Currently, all readyforwhatsnext model software is written in R (for model module libraries), R Markdown (for analysis programs and reporting sub-routines) and JavaScript (for the user interface component of Shiny applications) using the ready4 framework.

Therefore:

to use readyforwhatsnext model module libraries and programs / subroutines you must have an up to date version of R and the ready4 R library installed on your machine and it is recommended that you install the RStudio IDE; and
the requirements for using readyforwhatsnext model user interfaces depend on whether you are running a version we have deployed to the web (in which case you just need a supported browser) or whether you are running the app on your local machine (in which case you will need R, the ready4 library and RStudio).

2 - Installing readyforwhatsnext model modules

To implement a modelling analysis with readyforwhatsnext you need to install model module R libraries.

Before you install

If you plan on using existing readyforwhatsnext modules for a modelling project, you can review currently available module libraries, to identify which libraries are relevant to your project.

However, please note that no readyforwhatsnext module library is yet available as a [production release](https://www.ready4-dev.com/docs/software/status/production-releases/. You should therefore understand the limitations of using readyforwhatsnext model software development releases before you make the decision to install this software.

Installation

readyforwhatsnext module libraries are currently only available as development releases, so you will need to use a tool like devtools to assist with installing readyforwhatsnext R packages directly from our GitHub organisation. If you do not have devtools on your machine you can install it with the following command.

install.packages("devtools")

The command to install each readyforwhatsnext module takes the following format.

devtools::install_github("ready4-dev/PACKAGE_NAME")

For example, if you are planning to predict health utility using some of the mapping algorithms that we have previously developed, you can install the youthu library with the following command.

devtools::install_github("ready4-dev/youthu")

Configuration

A small number of readyforwhatsnext modules require that you configure some of the dependencies installed with them before they can be used. In particular:

if you are using modules from the TTU package to undertake a utility mapping study, you will need to have both installed and configured the cmdstanr R package as per the instructions on that package’s documentation website; and
if you are using the mychoice package to undertake a discrete choice experiment study and are using a Mac, you need to ensure that you have a Fortran compiler installed. Some relevant advice on this: https://mac.r-project.org/tools/ .

Try it out!

Before you apply readyforwhatsnext modules to your own project, you should make sure you can run some or all of the example code included in relevant library vignette articles. The package website URL takes the form of https://ready4-dev.github.io/PACKAGE_NAME/articles/ (e.g. the vignettes for the youthvars package are available at https://ready4-dev.github.io/youthvars/articles/).

3 - Terms of use

readyforwhatsnext is distributed without warranties under open source licenses - we just ask you to appropriately cite it.

3.1 - Open source licensing

readyforwhatsnext is freely available to all under copy-left licensing arrangements.

To help ensure the models we develop are as transparent as possible and to make their algorithms as useful to others as possible, all readyforwhatsnext software is free and open-source. You are encouraged to make as widespread use of our software, including the creation of derivative works, as you see fit, so long as it is consistent with each item’s license. Our software is typically licensed under GPL-3, a copy-left open-source licensing regime.

3.2 - Citing readyforwhatsnext

If you find readyforwhatsnext useful, please cite it appropriately - it is easy to do!

To make it easier to cite our software, each software item bundle includes a CITATION.cff file. Inclusion of this file means that the repositories storing our software can generate appropriate citations in the format of most relevance to you.

Currently:

Zenodo provides a free text field under the heading “Cite as” which enables you to generate a wide range of citation manager and journal specific citation outputs. There is also an “Export” tool that will generate citation metadata in multiple output formats;
OpenAire Explore has a “Cite this software” button that allows you to generate a citation in multiple journal formats or to download BibTeX or RIS files;
Github repositories have a “Cite this repository” button that can generate both BibTeX and APA output as well as link to the Citation.cff file.

Additionally, we have included a CITATION file in each of our R libraries so that you can generate a citation from within an R session using the citation function (for example: citation("ready4").

3.3 - Disclaimer

readyforwhatsnext is distributed without any warranties.

All readyforwhatsnext model software is distributed in the hope that it will be useful, but WITHOUT ANY WARRANTY; without even the implied warranty of MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.

Furthermore, no readyforwhatsnext model software is yet sufficiently well documented and tested to be given a production release. All readyforwhatsnext model software should therefore viewed as experimental development releases.

4 - The readyforwhatsnext model

readyforwhatsnext is an in-development modular economic model of the systems shaping the mental helath of young people. It is comprised of four sub-models.

4.1 - Modules for modelling people

Modules to model the characteristics, relationships, behaviours, risk factors and outcomes of young people and individuals who interact with young people are collectively referred to as the “Spring To Life” sub-model. A table summarising Spring To Life module libraries for modelling people is available. Additional information (e.g. tutorials and blog articles) about currently available Spring To Life modules is labelled with the “model-modules-people” tag. Resources about Spring To Life datasets are tagged with “data-datasets-people”. Brief information about additional unreleased Spring To Life modules that are in development is also available.

4.2 - Modules for modelling places

Modules for spatio-temporal modelling of the environments that shape young people’s mental health are collectively referred to as the “Springtides” sub-model. Both Springtides module libraries for modelling places that are available are highly preliminary and lack tutorials to demonstrate their use. A deprecated app built using these libraries is available for illustration purposes. Resources relating to preliminary and unreleased modules for the Springtides model is tagged with the “model-modules-places” tag and those relating to compatible datasets are tagged with “data-datasets-places”. Brief information on unreleased work in progress module libraries are also available.

4.3 - Modules for modelling platforms

Modules that model the processes, eligibility requirements, staffing and configurations of youth service platforms are collectively referred to as the “First Bounce” sub-model. No First Bounce modules are yet available - see details on unreleased work in progress.

4.4 - Modules for modelling programs

Modules for modelling the efficacy, cost-effectiveness and budget impact of youth mental health programs (e.g. interventions for prevention, treatment and wellbeing) are collectively referred to as the “On Target” sub-model. There are currently two development releases of On Target module libraries for modelling programs but both are highly preliminary. Resources (including tutorials) relating to these module libaries is tagged with “model-modules-programs”.

5 - Modules pipeline

Unreleased software and other preliminary work is currently being developed into readyforwhatsnext model modules.

5.1 - Pipeline of people modules

Current unreleased work to develop modules for modelling the characteristics, relationships, behaviours, risk factors and outcomes of young people and those important to them.

Our current pipeline of modules for modelling people is principally focused on developing tools for:

creating synthetic household datasets from multiple longitudinal datasets of varying structure, including modules specifically designed to streamline wrangling data from the HILDA and LSAC datasets (both from Australia); and
implementing agent based model simulations.

A significant amount of work has already been completed on the first project and initial development releases of each, along with one scientific manuscript, are planned for late 2024.

5.2 - Pipeline of places modules

Current unreleased work to develop modules for modelling the demographic, environmental and proximity drivers of access, equity and outcomes in youth mental health.

Our current pipeline of modules for modelling places (from the Springtides sub-model) will extend the libraries listed in summary table of module libraries for modelling places to:

predict prevalence and incidence by area; and
provide a user-interface (i.e. software to implement an updated version of the currently deprecated Springtides app).

Although unreleased, the source code for the above projects has been used to generate analysis during the early phase of the COVID-19 pandemic. Initial development releases of places module libraries, along with an updated app, are anticipated in the second half of 2024.

5.3 - Pipeline of platforms modules

Current unreleased work to develop modules for modelling the optimal staffing and configuration of support services for young people.

Our current pipeline of modules for modelling platforms includes code for implementing:

a discrete event simulation of primary mental health services for young people;
a simple cohort model of early psychosis services; and
a blended (systems dynamics / discrete event simulation) model for optimising eligibility and referral policies across multiple services.

The first two of the above models are currently implemented in R and are sufficiently advanced to produce exploratory analysis. However, neither are adequately documented or tested and need to be redeveloped as First Bounce sub-model modules and re-validated prior to development releases. The optimisation model was implemented in Java and was populated with toy data - this will require more substantial development prior to public release.

5.4 - Pipeline of programs modules

Current preliminary work to develop modules for modelling the affordability, value for money and appropriate targeting of interventions for young people.

We have no current pipeline of new module libraries for modelling programs (ie for the On Target sub-model). The currently released On Target libraries modules itemised in the summary table of module libraries for modelling programs are highly preliminary and are therefore our focus for future development in this area.