From Register to Electronic Medical Records: A Data-Quality Audit in a Bangladeshi Orthopaedic Outpatient Department
Discuss this preprint
Start a discussion What are Sciety discussions?Listed in
This article is not in any list yet, why not save it to one of your lists.Abstract
Background Electronic medical records (EMRs) are promoted to improve data quality, yet most hospitals in low- and middle-income countries (LMICs) still rely on paper or semi-electronic registers. Orthopaedic outpatient departments face unique documentation challenges due to complex coding, high caseloads, and limited digital integration. Evidence from empirical audits that quantify register deficiencies and map them to EMR safeguards remains sparse in Bangladesh and comparable LMIC settings. Methods We conducted a retrospective audit of all orthopaedic outpatient records (N = 4,085) at Khulna Medical College Hospital, Bangladesh, from 1 January to 30 June 2025. Three dimensions of data quality were assessed: completeness, validity (format, plausibility, categorical conformance), and internal consistency (duplicate detection, age–age group concordance). Missingness and violation rates were quantified with 95% Wilson confidence intervals. Identified deficiencies were translated into EMR configuration recommendations, including mandatory fields, constrained inputs, ICD-10 picklists, and database-enforced unique identifiers. Results Core demographic fields demonstrated high completeness: sex 98.6%, numeric age 98.7%, and occupation 94.9%. Presenting complaint, anatomical site, and painkiller history exceeded 96% completeness. Validity was strong, with no implausible ages (0–120 years) and only 0.5% unparsable dates. Identifier integrity was the principal weakness: 131 duplicate serial numbers (3.2%) and seven complete-row duplicates were detected. Age–age group mismatches were rare (0.2%). Conclusion Orthopaedic outpatient registers in this high-volume tertiary hospital are broadly EMR-ready, with most fields accurate and complete. Key vulnerabilities—duplicate identifiers and free-text complaints—can be addressed by unique visit IDs, constrained input controls, and ICD-10–aligned coding. Findings provide a practical blueprint for EMR configuration in resource-constrained settings, offering immediate pathways to safer care, credible service metrics, and scalable digital health analytics.