Eita Goto1. 1. Nagoya Medical Science Research Institute, Meitou-ku, Nagoya, Japan. Email: egoto1@nifty.com.
Abstract
OBJECTIVE: Weighing scales are often lacking at home birth in developing countries. Therefore, simple, reliable, and inexpensive methods for detecting low birth weight especially before birth would be useful. This study was performed to evaluate the diagnostic value of maternal anthropometric measurements for predicting low birth weight. METHODS: Bivariate diagnostic meta-analysis was conducted to construct hierarchical summary receiver operating characteristic curves. All English language studies included in the meta-analysis enrolled apparently healthy pregnant women and provided the data necessary to construct two-by-two tables (i.e., true positive, false positive, false negative, and true negative values). Ten data bases, including PubMed, were searched to identify these studies. RESULTS: A sufficient number of studies involving 309,419 women paired with their newborns in Africa, Asia, Europe, Latin America, the Middle East, and Oceania, included data on maternal height, weight, arm circumference, body mass index, and weight gain during pregnancy (n=85, 80, 23, 51, and 16, respectively) to provide generalizable findings. However, sensitivity of 0.46 (95% confidence interval (CI)=0.35-0.56) to 0.63 (95% CI=0.54-0.71), specificity of 0.55 (95% CI=0.42-0.67) to 0.71 (95% CI=0.61-0.80), and diagnostic odds ratios of 2 (95% CI=2-2) to 4 (95% CI=3-5) were not sufficiently high for primary screening. The generalizability of abdominal circumference data could not be guaranteed due to the limited sample (one article). CONCLUSIONS: Maternal anthropometric measurements are unsuitable for predicting low birth weight.
OBJECTIVE: Weighing scales are often lacking at home birth in developing countries. Therefore, simple, reliable, and inexpensive methods for detecting low birth weight especially before birth would be useful. This study was performed to evaluate the diagnostic value of maternal anthropometric measurements for predicting low birth weight. METHODS: Bivariate diagnostic meta-analysis was conducted to construct hierarchical summary receiver operating characteristic curves. All English language studies included in the meta-analysis enrolled apparently healthy pregnant women and provided the data necessary to construct two-by-two tables (i.e., true positive, false positive, false negative, and true negative values). Ten data bases, including PubMed, were searched to identify these studies. RESULTS: A sufficient number of studies involving 309,419 women paired with their newborns in Africa, Asia, Europe, Latin America, the Middle East, and Oceania, included data on maternal height, weight, arm circumference, body mass index, and weight gain during pregnancy (n=85, 80, 23, 51, and 16, respectively) to provide generalizable findings. However, sensitivity of 0.46 (95% confidence interval (CI)=0.35-0.56) to 0.63 (95% CI=0.54-0.71), specificity of 0.55 (95% CI=0.42-0.67) to 0.71 (95% CI=0.61-0.80), and diagnostic odds ratios of 2 (95% CI=2-2) to 4 (95% CI=3-5) were not sufficiently high for primary screening. The generalizability of abdominal circumference data could not be guaranteed due to the limited sample (one article). CONCLUSIONS: Maternal anthropometric measurements are unsuitable for predicting low birth weight.