Background/Purpose:

Gout is the most prevalent inflammatory arthritis globally. Despite treatment advances, it still has a significant effect on quality of life and healthcare costs. Studies using administrative coding as a marker of accurate diagnoses have shown inconsistencies due to diagnostic criteria differences or gout misdiagnosis. Although gout can be solely managed by primary care physicians (PCPs), complex cases often require rheumatology consultation. The wait time for an initial rheumatology clinic visit ranges from 38 days to 47 weeks. However, electronic consults (e-consults) allow for swift two-way communication between PCPs and rheumatologists (pre-consult exchange) to facilitate coordination of care among providers.

Objectives:

To determine the accuracy of gout diagnosis based on the International Classification of Diseases, ninth (ICD 9) and tenth (ICD 10) revision and the differences in gout outcomes depending on PCP management, e-consult or rheumatology clinic visits at the VA Medical Center in Long Beach, CA.

Methods:

A retrospective cohort study of 81 e-consult patients was created with a control group of 176 patients from 2009-2014. In the e-consult group, 58 patients were ICD 9 or 10 coded for gout and 23 were not; in the control group, 116 were ICD 9 or 10 coded for gout and 60 were not. A blinded abstractor determined the accuracy of gout coding based on chart review and EULAR criteria.

Additionally, a second sample of 163 gout patients from 2009-2014 was identified and stratified to 3 modes of management: PCP only (48), e-consult (48), and rheumatology clinic visit (67). Data was reviewed for 24 months following initial gout diagnosis or e-consult. Management was evaluated based on frequency of flares and related ED visits, creatinine clearance, and serum uric acid levels (sUA).

Results:

The sensitivity and specificity of ICD coding for accurate diagnosis of gout was 94% and 79% in the control (positive PPV and negative predictive values NPV were 88% and 90%). For e-consult patients, the sensitivity and specificity of accurate diagnosis coding was 100% and 70% (PPV 83%, NPV 100%). E-consult patients were more accurately diagnosed with gout by PCPs than in the control group (p= 0.03).

Of e-consults, 77% were resolved electronically and 23% were converted to rheumatology clinic visits. The mean wait time for e-consult recommendations was 2.1 days. The mean clinic visit wait after pre-consult exchange was 22.9 days compared to an average of 43.1 days for direct rheumatology clinic consults. Both e-consult and rheumatology clinic patients had more gout flares and related ED visits at baseline; however, at 12 months, both groups had significantly fewer gout-related ED visits, decreased sUA, and improved creatinine clearance.

Conclusion:

VA databases are an accurate source of gout patients based on ICD 9 and 10 coding. When viewing e-consults, rheumatologists can rely on accurate PCP gout diagnoses, confidently answer clinical questions, and triage to clinic more quickly. E-consult serves as an effective alternative in managing gout with shorter wait times for recommendations and appointments. Therefore, complex gout management can be enhanced by e-consults to decrease gaps in care and optimize healthcare resources.

« Back to 2018 ACR/ARHP Annual Meeting

ACR Meeting Abstracts - https://acrabstracts.org/abstract/diagnostic-accuracy-of-gout-in-electronic-health-records-and-the-role-of-rheumatology-electronic-consults/