Our overarching aim has been to bring better dosing methods to the ward, meaning we aim not only to make pharmacometric model-based calculations useable by clinicians, but to actually get them to use it. We want doses to be determined by the best possible method, meaning real patients receiving doses that were determined (at least in part), by the most robust and accurate calculation method possible – not just the easiest to understand or most popular protocol.
The project started with FirstDose – an antibiotic dose calculator for paediatric patients. The software went through several iterations, starting as a console-based Java app, then a flash version, then a more-compatible HTML and JavaScript version that would even run on Internet Explorer 6.
FirstDose was an HTML and JavaScript, fully client-side dose calculator used to recommend first and subsequent doses before concentration measurements become available for dose adjustment. It used published models for vancomycin, amikacin and gentamicin accounting for covariates such as height, weight, post-menstrual age, renal function and certain concurrent treatments. It was designed to be simple to use, simple to understand, and easily accessible from any computer on the ward.
We conducted a small clinical trial in paediatric and neonatal intensive care units in Auckland, and learnt valuable lessons regarding useability, software compatibility and communication with hospital staff. People use software differently and there was a need to anticipate unexpected input and guide users accordingly. Of significance, users were more interested in learning how to improve dosing once concentration measurements became available, so we began work on NextDose.
NextDose would use Bayesian methodology to make the best use of information available for target concentration intervention. It would need a database to keep track of patients, plus a focus on security, stability and collaboration between team members from different locations. Thanks to lessons learnt from FirstDose, initial development took place over just a few months before bringing to the hospital for initial evaluation.
NextDose launched as a browser-based web app with published models of busulfan, methotrexate and tacrolimus. It provides a relatively intuitive graphical user interface that drives a PHP server-side backend that communicates with NONMEM.
Dosing in the seriously ill is usually involves a multidisciplinary team, so NextDose is a collaborative tool, providing a common interface for all those involved in target concentration intervention and allows input from different locations at different times.
Today the development version of NextDose includes 11 medicines (busulfan, methotrexate, tacrolimus, warfarin, linezolid, voriconazole, gentamicin, amikacin, vancomycin, caffeine, mycophenolate) and has academic users around the world (NextDose n=295 , dev.NextDose n=122). The clinical version of NextDose has been used to guide dosing of busulfan in children (n=77) and adults (n=90) in Auckland Hospital since 2012.