Non-technical error leading to patient mortality in the Australian surgical population

Despite an increasing emphasis on patient safety, modern surgical practice is still plagued by the occurrence of serious Adverse Events (AEs). A large proportion of these AEs are caused not by technical errors, but short comings in non-technical skills (NTS) or non-technical errors.1 NTS are the cognitive and interpersonal components of surgical professionalism including communication, decision-making, situational awareness and leadership.2 Few studies have attempted to quantify the significance of non-technical errors leading to patient harm. Those that have, only include small, non-representative cohorts, or have assessed for NTS shortcomings using heterogeneous, non-standardised, and non-comprehensive assessment methods.3-6 As a result, the true impact of non-technical errors, and information about how, when, and why these errors occur are poorly understood. Evidence to guide NTS improvement is lacking.

This study7 aimed to investigate the incidence of non-technical errors linked to patient death in a large representative Australian cohort, investigate factors associated with fatal non-technical errors, and whether the incidence of non-technical errors had changed over time. To achieve this, an 8-year retrospective audit using surgical mortality cases was conducted. Data were derived from the Australian and New Zealand Audit of Surgical Mortality (ANZASM) a mandatory, national surgical mortality audit overseen by the Royal Australasian College of Surgeons. In Australia, every surgical death is reported to ANZASM for external peer review for the purpose of identifying clinical management issues (CMI’s) that could be improved in future. CMI’s are rated on a three-tiered system with ‘area of concern’ or ‘Adverse Event’ representing the two most serious. All surgical deaths between 2012-2019 (excluding New South Wales), flagged with an area of concern or AE were included with no further exclusions. Each case was assessed using the System for Identification and Categorization of Non-technical Errors in Surgical Settings (SICNESS), a validated tool, developed by the study authors.8 The SICNESS enabled assessors to identify if a non-technical error linked to patient death had occurred, and if so, to which NTS domain it belonged.

There were 30,971 surgical deaths reported to ANZASM during the study. Of these, 3829 were flagged with an AE or Area of concern. Some 134 had insufficient information for assessment, leaving 3695 cases for review. A non-technical error linked to patient death was identified in 2354 cases (63.7%). Of the cases with non-technical errors, 1375 (58.4%) had decision making errors, 1328 (56.4%) had situational awareness errors, 357 (15.2%) had communication/teamwork errors, and 128 (5.44%) had leadership errors.

Patients admitted to private hospitals were half as likely to suffer from a fatal communication/teamwork error than those admitted to public hospitals (p=0.04). Electively admitted patients were 1.32 time more likely to experience a fatal decision-making error compared to patients admitted under emergency conditions (p=0.01). Patient age was also a significant factor, with increasing age associated with increased likelihood of fatal decision-making errors but decreased likelihood of fatal situational awareness errors.

Overall, non-technical errors significantly decreased over the study interval but no improvement was demonstrated for the two most common non-technical error types – decision making or situational awareness.

This study demonstrates that inadequate NTS contribute significantly to surgical patient mortality. These results highlight decision making and situational awareness errors as clear NTS improvement priorities. Furthermore, the study provides evidence of the value of retrospective audit for NTS assessment when using a validated and standardised tool.

What is unknown, is how specialty-specific factors impact on the rate of fatal non-technical errors. Each surgical specialty has unique challenges including differences in patient demographics, acuity, risk profile and procedural difficulty. It is possible that a ‘one size fits all’ approach to improvement may not be appropriate and individualised NTS improvement may be warranted. If significant improvements in NTS can be achieved, the impact will not only be on mortality but also morbidity experienced by all surgical patients. Future research is needed to investigate the incidence and characteristics of fatal non-technical errors within specific surgical specialties and to explore clinical problems that are over represented in cases of surgical mortality. In doing so, we can gain a deeper understanding of how, when, and why fatal non-technical errors occur, to guide future resources and improvement efforts with the ultimate aim of reducing patient mortality.