This page is running a macro package for R hosted on a Shiny server. It calculates Z-scores and percentiles for children with Prader-Willi syndrome, based on LMS tables developed by Butler et al, 127(4):687-95, 2011. These authors developed separate tables for ages 0-36 months (length, weight, head circumference) and 3-18 years (height, weight, head circumference, and body mass index). You must therefore choose the appropriate tables via the selector in the sidebar. Using the published LMS tables, 'exact' LMS parameters are interpolated for age in months prior to calculating Z-scores. Observations with ages that are out-of-range for the selected tables may be submitted, but will simply return Z = NA (Z-scores will be automatically re-calculated when you revise the LMS table selection).
The sample.csv file shows the expected variable names and formats. Column order is immaterial, and additional columns are permitted. For assistance, please consult our step-by-step guide: Creating .csv spreadsheets . The identifier variable id must be unique for each observation. The LMS algorithm is based on exact age in months (agemons). The variable sex may be coded as M/F or m/f or 1/2 (1 = male); height and head circumference are in cm, and weight is in kg. For 0-36 months, height refers to recumbent length (standing height + 0.7cm). For children ≥ 36 months, height refers to standing height (recumbent length - 0.7cm). As recommended by the WHO for skew distributions, Z-scores for weight and BMI outside of the range [-3, 3] are calculated in units of SD23, the distance between Z=2 and Z=3 (or in the lower tail of the distribution, Z = -2 and Z = -3). In contrast, CDC macros typically report these extreme values as > 3 or < -3, for percentiles > 99.9 or < 0.1%.
A spreadsheet with comma separated variables (.csv) may be created using the 'Save As' .csv option in Excel and uploaded using the sidebar on the left. Once results are displayed, download them by clicking the <Download> button, which will typically save them to your Download folder with 'out_' prepended to the original dataset name.