Calculating Sensitivity, Specificity, and Predictive Values for Correlated Eye Data

Abstract

Purpose To describe and demonstrate appropriate statistical approaches for estimating sensitivity, specificity, predictive values and their 95% confidence intervals (95% CI) for correlated eye data. Methods We described generalized estimating equations (GEE) and cluster bootstrap to account for inter-eye correlation and applied them for analyzing the data from a clinical study of telemedicine for the detection of retinopathy of prematurity (ROP). Results Among 100 infants (200 eyes) selected for analysis, 20 infants had referral-warranted ROP (RW-ROP) in both eyes and 9 infants with RW-ROP only in one eye based on clinical eye examination. In the per-eye analysis that included both eyes of an infant, the image evaluation for RW-ROP had sensitivity of 83.7% and specificity of 86.8%. The 95% CI's from the naïve approach that ignored the inter-eye correlation were narrower than those of the GEE approach and cluster bootstrap for both sensitivity (width of 95% CI: 22.4% vs. 23.2% vs. 23.9%) and specificity (11.4% vs. 12.5% vs. 11.6%). The 95% CIs for sensitivity and specificity calculated from left eyes and right eyes separately were wider (35.2% and 30.8% respectively for sensitivity, 25.4% and 17.3% respectively for specificity). Conclusions When an ocular test is performed in both eyes of some or all of the study subjects, the statistical analyses are best performed at the eye-level and account for the inter-eye correlation by using either the GEE or cluster bootstrap. Ignoring the inter-eye correlation results in 95% CIs that are inappropriately narrow and analyzing data from two eyes separately are not efficient.

The Indicators of Clinical and Subclinical Mastitis in Equine Milk

“…The decrease of the Se values refers to the SCC’s and ECM’s ability to detect the mastitis when the mastitis is present [ 43 , 61 ] in selected groups caused by more and more false-negative samples being below a threshold. In the case of the CM group, for both evaluated indicators, the samples values were much higher than each of the considered thresholds.…”

Sensitivity and Specificity of Sheard and Percival’s Criteria for the Diagnosis of Young People with Near-heterophoria

“…The mean age of participants was 19.34 ± 3.325 years. The majority of subjects fell within the (15-20) year age group 154 (67%) followed by 65 (21)(22)(23)(24)(25) year age group 11 (28.3%), whereas the least number of participants were in the age group of 26-30 years (4.8%).…”

“…Two basic measures of test accuracy, sensitivity, and specificity, were estimated from the values in Table 1. Sensitivity is the test's ability to detect the disease when the disease is present The positive predictive value or precision rate is defined as a proportion of people with a positive test result (24,25) in this study subjects with decompensated heterophoria (A + B). It is calculated by the formula: true positive A = true positive +false positive A +B = The negative predictive value is defined as a proportion of people with a negative test result (20,21) in this study subjects with compensated heterophoria (C+D).…”

Tradução, adaptação cultural e validação do Frail Non-Disabled Questionnaire para o contexto brasileiro

“…Para a análise das propriedades psicométricas do FiND, verificou-se: a consistência interna do instrumento, por meio do Kuder-Richardson 20 (KR-20) 12 ; a estabilidade do escore do FiND, por meio do Coeficiente de Correlação Intraclasse 13 ; a validade de critério concorrente do FiND, por meio do Coeficiente de Correlação de Spearman, relacionando o seu escore obtido com o Fenótipo de Fragilidade de Fried 4 , em que a magnitude das correlações foi classificada conforme proposição de Levin e Fox 14 . Não somente, para verificar a validade de critério concorrente do instrumento realizouse ainda a análise dos valores de sensibilidade, especificidade, acurácia e valores preditivos positivos e negativos, com respectivos intervalos de 95% de confiança, além da relação linear quanti-qualitativa entre a área sob a curva ROC (AUC) 15 ; a validade de construto convergente do instrumento FiND, por meio do Coeficiente de Correlação de Spearman, relacionando seu escore obtido com os escores dos domínios do componente físico do SF-36, do ACE-R, da MAN e do GDS-15 16 .…”