Table 3.
Quality of the included concordance studies based on the Quality Assessment of Diagnostic Accuracy Studies 2 Tool.
| Author and year | Risk of bias | Applicability concerns | ||||||
|
|
Patient selection | Index test | Reference standard | Flow and timing | Patient selection | Index test | Reference standard | |
| Abboud et al [57], 2005 | Low | Low | Low | Low | Low | Low | Low | |
| Dias et al [56], 2023 | Low | Low | Low | Low | Low | Low | Low | |
| Wang et al [55], 2022 | Low | Low | Low | Low | Low | Low | Low | |
| Bovonratwet et al [58], 2022 | Higha | Low | Unclear | Low | Higha | Low | Low | |
| Bradley et al [47], 2021 | Highb | Low | Low | Low | Low | Low | Low | |
| Cottrell et al [59], 2018 | Highc | Low | Low | Low | Low | Low | Low | |
| Demaerschalk et al [48], 2022 | Low | Low | Unclear | Low | Low | Low | Low | |
| Exposto et al [42], 2022 | Highb | Low | Low | Low | Low | Low | Low | |
| Lade et al [49], 2012 | Unclear | Low | Highd | Low | Low | Low | Low | |
| Lovo et al [41], 2022 | Unclear | Low | Low | Low | Low | Low | Low | |
| Melnick et al [65], 2023 | Low | Low | Highe | Low | Low | Low | Low | |
| Michaelson et al [62], 2023 | Low | Low | Highe | Unclear | Low | Unclear | Low | |
| Peterson et al [63], 2019 | Highf,g | Low | Low | Low | Highg | Low | Low | |
| Rabin et al [61], 2022 | Low | Unclear | Unclear | Low | Highh | Low | Low | |
| Richardson et al [50], 2017 | Low | Low | Low | Low | Highi | Low | Low | |
| Russell et al [52], 2010 | Unclear | Low | Highd | Low | Highi | Low | Low | |
| Russell et al [51], 2010 | Unclear | Low | Low | Low | Highi | Low | Low | |
| Steele et al [53], 2012 | Unclear | Low | Highd | Low | Highi | Low | Low | |
| Tachakra et al [54], 2022 | Unclear | Low | Highj | Low | Low | Low | Low | |
| Turner et al [60], 2019 | Low | Low | Low | Low | Unclear | Low | Low | |
| Worboys et al [64], 2018 | Low | Low | Low | Low | Highk | Highk | Highk | |
aPatients who did not have a specific diagnosis and treatment plan during the remote assessment were excluded, which could increase agreement by including only patients who could be easily assessed remotely.
bCase-control design was used in the study.
cOnly a convenience sampling was used for recruitment.
dAssessments were conducted by physiotherapy students, which could reduce the validity of the diagnoses made by evaluators.
eDuring the in-person assessment, evaluators had access to the treatment plan proposed after the remote assessment.
fOnly a convenience sampling was used for recruitment.
gPatients with surgical history of the lumbosacral spine, visible deformity, or no reproduction of symptoms with certain orthopedic tests were excluded.
hAlmost half of the patients (22/47, 47%) were evaluated during a postoperative or nonoperative follow-up consultation, which could increase agreement on diagnosis and treatment plan.
iThis cohort was composed of younger patients recruited in the university clinic with only acute or subacute injuries, without degenerative pathologies.
jThe same evaluator conducted the remote and in-person assessments.
kInitial assessments were excluded, and a third party was present with the patient to assist with data collection, which could have increased the agreement.