Selected Publications

A list of all my publications can be found here, with some recent ones described below:

In this paper I describe some substantial extensions to the survsim command for simulating survival data. survsim can now simulate survival data from a parametric distribution, a custom/user-defined distribution, from a fitted merlin model, from a specified cause-specific hazards competing risks model, or from a specified general multi-state model. I illustrate the command with some examples from each setting, demonstrating the huge flexibilty that can be used to better evaluate statistical methods.
Pre-print, 2020
PDF Stata package

In this article, I present the community-contributed stmixed command for fitting multilevel survival models. It serves as both an alternative to Stata’s official mestreg command and a complimentary command with substantial extensions. stmixed can fit multilevel survival models with any number of levels and random effects at each level, including flexible spline-based approaches (such as Royston–Parmar and the log-hazard equivalent) and user-defined hazard models. Simple or complex time-dependent effects can be included, as can expected mortality for a relative survival model. Left-truncation (delayed entry) is supported, and t-distributed random effects are provided as an alternative to Gaussian random effects. I illustrate the methods with a commonly used dataset of patients with kidney disease suffering recurrent infections and a simulated example illustrating a simple approach to simulating clustered survival data using survsim (Crowther and Lambert 2012, Stata Journal 12: 674–687; 2013, Statistics in Medicine 32: 4118–4134). stmixed is part of the merlin family (Crowther 2017, arXiv Working Paper No. arXiv:1710.02223; 2018, arXiv Working Paper No. arXiv:1806.01615).
Stata Journal, 2019
PDF Stata package

Simulation studies are computer experiments which involve creating data by pseudorandom sampling. The key strength of simulation studies is the ability to understand the behaviour of statistical methods because some `truth’ is known from the process of generating the data. This allows us to consider properties of methods, such as bias. While widely used, simulation studies are often poorly designed, analysed and reported. This article outlines the rationale for using simulation studies and offers guidance for design, execution, analysis, reporting and presentation. In particular, we provide: a structured approach for planning and reporting simulation studies; coherent terminology for simulation studies; guidance on coding simulation studies; a critical discussion of key performance measures and their computation; ideas on structuring tabular and graphical presentation of results; and new graphical presentations. With a view to describing current practice and identifying areas for improvement, we review 100 articles taken from Volume 34 of Statistics in Medicine which included at least one simulation study.
Stat Med, 2019
PDF

merlin can do a lot of things. From simple stuff, like fitting a linear regression or a Weibull survival model, to a three-level logistic mixed effects model, or a multivariate joint model of multiple longitudinal outcomes (of different types) and a recurrent event and survival with non-linear effects…the list is rather endless. merlin can do things I haven’t even thought of yet. I’ll take a single dataset, and attempt to show you the full range of capabilities of merlin, and discuss some future directions for the implementation in Stata.
2018
PDF

Multivariate data occurs in a wide range of fields, with ever more flexible model specifications being proposed, often within a multivariate generalised linear mixed effects (MGLME) framework. In this article, we describe an extended framework, encompassing multiple outcomes of any type, each of which could be repeatedly measured (longitudinal), with any number of levels, and with any number of random effects at each level. Many standard distributions are described, as well as non-standard user-defined non-linear models. The extension focuses on a complex linear predictor for each outcome model, allowing sharing and linking between outcome models in an extremely flexible way, either by linking random effects directly, or the expected value of one outcome (or function of it) within the linear predictor of another. Non-linear and time-dependent effects are also seamlessly incorporated to the linear predictor through the use of splines or fractional polynomials. We further propose level-specific random effect distributions and numerical integration techniques to improve usability, relaxing the normally distributed random effects assumption to allow multivariate $t$-distributed random effects. We consider some special cases of the general framework, describing some new models in the fields of clustered survival data, joint longitudinal-survival models, and discuss various potential uses of the implementation. User friendly, and easily extendable, software is provided.
arXiv, 2017
Project

Multi-state models are increasingly being used to model complex disease profiles. By modelling transitions between disease states, accounting for competing events at each transition, we can gain a much richer understanding of patient trajectories and how risk factors impact over the entire disease pathway. In this article we concentrate on parametric multi-state models, both Markov and semi-Markov, and develop a flexible framework where each transition can be specified by a variety of parametric models including exponential, Weibull, Gompertz, Royston-Parmar proportional hazards models or log-logistic, log-normal, generalised gamma accelerated failure time models, possibly sharing parameters across transitions. We also extend the framework to allow time-dependent effects. We then use an efficient and generalisable simulation method to calculate transition probabilities from any fitted multi-state model, and show how it facilitates the simple calculation of clinically useful measures, such as expected length of stay in each state, and differences and ratios of proportion within each state as a function of time, for specific covariate patterns. We illustrate our methods using a dataset of patients with primary breast cancer. User friendly Stata software is provided.
In Stats in Med, 2017
PDF Slides Stata package

Software

As part of my research I have developed a range of software packages in Stata and R. More details, including tutorials, can be found on the package-specific pages:

merlin ~ mixed effects regression for linear and non-linear models
Tutorials in Stata, Stata version history (stable release), Stata version history (development release)

multistate ~ multi-state survival analysis
Stata version history (stable release), Stata version history (development release)

survsim ~ simulation of simple and complex survival data
Tutorials in Stata, Stata version history (stable release), Stata version history (development release)

stmixed ~ multilevel parametric survival models
Stata version history (stable release), Stata version history (development release)

staft ~ flexible parametric accelerated failure time models
Stata version history (stable release), Github repo.

sankey ~ Sankey graphs in Stata using Python
Stata version history (stable release)

stjm ~ joint models of longitudinal and survival data

stgenreg ~ general parametric survival models

stmix ~ two-component mixture parametric survival models

extfunnel ~ extended funnel plots for meta-analysis

metapow ~ simulation-based sample size calculations for designing trials based on an existing meta-analysis

Recent Posts

More Posts

survsim can now simulate survival data from a semi-Markov or multiple timescale multi-state model, and is about 70% faster

CONTINUE READING

survsim can now simulate survival data from a Markov multi-state model, defined by general transition-specific hazard functions

CONTINUE READING

survsim can now simulate survival data from a parametric distribution, a user-defined distribution, from a fitted merlin model, or from a general cause-specific hazards competing risks model

CONTINUE READING

Bringing together the strengths of stset and merlin

CONTINUE READING

How merlin makes modelling of non-linear effects a whole lot simpler

CONTINUE READING

Teaching

My core teaching is on the MSc Medical Statistics course at the University of Leicester.

I teach a number of short courses, some teaching material is made freely available on the course pages:

Selected Talks

More Talks

Simulating time-to-event data from parametric distributions, custom distributions, competing risk models and general multi-state models
Sep 17, 2020
PDF Video
Stata Happy Hour with Michael J. Crowther
Aug 12, 2020
Stata Happy Hour
Video
Modelling complex disease trajectories using multi-state survival models: Estimation, prediction and software
Jul 7, 2020
International Biometric Conference - Invited Session
Video
Simple and complex survival analysis: New developments in merlin
Aug 30, 2019
Stata Nordic and Baltic Conference 2019
PDF
Flexible parametric inference for cause-specific hazard competing risks models under interval censoring
Jul 15, 2019
ISCB 2019
PDF

Group

My research group currently consists of:

Research Staff

  • Jonathan Broomfield, NIHR Methods Fellow, University of Leicester.

PhD students

As main supervisor:

  • Nuzhat Ashra, University of Leicester (25th September 2017 - Present). Nuzhat is funded by an MRC IMPACT studentship and SPD Development Company, to work on joint modelling of biomarkers to predict miscarriage. She will also be working on some extensions to the stjm command in Stata, including dynamic predictions.

  • Micki Hill, University of Leicester (1st April 2019 - Present). Micki is funded by a Department of Health Sciences studentship, to work on methods in multi-state survival analysis with a focus on interval-censored data.

  • Hayley Smith, University of Leicester (1st October 2019 - Present). Hayley is funded by a HDRUK studentship, to work on methods in multi-state survival analysis and machine learning.

As co-supervisor:

  • Elinor Curnow, University of Bristol Ellie has been working on methods for imputing survival times from interval censored data. She’s studying part time, and I joined her supervisory team in November 2018.
  • Nikolaos Skourlis, Karolinska Institutet (1st November 2018 - Present) Nik is working on projects within the fields of relative survival and multi-state models, with applications in cancer epidemiology.

Previous research staff

  • Dr Emma Martin, Post-doctoral Research Associate in Biostatistics, University of Leicester. Emma was funded by my MRC New Investigator Research Grant to work on a variety of projects in multi-state survival models and joint models.

Previous PhD students

As main supervisor:

  • Alessandro Gasparini, University of Leicester (Awarded September 2019) [GitHub]
    Alessandro worked on frailty survival models and a RShiny app for use in summarising simulation studies. His main project centred on informative observations in joint modelling of longitudinal and survival data. More details on his PhD can be found here.

As co-supervisor:

  • Sam Brilleman, Monash University (Awarded 2018)
    [Homepage] [GitHub]
    Sam’s worked on a variety of projects, but a core project was the development of an R package to fit an extensive array of Bayesian joint models using Stan. More details on his PhD can be found here.

Contact