Spironello C, Hay J, Missiuna C, Faught B E, Cairney J
Department of Clinical Epidemiology and Biostatistics, McMaster University, Hamilton, ON, Canada.
Child Care Health Dev. 2010 Jul;36(4):499-507. doi: 10.1111/j.1365-2214.2009.01066.x. Epub 2010 Mar 9.
Among the most widely used instruments to assess developmental co-ordination disorder (DCD) in children are the Bruininks-Oseretsky Test of Motor Proficiency (BOTMP) and the Movement Assessment Battery for Children (M-ABC). However, there is little research on agreement between these tests, when administered to children in field-based settings by trained non-clinicians.
Ten of 75 schools participating in a larger study were randomly selected. All children in grade 4 (n= 340) in each of these schools were assessed at the same time using both the BOTMP-SF and the M-ABC in May of 2005. The order of tests was balanced, with an average gap in time between tests of 10-15 min. All tests were administered by trained research assistants.
The correlation between tests was moderate (r= 0.50, P < 0.01). Kappas were low at the fifth (k= 0.19) and 15th (k= 0.29) percentile cut-points, which are generally used to identify cases of DCD. Re-analysis using the relative improvement over chance (RIOC) statistic, however, revealed slightly better agreement at both cut-points (fifth percentile, RIOC = 0.29; 15th percentile, RIOC = 0.47). Children who scored as probable for DCD on both motor tests, as well as on only the BOTMP-SF, had higher body mass index, poorer physical fitness and lower levels of teacher-reported physical ability than those positive for DCD on the M-ABC only or those who scored negatively on both tests.
In general, the agreement between tests, even after adjustment for RIOC, was poor. Children identified with poor motor competence by both tests or by the BOTMP-SF only are at particular risk for poor physical fitness, overweight/obesity and physical inactivity. It appears that each assessment measures different dimensions of motor ability but that under field-based conditions the M-ABC may be less useful when applied by non-clinicians.
用于评估儿童发育性协调障碍（DCD）的最广泛使用的工具包括布鲁宁克斯-奥塞瑞斯基运动能力测试（BOTMP）和儿童运动评估量表（M-ABC）。然而，对于由经过培训的非临床医生在现场环境中对儿童进行这些测试时，它们之间的一致性研究很少。
从参与一项更大规模研究的75所学校中随机选择10所。2005年5月，对这些学校中每所学校的所有四年级学生（n = 340）同时使用BOTMP-SF和M-ABC进行评估。测试顺序是平衡的，两次测试之间的平均时间间隔为10 - 15分钟。所有测试均由经过培训的研究助理进行。
测试之间的相关性为中等（r = 0.50，P < 0.01）。在通常用于识别DCD病例的第五百分位（κ = 0.19）和第十五百分位（κ = 0.29）切点处，kappa值较低。然而，使用机会相对改善（RIOC）统计量进行重新分析后，发现在两个切点处的一致性略好一些（第五百分位，RIOC = 0.29；第十五百分位，RIOC = 0.47）。在两项运动测试中都被判定可能患有DCD的儿童，以及仅在BOTMP-SF测试中被判定可能患有DCD的儿童，与仅在M-ABC测试中被判定为DCD阳性的儿童或在两项测试中得分均为阴性的儿童相比，具有更高的体重指数、更差的身体素质以及更低的教师报告的身体能力水平。
总体而言，即使在对RIOC进行调整之后，测试之间的一致性仍然很差。通过两项测试或仅通过BOTMP-SF被识别为运动能力差的儿童，在身体素质差、超重/肥胖和身体活动不足方面具有特别的风险。似乎每种评估都测量了运动能力的不同维度，但在现场环境条件下，由非临床医生应用时，M-ABC可能不太有用。
