No CrossRef data available.
Published online by Cambridge University Press: 23 July 2021
Dispatchers should be trained to interrogate bystanders with strict protocols to elicit information focused on recognizing cardiac arrest and should provide telephone cardiopulmonary resuscitation (CPR) instructions in all cases of suspected cardiac arrest. While an objective assessment of training outcomes is needed, there is no performance assessment scale for simulated dispatcher-assisted CPR.
The aim of the study was to create a valid and reliable performance assessment scale for simulated dispatcher-assisted CPR.
In this prospective, randomized, controlled, multi-centric simulation-based trial (registration number TCTR20210130002), the scale was developed according to the European Resuscitation Council (ERC) and American Heart Association (AHA) Guidelines 2015 and revised by experts. The performance of 48 dispatchers’ telephone-CPR and of 48 bystanders carrying out CPR on a manikin was assessed by two independent evaluators using the scale and using a SkillReporter (PC) software to provide CPR objective performance. Continuous variables were described as mean (SD) and categorical variables as numbers and percentage (%). Comparative analysis between two groups used a Student t-test or a non-parametric test of Mann-Whitney. The internal structure of the scale was evaluated, including internal consistency using α Cronbach coefficient, and reproducibility using intraclass correlation coefficient (ICC) and linear correlation coefficient (R2) calculation.
The scale included three different parts: two sections for dispatchers’ (32 items) and bystanders’ CPR performance (15 items) assessment, and a third part recording times. There was excellent internal consistency (α Cronbach coefficient = 0.77) and reproducibility (ICC = 0.93; R² = 0.86). For dispatchers’ performance assessment, α Cronbach coefficient = 0.76; ICC = 0.91; R2 = 0.84. For bystanders’ performance assessment, α Cronbach coefficient = 0.75; ICC = 0.93; R2 = 0.87. Reproducibility was excellent for nine items, good for 19 items, and moderate for 19 items. No item had poor reproducibility. There was no significant difference between dispatch doctors’ and medical dispatch assistants’ performances (33.0 [SD = 4.7] versus 32.3 [SD = 3.2] out of 52, respectively; P = .70) or between trained and untrained bystanders to follow the instructions (14.3 [SD = 2.0] versus 13.9 [SD = 1.8], respectively; P = .64). Objective performance (%) was significantly higher for trained bystanders than for untrained bystanders (67.4 [SD = 14.5] versus 50.6 [SD = 19.3], respectively; P = .03).
The scale was valid and reliable to assess performance for simulated dispatcher-assisted CPR. To the authors’ knowledge, no other valid performance tool currently exists. It could be used in simulated telephone-CPR training programs to improve performance.