Patients can accurately self-report their elbow range of motion following surgery for trauma

Abstract

Introduction

There is an increased demand of telemedicine in the recent century, especially with the outbreak of Covid-19. The aim of this study was to investigate patients' reliability in self-assessing own elbow range of motion following surgery for trauma.

Methods

All patients of age ≥16 years who underwent surgery for elbow trauma at the local trauma unit between March 2015 to Aug 2018 were reviewed retrospectively. Identified cohort was invited to self-assess their elbow range of motion (ROM) using questionnaire with image instruction. They were then followed up with a clinical review for objective measurements by the lead clinician. Independent T-test was used to compare the measurements between patients and clinician. The power of the study was calculated using G∗Power software.

Results

Thirty-five patients were enrolled in the study with mean age of 41 years. 11 of 35 patients had an associated elbow fracture dislocation associated. Mean patient reported total ROM was 105.7° ± 32.8°, with mean extension of 24.6° ± 18.9° to mean flexion of 130.3° ± 18.2°. Mean objective ROM measured by lead clinician was 112.6° ± 18.3°, with mean extension 22.4° ± 10.9° to mean flexion 135.0° ± 10.8°. No statistical significance was found between self-reported and clinician-based extension (p = 0.36), flexion (p = 0.076), and overall range of motion (p = 0.12).

Conclusion

Patients can self-assess their elbow range of motion following surgery for trauma accurately. In the midst of increasing demands for telemedicine, we suggest the application of patients' self-reporting outcome in clinical settings.

1. Introduction

Outcomes in healthcare was first described by Ernest Codman in 1913 who laid the foundation of follow-up of patients to ascertain diagnostic and treatment complications in an effort to improve quality and drive research. Usage of subjective patient-reported outcome measures (PROMs) has shown an increasing trend as valid and reproducible method in monitoring post intervention outcome.^{1, 2, 3} The American Academy of Orthopaedic Surgeons (AAOS), in its position paper ‘Public Reporting of Provider Performance’ states ‘the systems for measuring and reporting quality in health care should continue to evolve and expand. The current generation of quality measures, which primarily rely on process measures and administrative data, have not yet been proven to accurately correlate with improved functional outcomes, which are the primary outcomes of interest to patients who undergo orthopedic procedures’.⁴

Several outcome measures in orthopedics still rely, in part, on physician-based assessments. The burden of patients being asked to attend for clinical evaluation can be high and for the purposes of research, which lead to non-attendance and attrition of participants. A few studies have assessed whether patients can themselves report their range of motion for a variety of joints and pathologies with mixed conclusions. Schnetzke et al., 2016 developed elbow motion assessment score (E-MAS) and shown good reliability and patient-physician agreement.⁵ On the other hand, Li et al. evaluated the accuracy of patient-reported range of motion (ROM) compared to clinician measured ROM, and concluded a great disparity between both.⁶

With the increasing trend of telemedicine and virtual clinic to sustain healthcare demands, the ability to assess outcomes through patients’ perspectives is important. In view of the disparity of literature, we aimed to compare patient-reported elbow ROM and clinician geometer measurement of elbow ROM post-intervention for trauma pathology using a more robust statistical analysis.

2. Materials and methods

Patients sustaining trauma to the elbow requiring surgical intervention between March 2015 and August 2018 were retrospectively reviewed. Patient demographics including date of birth, gender, underlying trauma, and surgery performed were obtained. Exclusion criteria included any patient aged less than 16 years old, injury sustained less than 1 year period, wearing devices restricting movements such as casts or splints, and underlying neurological or cognitive disorder.

Identified patients were requested to fill in a questionnaire and to self-assess the range of motion of their elbow. To self-assess their range of motion, patients were instructed to hold their shoulder at 90° of abduction and forearm in maximum possible supination as there may be patients who have restriction of forearm movements due to associated proximal radio-ulnar injuries. They were then instructed to maximally extend and flex their elbow. They were then asked to record these maximal values where the range of motion could then be derived. A series of pictures of a control subject were included with the questionnaire. A picture of the subject with the shoulder at 90-degrees of abduction and the elbow in full extension followed by flexion in 30-degree increments (measured with a goniometer) to a maximal value of 145° were displayed to act a visual representative of accurate degrees of movement (Fig. 1).

Fig. 1.

Pictures of control subject to aid self-reported measurements.

Patients who returned their questionnaire were invited for clinical review for objective range of motion measurements. The lead clinician (VP) took measurements using a 12-inch goniometer. In flexion and extension, the stationary arm of the goniometer was positioned in line with the mid-humeral line with the rotating axis on the lateral epicondyle, and the mobile arm was positioned on the mid-forearm line. The second clinician (HHC) was responsible in gathering the patients' questionnaires so that the lead clinician could be blinded prior to the measurements.

All statistical analyses were conducted using SPSS software (version 26.0; SPSS Inc., Chicago, II, USA). Descriptive (mean±standard deviation (SD)) and 95% confidence intervals (CI) were calculated for continuous variables. Independent T-test was used to compare self-reported and clinician-based measurements. The sample size calculation was performed using G∗Power 3.1.9.6 statistical power analysis software. We used the findings of Li et al., 2016 to calculate the required sample size.⁶ The type 1 and 2 errors were set at 0.05 and 80%, a p-value of <0.05 was considered to be significant. Our power analysis suggested that finding a significant difference in measurement would require a total sample size of 34 subjects.

3. Results

A total of 63 patients were identified and invited to fill in a questionnaire. 35 patients (56%) returned their questionnaire and agreed to return for clinical evaluation of their elbow. The average patient age was 41 years (range 26–87), 20 (57%) of whom were female. 11 of the 35 had an elbow dislocation associated with a radial head fracture. 5 of these also had an associated coronoid fracture. The remaining 24 fractures of the radial head were in isolation. Five of the patients with an elbow dislocation were treated with fixation of the radial head and ligament repair, five patients had a radial head replacement and the remaining one had a radial head excision. Of the patients presenting with an isolated radial head fracture, 10 had excision, 6 had fixation and 8 had a replacement. Injuries and treatments are summarized in Table 1.

Table 1.

Injuries and treatment by category.

Elbow dislocation with radial head fracture (11)	Isolated radial head fracture (24)
Radial head fixation (5)	Radial head fixation (6)
Radial head replacement (5)	Radial head replacement (8)
Radial head excision (1)	Radial head excision (10)

The subjective and objective range of motion of the elbows were recorded. Normal distributions for extension, flexion and overall range of motion were observed for both subjective and objective measurements. Skewness measurements were between −0.64 and +0.15. Mean values for subjective and objective extension and flexion was assessed (summarized in Table 2). Measurements favored scores done by a clinician with mean values of self-reported extension being 24.6° and flexion 130.6° whereas the mean clinical measurement of extension being 22.4° and flexion 135°. Independent T-test was used to correlate data sets for self-reported and clinician-based extension, flexion, and overall range of motion. No significance between means was found within all groups (flexion p = 0.076, extension p = 0.36 and range of motion p = 0.12).

Table 2.

Averages of patient reported, and clinician assessed range of motion.

	Patient Extension	Clinical Extension	Patient Flexion	Clinician Flexion	Patient ROM	Clinician ROM<
Mean	24.60	22.43	130.30	135.00	105.70	112.57
SD	18.87	10.94	18.17	10.78	32.78	18.29
Kurtosis	−0.47	−0.56	−0.06	−0.57	0.22	−1.32
Skewness	0.15	0.03	−0.10	−0.11	−0.64	0.03
Independent T-test, p =	0.36		0.076		0.12

ROM, range of motion; SD, standard deviation; Significance, p < 0.05.

4. Discussion

Our study confirms patients can accurately record their elbow range of motion following surgery for trauma with a moderate internal consistency compared to the gold standard clinician-based measurements. Although patients tended to underestimate their range of motion, this was not statistically significant compared to objective measurements.

The significance of this finding is of particular benefit in the current increasing trend for medical teleconferencing. Our findings should reassure the clinician and empower patients to provide their elbow range of motion. Patients' ability to reliably quantify their elbow range of motion permits outcome measurements to be employed for research purposes without the requirement for patient contact, in addition to clinical implications. The financial and logistical challenges in face-to-face contact can therefore be overcome and the attrition of patients in a research setting is likely lowered. Another possible advantage is patients self-reporting their elbow range of motion is not subject to the effect of a present physician. In our opinion, patients are likely to report their functional, pain-free range of motion when not influenced by a physician, whereas patients may feel they need to reassure the clinician by displaying a range of motion not usual or comfortable for them in normal circumstances. This may explain the tendency of the results obtained in our study.

Self-reported range of motion in orthopedics has already been widely studied in post total knee arthroplasty with several studies showing good reliability between patients self-reported ROM and clinician observed ROM.^{7, 8, 9} Several studies have similarly reported good agreement in elbow range of motion. Schnetzke et al.⁵ published their results for a validated questionnaire for self-determination of wrist and elbow ROM. Their heterogenous group of patients with a mixture of humeral, elbow and wrist pathologies showed a high level of patient-physician agreement. Their similar method of using pictures of a control subject with incremental movements was employed. The potential weakness of the patient selecting the pictures best representing their movement was addressed in our study. Our pictures were used only as a reference for which the patient could assess their movement in a continuous rather than ordinal way which is representative of how objective measurements are taken. The results of our study add to this body of evidence and even raises the questions of whether attitudes in objective measurements are dogmatic. Just as patient reported outcome measures are of more importance than clinician-based scores, so to may this be the case for patient reported range of motion, as a patient's perceived range of motion, or lack of, would be of more interest to the patient themselves.

There are some limitations to our study. The group of patients assessed was homogeneous, with all the patients having an elbow injury that had to be operated on. As a result, extrapolating our findings to other situations, particularly those of a degenerate nature, is questionable. Secondly, there was a timeline difference between the patient-reporting questionnaire and clinician-measurement in a clinical appointment. This may affect the measurement taken at two different settings as the elbow pathology might have evolved by the time of clinical review. However, we believe the risk of skewness is low in view of the time of injury sustained to our assessment period is more than 1 year period. Thirdly, the lack of an assessment of inter-observer and intra-observer variability may have an impact on the weighting of the measured observations. Lastly, we did not assess the supination and pronation movement in our sample group nor have a ‘normal’ control group as a benchmark.

We conclude that our analysis supports the use of patient self-reported visual ROM as a potential outcome measure in a virtual clinic setting, especially during the reign of a worldwide pandemic. This will not only save patients from the need of travelling for a clinic appointment, but also reduce the financial burden and logistic need of current healthcare service towards virtual clinic review. We recommend future study to concentrate on patient's perspective of self-reported functional outcome in elbow pathology.