Content area
Abstract
The aim of our study is to review research proposals submitted to the Institutional Review Board (IRB) at one academic medical center to 1) determine the proportion of research protocols that use data exclusively from chart reviews; and 2) assess and quantify the quality of these protocols using published methodologic criteria. Additional exclusion criteria included retrospective studies based on aggregate patient data and computerized databases, case reports and case series, systematic reviews, studies withdrawn by investigator, and those studies categorized as not human subjects research (NHSR). Proportion 95% CI Clear hypothesis or aim 95.2% 91.9–98.5 Confidentiality measures 87.9% 82.9–92.9 Intervention described 87.7% 81.5–94.0 Case selection criteria 79.9% 73.8–86.0 Abstraction forms included 78.2% 71.9–84.5 Diagnostic methods explained 66.1% 57.9–74.3 Comparison group defined 60.0% 48.5–71.5 Medical database identified 54.0% 46.4–61.7 Definition of study variables 49.7% 42.1–57.3 Participating institutions listed 49.4% 41.6–57.2 Sampling method described 45.6% 37.8–53.4 Sample size calculated 35.6% 28.2–43.0 Training of abstractors described 13.4% 8.1–18.7 Monitoring of abstractor performance 8.4% 4.0–12.8 Interrater reliability tested 7.2% 3.1–11.3 Missing-data plan 3.7% 0.8–6.6 Abstractors blind to hypothesis 2.6% 0.1–5.1 Table 1 Retrospective protocols that adhered to methodologic standards.