The story of this blog either starts a month ago, or at the beginning of 2020, or sometime during the year 2014, depending on your definition of ‘starts’. Back in 2014 I wrote a guide to building Health State Transition models in R, because I was getting my start doing it and found the available guidance to be hard to follow. I didn’t have a background in coding at all, and all the resources I could find seemed to be based on the assumption that I already knew what I was doing and just needed some help with syntax. I figured other people were likely to be in similar shape so I thought I’d take some time to spell out some of the basics that I’d pieced together over a million Stack Overflow pages and conversations with my buddy Stavros.
And then… I did nothing with it. The damn thing sat on my hard drive for years. I’d had it in my head that I would post the thing as a blog, and keep updating it as I worked. But I was also two years into my PhD and a budding music career and a new relationship and little by little the blog slipped down the priorities list. Because my anxiety tells me that things done imperfectly are not worth doing at all, I shelved those hours of work until “the right time”. Holla if you hear me. I shared it with a few colleagues, who said they found it helpful, but I never bothered publishing it anywhere.
When I left my job at the beginning of this year, I thought it was the right time to revisit the idea of the blog. After all, I had time now. I re-wrote the whole guide in Markdown, updated the code, and got it ready to publish. But around that same time I managed to find new distractions, including a new position at the Canadian Agency for Drugs and Technologies in Health. One of the opportunities that position afforded me was the chance to lead a small team in developing programs in R to help increase our productivity and our familiarity with what is rapidly becoming the standard approach to statistical and health economic analysis. Once again, more urgent issues took their place atop my ‘To Do’ list.
Then, about a month ago, I got really fed up with one of the principal “distractions” in my life, a side project I’d been developing since 2016, and decided to step away from it for a while. This, plus the fact that I was home all the time because of the COVID-19 pandemic, opened up some mental and calendar space for me to finally get this darn thing online.
So what is this blog for?
I know there are a lot of people, trainees and senior researchers alike, who find R intimidating. Over the years, I’ve developed some tricks for doing health economic stuff in R. I’ve also been paying casual attention to the broader conversation that is happening about Open Source methods in science broadly, and health economics specifically. I believe that I am in a position to make some people’s lives less frustrating so they can get on with the work of making the world a better place. That’s what this blog is about.
I’m going to take time to write up some ‘explainers’ of some of the various programs I’ve written to do economic analysis and related computational work. The intended audience will be people who are interested in R but aren’t trained as developers, and are trying to do health economics research. I also have a bunch of unpublished work that I did for my PhD that uses Python. Basically I developed a relatively simple(-ish) way of doing Discrete Event Simulation (DES) that is free and shareable. I’d like to talk a bit about that as well.
A hopefully unnecessary disclaimer: everything written here reflects my own personal opinion, and is not reflective of any position taken by my employer. I can’t imagine my employer has a position on decision models in R, but if they do you won’t find it on this blog. K? Mmm-k.