In this tutorial, we will explore the combined approach to clinical
significance using R. The combined approaches use the anchor- or
distribution-based approach in addition to the statistical approach.
This is obviously stricter than the two methods on their own, but offers
a more detailed interpretation. We will be working with the
antidepressants
and the claus_2020
datasets,
and the cs_combined()
function to demonstrate various
aspects of this approach.
Before we begin, ensure that you have the following prerequisites in place:
First, let’s have a look at the datasets, which come with the package.
library(clinicalsignificance)
antidepressants
#> # A tibble: 1,110 × 4
#> patient condition measurement mom_di
#> <chr> <fct> <fct> <dbl>
#> 1 S001 Wait List Before 50
#> 2 S001 Wait List After 36
#> 3 S002 Wait List Before 40
#> 4 S002 Wait List After 32
#> 5 S003 Wait List Before 38
#> 6 S003 Wait List After 41
#> 7 S004 Wait List Before 29
#> 8 S004 Wait List After 44
#> 9 S005 Wait List Before 37
#> 10 S005 Wait List After 45
#> # ℹ 1,100 more rows
claus_2020
#> # A tibble: 172 × 9
#> id age sex treatment time bdi shaps who hamd
#> <dbl> <dbl> <fct> <fct> <dbl> <dbl> <dbl> <dbl> <dbl>
#> 1 1 54 Male TAU 1 33 9 0 25
#> 2 1 54 Male TAU 2 28 6 3 17
#> 3 1 54 Male TAU 3 28 9 7 13
#> 4 1 54 Male TAU 4 27 8 3 13
#> 5 2 52 Female PA 1 26 11 2 15
#> 6 2 52 Female PA 2 26 10 0 16
#> 7 2 52 Female PA 3 25 10 0 7
#> 8 2 52 Female PA 4 19 9 3 11
#> 9 3 54 Male PA 1 15 2 0 28
#> 10 3 54 Male PA 2 13 5 9 17
#> # ℹ 162 more rows
This approach combines the statistical and anchor-based approach. The
cs_combinedl()
function is a tool for assessing clinical
significance in this way. It allows you to determine if changes in
patient outcomes are practically significant. Let’s go through the basic
usage and some advanced features of this function.
Let’s start with a basic statistical clinical significance analysis
using the antidepressants
dataset. We are interested in the
Mind over Mood Depression Inventory (mom_di
) measurements.
For the statistical approach, a functional population must be defined.
Suppose, we collected data from a non-clinical sample and determined a
mean of 7 points and a standard deviation of also 7 points. Furthermore,
an MID of 8 points will be regarded as minimally important.
combined_results <- antidepressants |>
cs_combined(
id = patient,
time = measurement,
outcome = mom_di,
m_functional = 7,
sd_functional = 7,
cutoff_type = "c",
mid_improvement = 8
)
#> ℹ Your "Before" was set as pre measurement and and your "After" as post.
#> • If that is not correct, please specify the pre measurement with the argument
#> "pre".
Sometimes, as in the example above, you may encounter warnings when using this function. You can turn off the warning by explicitly specifying the pre-measurement time point using the pre parameter. This can be helpful when your data lacks clear pre-post measurement labels.
# Turning off the warning by specifying pre-measurement time
combined_results <- antidepressants |>
cs_combined(
id = patient,
time = measurement,
outcome = mom_di,
pre = "Before",
m_functional = 7,
sd_functional = 7,
cutoff_type = "c",
mid_improvement = 8
)
Here’s a breakdown of the code:
patient
, measurement
, and
mom_di
represent the patient identifier, assessment time
points, and HAM-D scores, respectively.pre
and post
specify the time points for
the pre and post-assessment.m_functional
and sd_functional
define the
functional population’s mean and standard deviation. This information is
used to calculate the population cutoff."c"
specifies the population cutoff of choice.mid_improvement
sets the for improvement to 8.# Print the results
combined_results
#>
#> ── Clinical Significance Results ──
#>
#> Combined approach using the anchor-based and statistical approach.
#> Category | n | Percent
#> ----------------------------
#> Recovered | 166 | 29.91%
#> Improved | 150 | 27.03%
#> Unchanged | 175 | 31.53%
#> Deteriorated | 52 | 9.37%
#> Harmed | 12 | 2.16%
# Get a summary
summary(combined_results)
#>
#> ── Clinical Significance Results ──
#>
#> Combined analysis of clinical significance using the anchor-based and
#> statistical approach method for calculating the RCI and population cutoffs.
#> There were 555 participants in the whole dataset of which 555 (100%) could be
#> included in the analysis.
#> The outcome was mom_di and the MID was set to 8.
#> The cutoff type was c with a value of 19.72 based on the following sumamry
#> statistics:
#>
#> ── Population Characteristics
#> M Clinical | SD Clinical | M Functional | SD Functional
#> -------------------------------------------------------
#> 34.73 | 8.26 | 7 | 7
#>
#> ── Individual Level Results
#> Category | n | Percent
#> ----------------------------
#> Recovered | 166 | 29.91%
#> Improved | 150 | 27.03%
#> Unchanged | 175 | 31.53%
#> Deteriorated | 52 | 9.37%
#> Harmed | 12 | 2.16%
Visualizing the results can help you better understand the clinical significance of changes in patient outcomes.
When working with data that has more than two measurements, you must
explicitly define the pre and post measurement time points using the
pre
and post
parameters.
# Clinical significance distribution analysis with more than two measurements
cs_results <- claus_2020 |>
cs_combined(
id = id,
time = time,
outcome = bdi,
pre = 1,
post = 4,
m_functional = 7,
sd_functional = 7,
cutoff_type = "c",
mid_improvement = 8
)
# Display the results
cs_results
#>
#> ── Clinical Significance Results ──
#>
#> Combined approach using the anchor-based and statistical approach.
#> Category | n | Percent
#> ---------------------------
#> Recovered | 11 | 27.50%
#> Improved | 9 | 22.50%
#> Unchanged | 17 | 42.50%
#> Deteriorated | 3 | 7.50%
#> Harmed | 0 | 0.00%
summary(cs_results)
#>
#> ── Clinical Significance Results ──
#>
#> Combined analysis of clinical significance using the anchor-based and
#> statistical approach method for calculating the RCI and population cutoffs.
#> There were 43 participants in the whole dataset of which 40 (93%) could be
#> included in the analysis.
#> The outcome was bdi and the MID was set to 8.
#> The cutoff type was c with a value of 20.15 based on the following sumamry
#> statistics:
#>
#> ── Population Characteristics
#> M Clinical | SD Clinical | M Functional | SD Functional
#> -------------------------------------------------------
#> 35.48 | 8.16 | 7 | 7
#>
#> ── Individual Level Results
#> Category | n | Percent
#> ---------------------------
#> Recovered | 11 | 27.50%
#> Improved | 9 | 22.50%
#> Unchanged | 17 | 42.50%
#> Deteriorated | 3 | 7.50%
#> Harmed | 0 | 0.00%
plot(cs_results)
You can also perform a grouped analysis by providing a group column from the data. This is useful when comparing treatment groups or other categories.
cs_results_grouped <- claus_2020 |>
cs_combined(
id = id,
time = time,
outcome = bdi,
pre = 1,
post = 4,
m_functional = 7,
sd_functional = 7,
cutoff_type = "c",
mid_improvement = 8,
group = treatment
)
# Display and visualize the results
cs_results_grouped
#>
#> ── Clinical Significance Results ──
#>
#> Combined approach using the anchor-based and statistical approach.
#> Group | Category | n | Percent
#> -----------------------------------
#> TAU | Recovered | 4 | 10.00%
#> TAU | Improved | 2 | 5.00%
#> TAU | Unchanged | 10 | 25.00%
#> TAU | Deteriorated | 3 | 7.50%
#> TAU | Harmed | 0 | 0.00%
#> PA | Recovered | 7 | 17.50%
#> PA | Improved | 7 | 17.50%
#> PA | Unchanged | 7 | 17.50%
#> PA | Deteriorated | 0 | 0.00%
#> PA | Harmed | 0 | 0.00%
plot(cs_results_grouped)
In some cases, higher values of an outcome may be considered better.
You can specify this using the better_is
argument. Let’s
see an example with the WHO-5 score where higher values are considered
better.
# Clinical significance analysis for outcomes where higher values are better
cs_results_who <- claus_2020 |>
cs_combined(
id,
time,
who,
pre = 1,
post = 4,
m_functional = 7,
sd_functional = 7,
cutoff_type = "c",
mid_improvement = 8,
better_is = "higher"
)
# Display the results
cs_results_who
#>
#> ── Clinical Significance Results ──
#>
#> Combined approach using the anchor-based and statistical approach.
#> Category | n | Percent
#> ---------------------------
#> Recovered | 10 | 25.00%
#> Improved | 1 | 2.50%
#> Unchanged | 29 | 72.50%
#> Deteriorated | 0 | 0.00%
#> Harmed | 0 | 0.00%
The Jacobson & Truax approach combines the statistical with the distribution-based approach. For this, let’s suppose that the reliability of the MoM-DI is 0.80.
jt_results <- antidepressants |>
cs_combined(
id = patient,
time = measurement,
outcome = mom_di,
pre = "Before",
m_functional = 7,
sd_functional = 7,
cutoff_type = "c",
reliability = 0.80
)
# Summarize and visualize the results
summary(jt_results)
#>
#> ── Clinical Significance Results ──
#>
#> Combined analysis of clinical significance using the JT and statistical
#> approach method for calculating the RCI and population cutoffs.
#> There were 555 participants in the whole dataset of which 555 (100%) could be
#> included in the analysis.
#> The outcome was mom_di and the reliability was set to 0.8.
#> The cutoff type was c with a value of 19.72 based on the following sumamry
#> statistics:
#>
#> ── Population Characteristics
#> M Clinical | SD Clinical | M Functional | SD Functional
#> -------------------------------------------------------
#> 34.73 | 8.26 | 7 | 7
#>
#> ── Individual Level Results
#> Category | n | Percent
#> ----------------------------
#> Recovered | 159 | 28.65%
#> Improved | 115 | 20.72%
#> Unchanged | 237 | 42.70%
#> Deteriorated | 34 | 6.13%
#> Harmed | 10 | 1.80%
plot(jt_results)
In this tutorial, you’ve learned how to perform clinical significance
analysis using the cs_combined()
function in R. This
analysis may be crucial for determining the practical importance of
changes in patient outcomes. By adjusting thresholds and considering
grouped analyses, you can gain valuable insights for healthcare and
clinical research applications.