(305) Validating The Standardized and ICD-9 Code of Type 2 Diabetes Mellitus in a Saudi Common Data Model

Saturday, August 27, 2022

8:00 AM – 6:00 PM CEST

Location: Hall C

Publication Number: 979

Presenting Author(s)

Turki A.I. Althunian, PhD
– Director of the Saudi FDA Research Informatics Department, Saudi Food and Drug Authority, Riyadh, Saudi Arabia

Co-Author(s)

Meshael Alrasheed
– Riyadh, Saudi Arabia
Fatemah Alnofal, MSc
– Head of the Research Database Development Division, Saudi Food and Drug Authority, Saudi Arabia
Rawan Tafish, PharmD
– Kingdom Hospital and Consulting Clinics, Saudi Arabia
Mahmood A. Mira, MD
– Kingdom Hospital and Consulting Clinics, Saudi Arabia
Raseel A. Alroba, MSc
– Researcher, Saudi Food and Drug Authority, Saudi Arabia
Mohammed Kirdas, MD
– Kingdom Hospital and Consulting Clinics, Saudi Arabia
Thamir M. Alshammari, PhD
– Saudi Food and Drug Authority, Riyadh, Saudi Arabia

Background: Validating diagnostic codes of type 2 diabetes type 2 mellitus in the Saudi electronic health records (EHRs) is s a priority given its high prevalence and given the anticipated large, pharmacoepidemiologic studies on antidiabetics. However, no studies have been published to assess the validity of recording diagnostic codes in general, and that of type 2 diabetes mellitus in particular, in the Saudi EHRs.

Objectives: This study was conducted to assess the validity of the original, the extracted and the standardized diagnostic codes of type 2 diabetes mellitus of the EHRs that were imported from a hospital to a Saudi common data model (the centralized National Pharmacoepidemiologic Database [NPED]).

Methods: This study was a retrospective validation study. It was carried out using the EHRs that were imported and mapped from a private hospital in Riyadh to the NPED (a standardization was performed for these EHRs using the Observational Health Data Sciences and Informatics [OHDSI] common data model). A random sample of type 2 diabetic patients, who visited the hospital in the period between 01 January 2013 and 01 July 2018, was extracted from the standardized EHRs of the hospital and matched with a control group (non-diabetic patients) based on age and sex. The included participants were required to be ≥18 years and have at least one health record. The standardized coding of type 2 diabetes in the NPED was validated by comparing the presence of diabetes in the NPED vs. the original electronic records at the hospital, the recording in paper-based medical records, and the physician re-assessment of diabetes in the included cases and controls; respectively. Sensitivity, specificity, positive predictive value (PPV), and negative predictive value (NPV) values were estimated for each pairwise comparison using RStudio 1.4.1103.

Results: A total of 437 random diabetic patients were identified and were matched with 437 controls. Only 190 of 437 (43.0 %) had paper-based medical records. PPVs of NPED vs. original EHRs, paper-based records and physician re-assessment were 1.0 (95% confidence interval [CI] 0.99 to 1.0), 0.54 (95%CI 0.47 to 0.61), and 1.0 (95%CI 0.99 to 1.0); respectively. Sensitivities were 0.95 (95%CI 0.93 to 0.97), 0.93 (95%CI 0.86 to 0.97), and 0.95 (95%CI 0.93 to 0.97); respectively. NPVs were 0.95 (95%CI 0.92 to 0.97), 0.96 (0.92 to 0.98), and 0.95 (0.92 to 0.97); respectively. Specificities were 1.0 (95%CI 0.99 to 1.0), 0.68 (95%CI 0.62 to 0.73), and 1.0 (95%CI 0.99 to 1.0); respectively.

Conclusions: The results of our study substantiate the validity of coding, extracting, and standardizing type 2 diabetes mellitus in the NPED. A future multi-center study would help adding more emphasis to the study findings.