“嬰幼兒綜合發展測驗”動作分測驗與“皮巴迪動作發展量表第二版”的診斷準確度

Diagnostic Accuracy of the Motor Subtest of Comprehensive Developmental Inventory for Infants anti Toddlers anti the Peabody Developmental Motor Scales-Second Edition

10.6320/FJM.2005.9(3).02

吳雪玉(Hsueh-Yu Wu)；廖華芳(Hua-Fang Liao)；姚開屏(Grace Yao)；李旺祚(Wang-Tso Lee)；王天苗(Tien-Miau Wang)；謝正宜(Jeng-Yi Hsieh)

動作評估 ； 敏感度 ； 特異度 ； 診斷準確率 ； ROC曲線 ； 發展障礙兒童 ； child ； diagnostic accuracy ； motor assessment ； receiver operating characteristics curve ； sensitivity ； specificity

台灣醫學

9卷3期（2005 / 05 / 01）

312 - 322

繁體中文

Purpose: 1) Using clinical diagnosis as thee criterion to investigate the overall diagnostic accuracy of both the motor subtest of the Comprehensive Developmental Inventory for Infants and Toddlers (CDIIT) and the Peabody Developmental Motor Scales-Second Edition (PDMS-2).2) To investigate the sensitivity, specificity, positive likelihood ratio (LR+), and negative likelihood ratio (LR-), and then to use the prevalence rates of 4%, 6%, 9% to calculate the corresponding positive and negative predictive value (PPV & NPV) for clinical application of both scales. Methods: Eighty-five children with motor disabilities and 137 non-disabled children were recruited for this study. Both the CDIIT and PDMS-2 were administered by the same tester. The area under the receiver operating characteristics (ROC) curve was used to investigate the overall diagnostic accuracy of both tests. The cross-tabs ere used to calculate the diagnostic estimates, and then the estimated prevalence rate as used to calculate the PPR and NPR. Results: The overall diagnostic accuracy of the t o scales as high, with ROC area 0.97 in the CDIIT, and 0.98 in PDMS-2 (P＜0.001). The best cutoff point of the CDIIT was developmental quotient 70, with sensitivity 87%, specificity 97%, LR+29.0, LR-0.13. While a 6% prevalence rate as assumed, the PPV and NPV ere 65% and 99% respectively. And the best cutoff point of PDMS-2 was developmental quotient 85 with sensitivity 81%, specificity 99%, LR+81.0, and LR-0.19. The PPV of the PDMS-2 as 84%, and the NPV as 99%. The correlations between both scales and the clinical diagnoses were high with Kappa coefficients of 0.86 and 0.82 respectively (P＜0.001). Conclusions: Both scales had good and similar diagnostic accuracy. The diagnosis estimates of both scales ere slightly different but within an acceptable range. Both scales are useful for motor development evaluation in preschool children.

Purpose: 1) Using clinical diagnosis as thee criterion to investigate the overall diagnostic accuracy of both the motor subtest of the Comprehensive Developmental Inventory for Infants and Toddlers (CDIIT) and the Peabody Developmental Motor Scales-Second Edition (PDMS-2).2) To investigate the sensitivity, specificity, positive likelihood ratio (LR+), and negative likelihood ratio (LR-), and then to use the prevalence rates of 4%, 6%, 9% to calculate the corresponding positive and negative predictive value (PPV & NPV) for clinical application of both scales. Methods: Eighty-five children with motor disabilities and 137 non-disabled children were recruited for this study. Both the CDIIT and PDMS-2 were administered by the same tester. The area under the receiver operating characteristics (ROC) curve was used to investigate the overall diagnostic accuracy of both tests. The cross-tabs ere used to calculate the diagnostic estimates, and then the estimated prevalence rate as used to calculate the PPR and NPR. Results: The overall diagnostic accuracy of the t o scales as high, with ROC area 0.97 in the CDIIT, and 0.98 in PDMS-2 (P＜0.001). The best cutoff point of the CDIIT was developmental quotient 70, with sensitivity 87%, specificity 97%, LR+29.0, LR-0.13. While a 6% prevalence rate as assumed, the PPV and NPV ere 65% and 99% respectively. And the best cutoff point of PDMS-2 was developmental quotient 85 with sensitivity 81%, specificity 99%, LR+81.0, and LR-0.19. The PPV of the PDMS-2 as 84%, and the NPV as 99%. The correlations between both scales and the clinical diagnoses were high with Kappa coefficients of 0.86 and 0.82 respectively (P＜0.001). Conclusions: Both scales had good and similar diagnostic accuracy. The diagnosis estimates of both scales ere slightly different but within an acceptable range. Both scales are useful for motor development evaluation in preschool children.