Preventability of injury-related morbidity & mortality at four hospitals in Cameroon: A systematic approach to trauma quality improvement

1 BACKGROUND

Advances in the care of the injured could potentially save millions of lives each year.¹ Quality improvement (QI), the systematic evaluation of care processes and outcomes, reduces trauma morbidity and mortality.² QI programs implementing formal measures of quality and collaborative case reviews have strengthened trauma systems and led to improved outcomes in a variety of settings around the world.^3-5

In 2019, injury caused 10% of all deaths in Cameroon, a higher proportion than the average for global low- and middle-income countries (LMICs).⁶ The analysis of Cameroon trauma registry data drawn from multiple hospitals since 2015 displayed elevated rates of trauma-related mortality, particularly within the initial phases of care.^{7, 8} These findings have been corroborated by single-center studies from referral centers in the country's largest cities.^{9, 10}

A broad approach to QI includes both quantitative registry analysis and qualitative case evaluation of correctable deficiencies in care.¹¹ Techniques such as preventable death review, in which experts discuss individual cases to evaluate systems features contributing to potentially avoidable adverse outcomes, are valuable for QI efforts in LMICs. Prior work in western Africa and Asia has demonstrated the feasibility of performing trauma death reviews and has been associated with decreases in mortality,^{12, 13} but the overall implementation of trauma QI programs has been rare in LMIC settings, including Cameroon.¹⁴

There is a critical need to assess the applicability of comprehensive trauma QI methods in Cameroon as a means of lessening the country's burden of injury. In this study, we developed and implemented a system for the formal review of adverse events in Cameroonian trauma patients as a first step toward identifying modifiable care deficiencies and identify targets for systems modification.

2 METHODS

3 RESULTS

A total of 43 individual hospital M&M conferences and 7 committee meetings (involving all 4 institutions) were conducted over the 29-month study period. During the 50 total gatherings, 95 adverse events were reviewed (Table 1), including 58 deaths (61%). The other common types of adverse events were delays in diagnosis or treatment (22%) and surgical site infections (17%). The number of cases contributed by each hospital ranged from 19 (20%) to 30 (32%). The demographic profile of patients under review was predominantly male (86%) and age 18–35 years (45%).

Over the study period, 259 deaths were recorded in the trauma registry. Of these, 58 (22%) were formally reviewed (Table 2). The largest proportion (40%) originated from the tertiary referral center (Hospital A). The most common mechanism of injury was road traffic injury (67%), followed by blunt force (14%). The head (60%) and upper and lower extremities (57%) were the most frequently injured anatomical areas. Overall, 34 deaths (59%) were classified as preventable, 12 (21%) as potentially preventable, and the remaining 12 (21%) as not preventable.

Of the 46 preventable or potentially preventable deaths (Table 3), over half (52%) of the mortalities occurred in the ED, with the rest occurring mostly in the inpatient ward (30%). Preventable or potentially preventable deaths were commonly associated with traumatic brain injury (57%), respiratory failure (41%), and hemorrhage (39%). Nearly half (41%) of these patients failed to receive a recommended laboratory test or imaging study, most often because of financial reasons (22%) or medical instability (15%). Similarly, a recommended treatment was not delivered in 28% of the preventable/potentially preventable deaths, for reasons including the cost of treatment and communication failure among team members.

At the conclusion of the committee review, the group summarized the major factors identified during root cause analyses and generated a list of target areas for improvement relevant to each of the included hospitals (Table 4). Key recommendations included the development of emergency protocols, standardization of approaches to referral for higher level of care, and improvements in access to diagnostic imaging.

4 DISCUSSION

In this study, we developed and implemented a system for the formal review of adverse events occurring in injured patients treated at four Cameroonian hospitals. To our knowledge, this study represents the first documented effort at introducing a formal system for trauma QI in Cameroon. The majority of reviewed deaths were adjudicated as being preventable or potentially preventable. The structured analysis of 95 adverse events revealed critical targets for systems modification including the initial management of traumatic hemorrhage, respiratory emergencies, and head injury.

Although multiple adverse event types were evaluated in this study, our analysis focused on trauma mortality to improve the understanding of systemic factors contributing to preventable deaths. This decision was also driven by the lack of local electronic medical recordkeeping and variable quality of medical documentation, which prevented fellows from gathering sufficient data to perform comprehensive analyses of the hundreds of nonfatal adverse events occurring at their hospitals. Nonetheless, the 58 trauma deaths reviewed by the committee represent nearly a quarter of all mortalities recorded in the trauma registry during the study period, a significant proportion given local constraints. Of the deaths, 46 (79%) were classified as either preventable or potentially preventable; this rate was consistent across the four hospital study sites despite considerable variation in catchment and hospital type. Notably, the preventable death rate identified by this study is significantly higher than those established by trauma death reviews in high-income countries, which have ranged from 2% to 31%.^{18, 19} The study also differs from other LMICs, as seen in a review in Ghana of in-hospital trauma deaths where group discussions classified 60% of deaths as preventable or potentially preventable.²⁰

To facilitate root cause analysis, the discussion of each adverse event included the creation of a case narrative which was critical to understanding the clinical patterns and significant system contributors underlying preventable death in the Cameroonian context. Specifically, at all locations, deaths were associated with injuries to the head and the limbs, as were acute abnormalities of breathing or bleeding. Time-sensitive, potentially life-saving interventions such as definitive airway protection or tube thoracostomy were rarely performed. These observations corroborated and elaborated on patterns identified from the quantitative analysis of trauma deaths in the Cameroon trauma registry. A previous study identified variable completion rates of emergency interventions for primary survey abnormalities (i.e., airway, breathing, or circulation problems) among trauma non-survivors.²¹ The narrative-based QI approach employed in this study extended the understanding of current system factors underlying low rates of interventions early in trauma management. Per committee review, one potential target for trauma care improvement includes increasing training of systematic trauma evaluation. Specifically, this review highlighted opportunities for increasing early identification of signs and symptoms of potentially life-threatening injuries such as hemorrhagic shock by junior clinical staff.

A 2021 review of the American College of Surgeons Trauma QI Mortality Reporting System found that the most common location of errors contributing to death in the United States was the intensive care unit, and the individual most commonly involved was the attending physician, leading authors to emphasize the importance of systems-level interventions (e.g., automation and standardization) over individual-level strategies (e.g., education and training) for reducing errors in developed trauma systems.²² In contrast, the majority of preventable or potentially preventable deaths in our study occurred in the ED before patients could be admitted to the hospital. This comparison highlights the importance of tailoring approaches to QI to the local context, where deficiencies early in the care pathway may preclude the occurrence of downstream adverse events.

Interventions targeting both individuals and the care system are likely to be beneficial in the Cameroonian environment. Committee review identified several common patient trajectories of non-survivors, including patients presenting with multiple severe injuries with subsequent rapid deterioration, as well as patients who initially presented as stable and then expired unexpectedly. Critical systems contributors included lack of trauma-related knowledge and shortage of resources. For example, trauma care providers were often simultaneously unaware of how to diagnose a tension pneumothorax and poorly equipped to intervene, even if the correct diagnosis had been made. As a result, the committee recommended greater hospital-level emphasis on trauma training and care standardization through protocols and algorithms, which are relatively cost-effective interventions.²³

These recommendations are consistent with conclusions generated from reviews in other LMIC settings. Frequent causes of preventable death in Ghana included hemorrhage and injuries to the brain and spinal cord. Although recommendations originating from case review must be contextually designed, general methods of improvement for the evaluation and management of injuries have been outlined by the WHO.²⁴ Examples include training courses that consider the availability of local resources or checklists to improve recognition and treatment of life-threatening conditions. The implementation of standardized trauma protocols in settings such as Colombia has been associated with decreased mortality.²⁵ The cases reviewed in this study have underlined the need in Cameroon for context-specific initiatives to increase provider knowledge and strengthen capacity for the management of life-threatening issues in the ED. Work studying the effects of newly introduced trauma training and clinical tools at the trauma registry partner hospitals is currently underway in response to these findings.

The preventable deaths discussed in this study also indicated the importance of structural barriers to trauma care in Cameroon. Acutely injured patients were frequently unable to undergo diagnostic imaging or receive recommended treatment, with systems-level deficiencies cited as contributing factors. For example, ultrasound or computed tomography was not readily accessible outside of daytime hours due to staffing constraints. Financial limitations also served as a significant obstacle to patients receiving appropriate work-up, treatment, or referral to a higher level of care. Prior work has demonstrated that nearly all trauma patients in Cameroon pay out of pocket for emergency care.²⁶ Lack of financial protection especially influenced the preventability of head injury-related deaths, as the cost of imaging proved prohibitive for a large proportion of patients. Furthermore, discharge against medical advice was defined as a distinct category of the adverse event in this study because patients often chose to return home due to inability to afford hospital admission. Evidence regarding financial obstacles to trauma care in Cameroon continues to grow²⁷ as the country awaits the development of government-supported health insurance. Findings from all four hospitals in the study suggest that public policy supporting investment in emergency care is likely to significantly lessen the burden of preventable death in Cameroon.

AUTHOR CONTRIBUTIONS

Dennis J. Zheng: Data curation; formal analysis; investigation; writing – original draft; writing – review & editing. Lidwine Nsen Mbuh: Investigation; project administration; supervision; writing – review & editing. Rasheedat Oke: Data curation; formal analysis; methodology; project administration; supervision; writing – review & editing. Signe Mary Magdalene Tanjong: Investigation; project administration; writing – review & editing. Melissa Carvalho: Conceptualization; data curation; formal analysis; methodology; project administration; supervision; writing – review & editing. Banaken Louis Herman: Investigation; writing – review & editing. Débora Guidam: Investigation; writing – review & editing. Ndiformuche Zikirou Mbengawoh: Investigation; writing – review & editing. René Nlong Mang: Investigation; writing – review & editing. Fanny Nadia Dissak-Delon: Investigation; Supervision; writing – review & editing. S. Ariane Christie: Formal analysis; methodology; supervision; writing – review & editing. Alain Chichom-Mefire: Conceptualization; funding acquisition; investigation; methodology; supervision; writing – review & editing. Catherine Juillard: Conceptualization; formal analysis; funding acquisition; methodology; project administration; supervision; writing – review & editing.

CONFLICT OF INTEREST STATEMENT

The authors declare that they have no conflicts of interest.

ETHICS STATEMENT

This study protocol was approved by the Institutional Review Boards of the University of California Los Angeles and University of Buea, Cameroon.

MEETING PRESENTATION

This work was presented during the International Surgical Week at the 49th World Congress of the International Society of Surgery on August 18, 2022.

STATEMENT OF INFORMED CONSENT

Oral informed consent was obtained from all patients included in this study.