E-Comorbidity: Evaluation of the Electronic Comorbidity Assessment in Identifying Comorbid Conditions among Patients with Inflammatory Arthritis

Abstract

Objectives: 1. To assess the validity of an electronically comorbidity assessment strategy to identify comorbid conditions among inflammatory arthritis patients in standard practice. 2. To evaluate the impact of e-comorbidity assessment on the patients’ care and adherence to therapy.Methods: A cohort of 112 RA and 111 PsA subjects diagnosed according to RA ACR/EULAR criteria and PsA CASPAR criteria were followed longitudinally for 36 months. The patients were classified into a study group (112 patients) whose electronic patient-reported comorbidities were compared to a control group of 111 patients who were managed according to standard protocols. The sensitivity, specificity, positive and negative predictive values of the electronic data entry were compared to ICD-10 medical record (reference standard) and rheumatology clinic visits outcomes.Results: The sensitivity for identifying comorbidities using the electronic approach (median, 99.2%; interquartile range [IQR]: 96%-100%) outperformed those recorded using using ICD-10 codes (median, 66%; IQR: 50%-74%); and those recorded using clinic letters (median, 38%; IQR: 32%-54%). The median PPV and NPV were 97.7% (IQR: 96-100%) and 99.6% (IQR: 99-100%) for the e-comorbidity tool Vs 61.8% (IQR: 41%-76%) and 97.4% (IQR: 91%-98%) for the ICD-10 codes, physician recorded comorbidity respectively. The patients’ adherence to antirheumatic therapy was significantly (p<0.1) higher in the studied group.Conclusions: e-comorbidity assessment offered a specific and dynamic approach tailored to the patient’s needs over the 3-years study period, which is applicable in standard practice. Patient reported e-comorbidity outperformed the standard medical recording systems and can have a role in healthcare management and research.