Background Few studies on rheumatoid arthritis (RA) have generated machine learning models to predict biologic disease-modifying antirheumatic drugs (bDMARDs) responses; however, these studies included insufficient analysis on important features. Moreover, machine learning is yet to be used to predict bDMARD responses in ankylosing spondylitis (AS). Thus, in this study, machine learning was used to predict such responses in RA and AS patients. Methods Data were retrieved from the Korean College of Rheumatology Biologics therapy (KOBIO) registry. The number of RA and AS patients in the training dataset were 625 and 611, respectively. We prepared independent test datasets that did not participate in any process of generating machine learning models. Baseline clinical characteristics were used as input features. Responders were defined as those who met the ACR 20% improvement response criteria (ACR20) and ASAS 20% improvement response criteria (ASAS20) in RA and AS, respectively, at the first follow-up. Multiple machine learning methods, including random forest (RF-method), were used to generate models to predict bDMARD responses, and we compared them with the logistic regression model. Results The RF-method model had superior prediction performance to logistic regression model (accuracy: 0.726 [95% confidence interval (CI): 0.725–0.730] vs. 0.689 [0.606–0.717], area under curve (AUC) of the receiver operating characteristic curve (ROC) 0.638 [0.576–0.658] vs. 0.565 [0.493–0.605], F1 score 0.841 [0.837–0.843] vs. 0.803 [0.732–0.828], AUC of the precision-recall curve 0.808 [0.763–0.829] vs. 0.754 [0.714–0.789]) with independent test datasets in patients with RA. However, machine learning and logistic regression exhibited similar prediction performance in AS patients. Furthermore, the patient self-reporting scales, which are patient global assessment of disease activity (PtGA) in RA and Bath Ankylosing Spondylitis Functional Index (BASFI) in AS, were revealed as the most important features in both diseases. Conclusions RF-method exhibited superior prediction performance for responses of bDMARDs to a conventional statistical method, i.e., logistic regression, in RA patients. In contrast, despite the comparable size of the dataset, machine learning did not outperform in AS patients. The most important features of both diseases, according to feature importance analysis were patient self-reporting scales.

Background Previous studies have shown that the incidence and risk factors of gout differs according to sex. However, little research has been done on the association between reproductive factors and gout. We conducted an analysis of a large nationwide population-based cohort of postmenopausal women to determine whether there is an association between reproductive factors and the incidence of gout. Methods A total of 1,076,378 postmenopausal women aged 40–69 years who participated in national health screenings in 2009 were included in the study. The outcome was the occurrence of incident gout, which was defined using the ICD-10 code of gout (M10) in the claim database. Cox proportional hazard models were used for the analyses and stratified analyses according to body mass index (BMI) and the presence/absence of chronic kidney disease (CKD) were performed. Results The mean follow-up duration was 8.1 years, and incident cases of gout were 64,052 (incidence rate 7.31 per 1000 person-years). Later menarche, earlier menopause, and a shorter reproductive span were associated with a high risk of gout. No association between parity and gout incidence was observed. Use of oral contraceptives (OC) and hormone replacement therapy (HRT) were associated with an increased risk of gout. The association between reproductive factors and gout was not statistical significant in the high BMI group. The effects of OC and HRT usage on gout were not significant in the CKD group. Conclusion Shorter exposure to endogenous estrogen was associated with a high risk of gout. Conversely, exposure to exogenous estrogen such as OC and HRT was associated with an increased risk of gout.

Despite a growing burden posed by cardiovascular disease (CVD) in rheumatoid arthritis (RA) patients, large-scale studies on the association between the characteristics of RA patients and CVD risks and studies adjusted for various confounding factors are lacking. In this large-scale nationwide cohort study, we aimed to investigate the association between CVD risk and RA and factors that may increase CVD risk using a dataset provided by the Korean National Health Insurance Service (NHIS). We enrolled 136,469 patients with RA who participated in national health examinations within two years of RA diagnosis between 2010 and 2017 and non-RA controls matched by age and sex (n = 682,345). The outcome was the occurrence of myocardial infarction (MI) or stroke. MI was defined as one hospitalization or two outpatient visits with ICD-10-CM codes I21 or I22. Stroke was defined as one hospitalization with ICD-10-CM codes I63 or I64 and a claim for brain imaging (CT or MRI). The Cox proportional hazard model and Kaplan–Meier curve were used for analysis. The mean follow-up duration was 4.7 years, and the incidence rate of CVD was higher in the RA group than the control group (MI: 3.20 vs. 2.08; stroke: 2.84 vs. 2.33 per 1000 person-years). The risk of MI and stroke was about 50% and 20% higher, respectively, in RA patients. The association between RA and CVD was prominent in females after adjusting for confounding variables. The association between RA and risk of MI was significant in individuals without DM. Therefore, appropriate screening for CVD is important in all RA patients including females and younger patients.