators and data on population density.

RESULTS

The number of emergency calls (phone number 112 in Germany) increased by 9.1% from 2018 to 2021; however, the proportion of low-acuity calls did not increase. The regression model shows higher odds of low-acuity for young to medium age groups (especially for age 0-9, OR 1.50 [95% CI 1.45-1.55]; age 10-19, OR 1.77 [95% CI 1.71-1.83]; age 20-29, OR 1.64 [95% CI 1.59-1.68] and age 30-39, OR 1.40 [95% CI 1.37-1.44]; p < 0.001, reference group 80-89) and for females (OR 1.12 [95% CI 1.1-1.13], p < 0.001). Odds were slightly higher for calls from a neighbourhood with lower social status (OR 1.01 per index unit increase [95% CI 1.0-1.01], p < 0.05) and at the weekend (OR 1.02 [95% CI 1.0-1.04, p < 0.05]). No significant association of the call volume with population density was detected.

CONCLUSIONS

This analysis provides valuable new insights into pre-hospital emergency care. Low-acuity calls were not the primary driver of increased EMS utilisation in Berlin. Younger age is the strongest predictor for low-acuity calls in the model. The association with female gender is significant, while socially deprived neighbourhoods play a minor role. No statistically significant differences in call volume between densely and less densely populated regions were detected. The results can inform the EMS in future resource planning.

Dispatchers impression plus Medical Priority Dispatch System reduced dispatch centre times in cases of out of hospital cardiac arrest. Pre-alert--a prospective, cluster randomized trial

AIM OF THE STUDY

Dispatch centre processing times for out-of-hospital cardiac arrest or critically ill patients should be as short as possible. A modified 'pre-alert' dispatch workflow might be able to improve the processing time.

METHODS

Between October 2010 and May 2011 dispatch events, suspicious for cardiac arrest, were prospectively randomized in 24h clusters. The emergency medical service of the intervention group got, based on the dispatchers impression, a 'pre-alert' alarm-message followed by the standard Medical Priority Dispatch System query whereas the control group did not.

RESULTS

In 225 clusters 1500 events were eligible for analysis. Data are presented as median and 25-75 interquartile ranges. Per-protocol analysis demonstrated for the intervention group on 'pre-alert' days a median processing time of 143 s (109-187; n=256) versus 198 s (167-255; n=502) in the control group on non 'pre-alert' days, with a difference of 0.23 log-seconds (p<0.001; 95% CI 0.74-0.28). In critical ill patients, intention-to-treat analysis showed for the intervention group a median of 168 s (131-264; n=153) versus 239 s (176-309; n=164) in the control group, with a difference of 1.4 log-seconds (p<0.001; 95% CI 1.25-1.55).

CONCLUSION

Dispatch times can effectively be reduced in cases of out-of-hospital cardiac arrest or critical ill patients with a 'pre-alert' dispatch workflow in combination with the Medical Priority Dispatch System protocol. This might play an important role in improving patient care.