An analysis of data quality and population characteristics of the SIGN online surgical database: A 20 year review

Abstract

Background: There is a disproportionate burden of musculoskeletal injury in low- and middle-income countries (LMICs). The SIGN Online Surgical Database (SOSD) serves as one of the only robust global trauma registries, with over 150,000 cases recorded over the past two decades. The SOSD contains important demographic, injury-related, surgical, and follow-up data related to long-bone fractures treated with intramedullary nailing and has been used to answer important clinical questions in LMICs. However, the quality of data contained in the SOSD is largely unknown. The aim of the current study was to analyze the trend in data quality and follow-up rate of the SOSD over the past two decades. Methods: A de-identified SOSD dataset was imported into Stata for further data cleaning and analysis. Basic descriptive statistics were performed to analyze data completeness and follow-up rate. Fisher's exact test was used to compare between two decades of SOSD data (2002–2010 and 2010–2020) with significance set at 0.05. Results: As of August 2020, the SOSD contained 168,590 cases registered by 501 SIGN partners in 73 countries. Most cases were young males with femoral and/or tibial fractures treated with SIGN standard nails. Data completeness remained high for several variables in both decades, while the rate of follow-up nearly doubled from 23.1% in the first decade to 52.7% in the second decade (p < 0.001). Conclusion: The SOSD is a unique global trauma database for long-bone fractures treated with intramedullary nailing in LMICs. With increasing data quality standards supported by SIGN's well-established infrastructure, the SOSD was shown to be a dependable and sustainable trauma registry for studying outcomes of SIGN implants and addressing other relevant issues in musculoskeletal trauma care in LMICs.