Correctly diagnosing death

Published Date: 07th October 2016

Publication Authors: Webber A, ,

​Introduction

Around 50% of deaths in the UK occur in hospital, giving a total of around 250 000 in 2013 [1]. The responsibility of confirmation and documentation of these deaths rests across all specialties of hospital doctor. The Academy of Medical Royal Colleges (AMRC) released a consensus guideline in 2008 aiming to standardise this process [2]. Guidance is available for the diagnosis of death following cardiorespiratory arrest and death following cessation of brainstem function. The diagnosis should be transparent, consistent and instil public confidence in the process. This audit took a retrospective look at current practice and adherence to these guidelines.

Methods

A retrospective case notes examination was carried out between February and August 2015. 98 deaths were observed on Whiston ICU during this period. Doctors carrying out the reviews were not involved in the process of diagnosing death in order not to bias the study findings. Data was collected from entries made in the patient’s case notes and compared with recommended standards set out within AMRC guidelines. These include absence of heart and breath sounds, central pulse and corneal reflex. The patient should have been observed for 5 minutes and the time and date entered in the patient record, along with the name/grade of doctor.

Results

Of the 98 verifications of death examined, the % compliance was as follows:

Heart sounds 89.7, central pulse 82.6, respiratory effort 92.8, observed 5 minutes 9.1 (absent in 35 cases), pain response 72.1, pupillary response 89.7, corneal reflex 6.1, time correct 74.4, date correct 80.6, signed 97.9, printed name 97.9, grade 81.6. In the 63 cases where duration of observation was documented, mean time of was 2.1 minutes. There was only 1 instance of practice falling in line with current guidelines.

Conclusions

Despite AMRC consensus guidelines being available for best practice in diagnosis of death, there remains a great deal of variation in practice for this common procedure. Only 1% of the case notes examined met the guidelines, with the majority not observing the patient for long enough or documenting a corneal reflex. Furthermore, 20% of cases were found to have documented the time/date of death incorrectly. We propose that these wide variations in practice can easily be addressed with the introduction of a standardised proforma to be completed and signed by the clinician at the time of verification.

Webber, A; Ashe, M; Smith, K; Jeanrenaud, P. (2016). Correctly diagnosing death . Critical Care. 20 (Suppl 2), 177-178

« Back