Odds Ratio Estimation — odds

Compares bivariate responses between two groups in terms of odds ratios along with a confidence interval.

Usage

or_glm(data, conf_level)

or_clogit(data, conf_level)

s_odds_ratio(
  df,
  .var,
  .ref_group,
  .in_ref_col,
  .df_row,
  variables = list(arm = NULL, strata = NULL),
  conf_level = 0.95,
  groups_list = NULL
)

a_odds_ratio(
  df,
  .var,
  .ref_group,
  .in_ref_col,
  .df_row,
  variables = list(arm = NULL, strata = NULL),
  conf_level = 0.95,
  groups_list = NULL
)

estimate_odds_ratio(
  lyt,
  vars,
  ...,
  show_labels = "hidden",
  table_names = vars,
  .stats = "or_ci",
  .formats = NULL,
  .labels = NULL,
  .indent_mods = NULL
)

Arguments

data: (data frame)
with at least the variables rsp, grp, and in addition strata for or_clogit().
conf_level: (proportion)
confidence level of the interval.
df: (data frame)
data set containing all analysis variables.
.var: (string)
single variable name that is passed by rtables when requested by a statistics function.
.ref_group: (data frame or vector)
the data corresponding to the reference group.
.in_ref_col: (logical)
TRUE when working with the reference level, FALSE otherwise.
.df_row: (data frame)
data frame across all of the columns for the given row split.
variables: (named list of string)
list of additional analysis variables.
groups_list: (named list of character)
specifies the new group levels via the names and the levels that belong to it in the character vectors that are elements of the list.
lyt: (layout)
input layout where analyses will be added to.
vars: (character)
variable names for the primary analysis variable to be iterated over.
...: arguments passed to s_odds_ratio().
show_labels: label visibility: one of "default", "visible" and "hidden".
table_names: (character)
this can be customized in case that the same vars are analyzed multiple times, to avoid warnings from rtables.
.stats: (character)
statistics to select for the table.
.formats: (named character or list)
formats for the statistics.
.labels: (named character)
labels for the statistics (without indent).
.indent_mods: (named integer)
indent modifiers for the labels.

Details

This function uses either logistic regression for unstratified analyses, or conditional logistic regression for stratified analyses. The Wald confidence interval with the specified confidence level is calculated. Note that, for stratified analyses, there is currently no implementation for conditional likelihood confidence intervals, therefore the likelihood confidence interval as an option is not yet available. Besides, when rsp contains only responders or non-responders, then the result values will be NA, because no odds ratio estimation is possible.

Functions

or_glm(): estimates the odds ratio based on stats::glm(). Note that there must be exactly 2 groups in data as specified by the grp variable.
or_clogit(): estimates the odds ratio based on survival::clogit(). This is done for the whole data set including all groups, since the results are not the same as when doing pairwise comparisons between the groups.
s_odds_ratio(): Statistics function which estimates the odds ratio between a treatment and a control. Note that a variables list with arm and strata names needs to be passed if a stratified analysis is required.
a_odds_ratio(): Formatted Analysis function which can be further customized by calling rtables::make_afun() on it. It is used as afun in rtables::analyze().
estimate_odds_ratio(): Layout creating function which can be used for creating tables, which can take statistics function arguments and additional format arguments (see below).

Examples


# Data with 2 groups.
data <- data.frame(
  rsp = as.logical(c(1, 1, 0, 1, 0, 0, 1, 1)),
  grp = letters[c(1, 1, 1, 2, 2, 2, 1, 2)],
  strata = letters[c(1, 2, 1, 2, 2, 2, 1, 2)],
  stringsAsFactors = TRUE
)

# Odds ratio based on glm.
or_glm(data, conf_level = 0.95)
#> $or_ci
#>        est        lcl        ucl 
#> 0.33333333 0.01669735 6.65441589 
#> 
#> $n_tot
#> n_tot 
#>     8 
#> 

# Data with 3 groups.
data <- data.frame(
  rsp = as.logical(c(1, 1, 0, 1, 0, 0, 1, 1, 0, 0, 1, 1, 0, 1, 0, 0, 1, 1, 0, 0)),
  grp = letters[c(1, 1, 1, 2, 2, 2, 3, 3, 3, 3, 1, 1, 1, 2, 2, 2, 3, 3, 3, 3)],
  strata = LETTERS[c(1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2)],
  stringsAsFactors = TRUE
)

# Odds ratio based on stratified estimation by conditional logistic regression.
or_clogit(data, conf_level = 0.95)
#> $or_ci
#> $or_ci$b
#>        est        lcl        ucl 
#> 0.28814553 0.02981009 2.78522598 
#> 
#> $or_ci$c
#>       est       lcl       ucl 
#> 0.5367919 0.0673365 4.2791881 
#> 
#> 
#> $n_tot
#> n_tot 
#>    20 
#> 

set.seed(12)
dta <- data.frame(
  rsp = sample(c(TRUE, FALSE), 100, TRUE),
  grp = factor(rep(c("A", "B"), each = 50), levels = c("B", "A")),
  strata = factor(sample(c("C", "D"), 100, TRUE))
)

# Unstratified analysis.
s_odds_ratio(
  df = subset(dta, grp == "A"),
  .var = "rsp",
  .ref_group = subset(dta, grp == "B"),
  .in_ref_col = FALSE,
  .df_row = dta
)
#> $or_ci
#>       est       lcl       ucl 
#> 0.8484848 0.3831831 1.8788053 
#> attr(,"label")
#> [1] "Odds Ratio (95% CI)"
#> 
#> $n_tot
#> n_tot 
#>   100 
#> attr(,"label")
#> [1] "Total n"
#> 

# Stratified analysis.
s_odds_ratio(
  df = subset(dta, grp == "A"),
  .var = "rsp",
  .ref_group = subset(dta, grp == "B"),
  .in_ref_col = FALSE,
  .df_row = dta,
  variables = list(arm = "grp", strata = "strata")
)
#> $or_ci
#>       est       lcl       ucl 
#> 0.7689750 0.3424155 1.7269154 
#> attr(,"label")
#> [1] "Odds Ratio (95% CI)"
#> 
#> $n_tot
#> n_tot 
#>   100 
#> attr(,"label")
#> [1] "Total n"
#> 
a_odds_ratio(
  df = subset(dta, grp == "A"),
  .var = "rsp",
  .ref_group = subset(dta, grp == "B"),
  .in_ref_col = FALSE,
  .df_row = dta
)
#> RowsVerticalSection (in_rows) object print method:
#> ----------------------------
#>   row_name     formatted_cell indent_mod           row_label
#> 1    or_ci 0.85 (0.38 - 1.88)          1 Odds Ratio (95% CI)
#> 2    n_tot                100          0             Total n

dta <- data.frame(
  rsp = sample(c(TRUE, FALSE), 100, TRUE),
  grp = factor(rep(c("A", "B"), each = 50))
)

l <- basic_table() %>%
  split_cols_by(var = "grp", ref_group = "B") %>%
  estimate_odds_ratio(vars = "rsp")

build_table(l, df = dta)
#>                       B           A         
#> ————————————————————————————————————————————
#> Odds Ratio (95% CI)       0.72 (0.33 - 1.60)