Association between perinatal mortality and morbidity and customised and non-customised birthweight centiles in Denmark, Finland, Norway, Wales, and England: comparative, population based, record linkage study

Objectives To compare the risk of adverse perinatal outcomes according to infants who are born small for gestational age (SGA; <10th centile) or large for gestational age (LGA; >90th centile), as defined by birthweight centiles that are non-customised (ie, standardised by sex and gestational age only) and customised (by sex, gestational age, maternal weight, height, parity, and ethnic group). Design Comparative, population based, record linkage study with meta-analysis of results. Setting Denmark, Finland, Norway, Wales, and England (city of Bradford), 1986-2019. Participants 2 129 782 infants born at term in birth registries. Main outcome measures Stillbirth, neonatal death, infant death, admission to neonatal intensive care unit, and low Apgar score (<7) at 5 minutes. Results Relative to those infants born average for gestational age (AGA), both SGA and LGA births were at increased risk of all five outcomes, but observed relative risks were similar irrespective of whether non-customised or customised charts were used. For example, for SGA versus AGA births, when non-customised and customised charts were used, relative risks pooled over countries were 3.60 (95% confidence interval 3.29 to 3.93) versus 3.58 (3.02 to 4.24) for stillbirth, 2.83 (2.18 to 3.67) versus 3.32 (2.05 to 5.36) for neonatal death, 2.82 (2.07 to 3.83) versus 3.17 (2.20 to 4.56) for infant death, 1.66 (1.49 to 1.86) versus 1.54 (1.30 to 1.81) for low Apgar score at 5 minutes, and (based on Bradford data only) 1.97 (1.74 to 2.22) versus 1.94 (1.70 to 2.21) for admission to the neonatal intensive care unit. The estimated sensitivity of combined SGA or LGA births to identify the three mortality outcomes ranged from 31% to 34% for non-customised charts and from 34% to 38% for customised charts, with a specificity of 82% and 80% with non-customised and customised charts, respectively. Conclusions These results suggest an increased risk of adverse perinatal outcomes of a similar magnitude among SGA or LGA term infants when customised and non-customised centiles are used. Use of customised charts for SGA/LGA births—over and above use of non-customised charts for SGA/LGA births—is unlikely to provide benefits in terms of identifying term births at risk of these outcomes.


Birthweight centile customisation
Customisation was proposed by the group who lead the Perinatal Institute, with many previous studies using the Perinatal Institute methods and calculators to generate customised centiles.In Bradford data, we used the Perinatal Institute's global bulk calculator (version 8.0.4) [1] for deriving customised birthweight centiles.The calculator requires hosting participant data on the Institute's server whilst doing these calculations, and therefore, for information governance reasons, we could not use the calculator for customisation in other country data sets.The Perinatal Institute does not provide exact details of the coefficients used in their customization calculator, therefore we developed customised centiles within each dataset following the approach used by the Perinatal Institute as closely as possible from information on their website.[2] First, birthweight was regressed on gestational age, sex, maternal weight, height, parity, and ethnicity in observations with complete data on customisation variables.In the model, gestational age (in days) was centred at 280, and height and weight were centred at the population median, with quadratic and cubic terms added for weight.Parity was included as dummy variables (0 [reference category]/ 1 previous birth/ 2 previous births/ 3 previous births/ 4 or more previous births), and sex coded -1 for male infant and 1 for female.Dummy variables were included for all available ethnicity groups (with the majority group as the reference group).
We saved the coefficients, the constant and residual standard error (SE) from the regression (details in Table S4).The constant and the residual SE were used to derive the coefficient of variation (CV) (residual SE*100/constant).This was used to calculate the 10 th and 90th centile limits by multiplying the CV by the relevant z-score (i.e.-1.282 and 1.282, respectively).
For each observation (regardless of whether there was complete data on variables), we predicted birthweight at 280 days (term) based on the coefficients and the available information on infant sex and maternal height, weight, parity, and ethnic group.In the main analyses, we used all available data, assigning missing values with population median for height and weight and reference values for parity and ethnicity.In other words, when a birth observation had missing data on a customisation covariate such as weight, this coefficient would in effect drop from the equation for that observation, because the value would be 0 (the centred median value).We used the same equation as the Perinatal Institute (for gestational weeks ≥25) to derive the percentage birthweight expected for the specific gestational age compared to that at 280 days (term).Based on this, we calculated the "optimal" birthweight for the specific gestational age as a percentage of the term birthweight.Using the centile limits derived with the CV, we calculated whether the actual observed birthweight was above or below the 10 th (SGA) and 90 th (LGA) centiles of the optimal birthweight expected for an infant born at the same gestational age, with the same sex and maternal characteristics.

Data sets and sensitivity analyses
We conducted sensitivity analyses to assess the effect of the patterns of missing data on customisation variables in different countries, as described below.

FIGURE
FIGURE S1.Meta-analysis of risk ratios of perinatal adverse outcomes by SGA vs AGA (<10 th vs 10-90 th ) with non-customised and customised birthweight centiles (Bradford N=47,583, 2010-2019; Denmark N=384,885, 2004-2010; Finland N=576,758, 2004-2014; Norway N=276,078, 2012-2016; Wales N=844,478, 1986-2016) BMJ Publishing Group Limited (BMJ) disclaims all liability and responsibility arising from any reliance BMJ Publishing Group Limited (BMJ) disclaims all liability and responsibility arising from any reliance Supplemental material placed on this supplemental material which has been supplied by the author(s) FIGURE S4 -Meta-analysis of risk ratios of perinatal adverse outcomes by LGA (>90 th vs 10-90 th ) with non-customised and customised FIGURE S5 -Meta-analysis of risk ratios of perinatal adverse outcomes by SGA vs AGA (<10 th vs 10-90 th ) with non-customised and customised FIGURE S6 -Meta-analysis of risk ratios of perinatal adverse outcomes by LGA (>90 th vs 10-90 th ) with non-customised and customised placed on this supplemental material which has been supplied by the author(s) BMJMED doi: 10.1136/bmjmed-2023-000521 :e000521.2 2023; BMJMED , et al.Kilpi F

TABLE S2 -
Coefficients of birthweight centile customisation regression equations

TABLE S3A -
BRADFORD Risks of adverse perinatal outcomes with different birthweight centile customization methods and centile cut-offs in Bradford

TABLE S3B -
DENMARK Risks of adverse perinatal outcomes with different birthweight centile customization methods and centile cut-offs in Denmark

TABLE S3C -
FINLAND Risks of adverse perinatal outcomes with different birthweight centile customization methods and centile cut-offs in Finland

TABLE S3D -
NORWAY Risks of adverse perinatal outcomes with different birthweight centile customization methods and centile cut-offs in Norway

TABLE S3E -
WALES Risks of adverse perinatal outcomes with different birthweight centile customization methods and centile cut-offs in Wales

TABLE S4 -
Pooled estimates of sensitivity and specificity of SGA and LGA versus AGA (<10 th or >90 th vs 10-90 th ) for adverse perinatal outcomes by non-customised and customised birthweight centiles

TABLE S5A -
BRADFORD Sensitivity and specificity of different birthweight centile cut-offs in Bradford

TABLE S5B -
DENMARK Sensitivity and specificity of different birthweight centile cut-offs in Denmark

TABLE S5C -
FINLAND Sensitivity and specificity of different birthweight centile cut-offs in Finland

TABLE S5D -
NORWAY Sensitivity and specificity of different birthweight centile cut-offs in Norway

TABLE S5E -
WALES Sensitivity and specificity of different birthweight centile cut-offs in WalesBMJ Publishing Group Limited (BMJ) disclaims all liability and responsibility arising from any reliance Supplemental material placed on this supplemental material which has been supplied by the author(s)

TABLE S7 -
Denmark maternal country of origin frequencies and customisation coefficientsBMJ Publishing Group Limited (BMJ) disclaims all liability and responsibility arising from any reliance Supplemental material placed on this supplemental material which has been supplied by the author(s)