Classifying an incident as critical or non-critical
All staff who are involved in or witness an incident, including a staff assault, are required to enter a report on TOMS. Policy Directive 41 – Reporting of Incidents and Additional Notifications (PD41) provides guidelines and procedures for reporting incidents.
In order to record an incident on TOMS the author must first decide if the incident was critical or non-critical. Critical incidents are subject to additional reporting notifications not required of non-critical incidents. Immediate verbal advice must be provided to a number of people within the Department and other preliminary reports must be provided within an hour.
At the time of the review, PD41 referred to critical incidents as any event included in a list of provided examples. In addition to a serious assault, critical incidents included events such as escapes and attempted escapes, serious self-harm, attempted suicide incidents, security equipment being misplaced or stolen, bodily fluid contact, bomb threats, dangerous occurrences like natural disasters, and death. Between September 2010 and the end of 2012, 84 staff assault incidents were recorded as critical.
Non-critical incidents were described in PD41 as events that may result in charges being laid or a loss of privileges being applied to a prisoner. Non-critical incidents included: fighting, threatening behaviour, locating prohibited items, property damage, the discovery of unknown substances, and non-serious assaults. Between September 2010 and the end of 2012, 125 staff assault incidents were recorded as non-critical.
Classifying the type of assault: PD41 prior to 14 January 2014
During the period of this review, section 6.3 of PD41 stated that it was necessary to categorise all incidents accurately and specifically referred to a three-fold categorisation: ‘serious assaults’, ‘assaults’ and ‘other assaults’. However PD41 only defined ‘serious assaults’ and ‘non-serious assaults’ by way of examples and did not attempt define ‘other assaults’.
On the surface, PD41 aligned with the TOMS definitions (see table 1). However, there were discrepancies.
Table 1: Categories for assaults
Discrepancies may occur when the assault involved bodily-fluid to bodily-fluid contact. For example, if a prisoner spits and it came in contact with a staff member’s eye or mouth, PD41 categorised that event as a critical incident and an ‘other’ assault. Yet the TOMS definition of other assault was when no physical injuries occurred or the injuries did not require any medical treatment. This did not align with the ongoing medical treatment that is required with repeat blood borne virus testing which occurs months after the assault. Adding to this, further treatment may be required depending on the test results. An assault requiring ongoing medical treatment would otherwise be categorised as serious assault. Consequently defining and recording bodily-fluid to bodily-fluid contact was unnecessarily complicated.
It is not surprising that this confusion led to a significant number of data entry errors being detected during the review which included:
- incorrectly classifying an incident as either critical or non-critical;
- allocating the wrong assault type to the event; and
- inappropriately marrying the incident category with the wrong assault type.
Of the 84 staff assaults recorded as critical incidents, only 17 (20%) were correctly classified as critical incidents based on the information contained in the reports relating to the incident. The information regarding two assaults was wholly unknown from the reports and the remaining 65 should have been categorised as non-critical incidents according to the terms of PD41.
The 84 critical incidents were categorised as 12 serious assaults, 65 assaults and seven other assaults. From the information provided in the incident reports it appears that these should have been categorised as only two serious assaults, with 49 being assaults and 31 being other assaults.
Table 2: Incident reports classified as per PD41
Table 2 shows critical incidents and their recorded assault types compared to a reassessment based on the incident records and PD41 and departmental definitions. The highlighted figures represent those assault types which were categorised correctly. Of the 49 records correctly assigned an assault category, 46 were incorrectly logged as a critical incident, leaving only three out of 84 which were correctly classified as a critical incident and allocated the correct assault category.
It is the responsibility of the supervisor of the author of the incident report to review the report. This includes whether the incident is classified as critical or non-critical and the type of assault is correctly allocated. These findings demonstrate this quality assurance process is not working.