Evaluation of an automated phenotyping algorithm for rheumatoid arthritis.

To better understand the challenges of generally implementing and adapting computational phenotyping approaches, the performance of a Phenotype KnowledgeBase (PheKB) algorithm for rheumatoid arthritis (RA) was evaluated on a University of California, Los Angeles (UCLA) patient population, focusing on examining its performance on ambiguous cases. The algorithm was evaluated on a cohort of 4,766 patients, along with a chart review of 300 patients by rheumatologists against accepted diagnostic guidelines. The performance revealed low sensitivity towards specific subtypes of positive RA cases, which suggests revisions in features used for phenotyping. A close examination of select cases also indicated a significant portion of patients with missing data, drawing attention to the need to consider data integrity as an integral part of phenotyping pipelines, as well as issues around the usability of various codes for distinguishing cases. We use patterns in the PheKB algorithm's errors to further demonstrate important considerations when designing a phenotyping algorithm.

A Panel of Biomarkers Associates With Increased Risk for Cardiovascular Events in Women With Systemic Lupus Erythematosus.

OBJECTIVE: The increase in cardiovascular events (CVEs) in systemic lupus erythematosus (SLE) is not fully explained by traditional risk factors. We previously identified four biomarkers (proinflammatory high-density lipoprotein, leptin, soluble TNF-like weak inducer of apoptosis (sTWEAK), and homocysteine) that we combined with age and diabetes to create the predictors of risk for elevated flares, damage progression, and increased cardiovascular diseasein patients with SLE (PREDICTS) risk profile. PREDICTS more accurately identified patients with SLE at risk for progression of subclinical atherosclerosis than any individual variable. We examined whether PREDICTS can also identify patients with SLE at risk for future CVEs. METHODS: A total of 342 patients with SLE and 155 matched control subjects participated in this longitudinal prospective study. A high PREDICTS score was defined as three or more predictors or diabetes + one or more predictor. The biomarkers were measured at baseline using published methods. All major adverse CVEs (MACEs) were confirmed by medical record review. RESULTS: During 116 months of follow-up, 5% of patients with SLE died, 12% had a cerebrovascular event, and 5% had a cardiac event. Overall, 20% of patients with lupus experienced any new MACE compared with 5% of control subjects (P < 0.0001). More patients with SLE with a new MACE had high PREDICTS score at baseline (77%) versus patients with no new events (34%) (P < 0.0001). High baseline PREDICTS score also associated with cerebrovascular (P < 0.0001) and cardiac events (P < 0.0001) in SLE. Using Cox regression, a baseline high PREDICTS score associated with a 3.7-fold increased hazard ratio (HR) for a new MACE (P < 0.0001) in SLE. Hypertension (HR = 2.1; P = 0.006) was also a risk. CONCLUSION: A high PREDICTS score and hypertension confer increased risk for new MACEs in patients with SLE.