A clinical prediction model can be applied to several challenging clinical scenarios: screening high-risk individuals for asymptomatic disease, predicting future events such as disease or death, and assisting medical decision-making and health education. Despite the impact of clinical prediction models on practice, prediction modeling is a complex process requiring careful statistical analyses and sound clinical judgement. Although there is no definite consensus on the best methodology for model development and validation, a few recommendations and checklists have been proposed. In this review, we summarize five steps for developing and validating a clinical prediction model: preparation for establishing clinical prediction models; dataset selection; handling variables; model generation; and model evaluation and validation. We also review several studies that detail methods for developing clinical prediction models with comparable examples from real practice. After model development and vigorous validation in relevant settings, possibly with evaluation of utility/usability and fine-tuning, good models can be ready for the use in practice. We anticipate that this framework will revitalize the use of predictive or prognostic research in endocrinology, leading to active applications in real clinical practice.

Background

Risk assessment tools have been actively studied, and they summarize key predictors with relative weights/importance for a disease. Currently, standardized screening scores for type 2 diabetes mellitus (DM) and chronic kidney disease (CKD)-two key global health problems-are available in United States and Korea. We aimed to compare and evaluate screening scores for DM (or combined with prediabetes) and CKD, and assess the risk in contemporary United States and Korean populations.

Methods

Four (2×2) models were evaluated in the United States-National Health and Nutrition Examination Survey (NHANES 2015-2018) and Korea-NHANES (2016-2018)-8,928 and 16,209 adults. Weighted statistics were used to describe population characteristics. We used logistic regression for predictors in the models to assess associations with study outcomes (undiagnosed DM and CKD) and diagnostic measures for temporal and cross-validation.

Results

Korean adult population (mean age 47.5 years) appeared to be healthier than United States counterpart, in terms of DM and CKD risks and associated factors, with exceptions of undiagnosed DM, prediabetes and prehypertension. Models performed well in own country and external populations regarding predictor-outcome association and discrimination. Risk tests (high vs. low) showed area under the curve >0.75, sensitivity >84%, specificity >45%, positive predictive value >8%, and negative predictive value >99%. Discrimination was better for DM, compared to the combined outcome of DM and prediabetes, and excellent for CKD due to age.

Conclusion

Four easy-to-use screening scores for DM and CKD are well-validated in contemporary United States and Korean populations. Prevention of DM and CKD may serve as first-step in public health, with these self-assessment tools as basic tools to help health education and disparity.

Background

Non-alcoholic fatty liver disease (NAFLD) is a prevalent and rapidly increasing disease worldwide; however, no widely accepted screening models to assess the risk of NAFLD are available. Therefore, we aimed to develop and validate a self-assessment score for NAFLD in the general population using two independent cohorts.

Methods

The development cohort comprised 15676 subjects (8313 males and 7363 females) who visited the National Health Insurance Service Ilsan Hospital in Korea in 2008-2010. Anthropometric, clinical, and laboratory data were examined during regular health check-ups and fatty liver diagnosed by abdominal ultrasound. Logistic regression analysis was conducted to determine predictors of prevalent NAFLD and to derive risk scores/models. We validated our models and compared them with other existing methods using an external cohort (N = 66868).

Results

The simple self-assessment score consists of age, sex, waist circumference, body mass index, history of diabetes and dyslipidemia, alcohol intake, physical activity and menopause status, which are independently associated with NAFLD, and has a value of 0-15. A cut-off point of ≥ 8 defined 58% of males and 36% of females as being at high-risk of NAFLD, and yielded a sensitivity of 80% in men (77% in women), a specificity of 67% (81%), a positive predictive value of 72% (63%), a negative predictive value of 76% (89%) and an AUC of 0.82 (0.88). Comparable results were obtained using the validation dataset. The comprehensive NAFLD score, which includes additional laboratory parameters, has enhanced discrimination ability, with an AUC of 0.86 for males and 0.91 for females. Both simple and comprehensive NAFLD scores were significantly increased in subjects with higher fatty liver grades or severity of liver conditions (e.g., simple steatosis, steatohepatitis).

Conclusions

The new non-laboratory-based self-assessment score may be useful for identifying individuals at high-risk of NAFLD. Further studies are warranted to evaluate the utility and feasibility of the scores in various settings.