Background SLE, characterized by a heterogenous clinical phenotype, may present differently over time and between patients, creating challenges for identification and treatment of individuals seen at multiple facilities. Clinical data research networks (CDRN) aim to pool electronic health records (EHR) to provide more complete clinical information for patients shared across care centers. We determined whether algorithms to identify Systemic Lupus International Coordinating Clinics (SLICC) classification criteria attributes, using structured EHR data, could be applied to a CDRN to describe people with SLE.
Methods Published algorithms to identify SLICC classification criteria were adapted to the PCORnet Common Data Model used by the Chicago Area Patient Centered Outcomes Research Network (CAPriCORN). Initial patient selection required satisfying ≥1 criteria from the SLICC classification criteria as determined by diagnosis (ICD-9/10), procedure (CPT), medication (RX-NORM) and lab codes (LOINC) identified 1,231,130 unique patients. Next, persons with and without SLE were defined by 3+ or 0 instances of a SLE diagnosis by ICD 9/10 code, and the rates of each SLICC criterion was compared using Pearson Chi-squared test at the 95% confidence level. Patient records were assessed for the presence of SLICC criteria and whether they satisfied one SLICC rule for ‘Definite SLE’ fulfilling 4 total criteria with at least one from the clinical and immunologic domains.
Results The attribute identification frequency of SLICC criteria are represented in table 1. We identified 6,488 persons ≥ 3 SLE diagnostic codes: 1,201,999 persons with no diagnosis codes for SLE. All criteria items occurred significantly more often in persons with SLE compared with those without SLE and the greatest differences in SLICC attributes identified: Acute Cutaneous: 19% v. 0.17%; Chronic Cutaneous: 18% vs. 0.1%, Renal: 29% vs. 10%, Thrombocytopenia: 14% vs. 6.3%,, Antinuclear Antibodies: 26% vs. 2.2%, Anti-dsDNA Antibodies: 42% vs. 1.6%, Anti-Sm Antibodies: 6% vs. 0.08%, Antiphospholipid Antibodies: 7.1% vs. 0.26%, Low complement: 37% vs. 0.6%, Direct Coombs Test: 1.4% vs. 0.06%. There were 2770 persons (43%) among those ≥3 SLE diagnoses and 9770 persons (0.81%) without SLE satisfying the SLICC definition of ‘Definite SLE’.
Conclusions The results demonstrate identification of all SLICC classification criteria attributes in the CAPriCORN data set, an increased rate of attribute identification for all SLICC criteria, and an increased rate of definite SLE classification via SLICC in persons with ≥3 SLE diagnostic codes compared to those without SLE diagnostic codes. This suggests that SLE presentation can be characterized in CDRN data.
Acknowledgements The authors received funding support provided by grants from the National Institute of Arthritis and Musculoskeletal Disease (5R21AR072262 and P30AR072579) and the National Human Genome Research Institute (U01HG008657).
This is an open access article distributed in accordance with the Creative Commons Attribution Non Commercial (CC BY-NC 4.0) license, which permits others to distribute, remix, adapt, build upon this work non-commercially, and license their derivative works on different terms, provided the original work is properly cited, appropriate credit is given, any changes made indicated, and the use is non-commercial. See: http://creativecommons.org/licenses/by-nc/4.0/.
Statistics from Altmetric.com
If you wish to reuse any or all of this article please use the link below which will take you to the Copyright Clearance Center’s RightsLink service. You will be able to get a quick price and instant permission to reuse the content in many different ways.