Oommen Anu Mary, Abraham Vinod Joseph, Sathish Thirunavukkarasu, Jose V Jacob, George Kuryan
Department of Community Health, Christian Medical College, Vellore, India.
Centre for Population Health Sciences, Nanyang Technological University, Singapore, Singapore.
Diabetes Metab J. 2017 Oct;41(5):386-392. doi: 10.4093/dmj.2017.41.5.386.
The Achutha Menon Centre Diabetes Risk Score (AMCDRS), which was developed in rural Kerala State, South India, had not previously been externally validated. We examined the performance of the AMCDRS in urban and rural areas in the district of Vellore in the South Indian state of Tamil Nadu, and compared it with other diabetes risk scores developed from India.
We used the data from 4,896 participants (30 to 64 years) of a cross-sectional study conducted in Vellore (2010 to 2012), to calculate the AMCDRS scores using age, family history, and waist circumference. Sensitivity, specificity, positive predictive value (PPV), and negative predictive values (NPV), and the area under the receiver operating characteristic curve (AROC) were calculated for undiagnosed and total diabetes.
Of the 4,896 individuals surveyed, 274 (5.6%) had undiagnosed diabetes and 759 (15.5%) had total diabetes. The AMCDRS, with an optimum cut-point of ≥4, identified 45.0% for further testing with 59.5% sensitivity, 60.5% specificity, 9.1% PPV, 95.8% NPV, and an AROC of 0.639 (95% confidence interval [CI], 0.608 to 0.670) for undiagnosed diabetes. The corresponding figures for total diabetes were 75.1%, 60.5%, 25.9%, 93.0%, and 0.731 (95% CI, 0.713 to 0.750), respectively. The AROC for the AMCDRS was not significantly different from that of the Indian Diabetes Risk Score, the Ramachandran or the Chaturvedi risk scores for total diabetes, but was significantly lower than the AROC of the Chaturvedi score for undiagnosed diabetes.
The AMCDRS is a simple diabetes risk score that can be used to screen for undiagnosed and total diabetes in low-resource primary care settings in India. However, it probably requires recalibration to improve its performance for undiagnosed diabetes.
阿楚塔·梅农中心糖尿病风险评分（AMCDRS）是在印度南部喀拉拉邦农村地区开发的，此前尚未经过外部验证。我们在印度南部泰米尔纳德邦韦洛尔地区的城乡地区检验了AMCDRS的性能，并将其与印度开发的其他糖尿病风险评分进行比较。
我们使用了在韦洛尔（2010年至2012年）进行的一项横断面研究中4896名参与者（30至64岁）的数据，利用年龄、家族史和腰围来计算AMCDRS评分。计算了未诊断糖尿病和总体糖尿病的敏感性、特异性、阳性预测值（PPV）、阴性预测值（NPV）以及受试者工作特征曲线下面积（AROC）。
在4896名接受调查的个体中，274人（5.6%）患有未诊断糖尿病，759人（15.5%）患有总体糖尿病。AMCDRS的最佳切点为≥4，可识别出45.0%的个体进行进一步检测，对于未诊断糖尿病，其敏感性为59.5%，特异性为60.5%，PPV为9.1%，NPV为95.8%，AROC为0.639（95%置信区间[CI]，0.608至0.670）。总体糖尿病的相应数字分别为75.1%、60.5%、25.9%、93.0%和0.731（95%CI，0.713至0.750）。AMCDRS对于总体糖尿病的AROC与印度糖尿病风险评分、拉马钱德兰或查图尔维迪风险评分的AROC无显著差异，但对于未诊断糖尿病，其AROC显著低于查图尔维迪评分的AROC。
AMCDRS是一种简单的糖尿病风险评分，可用于在印度资源匮乏的初级保健环境中筛查未诊断糖尿病和总体糖尿病。然而，可能需要重新校准以提高其对未诊断糖尿病的性能。
