Comparative reliability of online assessment of elbow and wrist function in civilian and military populations

Sample

The two distinct cohorts were involved in this study: (1) MP, including 25 males from Ukraine (mean age 37.83 years, SD = 6.97) with combat-related upper limb injuries undergoing rehabilitation in Ukraine (UA), and (2) CP of 28 males (mean age 39.33 years SD = 8.85) with noncombatant injuries receiving rehabilitation in Lithuania (LT).

Sample size calculations were done according to the previous studies revealing that the examination of test-retest (on-site vs online) reliability of the instruments required at least 41 patients (two observations, power 80%, α = 0.05, expected ICC 0.5) (Bujang et al., 2017). ⁸ All participants were 18 years or older and provided signed consent priori to their involvement in the study. Table 1 presents characteristics of participants. Inclusion criteria were history of traumatic injuries requiring physiotherapy assessment and treatment. Exclusion criteria were an immediate surgical indication, significant cardiac or neurological disease, limited legal capacity, psychiatric disorders, dementia, or other cognitive impairments.

OPEN IN VIEWER

Table 1.

Characteristics of the study participants.

Variable	CP-LT (n = 28)	MP-UA (n = 25)
Age years (SD)	39.33 (SD = 8.85)	37.83 (SD = 6.97)
Dominant side (frequency (%))
Right	28	12 (48%)
Left	0	13 (52%)
Injured side
Right	19 (67.86%)	16 (64%)
Left	9 (32.14%)	7 (28%)
Both	0	2 (8%)
Localization of the injury
Elbow	15	8
Wrist and hand	11	6
Forearm	2	11

CP-LT: civilian patients from Lithuania; MP-UA: military patients from Ukraine.

Four qualified physiotherapists (two in Lithuania and two in Ukraine) with more than 10 years of working experience in traumatology and musculoskeletal physiotherapy assessment were trained and performed the MMWS and MEPS.

Instruments

The MEPS is a composite rating scale used to assess elbow function across four domains: (1) pain (45 points), (2) stability (10 points), (3) ROM (20 points), and (4) performance of daily functional tasks (25 points). The total score ranges from 0 (indicating worst function) to 100 (indicating normal elbow function). ¹⁰ The original version has been found to have good test-retest reliability (r = 0.82) and validity (r = 0.83). ⁸ Pain, stability, and ROM are each evaluated using a separate ordinal scale. The functional tasks domain is assessed based on the number of predefined activities the participant can perform.

The MMWS is a combined measure with a total score of 100 points, distributed across four domains: (1) pain (25 points), (2) active flexion/extension arc as a percentage of the contralateral side (25 points), (3) grip strength as a percentage of the contralateral side (25 points), and (4) the ability to return to regular employment or daily activities (25 points). Pain is rated by the evaluator based on the patient's subjective report as none (25 points), mild (20 points), moderate (10 points), or severe (0 points). The total MMWS ranges from 0 to 100, with higher scores reflecting better functional outcomes. A score of 90–100 is considered excellent, 80–89 good, 65–79 fair, and less than 65 poor (Green and Brien, 1978).

Testing procedure

Participants who met the eligibility criteria and agreed to participate in the study underwent two consecutive assessments (on-site and online) within a single clinic session with about 15 min rest between the two sessions. The sessions were conducted in the health care setting where the patients were treated. The EZ Read Jamar Goniometer, Patterson Medical, was used to measure the ROM at both locations. For the on-site session, the traditional physiotherapy room was used. The online physical assessments were conducted in the two separate rooms with the chair, desk, and computer using the ZOOM video-conferencing system. In preparation for conducting the online assessments, a pilot test was conducted in both research sites. This allowed physiotherapists from both countries to agree on the position of the videoconferencing platform and the assessment protocol alterations required to perform a physical examination via the Zoom platform. The MMWS and MEPS protocols were completed by physiotherapists on paper. The order of the assessments (on-site and online) was randomly assigned, and each examiner was blinded to the participants’ performance in the other assessment format and to the findings of other examiners. The researcher responsible for data entry was blinded to both the patients and the physiotherapists and entered the data into the designated data platform.

Data analyses

Data were analyzed using the SPSS software package, version 28 (IBM, Armonk, NY, USA).

The normality of the descriptive data, such as the participant demographics and the ROM values, was examined first using the Shapiro–Wilk test prior to conducting any other data analyses. ¹¹ Paired t-tests were used to compare the ROM values assessed by each examiner in the online and on-site formats. A P-value of <0.05 was considered statistically significant. Intraclass correlation coefficients (ICCs) were calculated to test reliability with ICC values based on established criteria: moderate (0.50–0.75), good (>0.75), and excellent (>0.90). ¹² Kappa statistics (κ) were used to assess the level of agreement between categorical ratings beyond chance. This statistic accounts for the proportion of agreement expected to occur randomly, providing a more accurate measure of inter-rater reliability. The strength of agreement, ranging from 0 to 1, is interpreted as follows: 0.00–0.20 = slight; 0.21–0.40 = fair; 0.41–0.60 = moderate; 0.61–0.80 = substantial; and 0.81–1.00 = almost perfect. ¹³ Values of k over 0.40 were considered to be clinically acceptable. The Spearman correlation coefficient between the total scores of MMWS and MEPS in online and on-site formats was used to analyze the association between the two contextual factors affecting joint recovery dynamics.

Reliability between the online and on-site assessment

The MEPS and MMWS have demonstrated acceptable inter and intra-rater reliability of on-site assessment in patients with impaired upper extremity joint functions.^8,14,15 However, none of these studies used MEPS and MMWS in online mode. The recent review reported the validity and reliability of telerehabilitation assessment tools: out of 10 selected studies only one reported outcomes of real-time digital physiotherapy assessment of elbow function. ¹⁶ Researchers employed structured clinical diagnostic assessments (e.g. ROM, special orthopedic maneuvers) to compare examination findings obtained through face-to-face and remote tele-examination. The results indicated >68% agreement between on-site, remote, and repeated remote evaluations, suggesting that remote examination of the elbow is generally valid and reliable for clinical use. The present study demonstrated high inter-rater reliability between on-site and online assessment in both study groups (military and civilian) with ICC exceeding 0.90. These findings are consistent with prior research supporting the reliability of remote assessment in musculoskeletal conditions. For example, Palacin-Marin et al. ¹⁷ reported ICC scores above 0.80 between on-site and telerehabilitation evaluations for 7 of the 9 outcome measures in a small group of individuals with low back pain (n = 15, mean age 37 years). Similarly, Mani et al. ¹⁸ found excellent inter-rater reliability (ICC = 0.93–0.99) between face-to-face and tele-rehabilitation measures across the five assessment outcomes in patients with non-specific neck pain (n = 11, mean age 32.7 years). To our knowledge, the present study is the first to report the reliability of MEPS and MMWS in a synchronous online format for both civilian and military populations. These results support the feasibility and clinical utility of remote upper extremity assessment using standardized instruments and further justify their integration into telerehabilitation protocols for broader accessibility and continuity of care.

Comparative assessment of combat and civilian study groups

Previous studies describing combat and civilian injury treatment have reported significant differences, particularly related to the mechanism of injury (e.g. blasts or ballistic forces), whereas civilian injuries derive more commonly from low- to moderate-energy mechanisms. ² Despite these differences, both cohorts experience a wide range of life-altering functional outcomes, and similar functional assessment instruments are frequently employed in their rehabilitation processes. In the present study, mean pain intensity outcomes were significantly lower in the military cohort from Ukraine, compared to the Lithuanian civilian cohort, indicating fewer pain complaints among combat-injured participants (P = 0.020 for MEPS, both online and on-site; P = 0.014 for MMWS on-site; P = 0.005 for MMWS online). In a recent observational study involving Ukrainian soldiers with extremity trauma, the average pain intensity on the Visual Analogue Scale was approximately 6/10, and neuropathic pain was reported in over half of the patients, indicating substantial symptom burden in this population. ¹⁹ No prior studies have directly compared pain intensity levels between combat and civilian upper-extremity injury patients. However, epidemiological evidence indicates that military veterans reported higher prevalence and severity of pain than civilians of comparable age and sex. For example, in U.S. veterans, the odds of suffering severe pain were approximately 40% higher. ²⁰ Furthermore, experimental pain studies have demonstrated that combat veterans exhibited higher thermal pain thresholds and lower pain sensitivity compared to CPs, suggesting possible alterations in pain perception among military-experienced individuals. ²¹ Moreover, a significant between-groups difference (P < 0.001) was observed in ROM and grip strength with CPs (LT) consistently exhibiting higher functional scores than MP (UA) in both online and on-site assessments. This aligns with military evidence showing that combat injuries frequently result in more complex, multi-structure damage to the upper extremity, leading to more pronounced impairments in ROM and strength post-injury ²²

Association between elbow and wrist function measures

The moderate and non-significant correlation observed in the LT study group between total MEPS and MMWS function scores demonstrated a relatively independent recovery course of these joints. This finding may be attributed to the heterogeneous nature of civilian injuries in this cohort, which likely included isolated or low-energy traumas (e.g. falls, workplace injuries, degenerative conditions), each requiring individualized treatment plans and rehabilitation protocols. Such injury-specific rehabilitation plans may result in variable recovery rates between the two joints, limiting functional interdependence. In contrast, the strong and statistically significant correlations between MEPS and MMWS scores in the Ukrainian cohort (r_s = 0.84–0.88, P < 0.001) that included only combat patients indicated a more unified pattern of upper limb functional recovery. Combat injuries typically involve high-energy trauma, often affecting multiple anatomical regions of the limb simultaneously (e.g. blast or ballistic injuries), leading to complex soft tissue, neurovascular, and skeletal involvement of both the elbow and wrist.^23,24 As a result, these patients may receive more intensive, multidisciplinary rehabilitation that targets the entire upper extremity, facilitating recovery of the elbow and wrist joint functions.

Furthermore, combat-injured patients may exhibit greater rehabilitation motivation and structured follow-up, often coordinated through military medical systems. ²⁵ This function-oriented approach likely promotes coordinated improvements in wrist and elbow function, as reflected in the high inter-joint function correlation coefficients in the study outcomes.