The Use of Movement-Based Tests for the Prediction of Non-Contact Low Back Injuries in Uninjured Military Personnel: A six-month prospective cohort study

• Author / Creator

  Crumback, Daniel J

• Abstract

  Background

  Low back pain (LBP) is the most reported musculoskeletal complaint in the Canadian Armed Forces (CAF). By identifying low-tech movement-based tests that predict risk of future non-contact low back injury (LBI), we could create an easy to administer screening test. It may then be possible to target intervention to high-risk individuals to affect the prevalence of future LBIs.
  The objective of this prospective cohort study was to (1) determine which personal characteristics, medical history, and movement-based test variables could predict non-contact LBI in a six-month follow-up in military personnel without LBI at baseline.
  Methods

  Volunteers without LBI for at least three months at baseline were recruited via multiple recruitment presentations to CAF personnel. At baseline, a standardized questionnaire was used to collect predictors variables and participants completed 19 movement-based tests including; the Functional Movement Screen (FMS), Lower and Upper Quarter Y-Balance Tests (LQYBT and UQYBT), LBP provocation tests, ankle dorsiflexion mobility, low back extensor endurance, side planks, and four spinal mobility tests. LBI was tracked for 6 months using monthly online surveys. An LBI was defined as a non-contact sudden onset or overuse injury to the spine, hip or pelvis causing ≥2/10 pain for at least three days, limiting ability to work/exercise for >24 hours, with self-reported function of ≤90%, and resulting in medical care. Independent t-tests for continuous variables and Chi-square tests for categorical variables were used to identify variables presenting univariate associations with LBI. Receiver operator characteristic (ROC) curves were used to dichotomize promising continuous variables. Dichotomous predictors with an odds ratio of 2.0 or more and with p<0.2 were used to develop a logistic regression model to predict future LBI. A clinical prediction rule was developed by examining the accuracy for presenting any number of predictors retained in the regression.
  Results

  Four hundred ninety-four personnel were enrolled. Data were available on 455 participants (92%): one withdrew, four retired, and 36 had incomplete data. Nineteen participants reported an LBI over the 6-month follow-up.
  The following seven dichotomized movement-based continuous variables presented significant univariate associations with future LBI: UQYBT inferolateral asymmetry 1.5cm, worst LQYBT anterior reach 55cm, LQYBT composite worst score ≤100%, fingertip-to-floor distance 16cm, side plank time asymmetry 8s, Modified Sorensen duration 86.0 seconds and Trunk Stability Push Up score 5.
  Three demographic and medical history categorical variables presented significant univariate associations with future LBI: smoker, more than one LBI episode in the last five years; and perceived low back baseline function score <90%.
  Five pain provocation tests predicted future LBI: side plank, ankle dorsiflexion, trunk stability push up, extension clearance and passive lumbar extension.
  A logistic regression prediction model for LBI was identified by combining five modifiable predictors: baseline perceived lumbar/hip function ≤90%, pain with extension clearance, UQYBT inferolateral asymmetry 1.5cm, side plank time asymmetry 8s and LQYBT composite worst score ≤100%. Using this model, 89.9% of the participants were correctly classified as injured/not injured during the 6-month follow-up.
  Participants with three or more predictors were 6.8 (CI 4.2-11.1) times more likely to have an LBI with 57.9% sensitivity, 91.5% specificity and 90.1% accuracy. Using three or more variables accurately predicted 11 of 19 cases.
  Conclusion

  History of LBI, current level of function, pain provocation and movement testing contributed to predict first episode or recurrent episode of LBI in CAF personnel without LBP at baseline at risk for future LBI. The proposed prediction rule is a moderately sensitive and highly specific test cluster to effectively identify people at higher risk of non-contact LBI. Presenting three or more of the five predictors represents a greater risk of LBI with a high specificity (91.5%). Such cases may benefit from a preventive training program. With two or fewer predictors, specificity decreased (59.8%), it may be impractical to offer preventative programming to the larger number of personnel thus identified. The benefits of early identification of “LBI risk” are potential for decreased costs, use of medical assets, and lost days. Modifying risk, if possible, may increase deployability of military personnel.

• Subjects / Keywords

  • prediction
  • low back
  • lumbar
  • musculoskeletal
  • pain
  • injury
  • tests
  • movement
  • military

• Graduation date

  Fall 2021

• Type of Item

  Thesis

• Degree

  Master of Science

• DOI

  https://doi.org/10.7939/r3-hv0h-gj26

• License

  This thesis is made available by the University of Alberta Libraries with permission of the copyright owner solely for non-commercial purposes. This thesis, or any portion thereof, may not otherwise be copied or reproduced without the written consent of the copyright owner, except to the extent permitted by Canadian copyright law.