Towards improved trauma care outcomes in India : studies of rates, trends and causes of mortality in urban Indian university hospitals

Abstract

Introduction and aim: Injury is a serious threat to global public health. Every six seconds someone in the world dies as a result of injury, adding up to five million people a year. This is more than the number of deaths due to HIV/AIDS, tuberculosis, malaria and maternal deaths combined. Injury is the top killer among the youth (aged 15-29 years), usually male who are physically fit individuals in their economically productive years. About 90% of all injury deaths occur in Low-and-Middle Income Countries (LMICs). Two million lives could be saved annually if the injury mortality rates in LMICs were reduced to the same level as in High-Income Countries (HICs). This would require implementation of robust injury prevention policies and improved post-injury care within hospitals. In India, injury kills one million people every year. More than half of these patients reach hospitals alive. There is a paucity of data on trauma care outcomes of the injured within Indian hospitals. The aim of this thesis was to explore the rates, trends and causes of in-hospital trauma mortality in urban university hospitals in India.

Methods: Four studies were conducted in urban university hospitals in India. Study I was a retrospective analysis of 24-hour in-hospital trauma mortality using three cohorts of admitted patients (1998, 2002, 2011) at a single hospital. Studies II-IV were prospective analyses of 30-day in-hospital trauma mortality in four hospitals. The variables collected by trained data collectors were mechanism of injury, transfer status, vital signs, injury to arrival time, arrival to investigation time, injury description by clinical, investigation and operative findings. Study IV used Delphi methods to define optimal trauma care within the urban university hospital context and peer review to evaluate each death for preventability. All patients were stratified by injury severity using the Injury Severity Score (ISS) into mild (1-8), moderate (9-15), severe (16-25), profound (26-75) ISS categories and by time to death into early (within 24 hours), delayed (between 24 hours and 7 days) or late mortality (between 8 and 30 days of in-hospital stay).

Results: A declining trend of 24-hour in-hospital mortality was observed in an urban Indian university hospital between the years 1998 and 2015 (I,II). The 30-day mortality rate was 21.4% among all trauma patients admitted to the studied hospitals (II). Simple physiological scoring systems using on-admission vital-signs were comparable in performance to more complex anatomical scoring systems in predicting mortality (II,III). All assessed trauma scoring systems predicted 24-hour early mortality better than 30-day late mortality (III). It is likely that 58% of all trauma deaths in studied hospitals were preventable and two-thirds of all deaths in mild or moderately injured patients with an ISS<16. Issues with airway management (14.3%) and resuscitation with haemorrhage control (16.3%) were identified as contributors to early mortality. Traumatic brain injury and burns accounted for the majority of non-preventable deaths (IV). System-related issues were a lack of protocols, lack of adherence to protocols, prehospital delays and delays in imaging (II,IV).

Conclusions: One in five trauma patients admitted to the urban university hospitals in India dies within 30-days and this rate is at least twice the mortality rate in HIC hospitals (II). The longitudinal trend in early in-hospital mortality shows a decline over 18 years (I,II). More than half of all in-hospital trauma deaths were preventable (IV). The steps towards improved trauma care outcomes are triage using vital signs (II,III), improved airway management, early haemorrhage control and resuscitation, establishing treatment protocols (IV), maintaining a trauma registry (II) and timely delivery of trauma care (II,IV). More research is needed to understand the causes of late mortality in trauma patients (IV).