update vignette meta-analysis of prognostic models

vignettes/ma-pf.Rmd

@@ -13,13 +13,15 @@ bibliography: "https://api.citedrive.com/bib/093a08ac-0337-479e-a147-17929fa7f7b
 
 ```{r, include = FALSE}
 knitr::opts_chunk$set(
-  collapse = TRUE,
-  comment = "#>"
+  collapse = FALSE,
+  message = FALSE,
+  warning = FALSE,
+  tidy = TRUE
 )
 ```
 
 
-```{r, echo = F, message = F, warning = F}
+```{r, echo = F}
 library(knitr)
 library(dplyr)
 
@@ -33,9 +35,7 @@ An important task in medical research is the identification and assessment of pr
 
 Over the past few decades, numerous prognostic factor studies have been published in the medical literature. For example, @riley_reporting_2003 identified 260 studies reporting associations for 130 different tumour markers for neuroblastoma. More recently, @tzoulaki_assessment_2009 identified 79 studies reporting the added value of 86 different markers when added to the Framingham risk score. Despite this huge research effort, the prognostic value of most traditional factors under discussion is uncertain and the usefulness of many specific markers, prognostic indices, and classification schemes is still unproven [@sauerbrei_evidence-based_2006].
 
-The aim of this introduction is to illustrate how to summarize the results from multiple prognostic factor studies and how to investigate sources of between-study heterogeneity. 
-
-In this practical we will make use of the R packages `metamisc` and `metafor`. The [https://cran.r-project.org/package=metafor](metafor) package provides a comprehensive collection of functions for conducting meta-analyses in R. The [https://cran.r-project.org/package=metamisc](metamisc) package provides additional functions to facilitate meta-analysis of prognosis studies. We can load these packages as follows:
+This vignette aims to illustrate how the results from multiple prognostic factor studies can be summarized and how sources of between-study heterogeneity can be examined. Hereto, we will use the R packages  **metamisc** and **metafor**. The [https://cran.r-project.org/package=metafor](metafor) package provides a comprehensive collection of functions for conducting meta-analyses in R. The [https://cran.r-project.org/package=metamisc](metamisc) package provides additional functions to facilitate meta-analysis of prognosis studies. We can load these packages as follows:
 
 ```{r, message = F, warning = F}
 library(metafor)
@@ -274,10 +274,12 @@ The probability that HER2 will yield a hazard ratio for overall survival of at l
 Nsim <- 1000000
 
 # Random draws from a Student T distribution
-rnd_t <- rt(Nsim, df= (numstudies - 2))
+rnd_t <- rt(Nsim, df = (numstudies - 2))
 
 # Generate 1,000,000 hazard ratios
 HRsim <- exp(c(mu) + rnd_t*sqrt(tau2 + sigma2))
+
+# Calculate the proportion of hazard ratios greater than  1.5
 mean(HRsim > 1.5)
 ```