Skip to content

Commit 74582a8

Browse files
committed
Minor text update
1 parent db500e5 commit 74582a8

File tree

9 files changed

+8
-8
lines changed

9 files changed

+8
-8
lines changed

manuscript/cSTM_Tutorial_Intro.Rmd

Lines changed: 2 additions & 2 deletions
Original file line numberDiff line numberDiff line change
@@ -818,14 +818,14 @@ In this tutorial, we provided a mathematical conceptualization of time-independe
818818

819819
The parameterization of our example model assumes all parameters are known, or at least, the characterization of their uncertainty is known (i.e., we know their distributions). However, to construct a real-world cSTM, modelers must conduct a thorough synthesis of current evidence to determine these appropriate structures and inform all parameters based on the current evidence. For example, literature must be carefully considered when determining whether transitions between non-death health states are estimated conditional on being alive or are estimated as competing risks along with mortality risks.[@Briggs2012] Similarly, our PSA analysis simplifies reality where all model parameters are assumed to be independent of each other. However, parameters could be correlated with each other or have a rank order, and appropriate statistical methods that simulate these correlations or rank order might be needed.[@Goldhaber-Fiebert2015] We encourage modelers to use appropriate statistical methods to synthesize and quantify model parameters uncertainty accurately. In addition, modelers should appropriately specify all model parameters for the cycle length of the model.[@Hunink2014]
820820

821-
In general, cSTMs are recommended when the number of states is considered "not too large".[@Siebert2012c] This recommendation arises because as the number of states increases, it becomes more challenging to keep track of their construction. It is possible to build reasonably complex cSTMs in R as long as the RAM of the computer running the analysis can store the size of the transition probability matrix and outputs of interest. For time-independent cSTMs, in general, this should not be a problem with the capacity of current RAM in personal computers. With increasing model complexity and interdependency of functions to conduct various analyses like PSA, it is essential to ensure all code and functions work as expected and all elements of the cSTM are valid. We can achieve this by creating functions that help with model debugging, validation, and thorough unit testing. In the accompanying GitHub repository, we provide functions to check that transition probability matrices and their elements are valid. However, unit testing is beyond the scope of this tutorial. We refer the reader to a previously published manuscript that describes unit testing in more detail and provides accompanying code.[@Alarid-Escudero2019e]
821+
In general, cSTMs are recommended when the number of states is considered "not too large".[@Siebert2012c] This recommendation arises because it becomes more challenging to keep track of their construction as the number of states increases. It is possible to build reasonably complex cSTMs in R as long as the RAM of the computer running the analysis can store the size of the transition probability matrix and outputs of interest. For time-independent cSTMs, in general, this should not be a problem with the capacity of current RAM in personal computers. With increasing model complexity and interdependency of functions to conduct various analyses like PSA, it is essential to ensure all code and functions work as expected and all elements of the cSTM are valid. We can achieve this by creating functions that help with model debugging, validation, and thorough unit testing. In the accompanying GitHub repository, we provide functions to check that transition probability matrices and their elements are valid. However, unit testing is beyond the scope of this tutorial. We refer the reader to a previously published manuscript that describes unit testing in more detail and provides accompanying code.[@Alarid-Escudero2019e]
822822

823823
In this tutorial, we implemented a cSTM using a (discrete-time) transition matrix, but cSTM can also be implemented via a set of (discrete-time) difference equations or (continuous-time) differential equations in R.[@Grimmett2014; @Axler2005] We refer readers interested in learning more on continuous-time cSTMs to previously published manuscripts[@Cao2016;@VanRosmalen2013;@Begun2013;@Soares2012] and a tutorial using R.[@Frederix2013a] Finally, the variable names used in this paper reflect our coding style. While we provide standardized variable names, adopting these conventions is ultimately a personal preference.
824824

825825
In summary, this tutorial provides a conceptualization of time-independent cSTMs and a step-by-step guide to implement them in R. We aim to add to the current body of literature and material on building this type of decision model so that health decision scientists and health economists can develop cSTMs in a more flexible, efficient, open-source manner and to encourage increased transparency and reproducibility. In the accompanying tutorial, we explore generalizing this framework to time-dependent cSTM, generating epidemiological outcomes, and incorporating transition rewards.
826826

827827
# Acknowledgements
828-
Dr Alarid-Escudero was supported by grants U01-CA199335 and U01-CA253913 from the National Cancer Institute (NCI) as part of the Cancer Intervention and Surveillance Modeling Network (CISNET), and a grant by the Gordon and Betty Moore Foundation. Miss Krijkamp was supported by a fellowship from a grant by the Gordon and Betty Moore Foundation (GBMF7853) through the Society for Medical Decision Making (SMDM). Dr. Enns was supported by a grant from the National Institute of Allergy and Infectious Diseases of the National Institutes of Health under award no. K25AI118476. Dr. Hunink received research funding from the American Diabetes Association, the Netherlands Organization for Health Research and Development, the German Innovation Fund, Netherlands Educational Grant ("Studie Voorschot Middelen"), and the Gordon and Betty Moore Foundation. Dr. Jalal was supported by a grant from the National Institute on Drug Abuse of the National Institute of Health under award no. K01DA048985. The content is solely the authors' responsibility and does not necessarily represent the official views of the National Institutes of Health. The funding agencies had no role in the study's design, interpretation of results, or writing of the manuscript. The funding agreement ensured the authors’ independence in designing the study, interpreting the data, writing, and publishing the report. We also want to thank the anonymous reviewers of *Medical Decision Making* for their valuable suggestions and the students who have taken our classes and provided invaluable feedback to improve the clarity and usability of our materials.
828+
Dr. Alarid-Escudero was supported by grants U01-CA199335 and U01-CA253913 from the National Cancer Institute (NCI) as part of the Cancer Intervention and Surveillance Modeling Network (CISNET), and a grant by the Gordon and Betty Moore Foundation. Miss Krijkamp was supported by the Society for Medical Decision Making (SMDM) fellowship through a grant by the Gordon and Betty Moore Foundation (GBMF7853). Dr. Enns was supported by a grant from the National Institute of Allergy and Infectious Diseases of the National Institutes of Health under award no. K25AI118476. Dr. Hunink received research funding from the American Diabetes Association, the Netherlands Organization for Health Research and Development, the German Innovation Fund, Netherlands Educational Grant ("Studie Voorschot Middelen"), and the Gordon and Betty Moore Foundation. Dr. Jalal was supported by a grant from the National Institute on Drug Abuse of the National Institute of Health under award no. K01DA048985. The content is solely the responsibility of the authors and does not necessarily represent the official views of the National Institutes of Health. The funding agencies had no role in the design of the study, interpretation of results, or writing of the manuscript. The funding agreement ensured the authors’ independence in designing the study, interpreting the data, writing, and publishing the report. We also want to thank the anonymous reviewers of *Medical Decision Making* for their valuable suggestions and the students who took our classes where we refined these materials.
829829

830830
# References
831831

manuscript/cSTM_Tutorial_Intro.log

Lines changed: 3 additions & 3 deletions
Original file line numberDiff line numberDiff line change
@@ -1,4 +1,4 @@
1-
This is pdfTeX, Version 3.141592653-2.6-1.40.22 (TeX Live 2021) (preloaded format=pdflatex 2021.4.17) 10 AUG 2021 14:08
1+
This is pdfTeX, Version 3.141592653-2.6-1.40.22 (TeX Live 2021) (preloaded format=pdflatex 2021.4.17) 12 AUG 2021 08:39
22
entering extended mode
33
restricted \write18 enabled.
44
%&-line parsing enabled.
@@ -1164,7 +1164,7 @@ Here is how much of TeX's memory you used:
11641164
31945 multiletter control sequences out of 15000+600000
11651165
492395 words of font info for 197 fonts, out of 8000000 for 9000
11661166
1141 hyphenation exceptions out of 8191
1167-
85i,12n,88p,1680b,451s stack positions out of 5000i,500n,10000p,200000b,80000s
1167+
85i,12n,88p,1625b,451s stack positions out of 5000i,500n,10000p,200000b,80000s
11681168
{/usr/local/texlive/2021/texmf-dist/fonts/enc
11691169
/dvips/lm/lm-mathsy.enc}{/usr/local/texlive/2021/texmf-dist/fonts/enc/dvips/lm/
11701170
lm-ec.enc}{/usr/local/texlive/2021/texmf-dist/fonts/enc/dvips/lm/lm-ts1.enc}{/u
@@ -1191,7 +1191,7 @@ ocal/texlive/2021/texmf-dist/fonts/type1/public/lm/lmtk10.pfb></usr/local/texli
11911191
ve/2021/texmf-dist/fonts/type1/public/lm/lmtko10.pfb></usr/local/texlive/2021/t
11921192
exmf-dist/fonts/type1/public/lm/lmtt10.pfb></usr/local/texlive/2021/texmf-dist/
11931193
fonts/type1/public/lm/lmtti10.pfb>
1194-
Output written on cSTM_Tutorial_Intro.pdf (26 pages, 843233 bytes).
1194+
Output written on cSTM_Tutorial_Intro.pdf (26 pages, 843196 bytes).
11951195
PDF statistics:
11961196
558 PDF objects out of 1000 (max. 8388607)
11971197
485 compressed objects within 5 object streams

manuscript/cSTM_Tutorial_Intro.pdf

-37 Bytes
Binary file not shown.

manuscript/cSTM_Tutorial_Intro.tex

Lines changed: 3 additions & 3 deletions
Original file line numberDiff line numberDiff line change
@@ -154,7 +154,7 @@
154154

155155
\title{An Introductory Tutorial to Cohort State-Transition Models in R}
156156
\author{Fernando Alarid-Escudero, PhD\footnote{Division of Public Administration, Center for Research and Teaching in Economics (CIDE), Aguascalientes, AGS, Mexico} \and Eline Krijkamp, MSc\footnote{Department of Epidemiology and Department of Radiology, Erasmus University Medical Center, Rotterdam, The Netherlands} \and Eva A. Enns, PhD\footnote{Division of Health Policy and Management, University of Minnesota School of Public Health, Minneapolis, MN, USA} \and Alan Yang, MSc\footnote{The Hospital for Sick Children, Toronto} \and Myriam G.M. Hunink, PhD\(^\dagger\)\footnote{Center for Health Decision Sciences, Harvard T.H. Chan School of Public Health, Boston, USA} \and Petros Pechlivanoglou, PhD\footnote{The Hospital for Sick Children, Toronto and University of Toronto, Toronto, Ontario, Canada} \and Hawre Jalal, MD, PhD\footnote{University of Pittsburgh, Pittsburgh, PA, USA}}
157-
\date{2021-08-10}
157+
\date{2021-08-12}
158158

159159
\begin{document}
160160
\maketitle
@@ -854,7 +854,7 @@ \section{Discussion}\label{discussion}}
854854

855855
The parameterization of our example model assumes all parameters are known, or at least, the characterization of their uncertainty is known (i.e., we know their distributions). However, to construct a real-world cSTM, modelers must conduct a thorough synthesis of current evidence to determine these appropriate structures and inform all parameters based on the current evidence. For example, literature must be carefully considered when determining whether transitions between non-death health states are estimated conditional on being alive or are estimated as competing risks along with mortality risks.\textsuperscript{\protect\hyperlink{ref-Briggs2012}{26}} Similarly, our PSA analysis simplifies reality where all model parameters are assumed to be independent of each other. However, parameters could be correlated with each other or have a rank order, and appropriate statistical methods that simulate these correlations or rank order might be needed.\textsuperscript{\protect\hyperlink{ref-Goldhaber-Fiebert2015}{30}} We encourage modelers to use appropriate statistical methods to synthesize and quantify model parameters uncertainty accurately. In addition, modelers should appropriately specify all model parameters for the cycle length of the model.\textsuperscript{\protect\hyperlink{ref-Hunink2014}{22}}
856856

857-
In general, cSTMs are recommended when the number of states is considered ``not too large''.\textsuperscript{\protect\hyperlink{ref-Siebert2012c}{5}} This recommendation arises because as the number of states increases, it becomes more challenging to keep track of their construction. It is possible to build reasonably complex cSTMs in R as long as the RAM of the computer running the analysis can store the size of the transition probability matrix and outputs of interest. For time-independent cSTMs, in general, this should not be a problem with the capacity of current RAM in personal computers. With increasing model complexity and interdependency of functions to conduct various analyses like PSA, it is essential to ensure all code and functions work as expected and all elements of the cSTM are valid. We can achieve this by creating functions that help with model debugging, validation, and thorough unit testing. In the accompanying GitHub repository, we provide functions to check that transition probability matrices and their elements are valid. However, unit testing is beyond the scope of this tutorial. We refer the reader to a previously published manuscript that describes unit testing in more detail and provides accompanying code.\textsuperscript{\protect\hyperlink{ref-Alarid-Escudero2019e}{20}}
857+
In general, cSTMs are recommended when the number of states is considered ``not too large''.\textsuperscript{\protect\hyperlink{ref-Siebert2012c}{5}} This recommendation arises because it becomes more challenging to keep track of their construction as the number of states increases. It is possible to build reasonably complex cSTMs in R as long as the RAM of the computer running the analysis can store the size of the transition probability matrix and outputs of interest. For time-independent cSTMs, in general, this should not be a problem with the capacity of current RAM in personal computers. With increasing model complexity and interdependency of functions to conduct various analyses like PSA, it is essential to ensure all code and functions work as expected and all elements of the cSTM are valid. We can achieve this by creating functions that help with model debugging, validation, and thorough unit testing. In the accompanying GitHub repository, we provide functions to check that transition probability matrices and their elements are valid. However, unit testing is beyond the scope of this tutorial. We refer the reader to a previously published manuscript that describes unit testing in more detail and provides accompanying code.\textsuperscript{\protect\hyperlink{ref-Alarid-Escudero2019e}{20}}
858858

859859
In this tutorial, we implemented a cSTM using a (discrete-time) transition matrix, but cSTM can also be implemented via a set of (discrete-time) difference equations or (continuous-time) differential equations in R.\textsuperscript{\protect\hyperlink{ref-Grimmett2014}{31},\protect\hyperlink{ref-Axler2005}{32}} We refer readers interested in learning more on continuous-time cSTMs to previously published manuscripts\textsuperscript{\protect\hyperlink{ref-VanRosmalen2013}{21},\protect\hyperlink{ref-Cao2016}{33}--\protect\hyperlink{ref-Soares2012}{35}} and a tutorial using R.\textsuperscript{\protect\hyperlink{ref-Frederix2013a}{36}} Finally, the variable names used in this paper reflect our coding style. While we provide standardized variable names, adopting these conventions is ultimately a personal preference.
860860

@@ -863,7 +863,7 @@ \section{Discussion}\label{discussion}}
863863
\hypertarget{acknowledgements}{%
864864
\section{Acknowledgements}\label{acknowledgements}}
865865

866-
Dr Alarid-Escudero was supported by grants U01-CA199335 and U01-CA253913 from the National Cancer Institute (NCI) as part of the Cancer Intervention and Surveillance Modeling Network (CISNET), and a grant by the Gordon and Betty Moore Foundation. Miss Krijkamp was supported by a fellowship from a grant by the Gordon and Betty Moore Foundation (GBMF7853) through the Society for Medical Decision Making (SMDM). Dr.~Enns was supported by a grant from the National Institute of Allergy and Infectious Diseases of the National Institutes of Health under award no. K25AI118476. Dr.~Hunink received research funding from the American Diabetes Association, the Netherlands Organization for Health Research and Development, the German Innovation Fund, Netherlands Educational Grant (``Studie Voorschot Middelen''), and the Gordon and Betty Moore Foundation. Dr.~Jalal was supported by a grant from the National Institute on Drug Abuse of the National Institute of Health under award no. K01DA048985. The content is solely the authors' responsibility and does not necessarily represent the official views of the National Institutes of Health. The funding agencies had no role in the study's design, interpretation of results, or writing of the manuscript. The funding agreement ensured the authors' independence in designing the study, interpreting the data, writing, and publishing the report. We also want to thank the anonymous reviewers of \emph{Medical Decision Making} for their valuable suggestions and the students who have taken our classes and provided invaluable feedback to improve the clarity and usability of our materials.
866+
Dr.~Alarid-Escudero was supported by grants U01-CA199335 and U01-CA253913 from the National Cancer Institute (NCI) as part of the Cancer Intervention and Surveillance Modeling Network (CISNET), and a grant by the Gordon and Betty Moore Foundation. Miss Krijkamp was supported by the Society for Medical Decision Making (SMDM) fellowship through a grant by the Gordon and Betty Moore Foundation (GBMF7853). Dr.~Enns was supported by a grant from the National Institute of Allergy and Infectious Diseases of the National Institutes of Health under award no. K25AI118476. Dr.~Hunink received research funding from the American Diabetes Association, the Netherlands Organization for Health Research and Development, the German Innovation Fund, Netherlands Educational Grant (``Studie Voorschot Middelen''), and the Gordon and Betty Moore Foundation. Dr.~Jalal was supported by a grant from the National Institute on Drug Abuse of the National Institute of Health under award no. K01DA048985. The content is solely the responsibility of the authors and does not necessarily represent the official views of the National Institutes of Health. The funding agencies had no role in the design of the study, interpretation of results, or writing of the manuscript. The funding agreement ensured the authors' independence in designing the study, interpreting the data, writing, and publishing the report. We also want to thank the anonymous reviewers of \emph{Medical Decision Making} for their valuable suggestions and the students who took our classes where we refined these materials.
867867

868868
\hypertarget{references}{%
869869
\section*{References}\label{references}}

manuscript/figs/CE-scatter-1.pdf

0 Bytes
Binary file not shown.

manuscript/figs/CEAC-1.pdf

0 Bytes
Binary file not shown.

manuscript/figs/ELC-1.pdf

0 Bytes
Binary file not shown.
-17 Bytes
Binary file not shown.
0 Bytes
Binary file not shown.

0 commit comments

Comments
 (0)