Validity and Reliability of Video-Based Functional Physiotherapy Assessment of Subacute Knee Injuries

BACKGROUND Knee pain is one of the most common complaints in primary care. Traditional knee assessments include history-taking, evaluation of joint mobility and muscle strength, and specific tests targeting structures such as ligaments and menisci. However, functional tests-such as chair rise or single-leg stance-evaluate the integrated performance of multiple structures and may be more feasible in digital consultations.

Existing literature shows variable validity for specific knee tests, with only a few reaching acceptable thresholds for sensitivity and specificity. In video-based assessments, joint mobility evaluation is reliable, and functional tests are recommended over strength tests. While some knee tests are suggested for digital use, their measurement properties as self-administered tests under physiotherapist supervision remain unclear.

This study will investigate whether a function-based assessment without specific knee tests can provide sufficient diagnostic information and guide treatment decisions for subacute knee injuries. It will also compare the validity and reliability of digital versus in-person physiotherapy assessments in primary care.

METHODS Participants will be consecutively recruited from public and private rehabilitation clinics in the Västra Götaland region of Sweden. Individuals with subacute knee pain (onset between 1 week and 3 months) will be informed about the study through clinics, websites, and online booking systems. Interested individuals will complete a digital screening form to assess eligibility. Those who meet the inclusion criteria will receive written study information and will be able to provide informed consent digitally via Bank-ID.

Demographic data will be collected through the same form, including age, gender, origin, symptom experience, pain intensity and duration, prior video consultation experience, perceived balance, and estimated travel distance to the clinic. This information will support analysis of environmental and accessibility factors.

Eligible participants will be contacted to schedule two knee assessments-one in-person and one via video. Each visit is expected to last approximately 30 minutes, with additional time allocated for preparation and documentation by the physiotherapists.

Inclusion criteria will include age ≥18, subacute knee symptoms, ability to bear weight and flex the knee, no tenderness over patella or fibula, sufficient Swedish language skills, and access to digital technology.

Exclusion criteria will include prior assessment for current symptoms, recent knee or hip surgery, and inability to walk independently.

Sample size calculations indicate that at least 22 participants per diagnostic category (total n=110) will be needed to detect a kappa coefficient of 0.60. With an estimated 20% dropout rate, the target sample size will be 138. Recruitment will continue until minimum group sizes are met or for up to one year.

The study will evaluate the validity and reliability of function-based knee assessments without specific tests, comparing in-person and video-based formats.

Participants will undergo two consecutive knee assessments at a rehabilitation clinic. The first will be a video-based evaluation, including history-taking and functional tests, conducted in real time with the physiotherapist in a separate room. The second will be a traditional in-person assessment, including both functional and specific knee tests. Physiotherapists will use standardized protocols and receive training prior to data collection.

Data collection Participant background data (e.g., age, gender, pain intensity, duration, prior video visit experience, and travel distance) will be collected via a secure web form after consent. Each participant will undergo two knee assessments-first via video, then in person-conducted by the same physiotherapist. Diagnostic agreement between methods will be evaluated, along with physiotherapist confidence and patient satisfaction. Standardized tools include NRS for pain and KOOS-12 for function and quality of life. Secondary outcomes include adverse events, need for further testing or referral, and time required per visit.

Statistical Analysis:

Descriptive statistics will include means, medians, standard deviations, and quartiles. The in-person assessment with specific knee tests will serve as the gold standard. Diagnostic agreement between methods will be analyzed using Cohen's kappa, which will also be used to assess intra-rater agreement. Sensitivity and specificity will be calculated to evaluate the accuracy of assessments without specific tests. Confidence intervals for kappa values will be reported.

Differences in diagnostic certainty level will be analyzed using paired t-test if the data is normally distributed, and wilcoxon signed-rank test if using non-parametric test.

• Degree of diagnostic certainty [Time frame: Day 1]
• Number of Participants with Agreement on Specific Diagnostic Categories [Time frame: Day 1]
• Patient-Rated Satisfaction With the Assessment [Time frame: Day 1]
• Adverse events [Time frame: Day 1]
• Number of Participants Requiring Additional Specific Knee Tests Following Assessment [Time frame: Day 1]
• Number of Participants Requiring Supplementary In-Person Examinations After Video-Based Assessment [Time frame: Day 1]
• Number of Participants Referred to a Physician Following Assessment [Time frame: Day 1]
• Time required for each visit [Time frame: Day 1]
• Reliability of 30 seconds chair stand test [Time frame: Day 1]
• Reliability of Joint Mobility Assessment [Time frame: Day 1]
• Reliability of Sit-to-Stand Ability From Chair [Time frame: Day 1]
• Reliability of Lunge Performance [Time frame: Day 1]