Remove role attribute from vignettes

vignettes/archetypes.Rmd

@@ -35,7 +35,6 @@ data(fev_data, package = "mmrm")
 data <- fev_data |>
   brm_data(
     outcome = "FEV1",
-    role = "response",
     group = "ARMCD",
     time = "AVISIT",
     patient = "USUBJID",

vignettes/archetypes.Rmd.source

@@ -44,7 +44,6 @@ data(fev_data, package = "mmrm")
 data <- fev_data |>
   brm_data(
     outcome = "FEV1",
-    role = "response",
     group = "ARMCD",
     time = "AVISIT",
     patient = "USUBJID",

vignettes/bcva.Rmd

@@ -88,7 +88,6 @@ The rest of the pre-processing steps create factors for the study arm and visit
 +   mutate(AVISIT = gsub("VIS0*", "VIS", as.character(AVISIT))) |>
 +   brm_data(
 +     outcome = "BCVA_CHG",
-+     role = "change",
 +     group = "ARMCD",
 +     time = "AVISIT",
 +     patient = "USUBJID",

vignettes/bcva.Rmd.source

@@ -100,7 +100,6 @@ bcva_data <- bcva_data |>
   mutate(AVISIT = gsub("VIS0*", "VIS", as.character(AVISIT))) |>
   brm_data(
     outcome = "BCVA_CHG",
-    role = "change",
     group = "ARMCD",
     time = "AVISIT",
     patient = "USUBJID",

vignettes/fev1.Rmd

@@ -100,7 +100,6 @@ The rest of the pre-processing steps create factors for the study arm and visit
 > fev_data <- brm_data(
 +   data = fev_data,
 +   outcome = "FEV1_CHG",
-+   role = "change",
 +   group = "ARMCD",
 +   time = "AVISIT",
 +   patient = "USUBJID",

vignettes/fev1.Rmd.source

@@ -111,7 +111,6 @@ The rest of the pre-processing steps create factors for the study arm and visit
 fev_data <- brm_data(
   data = fev_data,
   outcome = "FEV1_CHG",
-  role = "change",
   group = "ARMCD",
   time = "AVISIT",
   patient = "USUBJID",

vignettes/subgroup.Rmd

@@ -88,7 +88,6 @@ When you create the special classed dataset for `brms.mmrm` using `brm_data()`,
 data <- brm_data(
   data = raw_data,
   outcome = "response",
-  role = "response",
   baseline = NULL,
   group = "group",
   subgroup = "subgroup",
@@ -108,7 +107,6 @@ str(data)
 #>  $ missing : logi [1:900] FALSE FALSE FALSE FALSE FALSE FALSE ...
 #>  $ response: num [1:900] 1.263 -0.326 1.33 1.272 0.415 ...
 #>  - attr(*, "brm_outcome")= chr "response"
-#>  - attr(*, "brm_role")= chr "response"
 #>  - attr(*, "brm_group")= chr "group"
 #>  - attr(*, "brm_subgroup")= chr "subgroup"
 #>  - attr(*, "brm_time")= chr "time"

vignettes/subgroup.Rmd.source

@@ -62,7 +62,6 @@ When you create the special classed dataset for `brms.mmrm` using `brm_data()`,
 data <- brm_data(
   data = raw_data,
   outcome = "response",
-  role = "response",
   baseline = NULL,
   group = "group",
   subgroup = "subgroup",

vignettes/usage.Rmd

@@ -71,7 +71,6 @@ Next, create a special classed dataset that the package will recognize. The clas
 data <- brm_data(
   data = raw_data,
   outcome = "response",
-  role = "response",
   group = "group",
   patient = "patient",
   time = "time",
@@ -99,13 +98,12 @@ data
 class(data)
 #> [1] "brms_mmrm_data" "tbl_df"         "tbl"            "data.frame"
 
-roles <- attributes(data)
-roles$row.names <- NULL
-str(roles)
+attributes <- attributes(data)
+attributes$row.names <- NULL
+str(attributes)
 #> List of 10
 #>  $ names              : chr [1:7] "patient" "time" "response" "group" ...
 #>  $ brm_outcome        : chr "response"
-#>  $ brm_role           : chr "response"
 #>  $ brm_group          : chr "group"
 #>  $ brm_time           : chr "time"
 #>  $ brm_patient        : chr "patient"
@@ -381,8 +379,8 @@ model$brms.mmrm_formula
 Regardless of the choice of fixed effects formula, `brms.mmrm` performs inference on the marginal distributions at each treatment group and time point of the mean of the following quantities:
 
 1. Response.
-2. Change from baseline, if you set `role` to `"change"` in `brm_data()`.
-3. Treatment difference, in terms of change from baseline.
+2. Change from baseline. Only reported if you originally declared a baseline time point with the `reference_time` argument of `brm_data()`.
+3. Treatment difference. If you declared a baseline in (2), then treatment difference is calculated in terms of change from baseline. Otherwise, it is calculated in terms of raw response.
 4. Effect size: treatment difference divided by the residual standard deviation.
 
 To derive posterior draws of these marginals, use the `brm_marginal_draws()` function.

vignettes/usage.Rmd.source

@@ -61,7 +61,6 @@ Next, create a special classed dataset that the package will recognize. The clas
 data <- brm_data(
   data = raw_data,
   outcome = "response",
-  role = "response",
   group = "group",
   patient = "patient",
   time = "time",
@@ -74,9 +73,9 @@ data
 
 class(data)
 
-roles <- attributes(data)
-roles$row.names <- NULL
-str(roles)
+attributes <- attributes(data)
+attributes$row.names <- NULL
+str(attributes)
 ```
 
 # Formula
@@ -183,8 +182,8 @@ model$brms.mmrm_formula
 Regardless of the choice of fixed effects formula, `brms.mmrm` performs inference on the marginal distributions at each treatment group and time point of the mean of the following quantities:
 
 1. Response.
-2. Change from baseline, if you set `role` to `"change"` in `brm_data()`.
-3. Treatment difference, in terms of change from baseline.
+2. Change from baseline. Only reported if you originally declared a baseline time point with the `reference_time` argument of `brm_data()`.
+3. Treatment difference. If you declared a baseline in (2), then treatment difference is calculated in terms of change from baseline. Otherwise, it is calculated in terms of raw response.
 4. Effect size: treatment difference divided by the residual standard deviation.
 
 To derive posterior draws of these marginals, use the `brm_marginal_draws()` function.