JMIR Publications

ABSTRACT

Background:

Psoriasis (PsO) is a chronic, systemic, immune-mediated disease with multiorgan involvement. Psoriatic arthritis (PsA) is an inflammatory arthritis that is present in 6%–42% of patients with PsO. Approximately 15% of patients with PsO have undiagnosed PsA. Predicting patients with a risk of PsA is crucial for providing them with early examination and treatment that can prevent irreversible disease progression and function loss.

Objective:

To develop and validate a prediction model for psoriatic arthritis based on chronological large-scale and multidimensional electronic medical records using a machine learning algorithm.

Methods:

This case-control study used 2 million randomly sampled data from the Taiwan’s National Health Insurance Research Database from January 1, 1999 to December 31, 2013. A total of 443 psoriasis patients with psoriatic arthritis (International Classification of Diseases, Ninth Revision, Clinical Modification code 696.0) were included in the analysis. The original dataset was split into training and holdout set in an 80:20 ratio. A convolutional neural network was used to develop a prediction model. This model used 2.5-year diagnostic and medical records (inpatient/outpatient) with temporal-sequential information to predict the risk of psoriatic arthritis for a given patient within the next 6 months. The model was developed and cross-validated using the training data and was tested using the holdout data. An occlusion sensitivity analysis was performed to identify the important features of the model.

Results:

The prediction model included a total of 443 psoriatic arthritis patients with earlier diagnosis of psoriasis (266 men [61.43%]; mean [SD] age, 42.66 [17.21]) and 1,772 psoriasis patients without psoriatic arthritis (989 men [57.10%]; mean [SD] age, 46.85 [20.18]) for the control group. The 6-month psoriatic arthritis risk prediction model that uses sequential diagnostic and drug prescription information as a temporal phenomic map yielded an AUROC of 0.70 (95% CI, 0.559- 0.833), a mean (SD) sensitivity of 0.80 (0.11), a mean (SD) specificity of 0.60 (0.04), a mean (SD) positive predictive value of 0.32 (0.05) and a mean (SD) negative predictive value of 0.93 (0.04).

Conclusions:

The findings of this study suggest that the risk prediction model can identify psoriasis patients with a high risk of psoriatic arthritis. This model may help healthcare professionals to prioritize treatment for target high-risk populations and prevent irreversible disease progression and functional loss.