R

More changes to survHE. Earlier this week, I gave a talk at the Virtual Ispor US conference (our panel was W11). In preparation for the panel (and other things I was doing anyway), I got to talk to colleagues and thought a bit more of how survival modelling in HTA is kind of weird and “special”… In particular, I realised how survHE doesn’t give by default any real information about the hazard and the cumulative hazard functions — this isn’t entirely true, because once the survival curve has been estimated (which is the whole point of survHE!

Philip Cooney has alerted me to a slight inconsistency in survHE. Basically, he was trying to work in a setting where the data contain no censoring, using the HMC/rstan models generated by survHE. All works OK when using some of the models implemented, but failed when trying to fit the Gamma, Generalised Gamma and Generalised F. The reason for this is that, for simplicity (and actually, our empirical tests show that computational run-time is much reduced) these models construct the log-likelihood by separating the set of individuals who are fully observed from the one of individuals who are censored.

A couple of colleagues have alerted us to some issues with the BCEA function evppi, under MS Windows. Basically, it seemed as if under MS Windows and with R 4.0.0 (which is a recent, though not the most recent release of R) the evppi function would break, when trying to make the analysis based on INLA (and the work discussed here).

Last Friday on Monday, we held our online R-HTA workshop. We will discuss the feedback and agree on ways forward in the consortium Scientific Committee soon, but I thought I’d write down some personal reactions to the very interesting discussions — of course, what come out of all this eventually may well be very different to my synthesis! Still… I thought the two days went very well — I have enjoyed them very much and I think so did the participants.

This comes as the results of two external forces that have prompted me to do some work on the website — specifically the section on books. The first one is the newest version of hugo-academic (which is the engine underlying the whole of my website, together with the R package blogdown). This has a new facility that can be used to format books or tutorial or documentation.

<img src=“refactoring.png”, width = “400”, height = “600” style=“float: left;margin-right: 20px;margin-top: 0px;margin-bottom:0px;"> This is quite exciting: since Nathan (this is his very interesting blog) has arrived to UCL a couple of months ago, we’ve started to work on quite a few of very interesting projects — including a major “refactoring” of the code for BCEA. I’m obviously very attached to BCEA — it’s basically my first R package and one I’ve spent lots of time thinking about and then working on.

Andrea has just released a major update for missingHE (this is my own fork of the main project — but now available on CRAN too), the package we (well, mostly he!) have (has) developed to handle missing data in Bayesian models for cost-effectiveness analysis. Additions to the basic structure of the package include multilevel models and new plotting facilities, revolving around posterior predictive checks, which are of course very helpful when assessing the performance of models with missing data.

THis is the second video I was mentioning here — took a while to get out but it’s available now. I think you need to register here and then you can see our panel discussion. Like I said earlier, it was good fun and I think the actual session we did at ISPOR last year was, I think, very well received and it’s a shame that we can’t build on the momentum in the next R-HTA (which, I think, we’re going to have to postpone, given the COVID-19 emergency…).

29-30 June 2020, University College London Training event (29 June): Torrington (1-19) 113 - Public Cluster, 1-19 in Torrington Place (https://goo.gl/maps/RtR3Ypug2Dq), University College London, United Kingdom Main workshop (30 June): Room G12, 1-19 in Torrington Place (https://goo.gl/maps/RtR3Ypug2Dq), University College London, United Kingdom. Background and objectives It is our pleasure to announce a workshop and training event on the use of R for trial and model-based cost-effectiveness analysis (CEA).

Boby Mihaylova has two exciting posts available at the Health Economics Research Centre at the University of Oxford. In particular, she is looking for two R-minded researchers/analysts to develop work on disease modelling/cost-effectiveness using large individual-patients databases. In fact, I think it’s really good that they are explicitly including knowledge of R as part of the job specification — and they even ask R code as part of the application!