Updated the PediTools AAP 2022 bilirubin management tool to:

• provide clinical decision support for when phototherapy might be discontinued (2 mg/dL below the threshold at the time when phototherapy was initiated, or lower if rebound hyperbilirubinemia risk factors exist)
• simplified and cleaned up the output by consolidating it into its own area, instead of split between the different threshold notifications

Thanks to the suggestion of a PediTools user, I've added support for the CDC 2-20y growth charts for extended BMI calculations, for use when calculation of BMI is above the 95%ile using the standard charts.

Since 2013, there had been the option to support PediTools via PayPal donations. In February 2018, I made the decision to include advertising support on PediTools to offset development and maintenance costs.

As of 2022, the advertising revenue has been sufficient to support PediTools, so I've decided to remove all links to PayPal from the site.

However, I did want to thank and acknowledge the 32 individuals who supported PediTools (including 3 who provided monthly donations!), and have added them to an updated About PediTools page.

Many thanks to all of you!

On August 5, 2022, the AAP released an updated "Clinical Practice Guideline Revision: Management of Hyperbilirubinemia in the Newborn Infant 35 or More Weeks of Gestation".

In the past 3 days, I've developed an updated PediTools clinical decision support tool to assist in following these recommendations. It's still under active development — let me know if you notice any issues.

Because it will likely take some time for clinicians to change practice, the tool supporting the older 2004 AAP recommendations will remain, and the link will remain unchanged, to hopefully not disrupt any current workflows.

"PediTools Electronic Growth Chart Calculators: Applications in Clinical Care, Research, and Quality Improvement" has been published in the Jan 30, 2020 issue of the Journal of Medical Internet Research.

• Chou JH et al. Journal of Medical Internet Research. 2020;22(1):e16204 [PDF]

The article describes not only the deployment of the PediTools website, but also some applications of using large-scale growth anthropometric calculations for clinical research and quality improvement.

I chose JMIR because it's an Open Access journal — I wanted to make sure that it was available to everyone, and not just those lucky enough to have an institutional journal subscription. This does result in additional costs to the authors (the article processing fee was $2,500), but I really believe that science should be open to benefit as many as possible.

Reordered the icons on the front page to roughly match frequency of usage. If more charts are added, will probably want to split the tools onto separate pages (although I will make sure that any links / bookmarks folks have made won't be broken).

Went through all the growth calculators and modified them to give a (rather ugly) warning when any results are calculated from extrapolation outside of the published growth chart data. In retrospect, it may have been better to just not calculate using any extrapolated data, but that would be a major rewrite of a lot of code.

Summarized from the abstract: mid-upper-arm circumference (MUAC)-for-age z score for 5-19 year olds that accord with the World Health Organization growth standards. Curves were created using the Health Examination Survey (HES) / National Health and Nutrition Examination Survey (NHANES) US population datasets (age 5-25 years), which were used to construct the 2007 WHO growth reference for body mass index in this age group, and were merged with an imputed dataset matching the distribution of the WHO 2006 growth standards age 2-6 years.

I've added a tool that generates a Fenton 2013 growth chart, plotting longitudinal data of weight, head circumference, and length, from data copy and pasted from a properly formatted Excel spreadsheet document. Care has been taken to make it possible to upload the data without sending protected health information.

For every point, it also calculates percentiles and Z-scores. There is color-coding to help recognize if the baby is losing weight percentiles over time. Weekly weight gain is reported, as well as the weekly weight gain that would be needed to maintain the current weight percentile.

This tool has been in clinical use at my institution for years now and has been very helpful as a facet of growth + nutrition quality improvement projects.

Regular visitors to PediTools have surely noticed a change in the last week or so — the addition of advertisements.

I like the idea of transparency, and although I haven't seen other websites do this, I'd like to share some of the motivations and behind-the-scenes issues regarding running a site like PediTools.

First off, I'm a clinician and the genesis of PediTools stemmed from a desire to improve clinical care delivery to our patients. I'm also a self-avowed geek, so back in ~2011 when I recognized that clinical care could be improved if there were an easier way to do certain chores (e.g., documenting anthropometric measures), I developed a tool for my own use. I then recognized that others might find the tool helpful and so the public PediTools website was launched around January 2012.

My first motivation is to help our patients. Second, I'm a geek, and I find this a fun hobby to do in my so-called "free time" :) Incidentally, PediTools has not contributed anything to me professionally: it is not part of my job, it is not recognized by my employer, nor have I received any academic recognition for its development. That's OK — I'm happy when I think of all the babies and kids out there whose care might be just a little bit better because of PediTools. And as I said, I'm a geek, and I find it fun.

But there are also costs involved. From ~2012:

• web server hosting, ~$1,175
• software / development tools, ~$785

Back in 2013, I added a PayPal donation option. I'd like to thank those who have donated, some of whom have been remarkably generous! And regardless, just knowing that someone appreciated PediTools enough to want to say 'thank you' was a great motivator! I know that I use a lot of free or open-source tools and don't often enough express my appreciation, but that being said, since 2013 there have been 10 donors, for ~$310.

With the imbalance in costs to income and after the Feb 13, 2018 near collapse of PediTools (probably from running very outdated software) that required emergent and time-consuming 'resuscitation', I had a brief moment of soul-searching when I wondered why I was doing this. Thankfully, it was a brief moment, and the answer was easy — it's about the kids.

But, I felt like I needed to justify the costs imbalance. Honestly, part of the reason why the development tools got so woefully out-of-date and then failed was because I didn't want my 'hobby' to cost even more, and updates can be costly. And so, in an effort to make sure I can keep PediTools up and running, I've decided to try out advertising.

Remember, I use PediTools, too. I've tried to set up the advertisements so that they do NOT interfere with the usability of the site, whether on a desktop browser or on a mobile device. If the advertisements are causing problems for you, please let me know and I'll try to adjust them.

Hopefully, this step will help PediTools run smoothly for the long term. And perhaps this post will provide some background as to why, because it wasn't a choice I made lightly.

For management of hyperbilirubinemia, the trend of bilirubin levels is much more useful than a single point in time.

The hyperbilirubinemia management tool has been updated to be able to handle multiple bilirubin measurements at different ages, and plot all the points on the Bhutani nomogram as well as on the AAP recommended thresholds for phototherapy and exchange transfusion.

Please let me know if you notice any problems!

It's been a tough couple of days at PediTools, but the situation is stabilizing. Many components of the site have been upgraded or replaced.

Some fixes included updating and installing:


• Web development software to RapidWeaver v7.5.5
• Migrated the entire website to new version of RapidWeaver; changed directory paths for top-level pages
• Stacks plugin to v.3.5.7
• Gravity theme to v5.0.0
• With excellent technical support from Will Woodgate, received a modified version of the theme that adapts to small screens again
• Modified the Gravity theme to be compatible with secure socket layers (SSL / https://)
• Web server now has secure encrypted web access via SSL / https:// activated, and required
• Updated PHP running on the web server, which fixed a weird crash on calculating time intervals with diff()

Thanks to everyone who reported problems and thanks for your patience.

I have been maintaining the PediTools website with some very old web design software which has finally become so obsolete as to cause formatting and display problems.

I updated the site today to use the most recent software and plugins, which seems to have fixed some issues, but is still causing some problems with formatting. For example, when displaying on a mobile device (e.g., iPhone), the page no longer reformats itself for the smaller screen.

Apologies, but I'm not sure I have the expertise to correct these issues. I'll keep trying, though.

At the extremes of the charts, predicted percentiles (and Z-scores) much over the 95th percentile using the CDC charts may be inaccurate.

The PediTools CDC 2 - 20y calculator has been updated to classify obesity using the BMI relative to the 95th percentile BMI for more severe obesity. In brief, categories of obesity for children or adolescents >2 years of age (per Styne 2017):

• Overweight, for BMI >= 85th but <95th percentile
• Obese, for BMI >= 95th percentile
• Extremely obese if the BMI is >=120% of the 95th percentile or >=35 kg/m2

Thanks to the dietician who provided this feedback!

Uses data from the United States CDC National Health and Nutrition Examination Survey (1999 - 2012) to calculate mid-upper arm circumference percentiles and Z-scores on children from 2 months to 18 years of age.

Calculates Z-scores and percentiles based on the Zemel 2015 study for children ages 0 - 36 months and 2 - 20 years in United States with Down syndrome. Includes weight, length, head circumference, weight for length (for 0 - 36 months), and BMI for age (for 2 - 20 years).

On testing, the results seem to match the published work pretty well, but there are definitely small deviations that I am attributing to the use of only 2 significant digits of precision in the reference parameters in the publication supplementary materials. So, please don’t be surprised if the percentiles are 1 - 2% off.

Calculates Z-scores and percentiles based on the Olsen 2015 reference curves (gender-specific) for BMI in prematurely born infants. Please let me know if there are any problems or bugs.

On the Fenton 2013 growth calculator page, the decision support was amended to include the 2014 AAP Policy Statement on RSV prophylaxis with palivizumab (reference link at the bottom of the page).

Because some units are continuing to use the 2009 guidelines, I have not removed the older policy recommendations (yet).

Implemented new features:

• WHO 0 - 24 month growth calculator, added predicted increases in weight / head circumference / length to maintain current percentile, for both chronologic and corrected age
• CDC 2 - 20 year growth calculator, added ability to also display calculations for corrected age, for children who were born premature
• Gestational age calculator, updated to properly display age when > 1 year old

Thanks to the PediTools users who pointed out the problems and made suggestions for improvement!

Added a hyperbilirubinemia management tool, based on the AAP 2004 recommendations.

In addition to specifying the Bhutani nomogram risk zone and displaying the phototherapy initiation thresholds, it also displays the exchange transfusion thresholds, and also shows the point plotted on the three different graphs (risk nomogram, phototherapy thresholds, exchange transfusion thresholds).

To consider doing: add date/time of birth and lab draw, rather than entering hours of age.

Easy to do (actually, already done), but probably not much demand: ability to plot multiple points on the graphs, to better visualize trend and effect of therapy.

Note: when plotted, the points on the graph are sometimes a fraction off. This is because the figures as published are not perfectly horizontally / vertically aligned. The error is probably too small to be clinically significant. The reported numeric thresholds should be more accurate.