Statistic
2025-03-05
NSAIDs and Dementia: Reassessing Their Role in Alzheimer’s Disease
The Potential of NSAIDs: What Have We Overlooked in Alzheimer’s Research?
NSAIDs and Alzheimer’s Risk: A Fresh Look at the Evidence
In this file, I have attempted to summarize the code and reproduce it. Below, you will find:
- A list of questions and considerations for discussion.
- Some technical issues encountered in RStudio.
- The full R code, in case @Victor wants to check specific parts.
NACCALZD (Dementia because of AD) ≠ demented (without etiology of dementia)
? I looked everywhere, but I couldnt find the criteria for AD. Have you?
Discussion Questions & Considerations
Q1) How was the education-corrected MOCA score of 25 (19, 27) calculated?
Q2) Fig 1
- When looking at Figure 1, I find it odd that we include patients with a MOCA score below a certain threshold as “normal cognition”, even if their low score is due to impairments unrelated to dementia. On the other hand, it also seems strange to include patients with a MOCA score above 26 (up to 30) in both the demented and NACCALZD groups.
What do you think?
Q3) Why was the following hierarchy applied in assigning NSAID_TYPE? And without Etodolac?
# Assign NSAID_TYPE = 1 where DICLOFENAC is used (Diclofenac supersedes Naproxen)
collapsed_data2$NSAID_TYPE[collapsed_data2$DICLOFENAC == 1] <- 1
# Assign NSAID_TYPE = 0 where only Naproxen is used (and not Diclofenac)
collapsed_data2$NSAID_TYPE[is.na(collapsed_data2$NSAID_TYPE) & collapsed_data2$NAPROXEN == 1] <- 0Q4) Fig 2 only includes demented patients.
- What about NACCALZD?
Etodolac (It also had a positive, though weak, effect on the risk)?
comparing to non-NSAID?
Q5) Fig 4 Shall we also include a box for demented non-NSAID and NACCALZD ?
Q6) NSAIDs, Arthritis, and Dementia Risk
- Off-topic, but the data also suggests a significant decrease in dementia risk among those with comorbid arthritis. Would it be too much to compare this cohort based on NSAID use?
Q7) Changes in MOCA scores over time: A comparison of Naproxen, Diclofenac, and non-NSAID groups of NACCALZD
- How does the time interval between assessments affect the calculations? The impact may differ if visits are spaced one week apart versus five years apart.
Analysis includes AD patients with multiple MOCA assessments.
Evaluating the change (Δ) in MOCA scores over time.
Technical Questions: R Code Issues
I have the following R code where I’ve marked the parts I struggled
with by writing "ERROR" in the header. Some of these errors
are consequential.
Would you like me to send the R script file instead of pasting the code here? Or should I just provide the code inline for you to review ?
1) Dataframes
I am unable to reproduce the model_data dataframe or
the match_obj object. Of course as a result, all subsequent
formulas are incorrect.
2) What does this error mean and why is it a whole page long?
Warning message:
“glm.fit: fitted probabilities numerically 0 or 1 occurred”
Warning message:Warning message:
“Removed 25522 rows containing non-finite outside the scale range
(`stat_smooth()`).”`geom_smooth()` using formula = 'y ~ x'
Warning message:
“Removed 25522 rows containing non-finite outside the scale range
(`stat_smooth()`).”Code
## ── Attaching core tidyverse packages ──────────────────────── tidyverse 2.0.0 ──
## ✔ dplyr 1.1.4 ✔ readr 2.1.5
## ✔ forcats 1.0.0 ✔ stringr 1.5.1
## ✔ ggplot2 3.5.1 ✔ tibble 3.2.1
## ✔ lubridate 1.9.4 ✔ tidyr 1.3.1
## ✔ purrr 1.0.4
## ── Conflicts ────────────────────────────────────────── tidyverse_conflicts() ──
## ✖ dplyr::filter() masks stats::filter()
## ✖ dplyr::lag() masks stats::lag()
## ℹ Use the conflicted package (<http://conflicted.r-lib.org/>) to force all conflicts to become errors
## Loading required package: lattice
##
##
## Attaching package: 'caret'
##
##
## The following object is masked from 'package:purrr':
##
## lift
##
##
## Loading required package: carData
##
##
## Attaching package: 'car'
##
##
## The following object is masked from 'package:dplyr':
##
## recode
##
##
## The following object is masked from 'package:purrr':
##
## some
##
##
## Loading required package: zoo
##
##
## Attaching package: 'zoo'
##
##
## The following objects are masked from 'package:base':
##
## as.Date, as.Date.numeric
##
##
##
## Attaching package: 'gridExtra'
##
##
## The following object is masked from 'package:dplyr':
##
## combineData collapsed
Data collapsed 1
# Define the columns to search for drugs (columns that start with "DRUG")
drug_columns <- grep("^DRUG", names(data), value = TRUE)
# Convert all drug columns to lowercase for case-insensitive comparison
data[drug_columns] <- lapply(data[drug_columns], tolower)
# Efficiently check for the presence of each drug across the relevant columns
# Create DICLOFENAC column
data$DICLOFENAC <- ifelse(rowSums(sapply(data[drug_columns], function(x) grepl("diclofenac", x))) > 0, 1, 0)
# Create NAPROXEN column
data$NAPROXEN <- ifelse(rowSums(sapply(data[drug_columns], function(x) grepl("naproxen", x))) > 0, 1, 0)
# Create ETODOLAC column
data$ETODOLAC <- ifelse(rowSums(sapply(data[drug_columns], function(x) grepl("etodolac", x))) > 0, 1, 0)
grep("NACCTBI",names(data))## [1] 147# Define the columns to search for diagnostic conditions
diagnosis_cols <- c("NACCTBI", "HXHYPER", "HXSTROKE", "DEP", "BIPOLDX", "SCHIZOP",
"ANXIET", "PTSDDX", "OTHPSY", "ALCABUSE", "CANCER", "DIABETES",
"MYOINF", "CONGHRT", "AFIBRILL", "HYPERT", "HYPCHOL", "VB12DEF",
"THYDIS", "ARTH", "SLEEPAP", "OTHCOND")
# Step 2: Create binary indicators for each diagnosis column indicating if the patient has the condition at any visit
data <- data %>%
mutate(across(all_of(diagnosis_cols), ~ ifelse(. %in% c(1, 2), 1, 0)))
# Now, aggregate these indicators by NACCID to get the total number of unique conditions per patient
diagnosis_summary <- data %>%
group_by(NACCID) %>%
summarise(across(all_of(diagnosis_cols), ~ max(.x, na.rm = TRUE))) %>%
mutate(DIAGNOSIS = rowSums(across(all_of(diagnosis_cols))))
# Merge the DIAGNOSIS summary back to the original data without collapsing to one row per NACCID
data <- data %>%
left_join(diagnosis_summary %>% select(NACCID, DIAGNOSIS), by = "NACCID")
save(data, diagnosis_summary, diagnosis_cols, drug_columns, file = "data2Oct2024.RData")
# Preliminary: Collapse DICLOFENAC, ETODOLAC, NAPROXEN, and CSFTAU columns across all visits for each unique patient (NACCID)
data <- data %>%
group_by(NACCID) %>%
mutate(DICLOFENAC = ifelse(any(DICLOFENAC == 1, na.rm = TRUE), 1, 0),
NAPROXEN = ifelse(any(NAPROXEN == 1, na.rm = TRUE), 1, 0),
ETODOLAC = ifelse(any(ETODOLAC == 1, na.rm = TRUE), 1, 0),
CSFTAU = ifelse(any(CSFTAU == 1, na.rm = TRUE), 1, 0))%>%
ungroup()
# Select the most recent visit for specific variables if NACCMOCA is between 0 and 30, otherwise keep data from the most recent visit
most_recent_data <- data %>%
group_by(NACCID) %>%
filter((NACCMOCA >= 0 & NACCMOCA <= 30) | is.na(NACCMOCA) | NACCVNUM == max(NACCVNUM)) %>%
slice_max(NACCVNUM, n = 1) %>%
ungroup()
# c. Combine diagnosis data with the most recent visit data
collapsed_data <- most_recent_data %>%
select(NACCID, NACCVNUM, NACCAGE, SEX, NACCNIHR, EDUC, NACCALZD, NACCALZP,
NACCMOCA, CDRGLOB, NACCMMSE, NORMCOG, DEMENTED, NACCUDSD,
CSFTAU, NACCTBI, HXHYPER, HXSTROKE, DEP, BIPOLDX, SCHIZOP,
ANXIET, PTSDDX, OTHPSY, ALCABUSE, CANCER, DIABETES, MYOINF,
CONGHRT, AFIBRILL, HYPERT, HYPCHOL, VB12DEF, THYDIS, ARTH,
SLEEPAP, OTHCOND, DICLOFENAC, NAPROXEN, ETODOLAC, DIAGNOSIS,
MOCATRAI, MOCACUBE, MOCACLOC, MOCACLON, MOCACLOH, MOCANAMI,
MOCAFLUE, MOCAREPE, MOCAREGI, MOCARECN, MOCAABST, MOCADIGI,
MOCALETT, MOCASER7, MOCAORDT, MOCAORMO, MOCAORYR, MOCAORDY,
MOCAORPL, MOCAORCT)
# List of columns to convert
columns_to_convert <- c("MOCATRAI", "MOCACUBE", "MOCACLOC", "MOCACLON", "MOCACLOH", "MOCANAMI",
"MOCAFLUE", "MOCAREPE", "MOCAREGI", "MOCARECN", "MOCAABST", "MOCADIGI",
"MOCALETT", "MOCASER7", "MOCAORDT", "MOCAORMO", "MOCAORYR", "MOCAORDY",
"MOCAORPL", "MOCAORCT")
# Convert the columns to numeric
collapsed_data[columns_to_convert] <- lapply(collapsed_data[columns_to_convert], as.numeric)
save(data, collapsed_data, file="data2Oct2024.RData")
names(collapsed_data)## [1] "NACCID" "NACCVNUM" "NACCAGE" "SEX" "NACCNIHR"
## [6] "EDUC" "NACCALZD" "NACCALZP" "NACCMOCA" "CDRGLOB"
## [11] "NACCMMSE" "NORMCOG" "DEMENTED" "NACCUDSD" "CSFTAU"
## [16] "NACCTBI" "HXHYPER" "HXSTROKE" "DEP" "BIPOLDX"
## [21] "SCHIZOP" "ANXIET" "PTSDDX" "OTHPSY" "ALCABUSE"
## [26] "CANCER" "DIABETES" "MYOINF" "CONGHRT" "AFIBRILL"
## [31] "HYPERT" "HYPCHOL" "VB12DEF" "THYDIS" "ARTH"
## [36] "SLEEPAP" "OTHCOND" "DICLOFENAC" "NAPROXEN" "ETODOLAC"
## [41] "DIAGNOSIS" "MOCATRAI" "MOCACUBE" "MOCACLOC" "MOCACLON"
## [46] "MOCACLOH" "MOCANAMI" "MOCAFLUE" "MOCAREPE" "MOCAREGI"
## [51] "MOCARECN" "MOCAABST" "MOCADIGI" "MOCALETT" "MOCASER7"
## [56] "MOCAORDT" "MOCAORMO" "MOCAORYR" "MOCAORDY" "MOCAORPL"
## [61] "MOCAORCT"dim(collapsed_data)## [1] 47165 61n_unique_NACCID <- length(unique(data$NACCID))
#collapsed_data
collapsed_data <- collapsed_data %>%
mutate(across(
.cols = where(is.numeric) & !all_of("NACCAGE"),
.fns = ~ na_if(., -4))) %>%
mutate(across(
.cols = where(is.numeric) & !all_of("NACCAGE"),
.fns = ~ na_if(., 88))) %>%
mutate(across(
.cols = where(is.numeric) & !all_of("NACCAGE"),
.fns = ~ na_if(., 95))) %>%
mutate(across(
.cols = where(is.numeric) & !all_of("NACCAGE"),
.fns = ~ na_if(., 96))) %>%
mutate(across(
.cols = where(is.numeric) & !all_of("NACCAGE"),
.fns = ~ na_if(., 97))) %>%
mutate(across(
.cols = where(is.numeric) & !all_of("NACCAGE"),
.fns = ~ na_if(., 98))) %>%
mutate(across(
.cols = where(is.numeric) & !all_of("NACCAGE"),
.fns = ~ na_if(., 99)))
# Identify the variables that need to be summed
columns_of_interestVS <- c("MOCATRAI", "MOCACUBE", "MOCACLOC", "MOCACLON", "MOCACLOH")
columns_of_interestLAN <- c("MOCANAMI", "MOCAFLUE", "MOCAREPE")
columns_of_interestMEM1 <- c("MOCAREGI")
columns_of_interestMEM2 <- c("MOCARECN")
columns_of_interestABS <- c("MOCAABST")
columns_of_interestATTN <- c("MOCADIGI", "MOCALETT", "MOCASER7")
columns_of_interestORI <- c("MOCAORDT", "MOCAORMO", "MOCAORYR", "MOCAORDY", "MOCAORPL", "MOCAORCT")
# Create new columns based on the sum of the variables
collapsed_data <- collapsed_data %>%
mutate(
VISUOSPATIAL = rowSums(select(., all_of(columns_of_interestVS)), na.rm = TRUE),
LANGUAGE = rowSums(select(., all_of(columns_of_interestLAN)), na.rm = TRUE),
MEMORY1 = rowSums(select(., all_of(columns_of_interestMEM1)), na.rm = TRUE),
MEMORY2 = rowSums(select(., all_of(columns_of_interestMEM2)), na.rm = TRUE),
ABSTRACTION = rowSums(select(., all_of(columns_of_interestABS)), na.rm = TRUE),
ATTENTION = rowSums(select(., all_of(columns_of_interestATTN)), na.rm = TRUE),
ORIENTATION = rowSums(select(., all_of(columns_of_interestORI)), na.rm = TRUE)
)
summary(collapsed_data)## NACCID NACCVNUM NACCAGE SEX
## Length:47165 Min. : 1.000 Min. : 18.00 Min. :1.00
## Class :character 1st Qu.: 1.000 1st Qu.: 68.00 1st Qu.:1.00
## Mode :character Median : 3.000 Median : 75.00 Median :2.00
## Mean : 3.704 Mean : 74.62 Mean :1.57
## 3rd Qu.: 5.000 3rd Qu.: 82.00 3rd Qu.:2.00
## Max. :18.000 Max. :110.00 Max. :2.00
##
## NACCNIHR EDUC NACCALZD NACCALZP
## Min. :1.000 Min. : 0.00 Min. :0.000 Min. :1.000
## 1st Qu.:1.000 1st Qu.:12.00 1st Qu.:1.000 1st Qu.:1.000
## Median :1.000 Median :16.00 Median :1.000 Median :7.000
## Mean :1.411 Mean :15.17 Mean :3.292 Mean :4.757
## 3rd Qu.:1.000 3rd Qu.:18.00 3rd Qu.:8.000 3rd Qu.:8.000
## Max. :6.000 Max. :31.00 Max. :8.000 Max. :8.000
## NA's :824 NA's :363
## NACCMOCA CDRGLOB NACCMMSE NORMCOG
## Min. : 0.00 Min. :0.0000 Min. : 0.00 Min. :0.0000
## 1st Qu.:19.00 1st Qu.:0.0000 1st Qu.:19.00 1st Qu.:0.0000
## Median :24.00 Median :0.5000 Median :26.00 Median :0.0000
## Mean :22.05 Mean :0.8002 Mean :22.76 Mean :0.3567
## 3rd Qu.:27.00 3rd Qu.:1.0000 3rd Qu.:29.00 3rd Qu.:1.0000
## Max. :30.00 Max. :3.0000 Max. :30.00 Max. :1.0000
## NA's :32487 NA's :26564
## DEMENTED NACCUDSD CSFTAU NACCTBI
## Min. :0.0000 Min. :1.000 Min. :0.00000 Min. :0.00000
## 1st Qu.:0.0000 1st Qu.:1.000 1st Qu.:0.00000 1st Qu.:0.00000
## Median :0.0000 Median :3.000 Median :0.00000 Median :0.00000
## Mean :0.4252 Mean :2.672 Mean :0.01662 Mean :0.07887
## 3rd Qu.:1.0000 3rd Qu.:4.000 3rd Qu.:0.00000 3rd Qu.:0.00000
## Max. :1.0000 Max. :4.000 Max. :1.00000 Max. :1.00000
##
## HXHYPER HXSTROKE DEP BIPOLDX
## Min. :0.000 Min. :0.00000 Min. :0.0000 Min. :0.00000
## 1st Qu.:0.000 1st Qu.:0.00000 1st Qu.:0.0000 1st Qu.:0.00000
## Median :0.000 Median :0.00000 Median :0.0000 Median :0.00000
## Mean :0.255 Mean :0.03515 Mean :0.1763 Mean :0.00299
## 3rd Qu.:1.000 3rd Qu.:0.00000 3rd Qu.:0.0000 3rd Qu.:0.00000
## Max. :1.000 Max. :1.00000 Max. :1.0000 Max. :1.00000
##
## SCHIZOP ANXIET PTSDDX OTHPSY
## Min. :0.0000000 Min. :0.00000 Min. :0.000000 Min. :0.00000
## 1st Qu.:0.0000000 1st Qu.:0.00000 1st Qu.:0.000000 1st Qu.:0.00000
## Median :0.0000000 Median :0.00000 Median :0.000000 Median :0.00000
## Mean :0.0008057 Mean :0.02837 Mean :0.002353 Mean :0.01427
## 3rd Qu.:0.0000000 3rd Qu.:0.00000 3rd Qu.:0.000000 3rd Qu.:0.00000
## Max. :1.0000000 Max. :1.00000 Max. :1.000000 Max. :1.00000
##
## ALCABUSE CANCER DIABETES MYOINF
## Min. :0.000000 Min. :0.00000 Min. :0.00000 Min. :0.00000
## 1st Qu.:0.000000 1st Qu.:0.00000 1st Qu.:0.00000 1st Qu.:0.00000
## Median :0.000000 Median :0.00000 Median :0.00000 Median :0.00000
## Mean :0.001399 Mean :0.09348 Mean :0.09492 Mean :0.01904
## 3rd Qu.:0.000000 3rd Qu.:0.00000 3rd Qu.:0.00000 3rd Qu.:0.00000
## Max. :1.000000 Max. :1.00000 Max. :1.00000 Max. :1.00000
##
## CONGHRT AFIBRILL HYPERT HYPCHOL
## Min. :0.00000 Min. :0.0000 Min. :0.0000 Min. :0.0000
## 1st Qu.:0.00000 1st Qu.:0.0000 1st Qu.:0.0000 1st Qu.:0.0000
## Median :0.00000 Median :0.0000 Median :0.0000 Median :0.0000
## Mean :0.01446 Mean :0.0444 Mean :0.2561 Mean :0.2704
## 3rd Qu.:0.00000 3rd Qu.:0.0000 3rd Qu.:1.0000 3rd Qu.:1.0000
## Max. :1.00000 Max. :1.0000 Max. :1.0000 Max. :1.0000
##
## VB12DEF THYDIS ARTH SLEEPAP
## Min. :0.00000 Min. :0.00000 Min. :0.0000 Min. :0.00000
## 1st Qu.:0.00000 1st Qu.:0.00000 1st Qu.:0.0000 1st Qu.:0.00000
## Median :0.00000 Median :0.00000 Median :0.0000 Median :0.00000
## Mean :0.03785 Mean :0.09844 Mean :0.2793 Mean :0.09354
## 3rd Qu.:0.00000 3rd Qu.:0.00000 3rd Qu.:1.0000 3rd Qu.:0.00000
## Max. :1.00000 Max. :1.00000 Max. :1.0000 Max. :1.00000
##
## OTHCOND DICLOFENAC NAPROXEN ETODOLAC
## Min. :0.0000 Min. :0.0000 Min. :0.00000 Min. :0.000000
## 1st Qu.:0.0000 1st Qu.:0.0000 1st Qu.:0.00000 1st Qu.:0.000000
## Median :0.0000 Median :0.0000 Median :0.00000 Median :0.000000
## Mean :0.1595 Mean :0.0219 Mean :0.06098 Mean :0.003922
## 3rd Qu.:0.0000 3rd Qu.:0.0000 3rd Qu.:0.00000 3rd Qu.:0.000000
## Max. :1.0000 Max. :1.0000 Max. :1.00000 Max. :1.000000
##
## DIAGNOSIS MOCATRAI MOCACUBE MOCACLOC
## Min. : 0.000 Min. :0.00 Min. :0.0 Min. :0.00
## 1st Qu.: 1.000 1st Qu.:0.00 1st Qu.:0.0 1st Qu.:1.00
## Median : 2.000 Median :1.00 Median :1.0 Median :1.00
## Mean : 2.692 Mean :0.66 Mean :0.5 Mean :0.94
## 3rd Qu.: 4.000 3rd Qu.:1.00 3rd Qu.:1.0 3rd Qu.:1.00
## Max. :15.000 Max. :1.00 Max. :1.0 Max. :1.00
## NA's :32208 NA's :32188 NA's :32192
## MOCACLON MOCACLOH MOCANAMI MOCAFLUE
## Min. :0.00 Min. :0.00 Min. :0.00 Min. :0.000
## 1st Qu.:1.00 1st Qu.:0.00 1st Qu.:2.00 1st Qu.:0.000
## Median :1.00 Median :1.00 Median :3.00 Median :1.000
## Mean :0.77 Mean :0.58 Mean :2.64 Mean :0.674
## 3rd Qu.:1.00 3rd Qu.:1.00 3rd Qu.:3.00 3rd Qu.:1.000
## Max. :1.00 Max. :1.00 Max. :3.00 Max. :1.000
## NA's :32193 NA's :32194 NA's :32125 NA's :29182
## MOCAREPE MOCAREGI MOCARECN MOCAABST
## Min. :0.000 Min. : 0.00 Min. :0.00 Min. :0.000
## 1st Qu.:1.000 1st Qu.: 8.00 1st Qu.:0.00 1st Qu.:1.000
## Median :2.000 Median : 9.00 Median :3.00 Median :2.000
## Mean :1.348 Mean : 8.33 Mean :2.35 Mean :1.526
## 3rd Qu.:2.000 3rd Qu.:10.00 3rd Qu.:4.00 3rd Qu.:2.000
## Max. :2.000 Max. :10.00 Max. :5.00 Max. :2.000
## NA's :29160 NA's :32144 NA's :29205 NA's :29171
## MOCADIGI MOCALETT MOCASER7 MOCAORDT
## Min. :0.00 Min. :0.000 Min. :0.000 Min. :0.000
## 1st Qu.:2.00 1st Qu.:1.000 1st Qu.:2.000 1st Qu.:1.000
## Median :2.00 Median :1.000 Median :3.000 Median :1.000
## Mean :1.72 Mean :0.843 Mean :2.365 Mean :0.764
## 3rd Qu.:2.00 3rd Qu.:1.000 3rd Qu.:3.000 3rd Qu.:1.000
## Max. :2.00 Max. :1.000 Max. :3.000 Max. :1.000
## NA's :29134 NA's :29158 NA's :29263 NA's :29152
## MOCAORMO MOCAORYR MOCAORDY MOCAORPL
## Min. :0.000 Min. :0.000 Min. :0.000 Min. :0.000
## 1st Qu.:1.000 1st Qu.:1.000 1st Qu.:1.000 1st Qu.:1.000
## Median :1.000 Median :1.000 Median :1.000 Median :1.000
## Mean :0.888 Mean :0.876 Mean :0.855 Mean :0.849
## 3rd Qu.:1.000 3rd Qu.:1.000 3rd Qu.:1.000 3rd Qu.:1.000
## Max. :1.000 Max. :1.000 Max. :1.000 Max. :1.000
## NA's :29152 NA's :29152 NA's :29150 NA's :29152
## MOCAORCT VISUOSPATIAL LANGUAGE MEMORY1
## Min. :0.00 Min. :0.000 Min. :0.000 Min. : 0.000
## 1st Qu.:1.00 1st Qu.:0.000 1st Qu.:0.000 1st Qu.: 0.000
## Median :1.00 Median :0.000 Median :0.000 Median : 0.000
## Mean :0.94 Mean :1.098 Mean :1.614 Mean : 2.654
## 3rd Qu.:1.00 3rd Qu.:2.000 3rd Qu.:4.000 3rd Qu.: 7.000
## Max. :1.00 Max. :5.000 Max. :6.000 Max. :10.000
## NA's :29148
## MEMORY2 ABSTRACTION ATTENTION ORIENTATION
## Min. :0.000 Min. :0.0000 Min. :0.000 Min. :0.000
## 1st Qu.:0.000 1st Qu.:0.0000 1st Qu.:0.000 1st Qu.:0.000
## Median :0.000 Median :0.0000 Median :0.000 Median :0.000
## Mean :0.895 Mean :0.5821 Mean :1.877 Mean :1.976
## 3rd Qu.:1.000 3rd Qu.:2.0000 3rd Qu.:5.000 3rd Qu.:6.000
## Max. :5.000 Max. :2.0000 Max. :6.000 Max. :6.000
## collapsed_data <- collapsed_data %>%
mutate(SUM_SCORE = VISUOSPATIAL + LANGUAGE + MEMORY2 + ABSTRACTION + ATTENTION + ORIENTATION)
ggplot(collapsed_data, aes(x = SUM_SCORE, y = NACCMOCA)) +
geom_point(color = "blue", alpha = 0.7) +
labs(title = "Scatterplot of Sum of Cognitive Scores vs NACCMOCA",
x = "Sum of VISUOSPATIAL, LANGUAGE, MEMORY2, ABSTRACTION, ATTENTION, ORIENTATION",
y = "NACCMOCA") +
theme_minimal()
# Convert categorical variables to factors
collapsed_data <- collapsed_data %>%
mutate(
SEX = as.factor(SEX),
NACCNIHR = as.factor(NACCNIHR),
NACCTBI = as.factor(NACCTBI),
DEP = as.factor(DEP),
BIPOLDX = as.factor(BIPOLDX),
SCHIZOP = as.factor(SCHIZOP),
ANXIET = as.factor(ANXIET),
PTSDDX = as.factor(PTSDDX),
CANCER = as.factor(CANCER),
DIABETES = as.factor(DIABETES),
CONGHRT = as.factor(CONGHRT),
HYPERT = as.factor(HYPERT),
HYPCHOL = as.factor(HYPCHOL),
VB12DEF = as.factor(VB12DEF),
THYDIS = as.factor(THYDIS),
ARTH = as.factor(ARTH),
SLEEPAP = as.factor(SLEEPAP),
DICLOFENAC = as.factor(DICLOFENAC),
NAPROXEN = as.factor(NAPROXEN),
ETODOLAC = as.factor(ETODOLAC),
DIAGNOSIS = as.numeric(DIAGNOSIS) # DIAGNOSIS is already numeric (sum of conditions)
)
names(collapsed_data)## [1] "NACCID" "NACCVNUM" "NACCAGE" "SEX" "NACCNIHR"
## [6] "EDUC" "NACCALZD" "NACCALZP" "NACCMOCA" "CDRGLOB"
## [11] "NACCMMSE" "NORMCOG" "DEMENTED" "NACCUDSD" "CSFTAU"
## [16] "NACCTBI" "HXHYPER" "HXSTROKE" "DEP" "BIPOLDX"
## [21] "SCHIZOP" "ANXIET" "PTSDDX" "OTHPSY" "ALCABUSE"
## [26] "CANCER" "DIABETES" "MYOINF" "CONGHRT" "AFIBRILL"
## [31] "HYPERT" "HYPCHOL" "VB12DEF" "THYDIS" "ARTH"
## [36] "SLEEPAP" "OTHCOND" "DICLOFENAC" "NAPROXEN" "ETODOLAC"
## [41] "DIAGNOSIS" "MOCATRAI" "MOCACUBE" "MOCACLOC" "MOCACLON"
## [46] "MOCACLOH" "MOCANAMI" "MOCAFLUE" "MOCAREPE" "MOCAREGI"
## [51] "MOCARECN" "MOCAABST" "MOCADIGI" "MOCALETT" "MOCASER7"
## [56] "MOCAORDT" "MOCAORMO" "MOCAORYR" "MOCAORDY" "MOCAORPL"
## [61] "MOCAORCT" "VISUOSPATIAL" "LANGUAGE" "MEMORY1" "MEMORY2"
## [66] "ABSTRACTION" "ATTENTION" "ORIENTATION" "SUM_SCORE"Data collapsed 2
#Step 1: Filter out patients with NACCALZD == 0
filtered_data <- collapsed_data %>%
filter(NACCALZD != 0)
# Step 2: Recode NACCALZD where 8 (normal cognition) is changed to 0
filtered_data <- filtered_data %>%
mutate(NACCALZD = case_when(
NACCALZD == 8 ~ 0, # Recode 8 to 0 (8 ~ 0 no cognitive impairment)
NACCALZD == 1 ~ 1 # Keep 1 as 1 (Alzheimer's disease)
))
collapsed_data2<-filtered_data
# Initialize NSAID_TYPE as NA
collapsed_data2$NSAID_TYPE <- NA
# Assign NSAID_TYPE = 1 where DICLOFENAC is used (Diclofenac supersedes Naproxen)
collapsed_data2$NSAID_TYPE[collapsed_data2$DICLOFENAC == 1] <- 1
# Assign NSAID_TYPE = 0 where only Naproxen is used (and not Diclofenac)
collapsed_data2$NSAID_TYPE[is.na(collapsed_data2$NSAID_TYPE) & collapsed_data2$NAPROXEN == 1] <- 0
table(collapsed_data2$DICLOFENAC, useNA = "always")##
## 0 1 <NA>
## 36625 844 0table(collapsed_data2$NAPROXEN, useNA = "always")##
## 0 1 <NA>
## 35084 2385 0Data collapsed 3 & model_data, Match_obj matched_data
Iam unable to reproduce the model_data dataframe or
the match_obj object. As a result, all subsequent formulas
are incorrect.
# model_data
model_data <- collapsed_data2[!is.na(collapsed_data2$NSAID_TYPE),]
dim(model_data)## [1] 3111 70#collapsed_data
collapsed_data3<-collapsed_data2[!is.na(collapsed_data2$NSAID_TYPE),]
dim(collapsed_data3)## [1] 3111 70#match_data
match_obj <- matchit(NSAID_TYPE ~ NACCAGE + SEX + NACCNIHR + EDUC + DIAGNOSIS,
data = model_data,
method = "full",
distance = "glm",
link = "probit")## Error in `matchit()`:
## ! Missing and non-finite values are not allowed in the covariates.
## Covariates with missingness or non-finite values: NACCNIHR, EDUCmatched_data <- match.data(match_obj)## Error: object 'match_obj' not foundInconsistencies MOCA Score / Cogn. Score to Demented / NACCALZD
When looking at Figure 1, I find it odd that we include patients with a MOCA score below a certain threshold as “normal cognition”, even if their low score is due to impairments unrelated to dementia. On the other hand, it also seems strange to include patients with a MOCA score above 26 (up to 30) in both the demented and NACCALZD groups.
# Create new dataframe ad_moca with selected columns
ad_moca <- collapsed_data2[, c("NACCID", "NACCMOCA", "DEMENTED", "NACCALZD")]
# Filter for cases where DEMENTED status or NACCALZD Status doesnt match Moca
inconsistent_cases <- collapsed_data2 %>%
filter(NACCMOCA > 25 & DEMENTED == 1)
dim(inconsistent_cases)## [1] 21 70inconsistent_cases1 <- collapsed_data2 %>%
filter(NACCMOCA < 20 & DEMENTED == 0)
dim(inconsistent_cases1)## [1] 643 70 ## nearly 1%
inconsistent_cases2 <- collapsed_data2 %>%
filter(NACCMOCA > 26 & NACCALZD == 1)
dim(inconsistent_cases2)## [1] 190 70 ## nearly 1%
inconsistent_cases3 <- collapsed_data2 %>%
filter(NACCMOCA < 20 & NACCALZD == 0)
dim(inconsistent_cases3)## [1] 172 70# Display the results
print(inconsistent_cases)## # A tibble: 21 × 70
## NACCID NACCVNUM NACCAGE SEX NACCNIHR EDUC NACCALZD NACCALZP NACCMOCA
## <chr> <int> <int> <fct> <fct> <int> <dbl> <int> <int>
## 1 NACC097130 5 73 1 1 19 1 1 26
## 2 NACC108978 2 78 1 1 20 1 1 26
## 3 NACC162469 1 77 1 1 12 1 1 28
## 4 NACC175064 1 63 1 1 16 1 2 30
## 5 NACC179619 2 66 1 1 18 1 2 27
## 6 NACC194582 1 63 2 2 14 1 1 26
## 7 NACC253699 2 65 1 1 18 1 1 28
## 8 NACC308810 1 73 2 1 18 1 1 30
## 9 NACC344354 4 73 2 1 16 1 1 26
## 10 NACC406578 1 64 2 1 13 1 1 28
## # ℹ 11 more rows
## # ℹ 61 more variables: CDRGLOB <dbl>, NACCMMSE <int>, NORMCOG <int>,
## # DEMENTED <int>, NACCUDSD <int>, CSFTAU <dbl>, NACCTBI <fct>, HXHYPER <dbl>,
## # HXSTROKE <dbl>, DEP <fct>, BIPOLDX <fct>, SCHIZOP <fct>, ANXIET <fct>,
## # PTSDDX <fct>, OTHPSY <dbl>, ALCABUSE <dbl>, CANCER <fct>, DIABETES <fct>,
## # MYOINF <dbl>, CONGHRT <fct>, AFIBRILL <dbl>, HYPERT <fct>, HYPCHOL <fct>,
## # VB12DEF <fct>, THYDIS <fct>, ARTH <fct>, SLEEPAP <fct>, OTHCOND <dbl>, …Linear regression models Demented/NACCALZD
# Logistic regression model 1 without filtering NACCALZD, linear regression on demented
set.seed(123)
logistic_model1 <- glm(
DEMENTED ~ NACCAGE + SEX + NACCNIHR + EDUC + NACCTBI + DEP + BIPOLDX + SCHIZOP + ANXIET +
PTSDDX + CANCER + DIABETES + CONGHRT + HYPERT + HYPCHOL + VB12DEF +
THYDIS + ARTH + SLEEPAP + DICLOFENAC + NAPROXEN + ETODOLAC + DIAGNOSIS,
data = collapsed_data,
family = binomial(link = "logit")
)
# View the summary of the model
summary(logistic_model1)##
## Call:
## glm(formula = DEMENTED ~ NACCAGE + SEX + NACCNIHR + EDUC + NACCTBI +
## DEP + BIPOLDX + SCHIZOP + ANXIET + PTSDDX + CANCER + DIABETES +
## CONGHRT + HYPERT + HYPCHOL + VB12DEF + THYDIS + ARTH + SLEEPAP +
## DICLOFENAC + NAPROXEN + ETODOLAC + DIAGNOSIS, family = binomial(link = "logit"),
## data = collapsed_data)
##
## Coefficients:
## Estimate Std. Error z value Pr(>|z|)
## (Intercept) -0.2288157 0.0894181 -2.559 0.010499 *
## NACCAGE 0.0244402 0.0009579 25.514 < 2e-16 ***
## SEX2 -0.4082967 0.0210270 -19.418 < 2e-16 ***
## NACCNIHR2 -0.6084231 0.0326240 -18.650 < 2e-16 ***
## NACCNIHR3 -0.4569022 0.1317157 -3.469 0.000523 ***
## NACCNIHR4 0.3826622 0.3513750 1.089 0.276135
## NACCNIHR5 -0.3097220 0.0628109 -4.931 8.18e-07 ***
## NACCNIHR6 -0.4540993 0.0598494 -7.587 3.26e-14 ***
## EDUC -0.0834979 0.0031702 -26.339 < 2e-16 ***
## NACCTBI1 -0.1484236 0.0377231 -3.935 8.33e-05 ***
## DEP1 0.5397809 0.0279950 19.281 < 2e-16 ***
## BIPOLDX1 0.0429715 0.1916503 0.224 0.822588
## SCHIZOP1 0.0629820 0.3724545 0.169 0.865718
## ANXIET1 0.1670566 0.0635791 2.628 0.008600 **
## PTSDDX1 -0.8818708 0.2618662 -3.368 0.000758 ***
## CANCER1 -0.2720226 0.0397499 -6.843 7.74e-12 ***
## DIABETES1 -0.1495084 0.0362394 -4.126 3.70e-05 ***
## CONGHRT1 0.2238859 0.0884959 2.530 0.011409 *
## HYPERT1 -0.1031722 0.0326610 -3.159 0.001584 **
## HYPCHOL1 -0.1200694 0.0309884 -3.875 0.000107 ***
## VB12DEF1 0.3626941 0.0544457 6.662 2.71e-11 ***
## THYDIS1 -0.0113423 0.0386633 -0.293 0.769246
## ARTH1 -0.7911988 0.0303579 -26.062 < 2e-16 ***
## SLEEPAP1 -0.0505020 0.0402077 -1.256 0.209106
## DICLOFENAC1 -0.3210074 0.0756263 -4.245 2.19e-05 ***
## NAPROXEN1 -0.2572108 0.0440678 -5.837 5.32e-09 ***
## ETODOLAC1 -0.3605477 0.1664991 -2.165 0.030352 *
## DIAGNOSIS -0.0348220 0.0092516 -3.764 0.000167 ***
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
##
## (Dispersion parameter for binomial family taken to be 1)
##
## Null deviance: 62680 on 46010 degrees of freedom
## Residual deviance: 57643 on 45983 degrees of freedom
## (1154 observations deleted due to missingness)
## AIC: 57699
##
## Number of Fisher Scoring iterations: 4# Logistic regression model 2 filtering NACCALZD, linear regression on demented
set.seed(123)
logistic_model2 <- glm(
DEMENTED ~ NACCAGE + SEX + NACCNIHR + EDUC + NACCTBI + DEP + BIPOLDX + SCHIZOP + ANXIET +
PTSDDX + CANCER + DIABETES + CONGHRT + HYPERT + HYPCHOL + VB12DEF +
THYDIS + ARTH + SLEEPAP + DICLOFENAC + NAPROXEN + ETODOLAC + DIAGNOSIS,
data = collapsed_data2,
family = binomial(link = "logit")
)
# View the summary of the model
summary(logistic_model2)##
## Call:
## glm(formula = DEMENTED ~ NACCAGE + SEX + NACCNIHR + EDUC + NACCTBI +
## DEP + BIPOLDX + SCHIZOP + ANXIET + PTSDDX + CANCER + DIABETES +
## CONGHRT + HYPERT + HYPCHOL + VB12DEF + THYDIS + ARTH + SLEEPAP +
## DICLOFENAC + NAPROXEN + ETODOLAC + DIAGNOSIS, family = binomial(link = "logit"),
## data = collapsed_data2)
##
## Coefficients:
## Estimate Std. Error z value Pr(>|z|)
## (Intercept) -0.863713 0.105864 -8.159 3.39e-16 ***
## NACCAGE 0.036396 0.001125 32.352 < 2e-16 ***
## SEX2 -0.402324 0.024249 -16.591 < 2e-16 ***
## NACCNIHR2 -0.449112 0.036489 -12.308 < 2e-16 ***
## NACCNIHR3 -0.231145 0.152121 -1.519 0.128643
## NACCNIHR4 0.179366 0.449004 0.399 0.689542
## NACCNIHR5 -0.298822 0.072385 -4.128 3.66e-05 ***
## NACCNIHR6 -0.310564 0.067980 -4.568 4.91e-06 ***
## EDUC -0.104785 0.003689 -28.404 < 2e-16 ***
## NACCTBI1 -0.097578 0.045352 -2.152 0.031433 *
## DEP1 0.798841 0.033535 23.821 < 2e-16 ***
## BIPOLDX1 0.221831 0.257139 0.863 0.388307
## SCHIZOP1 0.162300 0.516691 0.314 0.753434
## ANXIET1 0.355259 0.075861 4.683 2.83e-06 ***
## PTSDDX1 -0.960394 0.346725 -2.770 0.005607 **
## CANCER1 -0.287111 0.044382 -6.469 9.86e-11 ***
## DIABETES1 -0.072395 0.042244 -1.714 0.086576 .
## CONGHRT1 0.216954 0.099803 2.174 0.029719 *
## HYPERT1 -0.164898 0.036819 -4.479 7.51e-06 ***
## HYPCHOL1 -0.127070 0.034762 -3.655 0.000257 ***
## VB12DEF1 0.388951 0.061902 6.283 3.31e-10 ***
## THYDIS1 -0.002559 0.043131 -0.059 0.952688
## ARTH1 -0.820028 0.034007 -24.113 < 2e-16 ***
## SLEEPAP1 -0.046994 0.046432 -1.012 0.311487
## DICLOFENAC1 -0.316232 0.085022 -3.719 0.000200 ***
## NAPROXEN1 -0.319987 0.049793 -6.426 1.31e-10 ***
## ETODOLAC1 -0.348451 0.183136 -1.903 0.057081 .
## DIAGNOSIS -0.032523 0.010481 -3.103 0.001914 **
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
##
## (Dispersion parameter for binomial family taken to be 1)
##
## Null deviance: 49815 on 36622 degrees of freedom
## Residual deviance: 44463 on 36595 degrees of freedom
## (846 observations deleted due to missingness)
## AIC: 44519
##
## Number of Fisher Scoring iterations: 4# Logistic regression model 3 filtering NACCALZD, linear regression on NACCALZD
set.seed(123)
logistic_model3 <- glm(
NACCALZD ~ NACCAGE + SEX + NACCNIHR + EDUC + NACCTBI + DEP + BIPOLDX + SCHIZOP + ANXIET +
PTSDDX + CANCER + DIABETES + CONGHRT + HYPERT + HYPCHOL + VB12DEF +
THYDIS + ARTH + SLEEPAP + DICLOFENAC + NAPROXEN + ETODOLAC + DIAGNOSIS,
data = collapsed_data2,
family = binomial(link = "logit")
)
# View the summary of the model
summary(logistic_model3)##
## Call:
## glm(formula = NACCALZD ~ NACCAGE + SEX + NACCNIHR + EDUC + NACCTBI +
## DEP + BIPOLDX + SCHIZOP + ANXIET + PTSDDX + CANCER + DIABETES +
## CONGHRT + HYPERT + HYPCHOL + VB12DEF + THYDIS + ARTH + SLEEPAP +
## DICLOFENAC + NAPROXEN + ETODOLAC + DIAGNOSIS, family = binomial(link = "logit"),
## data = collapsed_data2)
##
## Coefficients:
## Estimate Std. Error z value Pr(>|z|)
## (Intercept) -1.389924 0.107053 -12.984 < 2e-16 ***
## NACCAGE 0.049491 0.001153 42.919 < 2e-16 ***
## SEX2 -0.509775 0.024509 -20.800 < 2e-16 ***
## NACCNIHR2 -0.319050 0.035193 -9.066 < 2e-16 ***
## NACCNIHR3 -0.146400 0.151231 -0.968 0.33302
## NACCNIHR4 0.461190 0.464436 0.993 0.32070
## NACCNIHR5 -0.218825 0.069955 -3.128 0.00176 **
## NACCNIHR6 -0.202888 0.066383 -3.056 0.00224 **
## EDUC -0.098621 0.003796 -25.979 < 2e-16 ***
## NACCTBI1 0.119274 0.046636 2.558 0.01054 *
## DEP1 1.129675 0.036812 30.688 < 2e-16 ***
## BIPOLDX1 0.447559 0.255568 1.751 0.07991 .
## SCHIZOP1 0.771827 0.550836 1.401 0.16116
## ANXIET1 0.498332 0.077439 6.435 1.23e-10 ***
## PTSDDX1 -0.775745 0.288313 -2.691 0.00713 **
## CANCER1 -0.170735 0.041267 -4.137 3.51e-05 ***
## DIABETES1 0.082835 0.043415 1.908 0.05640 .
## CONGHRT1 0.167094 0.097655 1.711 0.08707 .
## HYPERT1 0.022381 0.035189 0.636 0.52476
## HYPCHOL1 -0.019456 0.033258 -0.585 0.55854
## VB12DEF1 0.445637 0.061317 7.268 3.65e-13 ***
## THYDIS1 0.026937 0.040842 0.660 0.50955
## ARTH1 -0.658537 0.032294 -20.392 < 2e-16 ***
## SLEEPAP1 0.057278 0.043415 1.319 0.18707
## DICLOFENAC1 -0.247914 0.078063 -3.176 0.00149 **
## NAPROXEN1 -0.281654 0.047079 -5.983 2.20e-09 ***
## ETODOLAC1 -0.415376 0.176151 -2.358 0.01837 *
## DIAGNOSIS -0.084324 0.010216 -8.254 < 2e-16 ***
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
##
## (Dispersion parameter for binomial family taken to be 1)
##
## Null deviance: 50427 on 36622 degrees of freedom
## Residual deviance: 44424 on 36595 degrees of freedom
## (846 observations deleted due to missingness)
## AIC: 44480
##
## Number of Fisher Scoring iterations: 4# View the summary of the model
summary(logistic_model)## Error: object 'logistic_model' not foundwith(summary(logistic_model1), 1 - deviance/null.deviance)## [1] 0.08036398with(summary(logistic_model2), 1 - deviance/null.deviance)## [1] 0.1074406with(summary(logistic_model3), 1 - deviance/null.deviance)## [1] 0.1190576#tidysummary_model <- tidy(logistic_model2)
#write_csv(tidysummary_model, file = "tableS1.csv")
#logistic_coefficients<-exp(cbind(OR = coef(logistic_model2), confint(logistic_model2)))
#logistic_coefficients <- as.data.frame(logistic_coefficients)
#write_csv(logistic_coefficients, file = "tableS1_coef.csv")
# Perform a t-test to compare NACCAGE between DEMENTED = 0 and DEMENTED = 1
t_test_result <- t.test(NACCAGE ~ DEMENTED, data = collapsed_data)
# Print the result of the t-test
print(t_test_result)##
## Welch Two Sample t-test
##
## data: NACCAGE by DEMENTED
## t = -21.108, df = 43815, p-value < 2.2e-16
## alternative hypothesis: true difference in means between group 0 and group 1 is not equal to 0
## 95 percent confidence interval:
## -2.345538 -1.946947
## sample estimates:
## mean in group 0 mean in group 1
## 73.70854 75.85478Visual checks Error
Error in int_abline(a = a, b = b, h = h, v = v, untf = untf, …) : plot.new has not been called yet
#Visual checks
# Visual checks
par(mfrow = c(1, 2)) # Set up 2 plots side by side
hist(collapsed_data$NACCAGE[data$DEMENTED == 0], main = "DEMENTED = 0", xlab = "NACCAGE", breaks = 20)
hist(collapsed_data$NACCAGE[data$DEMENTED == 1], main = "DEMENTED = 1", xlab = "NACCAGE", breaks = 20)
# Q-Q plot to check normality
qqnorm(collapsed_data$NACCAGE[collapsed_data$DEMENTED == 0], main = "Q-Q Plot DEMENTED = 0")
qqline(collapsed_data$NACCAGE[collapsed_data$DEMENTED == 0])
qqnorm(collapsed_data$NACCAGE[collapsed_data$DEMENTED == 1], main = "Q-Q Plot DEMENTED = 1")
qqline(collapsed_data$NACCAGE[collapsed_data$DEMENTED == 1])
Mann-Whitney U test (Wilcoxon rank-sum test) AGE ~ DEMENTED
# Mann-Whitney U test (Wilcoxon rank-sum test)
wilcox.test(NACCAGE ~ DEMENTED, data = collapsed_data)##
## Wilcoxon rank sum test with continuity correction
##
## data: NACCAGE by DEMENTED
## W = 240365721, p-value < 2.2e-16
## alternative hypothesis: true location shift is not equal to 0# Mann-Whitney U test (Wilcoxon rank-sum test)
wilcox.test(NACCAGE ~ NACCALZD, data = filtered_data)##
## Wilcoxon rank sum test with continuity correction
##
## data: NACCAGE by NACCALZD
## W = 130858134, p-value < 2.2e-16
## alternative hypothesis: true location shift is not equal to 0Problem with propensity score matching
ERROR Call: matchit(formula = NSAID_TYPE ~ NACCAGE + SEX + NACCNIHR + EDUC + DIAGNOSIS, data = model_data, method = “full”, distance = “glm”, link = “probit”) Summary of Balance for All Data: Summary of Balance for Matched Data: Sample Sizes:
# Remove rows with missing values in the key covariates
model_data <- collapsed_data2 %>%
filter(!is.na(NSAID_TYPE), !is.na(NACCAGE), !is.na(NACCNIHR), !is.na(EDUC), !is.na(DIAGNOSIS))
# Convert NSAID_TYPE to a factor
model_data$NSAID_TYPE <- as.factor(model_data$NSAID_TYPE)
# Perform propensity score matching
set.seed(123)
match_obj <- matchit(NSAID_TYPE ~ NACCAGE + SEX + NACCNIHR + EDUC + DIAGNOSIS,
data = model_data,
method = "full",
distance = "glm",
link = "probit")
# Check if the matching worked
summary(match_obj)
# Summarize the matching results
summary(match_obj)
plot(match_obj, type = "jitter", interactive = FALSE)
plot(summary(match_obj), abs = FALSE)
# lines # dont work in RMD
#pdf(file = "FigS_matching.pdf", # The directory you want to save the file in
width = 4, # The width of the plot in inches
height = 4) # The height of the plot in inches
#plot(summary(match_obj), abs = FALSE)
#dev.off()## Error in parse(text = input): <text>:27:14: unexpected ','
## 26: #pdf(file = "FigS_matching.pdf", # The directory you want to save the file in
## 27: width = 4,
## ^Preform logistic regression on the matched data ACCALZD/DEMENTED ~ NSAID_TYPE (consequential error)
# Extract matched data
matched_data <- match.data(match_obj)## Error: object 'match_obj' not found# Perform logistic regression on the matched data
set.seed(123)
res1 <- glm(DEMENTED ~ NSAID_TYPE, data = matched_data, family = binomial)## Error in eval(mf, parent.frame()): object 'matched_data' not found# Perform logistic regression on the matched data
set.seed(123)
res2 <- glm(NACCALZD ~ NSAID_TYPE, data = matched_data, family = binomial)## Error in eval(mf, parent.frame()): object 'matched_data' not foundsummary(res)## Error: object 'res' not found# Summarize the logistic regression results
summary(res1)## Error: object 'res1' not foundsummary(res2)## Error: object 'res2' not found# in # for rmd file
#tidysummary_model1 <- tidy(res1)
#write_csv(tidysummary_model1, file = "tableS1_match.csv")
#logistic_coefficients1<-exp(cbind(OR = coef(res1), confint(res1)))
#logistic_coefficients1 <- as.data.frame(logistic_coefficients)
#write_csv(logistic_coefficients, file = "tableS1_coef_match.csv")
#tidysummary_model2 <- tidy(res2)
#write_csv(tidysummary_model2, file = "tableS2_match.csv")
#logistic_coefficients2<-exp(cbind(OR = coef(res2), confint(res2)))
#logistic_coefficients2 <- as.data.frame(logistic_coefficients)
#write_csv(logistic_coefficients, file = "tableS2_coef_match.csv")Create a contingency table of NSAID_TYPE and DEMENTED matched (error)
# Create a contingency table of NSAID_TYPE and DEMENTED
contingency_table <- table(matched_data$NSAID_TYPE, matched_data$DEMENTED)## Error: object 'matched_data' not found# Perform Fisher's exact test
fisher_test_result <- fisher.test(contingency_table)## Error: object 'contingency_table' not found# Print the result of the Fisher's exact test
print(fisher_test_result)## Error: object 'fisher_test_result' not foundprint(contingency_table)## Error: object 'contingency_table' not foundfoo<-contingency_table## Error: object 'contingency_table' not foundwrite.csv(foo, "Dementia_match_contingency.csv")## Error in eval(expr, p): object 'foo' not found#Moca Subscores -> weird outpot of boxplots
# List of cognitive variables
cognitive_variables <- c("VISUOSPATIAL", "LANGUAGE", "MEMORY2", "ABSTRACTION", "ATTENTION", "ORIENTATION")
#doenst work in RMD pdf(file = "Fig2.pdf", # The directory you want to save the file in
width = 8, # The width of the plot in inches
height = 4) # The height of the plot in inches
collapsed_data3<-collapsed_data2[!is.na(collapsed_data2$NSAID_TYPE),]
#doenst work in RMD pdf(file = "Fig_trial.pdf", # The directory you want to save the file in
width = 12, # The width of the plot in inches
height = 12) # The height of the plot in inches
# Create individual plots for each cognitive variable by NSAID_TYPE
plot_list <- lapply(cognitive_variables, function(var) {
ggplot(collapsed_data3, aes(x = as.factor(NSAID_TYPE), y = .data[[var]], fill = as.factor(NSAID_TYPE))) +
geom_boxplot() +
labs(title = paste("Distribution of", var, "by NSAID_TYPE"), x = "NSAID Type", y = "Score") +
scale_fill_manual(values = c("0" = "blue", "1" = "red"), labels = c("0" = "Naproxen", "1" = "Diclofenac")) +
theme_minimal())
# Arrange plots in a grid
do.call(grid.arrange, c(plot_list, ncol = 3))## Error in parse(text = input): <text>:5:14: unexpected ','
## 4: #doenst work in RMD pdf(file = "Fig2.pdf", # The directory you want to save the file in
## 5: width = 8,
## ^CSF
# Logistic regression model CSF
set.seed(123)
logistic_model_CSF <- glm(
CSFTAU ~ NACCAGE + SEX + NACCNIHR + EDUC + NACCTBI + DEP + BIPOLDX + SCHIZOP + ANXIET +
PTSDDX + CANCER + DIABETES + CONGHRT + HYPERT + HYPCHOL + VB12DEF +
THYDIS + ARTH + SLEEPAP + DICLOFENAC + NAPROXEN + ETODOLAC + DIAGNOSIS,
data = collapsed_data2,
family = binomial(link = "logit")
)
# View the summary of the model
summary(logistic_model_CSF)##
## Call:
## glm(formula = CSFTAU ~ NACCAGE + SEX + NACCNIHR + EDUC + NACCTBI +
## DEP + BIPOLDX + SCHIZOP + ANXIET + PTSDDX + CANCER + DIABETES +
## CONGHRT + HYPERT + HYPCHOL + VB12DEF + THYDIS + ARTH + SLEEPAP +
## DICLOFENAC + NAPROXEN + ETODOLAC + DIAGNOSIS, family = binomial(link = "logit"),
## data = collapsed_data2)
##
## Coefficients:
## Estimate Std. Error z value Pr(>|z|)
## (Intercept) -1.064752 0.319324 -3.334 0.000855 ***
## NACCAGE -0.049836 0.003345 -14.897 < 2e-16 ***
## SEX2 -0.369395 0.081605 -4.527 5.99e-06 ***
## NACCNIHR2 -1.069886 0.190576 -5.614 1.98e-08 ***
## NACCNIHR3 -0.896474 0.716740 -1.251 0.211020
## NACCNIHR4 -12.382824 278.421472 -0.044 0.964526
## NACCNIHR5 -1.170738 0.383010 -3.057 0.002238 **
## NACCNIHR6 -0.485020 0.265932 -1.824 0.068174 .
## EDUC 0.038777 0.013691 2.832 0.004623 **
## NACCTBI1 -0.493659 0.186419 -2.648 0.008094 **
## DEP1 -0.165008 0.116818 -1.413 0.157796
## BIPOLDX1 0.488456 0.484024 1.009 0.312899
## SCHIZOP1 -11.277195 317.258656 -0.036 0.971645
## ANXIET1 0.634077 0.164231 3.861 0.000113 ***
## PTSDDX1 -0.273404 0.734351 -0.372 0.709665
## CANCER1 0.314663 0.113336 2.776 0.005497 **
## DIABETES1 -1.322862 0.294439 -4.493 7.03e-06 ***
## CONGHRT1 -0.997635 0.460654 -2.166 0.030335 *
## HYPERT1 -0.116910 0.106604 -1.097 0.272784
## HYPCHOL1 0.490275 0.100841 4.862 1.16e-06 ***
## VB12DEF1 0.377673 0.151139 2.499 0.012460 *
## THYDIS1 0.092134 0.118005 0.781 0.434944
## ARTH1 0.049171 0.099298 0.495 0.620465
## SLEEPAP1 0.186360 0.114383 1.629 0.103256
## DICLOFENAC1 0.001704 0.252642 0.007 0.994618
## NAPROXEN1 -0.225072 0.163996 -1.372 0.169932
## ETODOLAC1 -0.279308 0.716635 -0.390 0.696722
## DIAGNOSIS 0.099229 0.030804 3.221 0.001276 **
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
##
## (Dispersion parameter for binomial family taken to be 1)
##
## Null deviance: 6911.1 on 36622 degrees of freedom
## Residual deviance: 6354.7 on 36595 degrees of freedom
## (846 observations deleted due to missingness)
## AIC: 6410.7
##
## Number of Fisher Scoring iterations: 14with(summary(logistic_model_CSF), 1 - deviance/null.deviance)## [1] 0.08050304# in # for rmd file
#tidysummary_model_CSF <- tidy(logistic_model_CSF)
#write_csv(tidysummary_model_CSF, file = "tableS3_tau.csv")
#logistic_coefficients_CSF<-exp(cbind(OR = coef(logistic_model_CSF), confint(logistic_model_CSF)))
#logistic_coefficients_CSF <- as.data.frame(logistic_coefficients_CSF)
#write_csv(logistic_coefficients_CSF, file = "tableS3_coef_tau.csv")
# Perform logistic regression on the matched data
set.seed(123)
res_CSF <- glm(CSFTAU ~ NSAID_TYPE, data = matched_data, family = binomial)## Error in eval(mf, parent.frame()): object 'matched_data' not foundsummary(res_CSF)## Error: object 'res_CSF' not found# in # for rmd file
#tidysummary_model_CSF <- tidy(res_CSF)
#write_csv(tidysummary_model, file = "tableS3_tau_match.csv")
#logistic_coefficients_CSF<-exp(cbind(OR = coef(res_CSF), confint(res_CSF)))
#logistic_coefficients_CSF <- as.data.frame(logistic_coefficients_CSF)
#write_csv(logistic_coefficients_CSF, file = "tableS3_coef_tau_match.csv")
# Create a contingency table of NSAID_TYPE and NACCALZD
contingency_table_CSF <- table(matched_data$NSAID_TYPE, matched_data$CSFTAU)## Error: object 'matched_data' not found# Perform Fisher's exact test
fisher_test_result_CSF <- fisher.test(contingency_table_CSF)## Error: object 'contingency_table_CSF' not found# Print the result of the Fisher's exact test
print(fisher_test_result_CSF)## Error: object 'fisher_test_result_CSF' not foundprint(contingency_table_CSF)## Error: object 'contingency_table_CSF' not foundfoo<-contingency_table_CSF## Error: object 'contingency_table_CSF' not foundwrite.csv(foo, "Csftau_match_contingency.csv")## Error in eval(expr, p): object 'foo' not foundFigures
Fig 1v2
pdf(file = "Fig1_v2.pdf", # The directory you want to save the file in
width = 12, # The width of the plot in inches
height = 8) # The height of the plot in inches
p1 <- ggscatter(collapsed_data2, x = "NACCAGE", y = "NACCMOCA",
add = "reg.line", # Add regression line
conf.int = TRUE, # Add confidence interval
add.params = list(color = "blue",
fill = "lightgray")
)+
stat_cor(method = "pearson", label.x = 0, label.y = 20,aes(label = paste0(..r.label.., sep = " ")))+
stat_cor(method = "pearson", label.x = 0, label.y = 18.5,aes(label = paste0(..p.label.., sep = " ")))
p1<-ggpar(p1, xlim = c(0, 100),ylim = c(0, 30),title = "A",ylab="MOCA Score", xlab="Age (years)")
p2 <- ggscatter(matched_data, x = "NACCAGE", y = "NACCMOCA",
add = "reg.line", # Add regression line
conf.int = TRUE, # Add confidence interval
add.params = list(color = "blue",
fill = "lightgray")
)+
stat_cor(method = "pearson", label.x = 3, label.y = 15)## Error: object 'matched_data' not foundp2<-ggpar(p2, xlim = c(0, 100),ylim = c(0, 30),title = "D",ylab="MOCA Score", xlab="Age (years)")## Error: object 'p2' not foundp3 <- ggboxplot(collapsed_data2, x = "DEMENTED", y = "NACCMOCA")
my_comparisons <- list( c("0", "1") )
p3<- p3+stat_compare_means(comparisons = my_comparisons,method = "wilcox.test")+scale_x_discrete(labels = c("0" = "Not Demented", "1" = "Demented"))
p3<-ggpar(p3,xlab="Demented Status", ylab="MOCA Score",title = "B")
p4 <- ggboxplot(matched_data, x = "DEMENTED", y = "NACCMOCA")## Error: object 'matched_data' not foundmy_comparisons <- list( c("0", "1") )
p4<- p4+stat_compare_means(comparisons = my_comparisons,method = "wilcox.test")+scale_x_discrete(labels = c("0" = "Not Demented", "1" = "Demented"))## Error: object 'p4' not foundp4<-ggpar(p4,xlab="Demented Status", ylab="MOCA Score",title = "E")## Error: object 'p4' not foundp5 <- ggboxplot(collapsed_data2, x = "NACCALZD", y = "NACCMOCA")
my_comparisons <- list( c("0", "1") )
p5<- p5+stat_compare_means(comparisons = my_comparisons,method = "wilcox.test")+scale_x_discrete(labels = c("0" = "Normal cognition", "1" = "AD"))
p5<-ggpar(p5,xlab="AD Status", ylab="MOCA Score",title = "C")
p6 <- ggboxplot(matched_data, x = "NACCALZD", y = "NACCMOCA")## Error: object 'matched_data' not foundmy_comparisons <- list( c("0", "1") )
p6<- p6+stat_compare_means(comparisons = my_comparisons,method = "wilcox.test")+scale_x_discrete(labels = c("0" = "Normal cognition", "1" = "AD"))## Error: object 'p6' not foundp6<-ggpar(p6,xlab="AD Status", ylab="MOCA Score",title = "F")## Error: object 'p6' not foundgrid.arrange(p1, p3, p5, p2,p4,p6, ncol = 3)## Error: object 'p2' not foundFig 1 v1
# doesnt work in RMD : pdf(file = "Fig1.pdf", # The directory you want to save the file in
width = 12, # The width of the plot in inches
height = 8) # The height of the plot in inches
# Filter out NACCMOCA values of 88 and 99 for both datasets
collapsed_data_filtered <- filtered_data %>%
filter(NACCMOCA != 88 & NACCMOCA != 99)
# A. Relationship of NACCMOCA vs NACCAGE scatter plot in collapsed_data with trend line and R-squared
p1 <- ggplot(filtered_data, aes(x = NACCAGE, y = NACCMOCA)) +
geom_point(alpha = 0.5) +
geom_smooth(method = "lm", se = FALSE, color = "blue") + # Add trend line
theme_minimal() +
labs(title = "A. NACCMOCA vs NACCAGE (collapsed_data)",
x = "Age",
y = "MOCA Score") +
stat_cor(method = "pearson", label.x = 60, label.y = 30, aes(label = after_stat(rr.label))) # Updated notation
# B. Relationship of NACCMOCA vs NACCAGE scatter plot in matched_data with trend line and R-squared
p2 <- ggplot(matched_data, aes(x = NACCAGE, y = NACCMOCA)) +
geom_point(alpha = 0.5) +
geom_smooth(method = "lm", se = FALSE, color = "blue") + # Add trend line
theme_minimal() +
labs(title = "B. NACCMOCA vs NACCAGE (matched_data)",
x = "Age",
y = "MOCA Score") +
stat_cor(method = "pearson", label.x = 60, label.y = 30, aes(label = after_stat(rr.label))) # Updated notation
# C. Boxplot distribution of NACCMOCA among DEMENTED = 0 and 1 (collapsed_data)
p3 <- ggplot(filtered_data, aes(x = as.factor(DEMENTED), y = NACCMOCA)) +
geom_boxplot() +
theme_minimal() +
labs(title = "C. NACCMOCA by DEMENTED (collapsed_data)",
x = "Demented Status",
y = "MOCA Score") +
scale_x_discrete(labels = c("0" = "Not Demented", "1" = "Demented"))
# D. Boxplot distribution of NACCMOCA among DEMENTED = 0 and 1 (matched_data)
p4 <- ggplot(matched_data, aes(x = as.factor(DEMENTED), y = NACCMOCA)) +
geom_boxplot() +
theme_minimal() +
labs(title = "D. NACCMOCA by DEMENTED (matched_data)",
x = "Demented Status",
y = "MOCA Score") +
scale_x_discrete(labels = c("0" = "Not Demented", "1" = "Demented"))
# E. Boxplot distribution of NACCMOCA among NACCALZD = 0 and 1 (collapsed_data)
p5 <- ggplot(filtered_data, aes(x = as.factor(NACCALZD), y = NACCMOCA)) +
geom_boxplot() +
theme_minimal() +
labs(title = "E. NACCMOCA by NACCALZD (collapsed_data)",
x = "AD Status",
y = "MOCA Score") +
scale_x_discrete(labels = c("0" = "Normal cognition", "1" = "AD"))
# F. Boxplot distribution of NACCMOCA among NACCALZD = 0 and 1 (matched_data)
p6 <- ggplot(matched_data, aes(x = as.factor(NACCALZD), y = NACCMOCA)) +
geom_boxplot() +
theme_minimal() +
labs(title = "F. NACCMOCA by NACCALZD (matched_data)",
x = "AD Status",
y = "MOCA Score") +
scale_x_discrete(labels = c("0" = "Normal cognition", "1" = "AD"))
# Combine all plots into a 2x2 grid
grid.arrange(p1, p3, p5, p2, p4, p6, ncol = 3)## Error in parse(text = input): <text>:2:15: unexpected ','
## 1: # doesnt work in RMD : pdf(file = "Fig1.pdf", # The directory you want to save the file in
## 2: width = 12,
## ^Fig 2 v 2
#doesnt work in rmd # pdf(file = "Fig2_v2.pdf", # The directory you want to save the file in
width = 8, # The width of the plot in inches
height = 4) # The height of the plot in inches
collapsed_data3<-collapsed_data2[!is.na(collapsed_data2$NSAID_TYPE),]
p1 <- ggboxplot(collapsed_data3, x = "NSAID_TYPE", y = "NACCMOCA")
my_comparisons <- list( c("0", "1") )
p1<- p1+stat_compare_means(comparisons = my_comparisons,method = "wilcox.test")+scale_x_discrete(labels = c("0" = "Naproxen", "1" = "Diclofenac"))
p1<-ggpar(p1,xlab="NSAID Type", ylab="MOCA Score",title = "A")
p2 <- ggboxplot(matched_data, x = "NSAID_TYPE", y = "NACCMOCA")
my_comparisons <- list( c("0", "1") )
p2<- p2+stat_compare_means(comparisons = my_comparisons,method = "wilcox.test")+scale_x_discrete(labels = c("0" = "Naproxen", "1" = "Diclofenac"))
p2<-ggpar(p2,xlab="NSAID Type", ylab="MOCA Score",title = "B")
grid.arrange(p1, p2, ncol = 2)## Error in parse(text = input): <text>:2:14: unexpected ','
## 1: #doesnt work in rmd # pdf(file = "Fig2_v2.pdf", # The directory you want to save the file in
## 2: width = 8,
## ^Fig 2 v1
# doesnt work in rmd pdf(file = "Fig2.pdf", # The directory you want to save the file in
width = 8, # The width of the plot in inches
height = 4) # The height of the plot in inches
collapsed_data3<-collapsed_data2[!is.na(collapsed_data2$NSAID_TYPE),]
# A. Boxplot distribution of NACCMOCA among NACCALZD = 0 and 1 (collapsed_data)
p1 <- ggplot(collapsed_data3, aes(x = as.factor(NSAID_TYPE), y = NACCMOCA)) +
geom_boxplot() +
theme_minimal() +
labs(title = "A. NACCMOCA by NSAID TYPE (collapsed_data)",
x = "NSAID TYPE",
y = "MOCA Score") +
scale_x_discrete(labels = c("0" = "Naproxen", "1" = "Diclofenac"))
my_comparisons <- list( c("0", "1") )
p1<-p1 + stat_compare_means(comparisons = my_comparisons, method = "wilcox.test")
# B. Boxplot distribution of NACCMOCA among NACCALZD = 0 and 1 (matched_data)
p2 <- ggplot(matched_data, aes(x = as.factor(NSAID_TYPE), y = NACCMOCA)) +
geom_boxplot() +
theme_minimal() +
labs(title = "B. NACCMOCA by NSAID TYPE (matched_data)",
x = "NSAID TYPE",
y = "MOCA Score") +
scale_x_discrete(labels = c("0" = "Naproxen", "1" = "Diclofenac"))
my_comparisons <- list( c("0", "1") )
p2<-p2 + stat_compare_means(comparisons = my_comparisons, method = "wilcox.test")
# Combine all plots into a 2x2 grid
grid.arrange(p1, p2, ncol = 2)## Error in parse(text = input): <text>:2:14: unexpected ','
## 1: # doesnt work in rmd pdf(file = "Fig2.pdf", # The directory you want to save the file in
## 2: width = 8,
## ^#Fig 3
#doesnt work in rmd pdf(file = "Fig3.pdf", # The directory you want to save the file in
width = 8, # The width of the plot in inches
height = 4) # The height of the plot in inches
collapsed_data3<-collapsed_data2[!is.na(collapsed_data2$NSAID_TYPE),]
# A. Boxplot distribution of NACCMOCA among NACCALZD = 0 and 1 (collapsed_data)
p1 <- ggplot(collapsed_data3, aes(x = as.factor(CSFTAU), y = NACCMOCA)) +
geom_boxplot() +
theme_minimal() +
labs(title = "A. NACCMOCA by CSF TAU positivity (collapsed_data)",
x = "CSF TAU positivity",
y = "MOCA Score") +
scale_x_discrete(labels = c("0" = "CSF Tau Negative (n=3054)", "1" = "CSF Tau Positive (n=57)"))
my_comparisons <- list( c("0", "1") )
p1<-p1 + stat_compare_means(comparisons = my_comparisons, method = "wilcox.test")
# B. Boxplot distribution of NACCMOCA among NACCALZD = 0 and 1 (matched_data)
p2 <- ggplot(matched_data, aes(x = as.factor(CSFTAU), y = NACCMOCA)) +
geom_boxplot() +
theme_minimal() +
labs(title = "B. NACCMOCA by CSF TAU positivity (matched_data)",
x = "CSF TAU positivity",
y = "MOCA Score") +
scale_x_discrete(labels = c("0" = "CSF Tau Negative (n=2999)", "1" = "CSF Tau Positive (n=57)"))
my_comparisons <- list( c("0", "1") )
p2<-p2 + stat_compare_means(comparisons = my_comparisons, method = "wilcox.test")
# Combine all plots into a 2x2 grid
grid.arrange(p1, p2, ncol = 2)## Error in parse(text = input): <text>:2:14: unexpected ','
## 1: #doesnt work in rmd pdf(file = "Fig3.pdf", # The directory you want to save the file in
## 2: width = 8,
## ^Fig 4 Boxplot distribution of NACCMOCA among DEMENTED = 0 and 1
#doesnt work in rmd pdf(file = "Fig4.pdf", # The directory you want to save the file in
width = 12, # The width of the plot in inches
height = 8) # The height of the plot in inches
# C. Boxplot distribution of NACCMOCA among DEMENTED = 0 and 1 (collapsed_data)
p1 <- ggplot(collapsed_data3, aes(x = as.factor(NSAID_TYPE), y = VISUOSPATIAL)) +
geom_boxplot() +
theme_minimal() +
labs(title = "A. VISUOSPATIAL by NSAID TYPE",
x = "NSAID Type",
y = "Score") +
scale_x_discrete(labels = c("0" = "Naproxen", "1" = "Diclofenac"))+
stat_summary(fun=mean, geom="point", shape=18, size=3, color="red", fill="red")
my_comparisons <- list( c("0", "1") )
p1<-p1 + stat_compare_means(comparisons = my_comparisons, method = "wilcox.test")
# C. Boxplot distribution of NACCMOCA among DEMENTED = 0 and 1 (collapsed_data)
p2 <- ggplot(collapsed_data3, aes(x = as.factor(NSAID_TYPE), y = LANGUAGE)) +
geom_boxplot() +
theme_minimal() +
labs(title = "B. LANGUAGE by NSAID TYPE",
x = "NSAID Type",
y = "Score") +
scale_x_discrete(labels = c("0" = "Naproxen", "1" = "Diclofenac"))+
stat_summary(fun=mean, geom="point", shape=18, size=3, color="red", fill="red")
my_comparisons <- list( c("0", "1") )
p2<-p2 + stat_compare_means(comparisons = my_comparisons, method = "wilcox.test")
# C. Boxplot distribution of NACCMOCA among DEMENTED = 0 and 1 (collapsed_data)
p3 <- ggplot(collapsed_data3, aes(x = as.factor(NSAID_TYPE), y = MEMORY2)) +
geom_boxplot() +
theme_minimal() +
labs(title = "C. MEMORY2 by NSAID TYPE",
x = "NSAID Type",
y = "Score") +
scale_x_discrete(labels = c("0" = "Naproxen", "1" = "Diclofenac"))+
stat_summary(fun=mean, geom="point", shape=18, size=3, color="red", fill="red")
my_comparisons <- list( c("0", "1") )
p3<-p3 + stat_compare_means(comparisons = my_comparisons, method = "wilcox.test")
# C. Boxplot distribution of NACCMOCA among DEMENTED = 0 and 1 (collapsed_data)
p4 <- ggplot(collapsed_data3, aes(x = as.factor(NSAID_TYPE), y = ABSTRACTION)) +
geom_boxplot() +
theme_minimal() +
labs(title = "D. ABSTRACTION by NSAID TYPE",
x = "NSAID Type",
y = "Score") +
scale_x_discrete(labels = c("0" = "Naproxen", "1" = "Diclofenac"))+
stat_summary(fun=mean, geom="point", shape=18, size=3, color="red", fill="red")
my_comparisons <- list( c("0", "1") )
p4<-p4 + stat_compare_means(comparisons = my_comparisons, method = "wilcox.test")
# C. Boxplot distribution of NACCMOCA among DEMENTED = 0 and 1 (collapsed_data)
p5 <- ggplot(collapsed_data3, aes(x = as.factor(NSAID_TYPE), y = ATTENTION)) +
geom_boxplot() +
theme_minimal() +
labs(title = "E. ATTENTION by NSAID TYPE",
x = "NSAID Type",
y = "Score") +
scale_x_discrete(labels = c("0" = "Naproxen", "1" = "Diclofenac"))+
stat_summary(fun=mean, geom="point", shape=18, size=3, color="red", fill="red")
my_comparisons <- list( c("0", "1") )
p5<-p5 + stat_compare_means(comparisons = my_comparisons, method = "wilcox.test")
# C. Boxplot distribution of NACCMOCA among DEMENTED = 0 and 1 (collapsed_data)
p6 <- ggplot(collapsed_data3, aes(x = as.factor(NSAID_TYPE), y = ORIENTATION)) +
geom_boxplot() +
theme_minimal() +
labs(title = "F. ORIENTATION by NSAID TYPE",
x = "NSAID Type",
y = "Score") +
scale_x_discrete(labels = c("0" = "Naproxen", "1" = "Diclofenac"))+
stat_summary(fun=mean, geom="point", shape=18, size=3, color="red", fill="red")
my_comparisons <- list( c("0", "1") )
p6<-p6 + stat_compare_means(comparisons = my_comparisons, method = "wilcox.test")
# Combine all plots into a 2x2 grid
grid.arrange(p1, p2, p3, p4,p5,p6, ncol = 3)## Error in parse(text = input): <text>:2:15: unexpected ','
## 1: #doesnt work in rmd pdf(file = "Fig4.pdf", # The directory you want to save the file in
## 2: width = 12,
## ^Fig 4v3 Boxplot distribution of NACCMOCA among DEMENTED = 0 and 1
# doesnt work in rmd pdf(file = "Fig4_v3.pdf", # The directory you want to save the file in
width = 12, # The width of the plot in inches
height = 8) # The height of the plot in inches
p1 <- ggboxplot(matched_data, x = "NSAID_TYPE", y = "VISUOSPATIAL")
my_comparisons <- list( c("0", "1") )
p1<- p1+stat_compare_means(comparisons = my_comparisons,method = "wilcox.test")+
scale_x_discrete(labels = c("0" = "Naproxen", "1" = "Diclofenac"))+
stat_summary(fun=mean, geom="point", shape=18, size=3, color="red", fill="red")
p1<-ggpar(p1,xlab="NSAID Type", ylab="MOCA Subscore",title = "A. Visuospatial")
p2 <- ggboxplot(matched_data, x = "NSAID_TYPE", y = "LANGUAGE")
my_comparisons <- list( c("0", "1") )
p2<- p2+stat_compare_means(comparisons = my_comparisons,method = "wilcox.test")+
scale_x_discrete(labels = c("0" = "Naproxen", "1" = "Diclofenac"))+
stat_summary(fun=mean, geom="point", shape=18, size=3, color="red", fill="red")
p2<-ggpar(p2,xlab="NSAID Type", ylab="MOCA Subscore",title = "B. Language")
p3 <- ggboxplot(matched_data, x = "NSAID_TYPE", y = "MEMORY2")
my_comparisons <- list( c("0", "1") )
p3<- p3+stat_compare_means(comparisons = my_comparisons,method = "wilcox.test")+
scale_x_discrete(labels = c("0" = "Naproxen", "1" = "Diclofenac"))+
stat_summary(fun=mean, geom="point", shape=18, size=3, color="red", fill="red")
p3<-ggpar(p3,xlab="NSAID Type", ylab="MOCA Subscore",title = "C. Memory")
p4 <- ggboxplot(matched_data, x = "NSAID_TYPE", y = "ABSTRACTION")
my_comparisons <- list( c("0", "1") )
p4<- p4+stat_compare_means(comparisons = my_comparisons,method = "wilcox.test")+
scale_x_discrete(labels = c("0" = "Naproxen", "1" = "Diclofenac"))+
stat_summary(fun=mean, geom="point", shape=18, size=3, color="red", fill="red")
p4<-ggpar(p4,xlab="NSAID Type", ylab="MOCA Subscore",title = "D. Abstraction")
p5 <- ggboxplot(matched_data, x = "NSAID_TYPE", y = "ATTENTION")
my_comparisons <- list( c("0", "1") )
p5<- p5+stat_compare_means(comparisons = my_comparisons,method = "wilcox.test")+
scale_x_discrete(labels = c("0" = "Naproxen", "1" = "Diclofenac"))+
stat_summary(fun=mean, geom="point", shape=18, size=3, color="red", fill="red")
p5<-ggpar(p5,xlab="NSAID Type", ylab="MOCA Subscore",title = "E. Attention")
p6 <- ggboxplot(matched_data, x = "NSAID_TYPE", y = "ORIENTATION")
my_comparisons <- list( c("0", "1") )
p6<- p6+stat_compare_means(comparisons = my_comparisons,method = "wilcox.test")+
scale_x_discrete(labels = c("0" = "Naproxen", "1" = "Diclofenac"))+
stat_summary(fun=mean, geom="point", shape=18, size=3, color="red", fill="red")
p6<-ggpar(p6,xlab="NSAID Type", ylab="MOCA Subscore",title = "F. Orientation")
summary(collapsed_data3[collapsed_data3$NSAID_TYPE==0,]$ORIENTATION)
dim(collapsed_data3[collapsed_data3$NSAID_TYPE==0,])
summary(collapsed_data3[collapsed_data3$NSAID_TYPE==1,]$ORIENTATION)
dim(collapsed_data3[collapsed_data3$NSAID_TYPE==1,])
summary(collapsed_data3[collapsed_data3$NSAID_TYPE==0,]$ATTENTION)
dim(collapsed_data3[collapsed_data3$NSAID_TYPE==0,])
summary(collapsed_data3[collapsed_data3$NSAID_TYPE==1,]$ATTENTION)
dim(collapsed_data3[collapsed_data3$NSAID_TYPE==1,])
# Combine all plots into a 2x2 grid
grid.arrange(p1, p2, p3, p4,p5,p6, ncol = 3)## Error in parse(text = input): <text>:2:15: unexpected ','
## 1: # doesnt work in rmd pdf(file = "Fig4_v3.pdf", # The directory you want to save the file in
## 2: width = 12,
## ^#pdf2
summary(collapsed_data3[collapsed_data3$NSAID_TYPE==0,]$ORIENTATION)## Min. 1st Qu. Median Mean 3rd Qu. Max.
## 0.00 0.00 0.00 2.52 6.00 6.00dim(collapsed_data3[collapsed_data3$NSAID_TYPE==0,])## [1] 2267 70summary(collapsed_data3[collapsed_data3$NSAID_TYPE==1,]$ORIENTATION)## Min. 1st Qu. Median Mean 3rd Qu. Max.
## 0.000 0.000 4.000 3.082 6.000 6.000dim(collapsed_data3[collapsed_data3$NSAID_TYPE==1,])## [1] 844 70summary(collapsed_data3[collapsed_data3$NSAID_TYPE==0,]$ATTENTION)## Min. 1st Qu. Median Mean 3rd Qu. Max.
## 0.000 0.000 0.000 2.421 6.000 6.000dim(collapsed_data3[collapsed_data3$NSAID_TYPE==0,])## [1] 2267 70summary(collapsed_data3[collapsed_data3$NSAID_TYPE==1,]$ATTENTION)## Min. 1st Qu. Median Mean 3rd Qu. Max.
## 0.000 0.000 3.000 2.902 6.000 6.000dim(collapsed_data3[collapsed_data3$NSAID_TYPE==1,])## [1] 844 70