The Ability of Medical Students in the Prediction of Birth Weight

Abstract

Objective: To assess the accuracy of clinical fetal weight estimation performed by medical students.
Design Cross-sectional comparative study.
Material and methods: The study sample included 1880 pregnant women admitted for delivery in
Thammasat hospital between March 1999 and February 2000. Estimated fetal weight (EFW) was
performed clinically by last-year medical students and lecturers. The accuracy of clinical EFW was
determined by errors using actual birth weight as the gold standard, consisted of percentage error,
absolute percentage error and proportion of accurate estimation within 10% of actual birth weight.
Comparison of the accuracy between examiners was assessed by paired t test, comparison of
correlated variances, Wilcoxon signed-rank test, Student t test and c2 test.
Results: The actual birth weight in the study population averaged 3077± 503 g (mean ± SD). In the
entire population and in the middle range of birth weight group (2500 - 3999 g), students predicted birth
weight within acceptable range. In the low birth weight (< 2500 g) and high birth weight (> 4000 g)
groups, students systematically over and underestimated the actual birth weight respectively. The rate
of estimates within 10% of birth weight in the entire population was 69%. When comparing with
lecturers, there were no significant differences in mean percentage and absolute percentage errors
between student’s and lecturer’s estimates in the entire population. In the low birth weight group (<
2500 g), all mean errors of EFW by lecturers were significantly smaller than those of EFW by students
and the proportion of estimates within 10% of birth weight was significantly higher for lecturers.
Conclusion: Last-year medical students can fairly predict birth weight except in the low birth weight
group. Lecturers should train medical students by obtaining feedback on their assessments and note
the range of accuracy of their own weight predictions especially at the extremes of the scales (< 2500 g
and > 4000 g).