Introduction
“Yellow flags” are psychosocial factors that play an important role in the development of chronic spinal disorders and are predictors of patient-reported outcomes. It is essential that such obstacles to recovery are identified and addressed when administering treatment to patients to ensure better outcomes (1). Physiotherapists typically screen patients through history-taking and clinical reasoning, using the flag system to detect barriers to recovery. This study aimed to: 1. Evaluate spine physiotherapists’ ability to detect the presence or absence of yellow flags as compared with the flag status reported by patients and 2. To assess if physiotherapist's characteristics (age, years of experience and gender) may have an impact on their ability to detect yellow flags.
Methods
From 03/21 to 09/23 we included 218 patients (59±18y, 54% female) referred to inpatient spine physiotherapy from within the tertiary care hospital. Before the start of treatment, patients completed the validated Core Yellow Flags Index (CYFI), which assesses catastrophizing (CAT), depression (DEP), anxiety (ANX) and fear avoidance beliefs (FAB) (2). The treating physiotherapist, blind to the patient’s questionnaire responses, indicated the presence/absence of yellow flags (binary variable: physioYF+ or physioYF-) when completing the Spine Tango Conservative documentation form. Statistical analyses included: t-tests for differences in mean CYFI scores in physioYF+ versus physioYF-, and correlation coefficients for their interrelationships; due to the small sample size a permutation test was used for the statistical analysis; multiple logistic regression to analyse the degree of association between CYFI and the likelihood of being physioYF+; sensitivity and specificity for the ability of physiotherapists with respect to a binarized version of CYFI. Moreover, physiotherapists were divided into two groups based on their agreement or disagreement with the self-assessment. The t-test was used to compare the age and years of experience between these two groups, while the chi-square test was used to compare the gender distribution. The significance was set at p<0.05.
Results
Scores for CAT, DEP, and ANX and the whole CYFI were significantly higher for physioYF+ than physioYF- (each p<0.001) and showed a weak but significant (r=0.16, p=0.02 to r=0.24, p<0.001) correlation with the physioYF+/- rating. FAB was slightly above the significance level (p = 0.051). In multiple logistic regression with all four CYFI items entered simultaneously, none of the predictors was found to be statistically significant. The ability of physiotherapists to detect yellow flags with respect to the binarized CYF had a sensitivity of 0.25 and a specificity of 0.77. No significant differences were found in terms of age, experience, or gender between physiotherapists who were in agreement with the self-assessment (N=142) and those who were not (N=71).
Conclusion
Physiotherapists demonstrated a low sensitivity in detecting yellow flags. However, their high specificity in accurately distinguishing the absence of yellow flags emphasizes the complexity of assessing and classifying psychosocial factors. Drawing conclusions based solely on physiotherapist ratings or patient outcomes may oversimplify the evaluation
process, emphasizing the necessity of considering both elements for a more comprehensive understanding.