knitr::opts_chunk$set(collapse = T, comment = "#>", root.dir = 'vignettes') options(tibble.print_min = 4L, tibble.print_max = 4L, rmarkdown.html_vignette.check_title = FALSE)
Last updated on: r format(Sys.Date(), '%d %B, %Y')
The package is not available at CRAN. It needs to be installed from GitHub. Please use this command to install it.
remotes::install_github("meta-analyses/drpa")
You may need to install the remotes
package first.
The first column disease
specifies the supported diseases in the package. The other three columns, which are outcome types (fatal
, non-fatal
and fatal-and-non-fatal
), show the last knot (based on the available data points) for the specific disease. We don't have confidence that our meta-analysis is reliable after this point, and assumes an onward flat relationship. Whenever a disease such as all-cause-cvd
doesn't have an outcome type non-fatal
, it will be NA
. More will follow in the next section.
library(magrittr) require(kableExtra) dtable <- readr::read_csv('../inst/extdata/75p_diseases.csv') kbl(dtable) %>% kable_styling(bootstrap_options = c("striped", "hover", "condensed", "responsive"))
Load the package
library(drpa)
Get response for dose 40 for all-cause-cvd with the default quantile 0.5 and censoring at doubling of WHO's recommended level of physical activity for adults (17.5 MMETs hours per week).
The quantile
's value is between 0 and 1 [excluding the boundary values]
drpa::dose_response(cause = 'all-cause-cvd', outcome_type = 'fatal-and-non-fatal', dose = 40, quantile = 0.5, confidence_intervals = T, censor_method = 'WHO-DRL')
Description of the arguments are mentioned below:
| Argument | Description |
|-----------------|-------------------------------------------------------|
| cause | All-cause-mortality or disease name |
| outcome_type | Either one of the three possibilities for outcome type: fatal, non-fatal or fatal-and-non-fatal |
| dose | Dose (all-cause or disease) |
| quantile | Numeric value between 0 and 1 - default is 0.5
. This captures the range of dose values between lower
and upper
confidence levels. |
| censor_method | This determines whether any censor method applies to the dose-response relationship. There are four possibilities: none
, 75thPercentile
(to use 75th percentile using person years), WHO-DRL
(WHO's recommended physical activity per week 8.75
when doubled, so 17.5
MMETs hours per week) or WHO-QRL
(WHO's recommended physical activity per week 8.75
when quadrupled, so 35
MMETs hours per week) . The default value is WHO-DRL
- a fixed censor point at 17.5
after which the relationship remains the same. |
| confidence_intervals | Boolean variable to determine whether confidence intervals are returned or not - default false |
Add the following code to your website.
For more information on customizing the embed code, read Embedding Snippets.