About

Birth weight classification errors carry clinical consequences. A neonate at the 3rd percentile may require NICU admission for growth restriction monitoring, while one at the 97th may indicate maternal gestational diabetes. This calculator applies the WHO Child Growth Standards (2006) LMS method across gestational ages 24 - 42 weeks. It computes a Z-score via the Box-Cox power transformation and maps it to an exact percentile using the cumulative normal distribution. The underlying L, M, S parameters are sex-specific and interpolated from WHO reference data. Results approximate population norms and do not replace clinical assessment. Altitude, maternal BMI, ethnicity, and parity affect birth weight distributions but are not captured in these reference curves.

Formulas

The WHO LMS method models growth data using three age-and-sex-specific smoothing curves: L (Box-Cox power), M (median), and S (generalized coefficient of variation). Given a measured birth weight X, the Z-score is computed as:

Z = (XM)^L − 1L ⋅ S

When L = 0, the formula reduces to:

Z = ln(X ÷ M)S

The percentile P is derived by passing Z through the cumulative distribution function Φ of the standard normal distribution:

P = Φ(Z) × 100

Where: X = measured birth weight (g). L = Box-Cox transformation power. M = median weight for the given gestational age and sex (g). S = coefficient of variation. Z = standard deviation score. P = percentile rank (0 - 100). Φ = cumulative normal distribution function, approximated using the Abramowitz & Stegun rational method (algorithm 26.2.17) with six-decimal accuracy.

Reference Data

Gestational Age (weeks)	Boys Median (g)	Boys 10th %ile (g)	Boys 90th %ile (g)	Girls Median (g)	Girls 10th %ile (g)	Girls 90th %ile (g)
24	705	530	890	660	490	845
26	900	690	1130	850	640	1075
28	1140	880	1430	1080	820	1365
30	1450	1120	1810	1380	1050	1730
32	1820	1420	2260	1740	1340	2170
34	2270	1800	2790	2180	1710	2690
36	2750	2220	3330	2660	2130	3230
37	2950	2400	3550	2860	2310	3450
38	3150	2580	3770	3050	2490	3660
39	3320	2740	3960	3220	2640	3840
40	3460	2860	4100	3360	2760	3990
41	3550	2940	4200	3450	2840	4090
42	3600	2980	4260	3500	2890	4150

Frequently Asked Questions

The Z-score indicates how many standard deviations a baby's weight falls from the median for their gestational age and sex. A Z-score of 0 means the baby is exactly at the median. A Z-score below −2 (below the 2.3rd percentile) is classified as very low birth weight for gestational age per WHO guidelines. A Z-score above +2 (above the 97.7th percentile) may indicate macrosomia. Clinical thresholds vary by country; many use the 10th and 90th percentiles as cutoffs for SGA (Small for Gestational Age) and LGA (Large for Gestational Age).

Below 28 weeks gestational age, WHO reference data becomes sparse. The LMS parameters at extreme preterm ages (24-27 weeks) are derived from smaller population samples and carry wider confidence intervals. For clinical decisions at these ages, neonatologists typically reference the Fenton growth charts (2013 revision) or INTERGROWTH-21st standards, which include more granular preterm data. This calculator uses smoothed WHO-aligned estimates and should be treated as an approximation in the extreme preterm range.

Sexual dimorphism in birth weight is well-documented. Male neonates average approximately 100-150 grams heavier than females at the same gestational age. This difference is driven by androgen-mediated growth, placental function differences, and Y-chromosome-linked growth factors. Using sex-specific LMS parameters prevents misclassification - a 2900 g female at 40 weeks is at the 20th percentile, while the same weight for a male is at the 12th percentile.

SGA (Small for Gestational Age) below the 10th percentile can result from placental insufficiency, preeclampsia, chronic maternal hypertension, maternal smoking, congenital infections (TORCH complex), chromosomal anomalies, or severe maternal malnutrition. Symmetric growth restriction (proportionally small head and body) suggests early-onset causes. Asymmetric restriction (small body, preserved head circumference) suggests late-onset placental compromise. Isolated low percentile without pathology is also possible - constitutionally small parents produce smaller babies.

WHO growth standards and the LMS method operate in grams. This calculator accepts both units and converts internally. When entering pounds and ounces, the conversion is: weight in grams = (pounds × 453.592) + (ounces × 28.3495). For medical records, grams are standard internationally. US hospitals often record in pounds/ounces but convert to grams for growth chart plotting. Always verify the unit when transcribing from hospital records - a digit error in pounds produces roughly a 450 g discrepancy.

No. WHO birth weight references are based on singleton pregnancies. Multiples are systematically lighter - twins average 500-800 g less than singletons at the same gestational age. Plotting a twin's weight on singleton curves will produce artificially low percentiles. For multiples, use twin-specific references such as the Gielen (2017) or Ananth (2018) standards. This limitation is noted because approximately 3% of births are multiples, and misclassification can trigger unnecessary interventions.