Article type
Year
Abstract
Background: Systematic reviews of diagnostic test accuracy (DTA) studies are often characterized by heterogeneous results originating from differences in the conduct of included studies. Careful assessment of the study validity is therefore essential.
Objectives: To evaluate the quality of studies for diagnosing age-related macular degeneration and diabetic macular edema using an adapted version of the tool for QUality Assessment of DTA Studies (QUADAS-2), tailored to the needs of this research. Methods: Articles evaluating optical coherence tomography (index test) vs the corresponding reference test were selected. Two reviewers assessed study quality using an adapted version of QUADAS-2, which consists of four key domains: 1) patient selection, 2) index test, 3) reference test, 4) flow and timing. For this project, QUADAS-2 was modified by adding an additional signalling question to consider the unit of analysis issue (domain 5, statistical issues). Each domain was assessed in terms of risk of bias (RoB); the first three domains were assessed in terms of applicability.
Results: A total of 19 articles including 1462 patients were evaluated. The proportion of studies with low/unclear/high RoB was as follows: patient selection (low 37%/unclear 52%/high 11%), index test (low 63%/unclear 32%/high 5%), reference test (low 79%/unclear 21%/high 0%), flow and timing (low 42%/unclear 26%/high 32%), statistical issues (low 32/unclear 0/high 68%). Applicability concerns were distributed as follows: patient selection (low 47/unclear 42/high 11%), index test (low 58%/unclear 0%/high 42%), reference test (low 100%). One of the main concerns was the unit of analysis issue as both eyes or multiple measurements in one eye of one patient were treated as if they were independent in 69% of the included patients.
Conclusions: QUADAS-2 needs to be adapted for medical areas where analyses on paired organs are possible. Another issue is the continuous improvement of the technical development of the index test which raises concerns of the applicability of former models. Overall, authors, editors, and reviewers should pay more attention to rigorous reporting in DTA studies.
Objectives: To evaluate the quality of studies for diagnosing age-related macular degeneration and diabetic macular edema using an adapted version of the tool for QUality Assessment of DTA Studies (QUADAS-2), tailored to the needs of this research. Methods: Articles evaluating optical coherence tomography (index test) vs the corresponding reference test were selected. Two reviewers assessed study quality using an adapted version of QUADAS-2, which consists of four key domains: 1) patient selection, 2) index test, 3) reference test, 4) flow and timing. For this project, QUADAS-2 was modified by adding an additional signalling question to consider the unit of analysis issue (domain 5, statistical issues). Each domain was assessed in terms of risk of bias (RoB); the first three domains were assessed in terms of applicability.
Results: A total of 19 articles including 1462 patients were evaluated. The proportion of studies with low/unclear/high RoB was as follows: patient selection (low 37%/unclear 52%/high 11%), index test (low 63%/unclear 32%/high 5%), reference test (low 79%/unclear 21%/high 0%), flow and timing (low 42%/unclear 26%/high 32%), statistical issues (low 32/unclear 0/high 68%). Applicability concerns were distributed as follows: patient selection (low 47/unclear 42/high 11%), index test (low 58%/unclear 0%/high 42%), reference test (low 100%). One of the main concerns was the unit of analysis issue as both eyes or multiple measurements in one eye of one patient were treated as if they were independent in 69% of the included patients.
Conclusions: QUADAS-2 needs to be adapted for medical areas where analyses on paired organs are possible. Another issue is the continuous improvement of the technical development of the index test which raises concerns of the applicability of former models. Overall, authors, editors, and reviewers should pay more attention to rigorous reporting in DTA studies.