An alternate perspective on creating NICU growth charts
Reese Clark, Pediatrix Medical Group
5 August 2013
We respect the important contributions that Dr. Fenton and colleagues have made in their recent papers (1, 2) and we submit this letter to make several comments.
We noted 2 technical errors of concern in these publications: the LMS method skewness value (measured in degrees of freedom) was not set to 1 when creating the Olsen (3) or the WHO (4) weighted curves as stated in Table 1 of the Fenton, et al., paper(1). We know this firsthand for our curves, and the WHO methods state "The weight data for both sexes were skewed, so in specifying the model, the parameter related to skewness was fitted in addition to the median and the approximate coefficient of variation. In modeling skewness the girls' curves required more degrees of freedom to fit a curve for this parameter."(4)
In addition, we respectfully disagree with several elements of the Fenton curves' design: We reformatted our curves following many requests for a version that would span the NICU stay for most preterm infants, similar to the Fenton curves. The revisions include gender-specific 1-page versions of the Olsen intrauterine growth curves (3) that also now include The WHO Child Growth Standards (commonly known as the WHO growth charts-- http://www.who.int/childgrowth/standards/en/). Our updated charts are available at the following link: http://www.pediatrix.com/workfiles/NICUGrowthCurves7.30.pdf.
However, there are two important distinctions between the reformatted version of the Olsen curves and the recently revised Fenton curves (1). First, we begin the WHO curves in our graphs at 39 weeks' postmenstrual age (instead of 40 weeks) because 39 weeks is the median age for the WHO definition of full-term (range 37 to 41 wk) and 39 weeks is the median age of U.S. births (http://wonder.cdc.gov/controller/datarequest/D66).
The second consideration is that curves based on independent sets of data should not be connected because the smoothing process distorts the data. An example of this distortion can be observed in Figure 5 of the Fenton, et al., paper(1); the Fenton smoothed curves pull the WHO percentiles up from the dotted lines (original WHO curves) to the bold, dashed lines (smoothed version of the WHO curves). For this reason, we did not connect the Olsen data and WHO data in the reformatted version of our curves. Our updated charts are available for review (http://www.pediatrix.com/workfiles/NICUGrowthCurves7.30.pdf). It is our belief that this representation is a more accurate depiction of the proper presentation of two different data sets.
Irene E. Olsen, Ph.D., R.D., L.D.N.
M. Louise Lawson, Ph.D.
Reese H. Clark, M.D.
Alan R. Spitzer, M.D.
Reference List
1. Fenton TR, Kim JH. A systematic review and meta-analysis to revise the Fenton growth chart for preterm infants. BMC Pediatr 2013;13:59.
2. Fenton TR, Nasser R, Eliasziw M, et al. Validating the weight gain of preterm infants between the reference growth curve of the fetus and the term infant. BMC Pediatr 2013;13(1):92.
3. Olsen IE, Groveman SA, Lawson ML, et al. New intrauterine growth curves based on United States data. Pediatrics 2010;125(2):e214-24.
4. WHO Multicentre Growth Reference Study Group. WHO Child Growth Standards: Length/height-for-age, weight-for-age, weight-for-length, weight-for-height and body mass index-for-age: Methods and development. In. Geneva: World Health Organization; 2006.
An alternate perspective on creating NICU growth charts
5 August 2013
We respect the important contributions that Dr. Fenton and colleagues have made in their recent papers (1, 2) and we submit this letter to make several comments.
We noted 2 technical errors of concern in these publications: the LMS method skewness value (measured in degrees of freedom) was not set to 1 when creating the Olsen (3) or the WHO (4) weighted curves as stated in Table 1 of the Fenton, et al., paper(1). We know this firsthand for our curves, and the WHO methods state "The weight data for both sexes were skewed, so in specifying the model, the parameter related to skewness was fitted in addition to the median and the approximate coefficient of variation. In modeling skewness the girls' curves required more degrees of freedom to fit a curve for this parameter."(4)
In addition, we respectfully disagree with several elements of the Fenton curves' design: We reformatted our curves following many requests for a version that would span the NICU stay for most preterm infants, similar to the Fenton curves. The revisions include gender-specific 1-page versions of the Olsen intrauterine growth curves (3) that also now include The WHO Child Growth Standards (commonly known as the WHO growth charts-- http://www.who.int/childgrowth/standards/en/). Our updated charts are available at the following link: http://www.pediatrix.com/workfiles/NICUGrowthCurves7.30.pdf.
However, there are two important distinctions between the reformatted version of the Olsen curves and the recently revised Fenton curves (1). First, we begin the WHO curves in our graphs at 39 weeks' postmenstrual age (instead of 40 weeks) because 39 weeks is the median age for the WHO definition of full-term (range 37 to 41 wk) and 39 weeks is the median age of U.S. births (http://wonder.cdc.gov/controller/datarequest/D66).
The second consideration is that curves based on independent sets of data should not be connected because the smoothing process distorts the data. An example of this distortion can be observed in Figure 5 of the Fenton, et al., paper(1); the Fenton smoothed curves pull the WHO percentiles up from the dotted lines (original WHO curves) to the bold, dashed lines (smoothed version of the WHO curves). For this reason, we did not connect the Olsen data and WHO data in the reformatted version of our curves. Our updated charts are available for review (http://www.pediatrix.com/workfiles/NICUGrowthCurves7.30.pdf). It is our belief that this representation is a more accurate depiction of the proper presentation of two different data sets.
Irene E. Olsen, Ph.D., R.D., L.D.N.
M. Louise Lawson, Ph.D.
Reese H. Clark, M.D.
Alan R. Spitzer, M.D.
Reference List
1. Fenton TR, Kim JH. A systematic review and meta-analysis to revise the Fenton growth chart for preterm infants. BMC Pediatr 2013;13:59.
2. Fenton TR, Nasser R, Eliasziw M, et al. Validating the weight gain of preterm infants between the reference growth curve of the fetus and the term infant. BMC Pediatr 2013;13(1):92.
3. Olsen IE, Groveman SA, Lawson ML, et al. New intrauterine growth curves based on United States data. Pediatrics 2010;125(2):e214-24.
4. WHO Multicentre Growth Reference Study Group. WHO Child Growth Standards: Length/height-for-age, weight-for-age, weight-for-length, weight-for-height and body mass index-for-age: Methods and development. In. Geneva: World Health Organization; 2006.
Competing interests
None