B. Gohy1,2, R. Van den Bergh3, N. Brodin2
1Handicap International - Humanity and Inclusion, Brussels, Belgium, 2Karolinska Institute, Stockholm, Sweden, 3Médecins Sans Frontières, Brussels, Belgium
Background: Care following trauma often focuses on saving life and limbs, while restoring function – though essential – frequently remains overlooked. The Activity Independence Measure for Trauma (AIM-T) is a clinician-rated outcome measure (with a score ranging from 0 to 60) to monitor independence in 12 daily life activities among patients after trauma. It was first developed in Afghanistan, specifically for use in resource-poor settings, and was recently revised after data structure and content validity assessment. Information on its construct validity and reliability are necessary to allow wider use.
Purpose: This study aimed to assess the construct validity and inter-rater reliability of the AIM-T in humanitarian settings.
Methods: This was a cross-sectional study conducted between July and November 2019 in four health facilities run or supported by Médecins Sans Frontières (MSF), in Bujumbura (Burundi), Bangui (Central African Republic), Maroua (Cameroon) and Baghdad (Iraq). To test its construct validity, 195 patients were assessed in parallel with the AIM-T, the Barthel Index (BI) and a Pain score (either Visual Analogue Scale (VAS) or Faces Rating Scale (FRS)). Differences in mean AIM-T scores of groups of patients according to the acuteness and number of injuries were tested using independent t-test. A mean difference of 10 points between patients with acute and post-acute trauma, and 7 points between patients with single and multiple injuries were hypothesised. Correlations of the AIM-T with BI and Pain scores were analysed using Pearson Correlation Coefficient (PCC). A strong and positive correlation (PCC > 0.5) was expected with the BI, while a low and negative correlation with the pain scores (PCC < 0.5). To investigate inter-rater reliability, 84 patients were evaluated independently by a second rater using the AIM-T. The Intraclass Correlation Coefficients (ICC) were calculated for each pair of raters. An ICC of less than 0.5, between 0.5 and 0.75, between 0.75 and 0.9 and greater than 0.9 are indicative of poor, moderate, good and excellent reliability, respectively.
Results: A significant difference of 5.190 was observed between the mean AIM-T scores of patients with acute compared to post-acute trauma and a significant difference of 6.683 between patients with single compared to multiple injuries. Though significant, those differences were below the initial hypotheses. The AIM-T had a low but significant inverse correlation with both pain scores (PCC = -0.264 for VAS and PCC = -0.472 for FRS ) and was positively and strongly correlated with the Barthel Index (PCC = 0.827), confirming the hypotheses. Based on the ICC two-way mixed effects model with a 95% confidence interval, the inter-rater reliability in the four health facilities was good to excellent (ICC between 0.833 to 0.952).
Conclusion(s): The AIM-T is able to discriminate between different types of trauma patients, is consistent with other validated measures, measures the construct of interest (i.e. activity independence), and is reliable in these four humanitarian settings.
Implications: This study further supports the evidence for the use of the AIM-T in humanitarian settings. We recommend its use in clinical practice, as well as for health facilities monitoring or for research purposes.
Funding, acknowledgements: This research was conducted through a R2HC-Elrha grant, supported by Karolinska Institute, MSF and HI.
Keywords: Outcome measure, Trauma, Humanitarian
Topic: Disaster management
Did this work require ethics approval? Yes
Institution: Médecins Sans Frontière
Committee: MSF ERB and National ERBs in CAR, Cameroon, Iraq and Burundi
Ethics number: MSF - ID1893
All authors, affiliations and abstracts have been published as submitted.
