# Set general settings for markdown file
knitr::opts_chunk$set(
message = FALSE,
warning = FALSE)
# Clear environment
rm(list = ls())
# Load packages
library(dplyr) # for data wrangling
library(knitr) # for integrating computing and reporting in markdown
library(kableExtra) # for customizing appearance of tables
library(foreign) # for reading in SPSS file
library(psych) # for describeBy function
library(tidyr) # for gather and spread function
# Load functions
source("./functions/my_table_template.R")
# Turn off scientific notation
options(scipen = 999)The final sample consisted of 28 patients with OCD and 28 healthy control (HC) participants.
Patients with OCD and healthy control participants were individually matched for age, gender, and educational level.
Patients were diagnosed using the Structured Clinical Interview for DSM-IV (SCID-I; Wittchen et al., 1997) and were recruited from the outpatient clinic for OCD at Humboldt-Universität zu Berlin.
All participants were native German speakers and had normal or corrected-to-normal vision.
Exclusion criteria were lifetime diagnosis of bipolar, psychotic, or substance-related disorders; neurological diseases; and use of neuroleptic medication in the last 3 months or benzodiazepines in the last week.
Additional exclusion criteria for control participants were any current or past psychiatric disorder or psychotherapeutic treatment.
This table corresponds to Table 1 in the manuscript.
# Load data
participant_data <- read.spss("./data/Probandendaten.sav", to.data.frame = TRUE)
load(file = "./data/Trait_Data.rda")
# Prepare dataframe
demographics <- participant_data %>%
# Prepare id variable for merging and add group variable
dplyr::mutate(participant_id = as.factor(paste0(substr(Code, 13, 13), "_", substr(Code, 14, 15))),
group = as.factor(ifelse(substr(participant_id, 1, 1) == "C", "HC", "OCD"))) %>%
# Exclude P_02 and C_02 and row with NA
dplyr::filter(participant_id != "P_02" & participant_id != "C_02" & !is.na(Alter)) %>%
# Rename columns
dplyr::rename(
"Age (years)" = Alter,
"Gender" = Geschlecht,
"Years of education" = Schuljahre,
"Y-BOCS total score" = YBOCS_gesamt,
"Y-BOCS obsessions" = YBOCS_Gedanken,
"Y-BOCS compulsions" = YBOCS_Handlungen,
"Verbal intelligence (WST)" = WST,
"Handedness" = "Händigkeit") %>%
# Merge trait data
dplyr::left_join(., traits, by = "participant_id") %>%
# Select and reorder columns
dplyr::select(c("group", "Age (years)", "Gender", "Years of education", "Verbal intelligence (WST)", "Handedness",
"BDI-II", "OCI-R", "PSWQ", "STAI trait", "Y-BOCS total score", "Y-BOCS obsessions", "Y-BOCS compulsions"))
# Transform WST scores (range 24-37) to IQ scores (according to WST manual)
demographics[demographics$"Verbal intelligence (WST)" == 24, ]$"Verbal intelligence (WST)" <- 89
demographics[demographics$"Verbal intelligence (WST)" == 25, ]$"Verbal intelligence (WST)" <- 90
demographics[demographics$"Verbal intelligence (WST)" == 27, ]$"Verbal intelligence (WST)" <- 93
demographics[demographics$"Verbal intelligence (WST)" == 28, ]$"Verbal intelligence (WST)" <- 95
demographics[demographics$"Verbal intelligence (WST)" == 29, ]$"Verbal intelligence (WST)" <- 97
demographics[demographics$"Verbal intelligence (WST)" == 30, ]$"Verbal intelligence (WST)" <- 99
demographics[demographics$"Verbal intelligence (WST)" == 31, ]$"Verbal intelligence (WST)" <- 101
demographics[demographics$"Verbal intelligence (WST)" == 32, ]$"Verbal intelligence (WST)" <- 104
demographics[demographics$"Verbal intelligence (WST)" == 33, ]$"Verbal intelligence (WST)" <- 107
demographics[demographics$"Verbal intelligence (WST)" == 34, ]$"Verbal intelligence (WST)" <- 110
demographics[demographics$"Verbal intelligence (WST)" == 35, ]$"Verbal intelligence (WST)" <- 114
demographics[demographics$"Verbal intelligence (WST)" == 36, ]$"Verbal intelligence (WST)" <- 118
demographics[demographics$"Verbal intelligence (WST)" == 37, ]$"Verbal intelligence (WST)" <- 122
# Rescore one handedness score (according to EHI score that deviates from self report)
demographics[demographics$"Handedness" == "beidhändig" &
demographics$"Age (years)" == 29, ]$"Handedness" <- "rechts"
### Numeric variables
# Extract demographics per group
demographics_groups <- list()
demographics_groups <- describeBy(demographics %>% dplyr::select(-c("Gender", "Handedness", "group")), demographics$group)
ocd <- as.data.frame(demographics_groups$`OCD`)
hc <- as.data.frame(demographics_groups$`HC`)
# Create table with demographics formatted as "M (SD)"
demographics_groups <- as.data.frame(cbind(
Characteristic = rownames(ocd),
ocd = paste0(format(round(ocd$mean, 2), nsmall = 2), " (", format(round(ocd$sd, 2), nsmall = 2), ")"),
hc = paste0(format(round(hc$mean, 2), nsmall = 2), " (", format(round(hc$sd, 2), nsmall = 2), ")")))
# Calculate t tests
t_tests <- demographics %>%
dplyr::select("Age (years)", "Years of education", "Verbal intelligence (WST)", "BDI-II", "OCI-R",
"PSWQ", "STAI trait", "group") %>%
tidyr::gather(key = Characteristic, value = value, -group) %>%
dplyr::group_by(group, Characteristic) %>%
dplyr::summarise(value = list(value)) %>%
tidyr::spread(group, value) %>%
dplyr::ungroup() %>%
dplyr::group_by(Characteristic) %>%
dplyr::mutate("Test statistic" = t.test(unlist(HC), unlist(OCD))$statistic,
df = t.test(unlist(HC), unlist(OCD))$parameter,
p = t.test(unlist(HC), unlist(OCD))$p.value) %>%
dplyr::select(-c(HC, OCD)) %>%
dplyr::ungroup()
# Combine demographics and t test results
demographics_numeric <- left_join(demographics_groups, t_tests, by = "Characteristic")
### Count variables (gender, handedness)
# Gender: calculate chi-squared test and extract counts formatted as "female:male"
gender_test <- chisq.test(demographics$Gender, demographics$group)
gender <- demographics %>% dplyr::group_by(group) %>% dplyr::count(Gender)
gender_ocd <- paste0(gender[gender$group == "OCD" & gender$Gender == "weiblich", ]$n, ":",
gender[gender$group == "OCD" & gender$Gender == "männlich", ]$n)
gender_hc <- paste0(gender[gender$group == "HC" & gender$Gender == "weiblich", ]$n, ":",
gender[gender$group == "HC" & gender$Gender == "männlich", ]$n)
# Handedness: calculate Fishers's exact test and extract counts formatted as "right:left:ambidextrous"
hand_test <- fisher.test(demographics$Handedness, demographics$group)
hand <- demographics %>% dplyr::group_by(group) %>% dplyr::count(Handedness)
hand_ocd <- paste0(hand[hand$group == "OCD" & hand$Handedness == "rechts", ]$n, ":",
hand[hand$group == "OCD" & hand$Handedness == "links", ]$n, ":",
hand[hand$group == "OCD" & hand$Handedness == "beidhändig", ]$n)
hand_hc <- paste0(hand[hand$group == "HC" & hand$Handedness == "rechts", ]$n, ":",
hand[hand$group == "HC" & hand$Handedness == "links", ]$n, ":",
length(hand[hand$group == "HC" & hand$Handedness == "beidhändig", ]$n))
# Combine demographics and chi-squared test/Fisher's exact test
demographics_factorial <- as.data.frame(cbind(
Characteristic = c("Gender (n female:male)", "Handedness (n right:left:ambidextrous)"),
ocd = c(gender_ocd, hand_ocd),
hc = c(gender_hc, hand_hc),
"Test statistic" = c(gender_test$statistic, "-"),
df = c(gender_test$parameter, "-"),
p = c(gender_test$p.value, hand_test$p.value)))
### Create and display table
# Re-order rows and format p values
table <- rbind(demographics_numeric[c(1), ], demographics_factorial,
demographics_numeric[c(2:nrow(demographics_numeric)), ])
table[, c("df", "Test statistic", "p")] <- as.numeric(unlist(table[, c("df", "Test statistic", "p")]))
table[, c("p")] <- format.pval(table[, c("p")], eps = 0.001, digits = 3)
table <- table %>% dplyr::rename("Patients with OCD" = ocd, "Healthy control participants" = hc)
# Display table
my_table_template(table,
caption = "Demographic and Clinical Characteristics") %>%
add_footnote(
"\n Note. Means are reported with standard deviations in parentheses (except for gender and handedness).
Welch’s t test was used for continuous variables. Age range was 18 to 55 years (control participants: 20 to 54 years;
patients with OCD: 18 to 55 years). Handedness was assessed with the Edinburgh Handedness Inventory (Oldfield, 1971).
For Fisher’s exact test, there is no test statistic to report. Years of education include primary and secondary education.
WST = Wortschatztest; BDI-II = Beck Depression Inventory-II; OCI-R = Obsessive-Compulsive Inventory-Revised;
PSWQ = Penn State Worry Questionnaire; STAI = State-Trait Anxiety Inventory; Y-BOCS = Yale-Brown Obsessive
Compulsive Scale.", notation = "none")| Characteristic | Patients with OCD | Healthy control participants | Test statistic | df | p |
|---|---|---|---|---|---|
| Age (years) | 33.29 ( 8.57) | 33.07 (8.20) | -0.10 | 53.90 | 0.924 |
| Gender (n female:male) | 17:11 | 17:11 | 0.00 | 1.00 | 1.000 |
| Handedness (n right:left:ambidextrous) | 26:1:1 | 27:1:0 | NA | NA | 1.000 |
| Years of education | 12.14 ( 1.46) | 12.14 (1.08) | 0.00 | 49.74 | 1.000 |
| Verbal intelligence (WST) | 104.93 ( 7.14) | 105.71 (8.30) | 0.38 | 52.81 | 0.706 |
| BDI-II | 14.14 (11.34) | 1.86 (2.69) | -5.58 | 30.03 | <0.001 |
| OCI-R | 25.75 ( 9.95) | 6.25 (5.65) | -9.02 | 42.75 | <0.001 |
| PSWQ | 61.21 (10.82) | 37.07 (9.86) | -8.73 | 53.54 | <0.001 |
| STAI trait | 53.29 (10.08) | 32.29 (7.22) | -8.96 | 48.92 | <0.001 |
| Y-BOCS total score | 23.36 ( 3.84) | NaN ( NA) | NA | NA | NA |
| Y-BOCS obsessions | 11.43 ( 1.81) | NaN ( NA) | NA | NA | NA |
| Y-BOCS compulsions | 11.86 ( 2.53) | NaN ( NA) | NA | NA | NA |
|
Note. Means are reported with standard deviations in parentheses (except for gender and handedness). Welch’s t test was used for continuous variables. Age range was 18 to 55 years (control participants: 20 to 54 years; patients with OCD: 18 to 55 years). Handedness was assessed with the Edinburgh Handedness Inventory (Oldfield, 1971). For Fisher’s exact test, there is no test statistic to report. Years of education include primary and secondary education. WST = Wortschatztest; BDI-II = Beck Depression Inventory-II; OCI-R = Obsessive-Compulsive Inventory-Revised; PSWQ = Penn State Worry Questionnaire; STAI = State-Trait Anxiety Inventory; Y-BOCS = Yale-Brown Obsessive Compulsive Scale. |
Wittchen, H., Zaudig, M., & Fydrich, T. (1997). Strukturiertes klinisches Interview für DSM-IV (SKID). Hogrefe.
sessionInfo()## R version 3.6.1 (2019-07-05)
## Platform: x86_64-w64-mingw32/x64 (64-bit)
## Running under: Windows 10 x64 (build 19044)
##
## Matrix products: default
##
## locale:
## [1] LC_COLLATE=German_Germany.1252 LC_CTYPE=German_Germany.1252
## [3] LC_MONETARY=German_Germany.1252 LC_NUMERIC=C
## [5] LC_TIME=German_Germany.1252
##
## attached base packages:
## [1] stats graphics grDevices datasets utils methods base
##
## other attached packages:
## [1] tidyr_1.2.1 psych_2.2.9 foreign_0.8-71 kableExtra_1.3.4
## [5] knitr_1.40 dplyr_1.0.10
##
## loaded via a namespace (and not attached):
## [1] tidyselect_1.2.0 xfun_0.33 bslib_0.4.0 purrr_0.3.5
## [5] lattice_0.20-45 colorspace_2.0-3 vctrs_0.5.0 generics_0.1.3
## [9] htmltools_0.5.3 viridisLite_0.4.1 yaml_2.2.1 utf8_1.2.2
## [13] rlang_1.0.6 jquerylib_0.1.4 pillar_1.8.1 glue_1.6.2
## [17] withr_2.5.0 DBI_1.1.3 lifecycle_1.0.3 stringr_1.4.1
## [21] munsell_0.5.0 rvest_1.0.3 evaluate_0.16 fastmap_1.1.0
## [25] parallel_3.6.1 fansi_1.0.3 highr_0.9 renv_0.12.0
## [29] scales_1.2.1 cachem_1.0.4 webshot_0.5.4 jsonlite_1.8.3
## [33] systemfonts_1.0.4 mnormt_2.1.1 digest_0.6.30 stringi_1.6.1
## [37] grid_3.6.1 cli_3.4.1 tools_3.6.1 magrittr_2.0.3
## [41] sass_0.4.0 tibble_3.1.8 pkgconfig_2.0.3 ellipsis_0.3.2
## [45] xml2_1.3.3 assertthat_0.2.1 rmarkdown_2.16 svglite_2.1.0
## [49] httr_1.4.4 rstudioapi_0.14 R6_2.5.1 nlme_3.1-140
## [53] compiler_3.6.1