A nonparametric method of comparing the partial areas under two correlated receiver operating characteristic curves

A non-parametric method was developed and tested to compare the partial areas under two correlated Receiver Operating Characteristic curves. Based on the theory of generalized U-statistics the mathematical formulas have been derived for computing ROC area, and the variance and covariance between the portions of two ROC curves. A practical SAS application also has been developed to facilitate the calculations. The accuracy of the non-parametric method was evaluated by comparing it to other methods. By applying our method to the data from a published ROC analysis of CT image, our results are very close to theirs. A hypothetical example was used to demonstrate the effects of two crossed ROC curves. The two ROC areas are the same. However each portion of the area between two ROC curves were found to be significantly different by the partial ROC curve analysis. For computation of ROC curves with large scales, such as a logistic regression model, we applied our method to the breast cancer study with Medicare claims data. It yielded the same ROC area computation as the SAS Logistic procedure. Our method also provides an alternative to the global summary of ROC area comparison by directly comparing the true-positive rates for two regression models and by determining the range of false-positive values where the models differ. ^

A review of cataract in stable and conflict zones in Africa

Aim. To review and explore cataract prevalence in stable and unstable countries by examining published and unpublished ocular literature about Africa from 1980 onwards.^ Methods. Searches using OVID, Proquest Dissertations, WHO, and Ebsco Host were done. The review was restricted to articles utilizing WHO definitions of blindness and low vision. Random cluster sampling technique with a minimum sample size of 1,500, and reporting causes of blindness categorized by age and gender were inclusion considerations in the selected articles. ^ Results. Blindness and low vision increased with conflict. Women and the elderly were more likely to have vision impairing cataract. Cataract was the leading cause of blindness; the prevalence range was 22%–81% for the reviewed nations.^ Conclusion. Instability was connected to higher cataract prevalence and worse visual outcome across all characteristics examined except cataract surgical rates. ^