Exploring Statistical Methodology in Spinal Care Research

Date: July 21, 2023 Reference: McDonald et al. Patterns of change in prehospital spinal motion restriction: a retrospective database review. AEM July 2023 Guest Skeptic: Dr. Chris Bond is an emergency medicine physician and assistant Professor at the University of Calgary. He is also an avid FOAM supporter/producer through various online outlets including TheSGEM. Case: A 42-year-old is struck in the face by a slowly moving I-beam at work. He has a brief loss of consciousness (LOC) and then awakens and is ambulatory on scene. Emergency Medical Services (EMS) is called and on arrival the patient is walking but has obvious facial trauma and is complaining of some neck pain. He has midline neck tenderness but no limb numbness or paresthesia. As an EMS crew member, you are tasked with deciding what method of spinal motion restriction to use. Background: We have covered head injuries including concussions multiple times on the SGEM. This has included looking at the Canadian CT Head Rules/Tools (SGEM#106, SGEM#266, and SGEM#272). We have also covered concussions (SGEM#112, SGEM#331, and SGEM#362). Another core element of emergency department (ED) and pre-hospital care is the assessment for potential spinal injuries [1,2].  Patient care and positioning has evolved over time, previously routine spinal immobilization (SI) was with a cervical spine collar, placement on a long, rigid backboard, and straps or head blocks. Over time this has evolved to spinal motion restriction (SMR) with more variable use of cervical collars, patient positioning, and accessories such as head rolls and tape [3-4]. This has evolved due to recognition of some of the adverse effects of immobilization as well as limitations to its benefits. The role of the cervical collar itself varies by jurisdiction and it is not entirely clear which devices and procedures are most effective at reducing potentially harmful spinal motion [5-11]. Existing research on SMR confirms decreases in the use of long backboards and increases in collar-only treatment [12-14].  Some of this research has observed substantial under-treatment among patients who met criteria for precautions, as well as some patients with confirmed injuries who received no treatment from EMS [15-16]. Other studies have observed no increase in the diagnosis of cervical spine injuries, however, variable practice and the possibility of patients not receiving appropriate treatment remains a concern [17-18]. In order for standards for acute management of spinal injuries to progress, we must optimize patient protection and limit harm [19]. Clinical Question: How has the rate of pre-hospital spinal immobilization/spinal motion restriction changed from 2009 to 2020? Reference: McDonald et al. Patterns of change in prehospital spinal motion restriction: a retrospective database review. AEM July 2023 Population: EMS patients with traumatic injuries Excluded: None Intervention: Spinal immobilization/spinal motion restriction Comparison: This is a retrospective review and includes several changes over time, thus the comparison is the change in rate over time of SMR Important Changes Were: In 2009, selective spinal immobilization using NEXUS criteria was implemented. In July 2012, changing documentation required paramedics to record the indications for SI/SMR in all cases. In November 2014, cases of isolated penetrating trauma were exempted from SMR. In April 2016, treatment guidelines were revised to allow for collar-only treatment in low-risk scenarios (patient ambulatory prior to paramedic arrival) Outcome: Primary Outcome: Rate of spinal immobilization/spinal motion restriction Secondary Outcomes: Rates of splinting and wound care as proxy measures of the incidence of trauma care over time. Patient and practice-related factors associated with potential changes over time. Patient related factors include age, sex, acuity, mechanism of injury and indications for treatment. Practice-related factors included cervical collar size, patient positioning, the proportion of collar only use, rate of treatment of penetrating trauma. Type of Study: Retrospective database review Neil McDonald This is an SGEMHOP and we are pleased to have the lead author on the show. Neil McDonald is an Advanced Care Paramedic in Winnipeg, MB, where he works as a Training Officer and Research Coordinator for the Winnipeg Fire Paramedic Service. He also holds a PhD in Applied Health Sciences and a cross appointment as Lecturer in the Department of Emergency Medicine within the Rady Faculty of Health Sciences at the University of Manitoba.   Authors’ Conclusions: “This study shows decreasing SI/SMR treatment and changing patient and practice characteristics. These patterns of care cannot be attributed solely to formal protocol changes. Similar patterns and their possible explanations should be investigated elsewhere.” Quality Checklist for Observational Study: Did the study address a clearly focused issue? Yes Did the authors use an appropriate method to answer their question? Yes Was the cohort recruited in an acceptable way? Yes Was the exposure accurately measured to minimize bias? Yes Was the outcome accurately measured to minimize bias? Yes Have the authors identified all-important confounding factors? No Was the follow up of subjects complete enough? Yes How precise are the results? Confidence intervals for reported statistics are generally narrow Do you believe the results? Yes Can the results be applied to the local population? Yes Do the results of this study fit with other available evidence? Yes Funding of the Study: University of Manitoba Pamela Hardisty Graduate Fellowship Results: A total of 25,747 patients had SI/SMR with 809 excluded due to incomplete information. The median age was 40 years old, 58% of patients were male and 20% were classified as high acuity. Key Result: The rate of SI/SMR decreased significantly in the 2009-2012 and 2012-2016 time periods, but not in 2016-2020. Primary Outcome: Rate of spinal immobilization/spinal motion restriction In July 2012, paramedics were required to record the indications for SI/SMR in all cases. The change was associated with a significant increase in the rate of SI/SMR of 5.8 treatments per 100 trauma calls (95% CI: 4.6 to 7.1). This then decreased over time until the 2016 protocol change. The 2016 protocol change allowing collar-only treatment was not associated with a significant change in rate of SI/SMR and the final time period. Secondary Outcomes: Neither wound care nor splinting showed any substantial changes over the study period. In terms of patient characteristics, age and sex did not change significantly over time. The proportion of female patients over age 65 decreased by -2.8% per year (95% CI: -4.0 to -1.5%). A significantly higher proportion were high acuity over time, increasing from 11% in 2009 to 31% in 2020, average annual percent change of 9.6% (95% CI: 8.7% to 10.0%) There were smalls decreases in the proportions of falls, MVCs and assaults over the study period, with corresponding increases in non-reporting. Regarding collar size, “no-neck” collars were used more frequently than any other size (65%) but their use decreased over time by -3.8% per year to 49% in 2020. This corresponded to a decrease in “short” collars and an increase in “regular” and “tall” size collars. Prior to the protocol change in 2016 all patients were treated with a collar and board, then immediately after the change 47% of eligible patients were treated with only a cervical collar. This increased by an average of 6.3% per year (95% CI: 3.2% to 9.5%), rising to 60% in 2020. Positioning changes occurred significantly in all categories, with supine positioning decreasing on average 3.1% per year while all others increased. Semi-fowlers positioning increased 47% per year, rising from 0.8% of all patients in 2009 to 25% in 2020. Listen to the SGEM podcast to hear Neil respond to our five nerdy questions. 1.Reporting Data: How do you think the lack of mechanism of injury reporting data (37% of cases) affect the results? 2.Big Drop in SI/SMR: One of the big questions from this study is why was there such a big decrease in SI/SMR over time that cannot be explained by protocol changes alone? What questions do you think future studies need to ask to identify the reason for this drop? 3.Multiple Statistical Analysis Tools: The methods section described several tools used to interrogate your data. Can you comment on the need for each of these tools? 4.Size and Type of Collar: There was a lot of discussion on collar sizes and types changing over time. We were a bit unclear from the study how these changes were decided upon and what this reflects. 5. Does it Matter: Where is the high-quality evidence that SI/SMR provides a net patient-oriented outcome? I’m specifically interested in c-spine collars. Those claiming these devices “work” have the burden of proof to provide evidence to support their position. This has been debated in the pages of AEM with Drs. Serigano and Riscinti give it a colour code of YELLOW(uncertainty) for their NNT blog post based upon the Cochrane 2001 SRMA. In contrast, Drs. Baron and Scalea argue it is not yet time to abandon cervical collars in blunt trauma patients. Where do you currently stand on the issue? Comment on Authors’ Conclusion Compared to SGEM Conclusion: We generally agree with the authors’ conclusions. SGEM Bottom Line: Patterns of spinal motion restriction are changing over time with reduced use of SMR and changing patient and practice characteristics. Case Resolution: You decide to place the patient in a cervical collar and transport lying at 30 degrees in semi-fowlers position,