HOME HELP FEEDBACK SUBSCRIPTIONS ARCHIVE SEARCH TABLE OF CONTENTS		QUICK SEARCH:   [advanced] Author: Keyword(s): Year:  Vol:  Page:

This Article Abstract Full Text (PDF) Alert me when this article is cited Alert me if a correction is posted Services Similar articles in this journal Similar articles in PubMed Alert me to new issues of the journal Download to citation manager Citing Articles Citing Articles via HighWire Citing Articles via Google Scholar Google Scholar Articles by SMITH, L. N. Articles by MONGA, M. Search for Related Content PubMed PubMed Citation Articles by SMITH, L. N. Articles by MONGA, M.

Prior Knowledge of Obstetric Gestational Age and Possible Bias of Ballard Score

From the Departments of Obstetrics, Gynecology & Reproductive Sciences and Pediatrics, University of Texas Houston Medical School, Houston, Texas.

Address reprint requests to: Manju Monga, MD, Department of Obstetrics, Gynecology, and Reproductive Sciences, University of Texas Houston Medical School, 6431 Fannin, 3.268, Houston, TX 77030, E-mail: mmonga{at}obg.med.uth.tmc.edu

	Abstract

Top Abstract Methods Results Discussion References

 
Objective: To determine whether prior knowledge of obstetric estimate of gestational age creates a bias in assignment of gestational age by the Ballard assessment score.

Methods: The Ballard assessment score was done on 82 infants weighing less than 2500 g by two examiners, one who had prior knowledge of best obstetric gestational age estimate and the other who was masked to that information. Obstetric gestational age was correlated with masked and unmasked score. Statistical analysis used Spearman rank correlation test, plotting the measurement means against the measurement differences, ², Wilcoxon signed-rank test, analysis of variance, and Student t test. Significance was assumed at P < .05.

Results: There was high reliability (r = 0.84–0.86) between obstetric gestational age estimates (33.3 ± 3.0 weeks) and the gestational age derived from masked (34.3 ± 3.0 weeks) and unmasked (34.0 ± 3.0 weeks) Ballard scores. The mean difference between unmasked scores and gestational age was 1.38 ± 1.15 weeks and that between masked scores and obstetric gestational age was 1.40 ± 1.15 weeks, a nonsignificant difference. There was no significant difference in the number of cases with 100% agreement between masked and unmasked scores and obstetric gestational age or in the number of cases in which the score (masked or unmasked) differed by more than 2 weeks from obstetric gestational age. This study had over 99% power to detect a 1-week difference between the accuracy of masked and unmasked scores.

Conclusion: Prior knowledge of obstetric gestational age did not bias the Ballard assessment score.

Standardized scoring systems such as the Dubowitz¹ and Ballard^2,3 examinations are used to estimate gestational age at birth. Postnatal estimation of gestational age is an important prognosticator in premature or low birth weight infants in whom obstetric gestational age is unknown (such as those born to women who receive limited prenatal care or are uncertain about their last menstrual period). The Dubowitz and Ballard examinations were validated previously by comparing each scoring method with known obstetric gestational ages. In most of those studies, examiners were aware of obstetric gestational ages.^1,4 In other studies,^2,3,5 which were held as criterion standards, examiners were masked to obstetric gestational ages. Prior knowledge of obstetric gestational age might bias the Ballard score, which would artificially lead to greater concordance between the Ballard estimation of gestational age and the obstetric gestational age. The objective of this study was to determine whether prior knowledge of obstetric dating criteria creates a bias in the assignment of gestational age by the Ballard assessment score.

	Methods

Top Abstract Methods Results Discussion References

 
This study was conducted at Hermann Hospital in Houston, Texas. Two authors (L.N.S. and V.H.D.) were formally trained to assign Ballard assessment scores. After receiving training, L.N.S. and V.H.D. assigned scores while masked to obstetric gestational ages of ten infants, and their scores were compared using the Spearman rank correlation test to ensure that no a priori bias existed between examiners.

Infants who weighed less than 2500 g were enrolled in the study. Each infant had a Ballard examination by L.N.S. and V.H.D. within 96 hours of birth. One examiner determined the best obstetric estimate of gestational age from the mother’s chart (based on last menstrual period and earliest ultrasound) before assigning the score. The other examiner was masked to obstetric gestational age estimates. For this study, best obstetric estimate of gestational age was determined by last menstrual period if no ultrasound was available. Eighty percent of women had ultrasound done in the first or second trimester. Ultrasound was used to assign obstetric gestational age if ultrasound measurements differed from last menstrual period data by more than 1 week in the first trimester or more than 2 weeks in the second trimester. Exclusion criteria included congenital malformations, pharmacologic sedation or paralysis, critical illness, or inability of both examiners to assign the Ballard assessment score within 96 hours of birth.

Statistics were analyzed by Spearman rank correlation test, plotting the measurement means against measurement differences,^{6 2}, Wilcoxon signed-rank test, analysis of variance (Neuman Keul multiple-comparison test), and the Student t test where appropriate. Significance was assumed at P < .05. Power calculations were done with an alpha of 0.05 and beta of 0.99.

	Results

Top Abstract Methods Results Discussion References

 
The rank correlation test of Ballard scores assigned by the two examiners on ten infants found no significant difference between examiners (r = 0.85), providing internal validation of Ballard scores. Between January 1, 1998 and April 30, 1998, 82 infants who weighed less than 2500 g were enrolled in the study. The mean (± standard deviation [SD]) birth weight was 1805 ± 423 g, and the mean obstetric gestational age was 33.3 ± 3.0 weeks. Eighty-five percent of infants were born before 37 weeks’ gestation. Table 1 shows pregnancy complications in the study group. There was no difference in postnatal age at which Ballard score was assigned in the known obstetric gestational age group (unmasked) versus the masked group (57.6 ± 29.3 versus 57.3 ± 29.0 hours).

The distribution of differences between Ballard scores and obstetric gestational ages is shown in Figure 1. The mean difference between the unmasked Ballard score and obstetric gestational age was 1.38 ± 1.15 weeks, and the mean difference between the masked Ballard score and the obstetric gestational age was 1.40 ± 1.15 weeks, not a significant difference. Agreement between obstetric estimates of gestational age and postnatal determinations of gestational age (using the masked or unmasked Ballard scores) was excellent. The 95% confidence interval (CI) was ± 2.3 weeks. Masked and unmasked Ballard scores did not differ significantly in the number of times in which there was 100% agreement with obstetric gestational age (23.8% versus 26.8%, respectively). There was also no significant difference in the number of cases in which Ballard scores (masked or unmasked) differed by more than 2 weeks from the obstetric gestational ages (15% versus 12.5%, respectively). More than 85% of Ballard scores agreed with the best estimate of obstetric gestational age within 2 weeks.

View larger version (12K): [in this window] [in a new window]   Figure 1. Differences between Ballard score estimations of gestational age (done with [unmasked] or without [masked] prior knowledge of obstetric estimate of gestational age) and best obstetric estimate of gestational age.

	Discussion

Top Abstract Methods Results Discussion References

On the basis of a MEDLINE search from 1966 to September 1998 using the search terms "Ballard" and "gestational age," we did not find any study that investigated the influence of prior knowledge of obstetric gestational age on the Ballard score. Gagliardi et al⁵ reviewed charts of 302 infants in whom neonatologists, who were aware of obstetric estimates of gestational age, recorded the Dubowitz estimates of gestational ages graphically with nomograms (score entered on the abscissa and estimated gestational age read from the ordinate). The authors found that conversion of the Dubowitz score to gestational age by using an algebraic equation resulted in lower correlation with obstetric gestational ages than Dubowitz-calculated gestational ages recorded by the neonatologists. They suggested that the Dubowitz neonatal assessment score might be influenced by concurrent information, such as knowledge of obstetric estimate of gestational age.⁵

Based on our literature search, this is the first prospective study to assess potential bias of assignment of neonatal gestational age by prior knowledge of the obstetric dating criteria. We hypothesized that prior knowledge of the obstetric estimate of gestational age would bias the neonatal assessment system, leading to greater concordance between the Ballard estimation of gestational age and the known obstetric gestational age. We chose to study only infants weighing less than 2500 g at birth because accurate assignment of scores is important for interpretation of gestational age-related perinatal events and outcomes in preterm infants. We found no significant difference in gestational age assessment by Ballard scores with or without prior knowledge of obstetric gestational ages; there was no discernible bias. With our pooled variance of 2.47 weeks, this study had over 99% power to detect a 1-week difference between the accuracy of masked and unmasked Ballard scores compared with obstetric gestational ages. Masked and unmasked Ballard scores were highly reliable when compared with obstetric gestational ages (confirmed by first or second trimester ultrasound in 80% of cases). The 95% CI of the difference between neonatal assessment and obstetrically derived gestational age was ± 2.3 weeks for masked and unmasked Ballard scores, which compared favorably with those of previously reported studies.^2–4

Neonatal assessment scores by masked or unmasked examiners more often resulted in overestimation than underestimation of gestational age, which was consistent with previous studies of systematic overestimation using neonatal assessment tests in up to 50% of preterm infants.^3,4,7,8 Possible reasons for overestimation include development of neonatal assessment tests in term and post-term infants with subsequent extrapolation to pre-term infants; inaccurate estimation of gestational age by last menstrual period alone in early studies; and medical complications such as hypertension, which increase gestational age assignment by neonatal assessment scoring tests.^7,8

	Footnotes

 
PII S0029-7844(98)00530-4

Received September 4, 1998. Received in revised form October 26, 1998. Accepted November 12, 1998.

	References

Top Abstract Methods Results Discussion References

 
1. Dubowitz LM, Dubowitz V, Golderg C. Clinical assessment of gestational age in the newborn infant. J Pediatr 1970;77:1–10.[Medline]

2. Ballard JL, Novak KK, Driver M. A simplified score for assessment of fetal maturation of newly born infants. J Pediatr 1979;95:769–74.[Medline]

3. Ballard JL, Khoury JC, Wedig K, Wang L, Eilers-Walsman BL, Lipp R. New Ballard Score, expanded to include extremely premature infants. J Pediatr 1991;119:417–23.[Medline]

4. Malan AF, Higgs SC. Gestational age assessment in infants of very low birthweight. Arch Dis Child 1975;50:322–4.[Abstract]

5. Gagliardi L, Scimone F, DelPrete A, Petecca C, Stival G, Pasinetti G, et al. Precision of gestational age assessment in the neonate. Acta Paediatr 1992;81:95–9.[Medline]

6. Bland JM, Altman DG. Statistical methods for assessing agreement between two methods of clinical measurement. Lancet 1986;1:307–10.[Medline]

7. Sanders M, Allen M, Alexander GR, Yankowitz J, Graeber J, Johnson TR, et al. Gestational age assessment in preterm neonates weighing less than 1500 grams. Pediatrics 1991;88:542–6.[Abstract/Free Full Text]

8. Alexander GR, de Caunes F, Hulsey TC, Tompkins ME, Allen M. Validity of postnatal assessments of gestational age: A comparison of the method of Ballard et al. and early ultrasonography. Am J Obstet Gynecol 1992;166:891–5.[Medline]

This article has been cited by other articles:

				M. H. Wyckoff, C. El-Turk, A. Laptook, C. Timmons, F. H. Gannon, X. Zhang, S. Mumm, and M. P. Whyte Neonatal Lethal Osteochondrodysplasia with Low Serum Levels of Alkaline Phosphatase and Osteocalcin J. Clin. Endocrinol. Metab., February 1, 2005; 90(2): 1233 - 1240. [Abstract] [Full Text] [PDF]

				I. P. Gray, P. A. Cooper, B. J. Cory, M. Toman, and N. J. Crowther The Intrauterine Environment Is a Strong Determinant of Glucose Tolerance during the Neonatal Period, Even in Prematurity J. Clin. Endocrinol. Metab., September 1, 2002; 87(9): 4252 - 4256. [Abstract] [Full Text] [PDF]

This Article Abstract Full Text (PDF) Alert me when this article is cited Alert me if a correction is posted Services Similar articles in this journal Similar articles in PubMed Alert me to new issues of the journal Download to citation manager Citing Articles Citing Articles via HighWire Citing Articles via Google Scholar Google Scholar Articles by SMITH, L. N. Articles by MONGA, M. Search for Related Content PubMed PubMed Citation Articles by SMITH, L. N. Articles by MONGA, M.